The TRL Overseas Road Notes are a series of manuals and guidelines prepared for use by professionals working

in the field of transport in developing countries and countries in transition. The information contained within the ORNs represents the most up to date thinking on good engineering, transport safety, environmental mitigation and transport planning practice applicable in such countries. The Overseas Road Note series dates from 1981. They are an ad hoc series covering a range of topics reflecting the types of work being carried out at a particular time. They do not conform to any master plan and are therefore numbered simply according to order of publication. The exception in the numbering system is Overseas Road Note 31. This originated in the TRL Road Note series as RN 31 and the number became firmly associated with the title. Thus, when the ORN series come into being and incorporated RN 31, the guide was allowed to retain its original number. Road Note 40, similarly, is an honorary ORN that still carries its original RN number (but see below). The Overseas Road Notes mostly remain current for many years and do not require regular updating. However, they are revised when necessary. ORN 3 was recently revised and is now published in its second edition. RN 40 is currently under revision and will shortly be reissued as ORN 40.

The titles are as follows, but the titles can also be displayed by clicking on the expand symbol next to Overseas Road Notes in the screen index. ORN ORN ORN ORN ORN ORN ORN ORN ORN ORN ORN ORN ORN ORN ORN ORN ORN ORN Maintenance management for district engineers Maintenance techniques for district engineers A guide to surface dressing in tropical and sub-tropical countries Field survey techniques and analysis for urban bus operators A guide to road project appraisal A guide to geometric design Volume 1. A guide to bridge inspection and data systems for district engineers Volume 2. Bridge inspectors handbook A users manual for a program to analyse Dynamic Cone Penetrometer data A design manual for small bridges Costing road accidents in developing countries Urban road traffic surveys Design guidelines for busway transit Use of traffic signals in developing countries Hydrological design manual for slope stability in the tropics Guidelines for the design and operation of road management systems Principles of low cost road engineering in mountainous regions Road safety education in developing countries. Guidelines for good practice in primary schools ORN 18 A guide to the pavement evaluation and maintenance of bitumen-surfaced roads in tropical and sub-tropical countries ORN 31 A guide to the structural design of bitumen-surfaced roads in tropical and sub-tropical countries ORN 40 A guide to the measurement of axle loads in developing countries using a portable weighbridge 1 2 3 4 5 6 7 7 8 9 10 11 12 13 14 15 16 17

Transport and Road Research Laboratory Overseas Unit Department of Transport Overseas Development Administration

Overseas Road Note 1

Maintenance management for district engineers (2nd edition)

Overseas Unit Transport and Road Research Laboratory Crowthorne Berkshire United Kingdom 1987

ACKNOWLEDGEMENTS
This Note was drafted by Dr R Robinson of the TRRL Overseas Unit. It is based on an original text produced by Scott Wilson Kirkpatrick and Partners. Certain details of the inspection procedures recommended are based on those developed for System BSM by John Burrow & Partners. The paved road intervention levels were devised by Dr J Rolt, the appendix illustrating typical defects was prepared by Dr G Morosiuk, and other contributions were made by Mr D M Brooks, all of the TRRL Overseas Unit. Final editing was carried out by Harold Lewis, Consultant Technical Editor.

First published 1981 Reprinted with minor revisions 1983 Second edition 1987 Reprinted with minor revisions 1995

OVERSEAS ROAD NOTES

Overseas Road Notes are prepared principally for road and road transport authorities in countries receiving technical assistance from the British Government. A limited number of copies is available to other organisations and to individuals with an interest in roads overseas, and may be obtained from: Overseas Centre Transport Research Laboratory Crowthorne, Berkshire, RG45 6AU United Kingdom

Crown copyright 1987 Limited extracts from the text may be produced provided the source is acknowledged. For more extensive reproduction, please write to Head of Overseas Centre, Transport Research Laboratory

ISSN 0951-8987

CONTENTS
page 1 1 I 1 1 1 2 2 2 2 3 3 4 4 4 6 6 6 7 11 11 12 12 12 16 16 16 16 17 19 20 22 22 22 23 23 23 24 24 24 26 26 26

1.

2.

3. 4. 5.

6.

7.

8.

9.

10.

11.

Introduction Purpose and scope Structure of the Note The role of the maintenance engineer Management responsibilities Involvement on site Delegation Training Using microcomputers Implementation Maintenance activities Management tasks Inventory Content and preparation Presentation Inspection Frequency of inspection Condition survey Recording results Determination of maintenance requirements Recurrent and periodic maintenance of unpaved roads Intervention levels on paved roads Diagnozing the cause of deterioration Specifying the work required Resource estimation Choice of work method Use of contractors Labour Equipment Materials Resource allocation Identification of priorities Maintenance activities by order of importance Roads by order of importance Priority matrix Adapting priorities to local conditions Determining the work programme Work scheduling and execution Schedules Worksheets Monitoring Site inspections Desk review

References Appendix A - Field procedures for inventory and condition measurement surveys A.l General procedure Recording Organization of teams Safety Transport Notes for inspectors Duties of team members Order of work Accuracy A.2 Condition measurement General considerations Side drains and turnouts (all roads) Loss of material (gravel roads) Deformation (all roads) Cracking (paved roads) Pot-holes (all roads) Edge drainage (paved roads) Edge step (paved roads) Appendix B - Management of grading for unpaved roads B.1 Determining optimum grading frequencies B.2 Monitoring of optimum frequencies B.3 Measurement methods Reference Appendix C - Traffic counting C.1 Frequency and duration of counts C.2 Estimation of ADT from counts C.3 Manual counts C.4 Automatic counters C.5 Moving observer counts C.6 Development of traffic counts Appendix D - Recommended standard forms Appendix E - Illustrations of typical defects

page 26 27 27 27 27 27 27 28 28 28 28 28 28 28 29 29 29 29 29 29 32 32 32 33 33 34 34 34 34 34 34 35 36 42

1.

INTRODUCTION

2. THE ROLE OF THE MAINTENANCE ENGINEER

2.1 Roads are expensive to build. They repay their initial investment only by means of long4erm care and maintenance. A road system that is well maintained brings important social and economic benefits: the transport links on which development depends are kept in good working order roads have a longer lifetime of service because their surfaces do not deteriorate so rapidly vehicle operating costs are reduced because traffic is able to run smoothly transport operations are safer and more reliable 2.2 The maintenance engineer responsible for operations at regional or district level has a key role to play in achieving these benefits. His success depends largely on the way he approaches the task of management.

Purpose and scope

1.1 This Note is a practical guide to the management of maintenance operations. It outlines a rational approach that will help maintenance engineers organize and control the activities for which they are responsible, so as to improve efficiency and make more productive use of maintenance resources. 1.2 The maintenance operations discussed here are those required to keep roads in good condition and repair. The Note does not deal with any form of road improvement works, pavement strengthening or reconstruction, even where these activities are undertaken by an organization responsible also for normal maintenance. 1.3 The Note offers advice on techniques basic to good maintenance practice, but it does not set out to define 'model' systems that should be copied generally or to describe all the management procedures an ideal maintenance organization would follow. This is because in any location the best management system will be one which is matched closely to the technical skills, human resources and equipment available to the individual maintenance organization; and the most effective procedures will be those that are appropriate to the experience and capabilities of its staff. Using this Note, engineers will be able to assess the range of management techniques applicable to road maintenance and so identity methods they can usefully put into practice within the context of their own organizations. Structure of the Note 1.4 Following this introduction, Section 2 summarizes the responsibilities of the maintenance engineer and comments on the approach he should adopt to key management tasks. Section 3 sets out a classification of maintenance activities, and Section 4 defines the sequence of management tasks which the engineer has to fulfil. These tasks are the subject of Sections 5 to 11, which explain each stage of the management process in turn. Appendices A to D provide details of technical procedures and illustrate standard forms and worksheets. Appendix E illustrates typical defects found on roads.

Management responsibilities
2.3 This task normally involves five main responsibilities: planning the annual programme of maintenance work for his area, assessing the resources needed and preparing an appropriate budget estimate arranging that funds are allocated fairly to the various parts of the road network, and deciding on priorities if the funds available do not allow him to undertake the full programme authorising and scheduling work making sure that his staff know how to carry out the work methodically and efficiently monitoring the quality and effectiveness of maintenance activities. 2.4 Each of these responsibilities forms a major step in the sequence of maintenance management explained in Sections 5 to 11 of this Note.

Involvement on site
2.5 In performing his management role, the maintenance engineer will, of course, have many hours of office work on matters such as planning and administration. But it is essential that he also gets out into the field as much as possible. The simple procedure of seeing things for themselves would help maintenance engineers overcome many of the problems that at present affect their operations. 2.6 There are several reasons why site visits are important:

they enable the maintenance engineer to become thoroughly familiar with road conditions in the area, and so recognize trouble spots and other places where difficulties are likely to occur he can gain first-hand knowledge of the extent and quality of the maintenance that has actually been carried out, instead of having to rely on what he reads in reports he can use this knowledge to assess maintenance priorities much more confidently his presence on the spot means that he can advise on problems as they arise seeing him regularly on site should boost the morale of road gangs and improve their standard of work and their output. His attitude will have more influence on their performance than any other factor, and site visits are the most effective way of demonstrating his commitment to getting the job done successfully.

Using microcomputers
2.10 Now that 'micros' are becoming widely available, their application to maintenance management is a subject where staff training may be particularly useful. A growing range of specialised software is available to help process data and analyze problems, and using a micro can save considerable time beside freeing the maintenance engineer for inspection and monitoring on site. But expenditure on computers can prove an expensive mistake if the system is not chosen with care and if suitable personnel are not available to make the best use of it. Maintenance can be managed efficiently without a micro; and even if there is one in the office, it does not lessen the need for regular and accurate data collection in the field.

Implementation
2.11 For most organizations, the management approach recommended in this Note will take some time to implement - perhaps a period of several years. The maintenance engineer should not try to put everything into practice at once. It is better to introduce new methods and procedures gradually, starting with straightforward measures that will produce early and positive results. Proceed step by step, and wait until one stage is working reasonably well before moving to the next. Concentrate first on the sections of road that carry the largest volumes of traffic and ensure these are adequately maintained before dealing with less busy roads.

Delegation
2.7 The maintenance engineer should use his knowledge of road conditions to decide which operations need his personal supervision and which he can safely delegate to staff. On lengths of road where maintenance is straightforward and easy to specify in advance, it should be possible to leave day-to-day work in the hands of suitably trained foremen or contractor's staff. On other roads where there are problems requiring complicated treatment and on-the-spot judgment, the maintenance engineer will have to become personally involved in determining what needs to be done and supervising the work. Thy key point is that he should not let his time be taken up by simple operations which less qualified staff are able to manage.

Training
2.8 Delegation will only succeed if staff have the knowledge and competence to fulfil the duties they are given. The maintenance engineer has to make sure that supervisors, foremen and other personnel receive the necessary training, and that there are enough trained staff to carry out his instructions. This means that training is an important part of his responsibilities. 2.9 Though training methods are outside the scope of this document, there are some basic points to remember. First, every member of staff should have appropriate training. Secondly, this training should be built into the work programme and include practical on-the-job experience as well as more formal courses. Thirdly, training should be an on-going feature of employment in the maintenance organization, so that competent staff are available to take over when more experienced personnel are promoted or transferred to other duties.

3.
3.1

MAINTENANCE ACTIVITIES

4. MANAGEMENT TASKS
4.1 As noted in para 2.3, the maintenance engineer has the
responsibility of structuring these activities into a programme of maintenance work throughout his area, with an appropriate allocation of resources. This responsibility involves a sequence of tasks, which are discussed in turn in Sections 5 to 11, and can be summarised as follows:

For the purposes of management, the most useful way to classify maintenance activities is in terms of their frequency. There are four categories:

routine maintenance, required continually on every

road whatever its engineering characteristics or traffic volume

(i)

recurrent maintenance, required at intervals during

the year with a frequency that depends on the volume of traffic using the road

inventory: recording the basic characteristics of each section of the road network inspection: examining the road and measuring its
condition

(ii)

periodic maintenance, required only at intervals of several years

(iii)

determination of maintenance requirements:

analyzing why defects are occurring; and specifying what maintenance activities are needed to put them right and delay further deterioration

urgent maintenance, needed to deal with emergencies and problems calling for immediate action when a road is blocked.

(iv)

resource estimation: costing the maintenance

programme in order to define an overall budget

3.2 Examples of activities within these categories are as follows:

(v)

identification of priorities: deciding the work that

has to take precedence if resources are limited

routine:
grass cutting; drain clearing; recutting ditches; culvert maintenance; road signs maintenance (vi)

work scheduling and performance:

controlling the work as it is carried out

recurrent on unpaved roads:

repairing pot-holes and ruts; dragging; grading

(vii)

monitoring: checking the quality and effectiveness of the work.

recurrent on paved roads:

repairing pot-holes; patching; repairing edges; sealing cracks

periodic on unpaved roads:

regravelling

periodic on paved roads:

resealing (surface dressing, slurry sealing, fog spray, etc.); regravelling shoulders; road surface marking

4.2 Because unpaved roads deteriorate more rapidly than paved roads, they require more frequent attention. The recurrent maintenance listed for unpaved roads in para 3.2 should be applied at regular and predetermined intervals throughout the year. These activities must not be deferred until defects show up in the course of an annual or seasonal inspection. Methods of calculating how frequently they need to be undertaken during the year are discussed in Section 7.

urgent:
removal of debris and other obstacles; placement of warning signs and diversion works. 3.3 Routine maintenance, by definition, has to be undertaken on a regular basis: its frequency does not depend on the character of the road or the amount of traffic. For this reason it is treated as a fixed-cost item in the maintenance budget. Recurrent and periodic maintenance, however; are treated as variable-cost activities, because the frequency with which they are required is dependent on the engineering and traffic characteristics of the particular road.

5. INVENTORY
Content and preparation
5.1 The inventory is a set of information about the basic engineering and traffic characteristics of the road network. It defines the key features of each section of road and indicates the level of traffic use. This information is an essential reference source for the subsequent stages of inspection and analysis. 5.2 The content of the inventory should be directly relevant to maintenance management. When it is first drawn up, it should be as simple as possible and need contain information only on the following items: type of surface and construction carriageway and shoulders cross-section width carriageway and shoulders traffic volume annual average daily traffic (numbers of vehicles per day).

Presentation
5.6 There are three useful ways of presenting the information recorded in the inventory: diagrammatic maps strip maps card index systems.

5.7 Fig. 5.1 shows a typical diagrammatic map. It is basically a road plan of the area marked to indicate traffic levels, categories of road surface and road widths. This kind of map is particularly helpful in giving an overview of the whole network, enabling the maintenance engineer to see at a glance how roads with differing features relate to each other. 5.8 Fig. 5.2 is an example of a strip map. This is a simple annotated drawing which records significant information about a section of road and its surroundings. Its principal use is in the field, where it provides a quick means of reference during inspections and surveys - especially for the location of chainage. It is often convenient to staple strip maps together so as to form a pocket-sized notebook for each road. 5.9 Card index systems are useful for registering details of items such as road structures and road signs. They can easily be updated when these items are repaired or replaced, so as to provide a continuous record of their condition and maintenance history. 5.10 Where the means are available, it may be practical to use computer-based inventory systems; but the database of a maintenance inventory - particularly in its initial stages of development - will rarely be complex enough to require the application of computer technology.

5.3 As the inventory is built up, information on the following items can be added: structures pipe culverts; box culverts: bridges junctions location road furniture road signs; road markings; guard rails.

5.4 It will also be helpful to include data on other factors influencing maintenance needs, such as rainfall and runoff, topography and soil conditions. These factors can influence the degree of priority given to various operations when the work programme is prepared (para 9.9-9.10). In addition, data about the distribution and engineering properties of soils will be useful in identifying possible sources of maintenance materials. 5.5 Appendix A of this Note describes the field procedures used in setting up an inventory, including the organization of teams and transport. The maintenance engineer can delegate the preparation and day-to-day supervision of the work to a senior technician who fully understands the procedures, if one is available. If not, he should undertake this task himself.

Fig.5.1 Diagrammatic map of maintenance district

6.

INSPECTION

6.1 Recommended procedures for field inspections are described in Appendix A: they are similar to inventory procedures, but their emphasis is on the condition of the road rather than its basic characteristics. They identify locations where deterioration is occurring, measure the extent of the problem and define the action needed to put matters right. It is likely that, while the management system is being developed, the early inspections will have to rely largely on a visual assessment of defects: but measurement techniques should be introduced as soon as it is practicable to do so, following the guidance in Appendix A. Frequency of inspection
6.2 The maintenance engineer should have the entire length of the road network inspected at least once a year - and he should aim to improve on this frequency if it is at all practicable. In regions where there are distinct wet and dry seasons, a total inspection ought to be made in each part of the year. The wet season inspection will be particularly useful in detecting cracking in bituminous surfaces (since this defect is more easily visible when the road surface is drying after rain) and in assessing the efficiency of drainage. 6.3 The day-to-day supervision of inspection work can be delegated to trained technicians. But the maintenance engineer should visit the inspection teams, especially at locations where detailed assessments are required (para 6.5). 6.4 The network inspection will need to be completed in time for its results to be fed into the preparation of the following year's budget estimates. Since most organizations prepare their estimates in the second half of the financial year, the maintenance engineer has to make sure that the inspection programme is undertaken early enough to produce the necessary input.

Condition surveys
6.5 In assessing the condition of the road, it is advisable to adopt a two stage process of inspection: (i) In the first stage a trained but relatively unskilled team uses standard procedures and simple equipment to measure and record defects in the road. It is led by a technician who determines routine and recurrent maintenance needs and identifies locations where further examination is necessary. Occasional monitoring of these surveys by the maintenance engineer is recommended. The second stage involves a more experienced team, led by an engineer, whose task is to

(ii)

Fig.5.2 Strip-map of road inventory 6

determine the requirements for periodic work by making such additional investigations, measurements and analyses as are required.

6.6 The advantages of this approach are that it provides a double check on the state of the road and the scale of maintenance requirements, and it uses professional resources in a cost-effective way by directing them specifically to locations that call for skilled inspections and treatment.

Recording results
6.7 Figs 6.1-6.3 show how standard pre-printed forms can be used to record the results of inspection surveys. Each road should be divided into sub-sections and a separate form used for each sub-section. Different forms are needed for paved roads (Fig. 6.1), unpaved roads (Fig. 6.2) and for structures and road furniture (Fig. 6.3). Note that the form for unpaved roads includes space for the results of monitoring surveys that should be carried out at a later date to gauge the effectiveness of the maintenance programme (para 7.5). Blank versions of the forms are supplied in Appendix D.

6.8 Pre-printed forms are especially useful in providing a check-list that tells the technician what items are to be examined during an inspection, and so reduces the possibility that significant information may be omitted. The examples of forms shown here are intended only as a guide. In some cases the maintenance engineer may receive an issue of standard forms from his organization. In other cases, he may find it more useful to draw up his own form, designed to suit the particular road conditions in his area.

6.9 Whatever form is used, it should be easy to understand and to complete. The technician responsible for the inspection should fill in the results on site, recording them accurately and legibly. The forms should then be retained in the office so as to provide a permanent record of inspection results. There is no need to make fair copies of forms completed on site: this wastes time and involves the risk of errors when information is transcribed.

6.10 It may, however, be useful to summarise key results in the form of statistical tables or diagrams - for example, graphs that show rates of deterioration over time.

8 Fig.6.1 Example of inspection form for paved road

9 Fig.6.2 Example of inspection form for unpaved roads

10

Fig.6.3 Example of inspection form for structures and road furniture

7.

DETERMINATION OF MAINTENANCE REQUIREMENTS

recommended frequency. which is the one that should normally be adopted. Grading the road more frequently than the mean will give road users a higher level of traffic service: correspondingly, a grading frequency less than the mean will result in a lower level of service. 7.4 Dragging. This needs to be done regularly where loose material lies on the road or where corrugations are liable to form. In the latter case the operation may have to be repeated every few days. The maintenance engineer should determine the necessary frequency from a series of practical tests, by seeing how long it takes the corrugations to return after dragging is carried out. The frequency will vary for different materials, design of drag, traffic volumes and physical conditions. 7.5 Monitoring recurrent maintenance. Appendix B describes procedures for monitoring the effectiveness of recurrent maintenance activities on unpaved roads. The maintenance engineer should identify several roads that are representative of the network and arrange for a series of inspection surveys to be performed immediately before the dates on which recurrent maintenance is scheduled. The inspection results will enable him to check whether he has gauged accurately the appropriate frequency of grading and dragging operations. If, for example, a high degree of deterioration is still evident, he should adjust the existing schedule so as to perform these operations more frequently. As noted in para 6.7, the inspection form should include space to record the results of these monitoring surveys.

Recurrent and periodic maintenance of unpaved roads

7.1 As noted in para 4.2, unpaved roads need a programme of regular and systematic attention. The requirements for recurrent maintenance on these roads have to be determined independently of the results of network inspections. 7.2 Grading. The maintenance engineer must decide how many times during the year each unpaved road will need grading so as to provide the best level of traffic service possible in the local conditions. In assessing this, he will have to take account of a range of factors including the type and size of the road material, the amount of traffic using the road, the local topography, climatic regime and other physical features. 7.3 If he has access to a computing facility such as a micro or a programmable calculator, and has roughness measuring equipment available, he should use the method outlined in Appendix B to determine the optimum grading frequency for the road. If not, he should use Fig. 7.1 to gauge the frequency appropriate to the annual average daily traffic on the road. Fig. 7.1 is derived from studies undertaken by TRRL. The solid line indicates the mean

Fig.7.1 Grading frequency chart 11

7.6 Regravelling. This is a periodic activity that will need to be performed whenever the existing layer of gravel becomes unacceptably thin. If this layer is found to be less than 50mm thick for more than 20 per cent of the length of the sub-section of road being inspected (Table 7.1), regravelling should take place. Appendix A includes advice on the inspection procedure.

measurements to be made, as explained in Appendix A. Others involve simply a visual assessment. While the management system is being introduced, it is likely that the maintenance engineer will have to rely entirely on the visual assessment of defects, but he should try to introduce the use of measurement techniques as skills and resources permit.

Intervention levels on paved roads

7.7 In dealing with paved roads, the maintenance engineer must interpret the inspection results (Section 6) so as to decide when and where repairs are needed and what form of maintenance activity is required. Like a doctor treating an illness, he has to recognize the symptoms that indicate it is time to take remedial action. These indications are termed 'intervention levels': i.e. they identify the stage or circumstance in which the maintenance engineer should intervene and the action he should take to stop farther deterioration. Intervention levels can be defined for unpaved roads too, as shown in Table 7.1, though the frequency of recurrent operations is determined in advance (paras 7.1-7.5). 7.8 It is advisable always to adopt intervention levels suited to the particular local conditions. As a guide, Tables 7.1-7.3 set out recommended intervention levels for unpaved roads, paved roads, and road structures and furniture, respectively. The recommended levels are based on the assumption that the road network is already being maintained to an adequate standard and that sufficient resources are available to keep up this standard. In other words, they represent a target that the maintenance engineer should aim eventually to achieve. If, as will often be the case, he has too few resources at his disposal to apply the recommended levels, he can adopt alternative levels that are more appropriate to the workload and capabilities of his organization. 7.9 Some of the intervention levels in Tables 7.1-7.3 are expressed in quantitative terms and require

Diagnozing the cause of deterioration

7.10 It is important to identify the cause of deterioration and to put this right if possible, rather than just treating the symptom. For example, there is little point in continually filling in pot-holes in a road if they keep occurring only because of poor drainage. Finding the real problem and focusing attention on its solution will produce a more costeffective use of maintenance resources. 7.11 Some problems, of course, may be outside the scope of maintenance. For example, a road across flat country with inadequate drainage outfalls may experience base failure as a result of the capillary rise of water in the wet season. The only solution to the problem is to raise the level of the road. This would be a road improvement, not a maintenance operation: as such, it may warrant a special allocation of resources in accordance with the organization's procedures for road improvement work.

Specifying the work required

7.12 The maintenance engineer should examine the completed inspection forms (para 6.7) and compare the inspection results with intervention levels of the type recommended in Tables 7.1-7.3. He can then determine the maintenance needs of each sub-section of the road network and specify the work required. The necessary action should be marked on the inspection forms, preferably in a different colour from that used for the survey results. The forms will provide a permanent record of maintenance requirements.

TABLE 7.1 MAINTENANCE INTERVENTION LEVELS: UNPAVED ROADS

Extent (% of Defect Gravel thickness Camber Roughness Rutting Corrugations Pot-holes (See paras 7.1-7.5) Grade/drag Recurrent Planned on a programmed basis Level <50mm Sub-section length) >20 Action Regravel Programme Periodic Notes

12

TABLE 7.2 MAINTENANCE INTERVENTION LEVELS: PAVED ROADS

13

TABLE 7.2 (Continued)

14

TABLE 7.3 MAINTENANCE INTERVENTION LEVELS: STRUCTURES AND ROAD FURNITURE

15

8.

RESOURCE ESTIMATION

8.5 Contracting work out can have the advantage of

relieving pressure on the organization's labour resources and it can offer a high level of efficiency at a competitive cost. On the other hand, the organization will have to prepare detailed contract documents, set up a tendering procedure and provide contract supervision - all of which may require a substantial amount of time and effort on the part of its staff. There is also a risk that contractors bidding for maintenance work on a regular basis may introduce 'price fixing' to increase their profitability, and so involve the organization in higher costs. The maintenance engineer will have to weigh up these factors in relation to the capacities of his organization and the performance of local contractors.

Choice of work method

8.1 Maintenance operations involve three types of resources materials, labour and equipment. It is easy to estimate directly the quantities of materials needed for an operation (para 8.15), but the amounts of labour and equipment required will depend on the method used in carrying out the work. There are some activities that can only be done by manual labour; there are others where plant and machinery are essential; but many activities offer the option of either method. Trying to combine manual labour and machinery on the same task will normally lead to inefficiency. The maintenance engineer will therefore have to choose between a method that largely makes use of manual labour and a method based on the use of machinery. 8.2 In making this choice he has to take account of several factors: the probable costs of each method in relation to the available budget the standard of work resulting from each method. Machinery normally produces a truer surface and a more consistent finish than manual labour. but this may not always be required. For example, it would be difficult to find any economic justification for giving low-volume roads the close tolerances of level and smoothness that machinery can achieve the implications for the way the work is organized. If large-scale works are involved, labour-based methods will require precise organization and management. Labour has to be available in sufficient numbers in the right place and at the right time. Transporting the labour force to and from worksites can cause problems; standardised, goodquality tools and equipment will be needed to achieve high levels of output; the health and nutrition of the men must be thought about, since improved standards help to increase productivity.

8.6 The following activities should present no

significant problems of administration or quality control if contracted out on the basis of competitive tenders:

Supply of materials Steel reinforcement Bitumen products Cement and lime Natural gravel Screened gravel Rock aggregate (for subsequent crushing by the highways department) Crushed aggregate (for surface dressing and concrete) Washed sand (for concrete or slurry sealing) Precast concrete blocks Precast concrete box culverts or pipes Gabion baskets.

Maintenance operations
Recutting of side drains and turnouts Reconstruction of culverts Regravelling Resealing: surface dressing, slurry sealing, fog spray, etc. Grass cutting and bush clearing.

8.7 Specialist contractors can also be brought in to undertake the supply and maintenance of machinery and vehicles. 8.8 The maintenance engineer should satisfy himself that any contractor invited to tender for a job is capable of completing it satisfactorily, has the necessary personnel and equipment available, and is sufficiently knowledgeable about estimating procedures and current market rates to submit realistic prices.

8.3 Table 8.1 gives an assessment of the potential offered by various maintenance activities for the use of labour-based and machinery-based methods.

Use of contractors
8.4 Though maintenance organizations normally employ their own labour force, there is a range of activities that can be undertaken for them by local contractors. If the policies of his organization allow the use of contractors, the maintenance engineer should examine this possibility.

8.9 It is vital that the maintenance engineer ensures that

all contracts are properly supervised and that all specifications are met before he authorises payment.

Labour
8.9 Table 8.2 indicates the outputs of work that can be

16

expected from teams engaged on normal maintenance activities, using manual labour or machinery as appropriate. Each output is expressed as a range: the amount of work that a team in fact achieves should lie within this range, depending on local conditions. For example, a labour-based approach to clearing side drains is likely to involve between 4 and 10 men using simple hand equipment, and the team can be expected to clear between 30 and 60 metres of drain per man-day. A machinery-based approach to the same task will reduce the labour requirement to 2 or 3 men using a grader and shovels, and they will be able to clear as a team between 4 and 7 km per day. 8.10 The maintenance engineer should use Table 8.2 as a

guide in assessing the levels of productivity associated with various operations, and estimating the amount of labour required for each activity. 8.11 The values given in the table refer only to working time. They make no allowance for time that is non-productive, i.e. not spent actually working. Non-productive time can build up significantly during maintenance operations, and it should be an aim of management to reduce it to a minimum.

Equipment
8.12 Many activities need only simple equipment such as picks and shovels. It is important to have enough of these available to meet day-to-day requirements.

TABLE 8.1 POTENTIAL FOR USE OF LABOUR-BASED & MACHINERY-BASED METHODS ACTIVITY Ditch cleaning and cutting Cleaning and minor repairs to culverts and bridges Building scour controls Repair of structures Grading unpaved surfaces Dragging and brushing of unpaved surfaces Patching sanding or local sealing of bituminous surfaces Filling on unpaved surfaces and slopes Grass cutting Manufacturing signs Repairing and replacing traffic signs Road line markings Stockpiling gravel Regravelling gravel surfaces Stockpiling chippings Surface dressing POTENTIAL FOR LABOUR Good (*) Good Good Good Impracticable Poor Good Good Good GoodN (Skilled) Good Fair Fair Fair Poor Fair (Skilled) Poor Good Good Good Good Good (Skilled) MACHINERY Good (*) Poor Poor Poor Good (Skilled) Good Poor Poor Good(**) FairN

NOTES: (*)

(**) N

The potential in these activities is dependent upon suitable design of the ditch cross-section. 'V'-shaped ditches are suitable for maintenance by grader, whereas flat bottomed ditches are suitable for maintenance by hand or mechanical shovel. The potential in this activity is dependent on the width of the shoulder and presence of obstructions such as road furniture and culvert headwalls. Some methods of manufacture may require the use of specialised plant (eg. vacuum application of reflective sheeting to sign plates). The expression 'skilled' implies that specific training of operatives is essential.

17

TABLE 8.2 OUTPUTS OF WORK

* 18

The unit of pass-km is the actual distance the grader travels while working. To determine the length of road graded, this figure must be divided by the number of passes necessary to cover the whole width of the road.

TABLE 8.2 (Continued)

8.13 The maintenance engineer will probably not be responsible for the supply of plant and vehicles, since this is usually handled separately within the organization. But he should keep aware of the supply situation and must do his best to make sure that equipment is systematically maintained in good working order. Shortage of working equipment or vehicles is almost always a major symptom of inefficiency in maintenance organizations. 8.14 The essential point is to establish a daily routine of equipment maintenance. This can be done through the following measures carried out by the maintenance engineer: ensuring that operators appreciate the purpose and benefits of preventative maintenance, understand the maintenance needs of the equipment they use, and are trained to look after this equipment on a daily basis arranging for a regular supply of oil and grease, etc, to be made available; this supply will need secure storage facilities under the supervision of a storeman

checking for himself that the daily maintenance routine is being carried out, and demonstrating his personal concern for maintenance standards using his site visits to observe how operators treat their equipment and discouraging its mishandling or misuse.

Materials
8.15 The procedures for estimating quantities of materials needed in maintenance operations are relatively straightforward. Where the extent of the work can be defined precisely - for example, in the case of surface dressing or regravelling - quantities are taken off in accordance with standard engineering practice. For other activities where requirements are less easy to define such as edge repairs to bituminous surfaces, patching pot-holes and repairing defects in culverts -quantities are gauged on the basis of inspection results, again following standard practice; but these estimates will of necessity be less precise. The quantities ordered should include an allowance for any emergency work that may be needed, for example in repairing storm damage to culverts or vehicle damage to bridge parapets.

19

8.16 In planning his programme of work, the maintenance engineer should take account of the delivery periods for materials. If deliveries are likely to be uncertain, it will be advisable to stockpile basic materials such as aggregates cement and bitumen. These will need to be stored securely and kept in good condition.

Resource allocation
8.17 Fig 8.1 shows how a standard form can be used to record the resources required for each activity and their estimated costs. A blank copy of the form is included in Appendix D. 8.18 The extent to which the maintenance engineer can influence how the available budget is spent will depend on the resource allocation procedures in his organization. Some authorities may allocate specific amounts of labour, equipment and materials whose costs are charged for the whole year regardless of use. Others may only define a total budget and leave the maintenance engineer free to purchase materials and hire whatever labour and equipment are needed. In most cases, the procedures will fall somewhere between these extremes. 8.19 The maintenance budget may include sums allocated by the organization to various activities other than actual maintenance operations. The maintenance engineer must take these sums into account when calculating the resources available for road works. The following items may be involved: establishment: staff and buildings for materials testing, surveys, training, records, accounts and other administration equipment: operation and maintenance of transport and equipment, including workshop facilities emergencies: some funds may be set aside to cover the costs of reopening roads after emergencies or accidents contracts: supply of materials or equipment. or contractors services as part of the maintenance programme technical assistance: e.g. advice to other organizations or local communities.

20

Fig.8.1 Example of resource requirements form 21

9.

IDENTIFICATION OF PRIORITIES

Periodic work
Regravelling of unpaved roads Surface dressing of paved roads Periodic work can be treated as a series of distinct projects that have to compete for the resources available and can be undertaken separately, deferred or brought forward as required.

9.1

If he is fortunate, the maintenance engineer may find that he has all the resources he needs to carry out the full programme of work. But it is more likely that resources will be limited and he will have to decide the most effective way of applying them. This means working out an order or priority, with the operations that have the strongest claim on resources placed at the head of the list and those that have least claim placed at the end. This section of the Note explains a method that will enable the maintenance engineer to identify priorities objectively and consistently. 9.2 The method is straightforward: it simply relates the importance of the maintenance activity to the importance of the road. It asks two basic questions: (i) how critical is a particular maintenance activity to the traffic performance of the road? (ii) how significant is the particular road as a transport link?

Other routine work

Filling on shoulders and slopes Grass cutting Cleaning, repainting, repairing and replacing road furniture

Special
Overlaying Reconstruction These activities should be treated as capital projects whose funding does not come out of the maintenance budget. If this principle is not established, there is a danger that a major overlay or reconstruction scheme will swallow up virtually the whole of the maintenance budget, leaving nothing for routine and recurrent work.

Maintenance activities by order of importance

9.3 Maintenance activities may be ranked in the following order of importance:

Urgent
Emergency repairs to blocked or impassable roads Removal of debris and stabilisation of side slopes By definition, urgent works of this type demand top priority: past experience will indicate the scale of the resources that need to be set aside for emergencies.

Roads by order of importance

9.4 The roads that carry the heaviest loads of traffic are normally the most important parts of the network from an economic standpoint, and they are the roads liable to deteriorate most rapidly from wear and tear. There may also be roads with relatively low levels of traffic which nevertheless have key strategic importance because of the places they link. Generally there will be only one or two such roads in any region. Since it is vital to keep these strategically important roads in good condition, it makes sense to give them top priority for maintenance work. 9.5 The remainder of the network should be classified by the level of traffic on each road. This level is usually expressed in terms of the estimated annual average daily traffic (ADT); i.e. the total traffic in both directions during the year. divided by 365. The estimate can only be reliable if it is derived from actual traffic counts. This means that traffic information is needed about every part of the network, including minor and unpaved roads. For the purposes of maintenance planning, a simple traffic count procedure as outlined in Appendix C will be adequate; but the maintenance engineer should note that counts lasting for less than 7 consecutive days are liable to very large errors, particularly where traffic levels are low.

Routine drainage work

Cleaning out and recutting ditches and turnouts Cleaning out bridges and culverts Filling scoured areas Building check-dams and scour controls Repair of drainage structures This work always deserves high priority because neglected drainage can rapidly lead to deterioration of the whole road. It should not be assumed that all drainage work must take precedence over all recurrent work. The key point is that repairing surface defects caused by poor drainage will be a waste of time and effort unless the drainage is put right first.

Recurrent work
Dragging, brushing, grading or filling of unpaved roads Patching or local sealing of paved roads

22

TABLE 9.1
Priority matrix
ROADS CLASSIFIED BY CATEGORY OF IMPORTANCE 9.8 Fig 9.1 is a matrix which shows how the classification of maintenance activities should be related to the classification of roads. Maintenance activities are numbered from 1 (highest priority - urgent maintenance on strategic roads) to 48 (lowest priority -special works on unpaved roads with very low levels of traffic). The matrix is designed to ensure that every road in the network receives at least the minimum maintenance needed to keep it operational, while at the same time focusing recurrent and periodic maintenance on the economically important roads with high traffic levels. As noted in para 9.4, strategically important roads have absolute priority for resources, even where special works are concerned.

Category 1 2 3 4 5 6 7 8

ADT (Strategic roads) Greater than 1000 500 1000 200 500 Greater than 200 Less than 200 50 200 Less than 50

Surface Type Paved Paved Paved Paved Unpaved Paved Unpaved Unpaved

Adapting priorities to local conditions

9.6 Table 9.1 sets out a simple classification of roads by category of importance based on traffic level. It is meant as an example which the maintenance engineer can adapt to reflect the character of his road network and the general levels of traffic. 9.7 An axle load survey, in addition to traffic counts, may be justified if the traffic composition on a particular road is believed to be significantly different from that on the network as a whole - for example, on a road that carries large numbers of heavy freight vehicles. The survey, which can be completed in a few days using a portable weighbridge, involves the preparation of a suitable site and usually will require the assistance of the police in enforcing the operation. The Transport and Road Research Laboratory (1978) has published a booklet (Road Note 40) which offers guidance on procedures for axle load surveys. 9.9 The maintenance engineer should use the matrix to produce a master list identifying in order of priority all the works that need to be done on his network. While the numbering scheme in Fig 9.1 is a rational and consistent order of priority, it is not meant to be inflexible. There may well be local conditions of soil type, topography and climate which influence maintenance requirements on different parts of the road network, and these may warrant an amended order of priority. The maintenance engineer should feel free to reorder the list of priorities to suit local conditions, where necessary. 9.10 The type of local factors involved may include, for example, the presence of expansive clay subgrades which are likely to cause maintenance problems, or the combination of steep slopes and high rainfall where runoff may accelerate the process of erosion.

Determining the work programme

9.11 The next step is to compare the resource costs of the proposed maintenance works with the resource funds available. The type of form shown in Fig 8.1 provides a useful means of doing this, since it summarises the resource requirements and estimated costs of each maintenance task (para 8.17). The forms should all be placed in a ring binder, initially following the order of priority defined in the master list, with the top priority tasks at the front and the lowest priority tasks at the end. The order of the forms in the file can be changed where necessary to take account of local conditions as described in para 9.9. By working through the file, the maintenance engineer can calculate how far the available resources will extend, and so he is able to identify which tasks he can include in his programme for the year and which tasks he will have to defer until the next financial year. 9.12 Budgeting procedures within his organization may mean that the funds available for routine, recurrent and

Fig.9.1 Matrix of maintenance priorities

23

periodic maintenance are not interchangeable. Some flexibility and adjustment of priorities may be necessary to make sure that all the funds available for a particular category of work are used. 9.13 It is often useful to have routine and recurrent maintenance work organized at a sub-district level. In this case each sub-district should be given a programme of work based on the priorities defined in the master list. Periodic maintenance work organized directly at district level should have its priorities assessed in the same way. In this situation, the maintenance engineer should give careful thought to the location and organization of his maintenance crews so as to check whether existing arrangements meet the requirements of the district as efficiently as possible. He should recommend to headquarters any necessary changes.

1O. WORK SCHEDULING AND EXECUTION

Schedules
10.1 The maintenance engineer's next task is to prepare detailed work schedules for the teams under his direction. The schedules are essentially sets of instructions which tell the foremen or technicians supervising an activity how much work is to be done each day, the time it should take and the labour, equipment and materials to be used. Schedules should cover periods of not less than 2 weeks and not more than 4 weeks: shorter periods would probably not warrant the effort of preparing a schedule, while longer periods would risk losing a sense of urgency. 10.2 In drawing up schedules, the maintenance engineer should first discuss the work with the foremen and supervisors who will have to put the schedules into practice. It is essential for these personnel to feel confident that the 'production target' the output of work required each day is pitched at a realistic level. One useful idea is to ask the foremen to draft their own schedules: the maintenance engineer can use these as a basis for discussion and the results can be helpful in developing the work schedules that are issued to teams.

Worksheets
10.3 Fig 10.1 shows an example of a completed worksheet. This provides a daily record of the progress made on a job and the resources used, allowing the actual output of work to be measured against the target output. The target is the first item to be recorded on the worksheet; the amounts of each resource to be used are entered next, in the top half of each line. The worksheet is then issued to the supervisor who enters in the bottom half of each line the progress actually made and the resources used day by day. At the end of the schedule period, the supervisor returns the completed forms to the maintenance engineer who calculates in percentage terms how much of the production target has been achieved and so is able to assess the productivity of the work. Any significant shortfalls that may occur, or instances of overachievement, can be discussed with the supervisory personnel when the next schedule is being prepared.

10.4 The execution of maintenance tasks at district level is the subject of Overseas Road Note 2 (TRRL Overseas Unit 1985).

24

25 Fig.10.1 Example of worksheet

11.

MONITORING

REFERENCES
TRANSPORT AND ROAD RESEARCH LABORATORY, 1978. A guide to the measurement of axle loads in developing countries using a portable weighbridge. Road Note 40. London: HMSO, available from TRRL Overseas Unit. TRRL OVERSEAS UNIT, 1985. Maintenance techniques for District Engineers. Overseas Road Note 2. Crowthorne: Transport and Road Research Laboratory, 2nd edition, available from TRRL Overseas Unit.

11.1 Monitoring serves two main purposes: it enables the maintenance engineer to check the quality and effectiveness of the work being done, as noted in para 4.1; and it provides data that can be used to improve the management and performance of future maintenance operations. There are two aspects to the task: site inspections desk review.

Site inspections
11.2 Section 2 of this Note emphasized the importance of site inspections in the work of the maintenance engineer and the advantages to be gained from taking a personal involvement in the inspection process (para 2.6). Because of the length of road to be covered, inspection visits will require the use of a motor vehicle, which should be driven slowly along the section being examined. Enough time should be allowed for the maintenance engineer to make a series of detailed inspections on foot. He should examine closely the road surface, side drains, culverts and soils. It is best if he travels as a passenger: he can then concentrate on his inspection, and when he goes off to look at the road his driver can follow with the vehicle.

Desk review
11.3 This is an office task which involves reviewing all the maintenance documentation - inspection reports, resource requirement forms, work schedules and completed worksheets - so as to assess the performance of the programme and the effectiveness of the management system. It provides an opportunity to check that adequate resources were allocated to each task and that maintenance problems were treated efficiently; cost estimates can be compared with expenditure and production targets with output; and the reasons for inconsistencies or shortfalls can be identified. In some cases, resources may have been insufficient; in others, targets may have been pitched too high. 11.4 The maintenance engineer should discuss the results of the desk review with this supervisors and try to find ways of remedying any deficiencies. It is likely that some of these may result from factors beyond the control of the maintenance staff, but it should be possible to identify those points that can be put right and to plan appropriately for the following year.

26

APPENDIX A

TABLE A1 EQUIPMENT FOR INSPECTIONS Item No. 1 2 3 4 5 6 7 8 9 10 11 12 Description Transport Warning signs and reflective jackets Maps of the area Field inspection instructions and codes where applicable Printed inspection forms Report sheets for 'urgent' defects Clip-boards Ballpoint pens or pencils Road marking crayons or chalk, road marking paint and brush 30m fibre measuring tape 3m steel measuring tape Measuring wheel with stand or, if the road is adequately provided with km posts a bicycle fitted with a distancemeasuring cyclometer 2m straight-edge and calibrated wedge Spirit level for measuring camber with a 2m straight edge

FIELD PROCEDURES FOR INVENTORY AND CONDITION MEASUREMENT SURVEYS

A.1 GENERAL PROCEDURE

Recording
A.1.1 Inventories and inspections use largely the same field procedures. Standard forms for recording the results of surveys are included in Appendix D. A.l.2 The road should first be divided into sections of 5 to 20 kilometres. These sections should be reasonably uniform in terms of their geometric characteristics, construction, terrain and traffic, but they need not all have the same length. Each section should then be divided into sub-sections of 100 to 200 metres in length. A.1.3 The aim is to measure and record all defects appearing in each sub-section. Because this is a timeconsuming process, it may only be possible initially to take representative measurements over perhaps one 25 metre length in each sub-section. As inspection teams become more skilled and as more teams are introduced, it should become practicable to increase the coverage of the surveys until measurements are taken over the entire sub-section. A.l.4 The extent of characteristics or defects should be recorded on the forms by simply marking the locations where defects occur. Defects whose extent is assessed in terms of length (eg cracking) are indicated by recording the assessment of extent (Fig 6.1). Table Al sets out a check list of the equipment required for the field work of inspections. At first, only items 1-8 will be needed since the inspection will be carried out largely by visual assessment, but as the maintenance system develops and as measurement methods are introduced, the remaining items can be added to the list.

13 14

Safety
A.1.6 This is a most important aspect of the work. The road being inspected may have vehicles travelling at high speed. Any person working on the road should therefore take simple and sensible precautions. Even where traffic is not heavy, safety procedures should not be relaxed, since there may then be a danger from unexpected vehicles. A.1.7 All members of the field survey team should wear high visibility fluorescent and reflective safety jackets at all times during the survey. In rural areas, warning signs should be placed at each end of the work area. If a vehicle is provided it should be fitted with warning signs or warning lights, and should be parked between the team and the oncoming traffic.

Organisation of teams
A.1.5 The basic duties of the team are the observation, measurement and recording of information. Composition of teams will depend upon local circumstances and the type and category of road, but normally the team will consist of a team leader and an assistant. In some cases, where traffic densities are high, an additional assistant will be needed for safety purposes and to help direct traffic. If a detailed structural survey of the road is to be carried out at the same time as the inspection, additional team members may also be needed.

Transport
A.1.8 Transport is needed to carry teams to and from their place of work, to carry equipment, to give temporary shelter during bad weather and to provide protection on heavily-trafficked roads. Either land rovers or vans can be used, but the land rover has a particular advantage as a protective vehicle in that warning signs can be easily attached to it. Where possible, vehicles should display hazard warning lights. If the vehicle is used simply to transport the survey team at the start and the end of the day's work, the team leader and the driver must be in no doubt as to the time and place of collection.

27

Notes for inspectors

A.1.9 The maintenance engineer should prepare notes for inspectors which include further guidance on the recommendations given in this Note and rules and interpretations suited to local conditions. The notes should include any additional detail necessary for assessing and recording individual items, together with a list of standard criteria. A.1.10 Clerical work should be done by the inspectors. This will give them closer involvement in the management process, and is likely to result in higher standards of inspection and recording.

A.1.13 Once on site. the inspector must establish the start of the length to be inspected. From this point, he and his team should methodically record defects or features. The start point should be a kilometre post or a structure, if possible.

Accuracy A.1.14 Correct, accurate and legible recording in the

field is essential to avoid unnecessary effort in checking errors later. It is important to record the correct chainages for the start and end of defective areas, in the case of condition measurement surveys, or changes in characteristics in the case of an inventory. Chainages should be recorded to the nearest metre, and widths to the nearest 0.1 metre. For condition measurement surveys, it is also useful at this time to paint the outline of areas of paved roads that need patching, since this will make them easier to find when the work is carried out later.

Duties of team members

A.1.Il Team leader He has the overall responsibility for the work of the
team. He should carry out a visual inspection of the carriageway, shoulders and drainage, assess structural condition and record all information on the relevant forms. With a team of two, the leader usually pushes a measuring wheel

A.2

CONDITION MEASUREMENT

General considerations
A.2.1 In order to establish maintenance needs, it is necessary to compare the measured condition of the road with maintenance intervention levels such as those set out in Tables 7.1-7.3. These levels may need to be modified by the maintenance engineer to suit local conditions as noted in para 7.8.
A.2.2 When the maintenance system is introduced, it will probably only be possible for the inspector to assess the severity of defects by observation. As the system is developed, physical measurement of defects should be gradually introduced into the inspection procedure; but the quality of visual assessment can be improved by having first-hand knowledge of the physical measurement techniques. A.2.3 Most defects require either the area or the length of deterioration to be measured. The area or length of deterioration must first be identified and classified using the method outlined in the following paragraphs. As each sub-section is approximately 100-200 metres long, it is probable that several occurrences of the same type of deterioration will be found within the subsection. These should be added together to give the total extent for each sub-section.

Assistant
When inspecting procedures are first introduced, the team leader will make his assessment on the basis of a visual inspection. As the maintenance system is developed, assessments will be based on the detailed measurement of defects. An assistant will be needed to carry out these measurements. On roads where traffic is heavy, additional assistants may be needed, in particular to make rutting measurements on the far side of the carriageway, to control traffic and generally to assist in measurement procedures

Driver When a vehicle is provided, a driver is necessary to

enable the inspector to concentrate on the job in hand. The driver should be responsible for erecting and moving advance warning signs where no other person is provided for this task. He can also assist in traffic control or in making measurements. Where possible he should, as a safety measure, park his vehicle on the road between the on-coming traffic and the inspection team and switch on his hazard warning lights.

Order of work
A.1.12 Before leaving the depot, the inspector should personally check that all of the equipment needed has been collected together and loaded in the vehicle. He should ensure that the vehicle has sufficient fuel or that arrangements have been made to refuel on the way.

Side drains and turnouts (all roads)

A.2.4 The level of the water table beneath the road has a major influence on the strength of the subgrade. To keep water out of the pavement structure, side drains should

28

be maintained at least one metre below the road surface. The depth of side drains should be measured at approximately 25 metre intervals using a straightedge and measuring tape as indicated in Fig. Al.

Cracking (paved roads)

A.2.10 A crack can have the appearance of a line in the surfacing (particularly for asphalt) without two clearly defined edges. This condition indicates an early stage in the deterioration process, when the surfacing still remains waterproof. For the purpose of condition measurement surveys, it is more practical to define a crack as being visible from standing height. Such a crack would, on close inspection, have two clearly defined edges at the surface. It should be remembered that this condition is much further along the deterioration path than that described above. Water may be penetrating into the pavement through cracks of this severity whether the road is surfaced in asphalt or surface dressing. A.2.ll In bright sunlight, it is often very difficult to see even quite wide cracks. A method of shading the area under observation should be used. The width of cracks in asphalt can change markedly between early morning and mid-day owing to thermal expansion. In some cases, cracks can close completely and not be seen at all. Surveys are best carried out in the morning or during cloudy weather conditions, particularly when the road is drying out after rain. A.2.12 During surveys, cracking should be recorded in two classes: wheeltrack cracking and non-wheeltrack cracking. In each case, the total length of road affected by cracking should be recorded regardless of the area or severity of cracking across the road.

Loss of material (gravel roads)

A.2.5 An estimate can be made of the thickness of gravel on the road by examining pot-holes or by digging a small hole in the road surface until subgrade is reached and probing the depth with a measuring tape. Material should then be replaced and compacted with a hand tamper or punner. A.2.6 Ideally, thicknesses should be measured immediately after grading. But in view of the difficulty of timing such measurements and the need to carry out surveys on a routine basis, the following approach may be more appropriate. The survey team should take measurements on each sub-section or at 200 metre intervals along the road. At each survey point, the team should identify any obvious ruts and should dig two holes: one in the rut and the other at the peak between ruts. The gravel thickness should be recorded as the average of the two measurements, as shown in Fig. A2.

Deformation (all roads)

A.2.7 Ruts are measured using the deformation gauge shown in Fig. A3. Initially, measurements will be made at only one point on each sub-section or at 200 metre intervals along the road, but as the survey procedure is improved and extended, more frequent measurements should be taken until recordings are made every 25 metres. A.2.8 Rut depths are measured by placing a two metre straightedge transversely to the road edge over the wheelpath. The deepest part of the depression beneath the straight edge is measured and recorded as the rut depth using a calibrated wedge as shown in Fig A3. The depth of rut indicated is the value recorded. In some cases, a bump may have formed at the edge of a wheelpath owing to shoving of material in the pavement. The condition should then be reported as a comment on the form and an indication of the seriousness of the condition obtained by placing one end of the straightedge on the bump and the measuring the maximum rut depth in the wheelpath. Measurements of rutting are made in the two wheeltracks nearest the edge on both sides of the road, and the higher of the two rutting values is recorded. A.2.9 Corrugations on unpaved roads are measured by placing the straightedge longitudinally in the wheelpaths spanning the corrugation crests. Corrugation depths are measured at the deepest point of the troughs using the calibrated wedge. The highest value of depth from any trough in any wheelpath is the value that should be recorded.

Pot-holes (all roads)

A.2.13 The number of pot-holes in the sub-section should be counted and recorded.

Edge damage (paved roads)

A.2.14 This is measured at the edge of the paved area of the road at the same chainage where rutting is measured. Excessive edge damage is recorded if the edge is eroded by more than l50mm from the original position. The measuring wedge, which is I50mm long, provides a simple means of measuring this defect.

Edge step (paved roads)

A.2.15 This is measured at the same chainages as the rutting. Measurements are taken on both sides of the road at the point where the carriageway meets the shoulder. The measuring wedge should be placed on the shoulder with its 'toe' against the edge of the carriageway. If the step is greater than the 50mm marker, an excessive edge step should be recorded. A.2.16 On certain roads, particularly where an overlay has been applied, there may be an edge step which is a construction feature. In this case, measurement of the step should still be made in the same way. If the measurement exceeds 50mm, a defect should be recorded.

29

Fig.A1 Measuring depth of side drains with straight edge and measuring tape

Fig.A2 Measurement of gravel thickness as the average of two measurements 30

Fig.A3 Deformation gauge (straight-edge and calibrated wedge)

31

APPENDIX B
MANAGEMENT OF GRADING FOR UNPAVED ROADS
B.1 DETERMINING OPTIMUM GRADING FREQUENCIES B.1.1 Ideally, an appropriate frequency of grading should be identified for each individual road. However, it is more practical to determine optimum grading frequencies for groups of roads with similar traffic levels, similar materials types and sizes, and sharing similar topographic and other physical features. The more often a road is graded, the smoother its surface will become and the less deterioration there will be between gradings. The end result will be lower vehicle operating costs on the road. The aim is to choose a frequency of grading which minimises the sum of maintenance costs and vehicle operating costs during the maintenance year. B.1.2 The easiest way to determine vehicle operating costs for particular grading frequencies is to use a road investment model such as RTIM3 or HDM-III. Details of how to obtain copies of RTIM3 are available from: Overseas Centre Transport Research Laboratory Crowthorne Berkshire RG45 6AU United Kingdom. B.1.3 from: Details about obtaining HDM-III are available

Transportation Department The World Bank 1818 H Street NW Washington DC 20433 USA. B.1.4 If computer facilities are not available, tables of vehicle operating costs can be obtained free of charge from TRL Overseas Centre at the above address. However, use of the tables for this particular application is very laborious and is recommended only if there is no other option. A better method is to program the relationships in the tables on a suitable hand calculator and to use these to determine vehicle operating costs. B.1.5 The analysis must be carried out for each selected road surface material type, climate, road geometry and traffic level, and vehicle operating costs determined for a range of grading frequencies. The results should be plotted in graph form. For each material type, climate and road geometry, curves such as those shown in Fig. B 1 should be obtained. The minimum cost plotted at each traffic level will indicate the optimum grading

Fig.Bl

Typical curves showing optimum grading frequency for different traffic levels on roads with similar climate, materials and geometry

frequency. The grading frequency should be applied to particular roads on the network by selecting the traffic level on the graph closest to that found in the field. Different optimum values will be obtained, not only for different traffic levels, but also for roads with different material types, with different geometries and in different climatic areas. It is necessary to undertake separate analyses for each of these cases.

B.2

MONITORING OF OPTIMUM FREQUENCIES

B.2.1 By whatever means the optimum grading frequencies are determined, they should be implemented in the field and a monitoring system should be established to check their effectiveness. A representative sample of unpaved roads in each group should be

32

monitored for a period of at least one year, and measurements should be taken of roughness and rut depth immediately before and after each scheduled maintenance grading. Traffic levels on these roads should also be monitored. B.2.2 Modifications to the grading frequencies can then be made to ensure that the mean values of roughness and rut depth measured in the field agree with those required by the analysis using the investment model. This ensures that the sum of road maintenance and vehicle operating cost is minimised. In this way, by the end of the year, grading frequencies will have been obtained that are suitable for local conditions and are close to the optimum. A sample of the road network should have its performance monitored on a continuing basis to make sure that the grading frequencies effectively keep roads in a condition that is close to the optimum. Adjustments can then be made to the grading frequencies as necessary. B.3 MEASUREMENT METHODS

an estimate of roughness to be obtained without complex calculation (Cundill 1991). Drawings of the MERLIN are available from the Overseas Centre, TRL. REFERENCES CUNDILL, M A, (1991). The MERLIN low-cost road roughness measuring machine. TRRL Research Report RR 301. Crowthorne: Transport Research Laboratory, available from TRL Overseas Centre. SAYERS, M W et a l, 1986. Guidelines for conducting and calibrating road roughness measurements. World Bank Technical Paper 46. Washington DC: International Bank for Reconstruction and Development.

B.3.1 Rut depths should be measured with a straightedge and wedge following the method described in Appendix A. B.3.2 Roughness will normally be measured with a 'response-type' instrument such as a 'bump integrator', 'NAASRA meter' or Mays ride meter'. It will be necessary to calibrate this instrument to ensure that the results obtained are consistent with either the International Roughness Index (IRI) or the 'towed fifth wheel bump integrator' values used for vehicle operating cost calculations in the road investment models. Equations now exist (Sayers 1986) for relating these standard roughness values to the absolute longitudinal profile of the road. The response-type instruments used can be calibrated by operating them over roads with various levels of roughness. The longitudinal profile can be determined by measuring the road with a staff and engineer's level. However, this method is laborious and the calculations needed to determine the calibrated values are time-consuming. Two instruments have been developed by TRL which provide alternative methods for calibrating response-type measuring devices more easily, the MERLIN and the Profile beam ('Abay' beam). The TRL Profile beam is a Class 1 instrument (Sayers 1986) which samples the profile of a road section at l00mm intervals. It has a microprocessor to calculate the standard roughness values and can provide an automatic readout of calibrated roughness. The MERLIN is a Class 2 instrument. It is a simple, robust and inexpensive device which is easily maintained, easy to use and can be made by local craftsmen. It measures surface undulations at intervals along the road section and has a graphical procedure for data analysis to enable

33

APPENDIX C
TRAFFIC COUNTING
C.1 FREQUENCY AND DURATION OF COUNTS

The ADT is taken as the weighted mean of the seasonal flows, calculated thus: ADT = (8/12 x 28) + (4/12 x 54) = 37 vehicles

C.3

MANUAL COUNTS

C.1.1 The purpose of the traffic counts is to classify roads in broad categories for maintenance planning purposes. A high level of accuracy is not required. On roads carrying more than about 1000 vehicles per day, a one-day manual count will usually be sufficient to provide an estimate of ADT for the maintenance classification. On roads with fewer vehicles, account should be taken of daily and seasonal variations: the effects of local market days, which may account for a high proportion of annual traffic on low-volume roads, may be allowed for by making two counts, one on a market day and one on a non-market day a count should be carried out during each main climatic and/or agricultural season so that the effect of seasonal rainfall and of variations in agricultural activity (especially during and just after harvest-time, when traffic flows usually show a marked increase) are taken into account. Counts extending over 16 hours (from 6 am to 10 pm) will usually be adequate, but, in some countries where traffic is heavy at night counts should extend over 24 hours.

C.3.1 The equipment required for manual counting comprises a supply of pre-printed forms, pencils and clip-boards. A shelter from sun or rain may be required. At low traffic volumes, the work is not demanding and the main problem is likely to be ensuring that the enumerators stay on the job. Unscheduled visits by the engineer or a senior supervisor may be necessary to secure discipline and attention to the work. If the proportion of heavy vehicles in the traffic flows is expected to be high, classified counts may be necessary.

C.4.

AUTOMATIC COUNTERS

C.4.1 It is now possible to obtain cheap and reliable automatic traffic counters recording total numbers of axle pairs which can be used to supplement manual counts or to count for periods of more than one day on low volume roads. The equipment normally consist of a small rubber tube fixed across the road and attached to a pneumatic counter. When vehicles drive over the tube. it is compressed and a pulse of air triggers the counter. The instrument records one count for the passage of two axles. These traffic counters need careful adjustment to ensure that bicycles, motorcycles and pedestrians are not counted and that heavy or fast vehicles are not 'double counted'. Provided that they are checked regularly and the rubber tube replaced when it is damaged, their use is recommended. C.4.2 Buried loop detectors can be used instead of pneumatic tubes and are generally more reliable. However, their permanent installation will not normally be justified solely for traffic counts for maintenance purposes. C.4.3 The maintenance engineer should select one of his staff trained in the installation and use of counters to be responsible for their checking and maintenance. He should also arrange for the daily inspection of the counter and recording of the counter reading. In some locations, automatic counters may be subject to theft or damage. The counters themselves can be protected by lockable cabinets attached to solid objects, but the tubes cannot be effectively protected from deliberate interference. When this is experienced and regular inspection is not sufficient to prevent it, automatic counters of this type should not be used.

C.2

ESTIMATION OF ADT FROM COUNTS

C.2.1 The estimation of ADT must take account of the factors noted above. The following example indicates the recommended procedure:

Example
C.2.2 Counts were taken near a village which has a market every eight days. The wet season lasts for about four months and the dry season eight months. 1. Two 16-hour counts were taken at the end of the dry season (low level of agricultural activity) Count on market day = 73 vehicles Count on non-market day = 21 vehicles Two 16-hour counts were taken at the end of the wet season (harvest time) Count on market day = 94 vehicles Count on non-market day = 48 vehicles

2.

C.5

MOVING OBSERVER COUNTS

C.2.3 For this example of an eight day traffic cycle, the average daily traffic is calculated for each season as follows: Average daily traffic (dry season) Tdry = (1/8 x 73) + (7/8 x 21) = 28 vehicles Average daily traffic (wet season) Twet = (1/8 x 94) + (7/8 x 48) = 54 vehicles

C.5.1 Moving observer counts can usually be carried out by the maintenance engineer or supervisor in the course of personal inspections, using a hand-tally to record the

34

number of vehicles. The flow can be estimated from the expression: (x + y) q = t

where

q = total flow in both directions in time t x = number of vehicles met (ie. travelling in the opposite direction) y = number of vehicles that overtake the observer minus the number he overtakes t = journey time

C.5.2 This expression assumes that flows in each direction are equal. If the observer can drive so that he passes as many vehicles as pass him in the direction of travel (this can usually be achieved without risk at low traffic volumes), the expression becomes:x q = _______ t C.5.3 Counts of this type are useful as a cross-check on static counts and to assess the extent of variations in flow from day-to-day or between one season and another.

C.6

DEVELOPMENT OF TRAFFIC COUNTS

C.6.1 When the maintenance system is first applied, it will probably not be possible to include traffic counts as part of the maintenance planning process. Initially, traffic levels on the roads within the maintenance district may have to be estimated. But there is no substitute for counting vehicles. As the maintenance system develops, the traffic counting methods described in this Appendix should gradually be introduced into the management process.

35

APPENDIX D
RECOMMENDED STANDARD FORMS

36

37

38

39

40

41

APPENDIX E
ILLUSTRATIONS OF TYPICAL DEFECTS
Photographs are provided that illustrate the most common, or important, defects described in Tables 7.17.3 of recommended maintenance intervention levels. For paved roads, more categories of cracking are shown than are described in the text to assist engineers in determining possible causes of failure. To aid clarity, the photographs, in most cases, show deterioration which is worse than that at which intervention is recommended.

42

Debris on road.

Broken down vehicle.

43

Side drain silted and too shallow.

Blocked side drain.

44

Scoured side drain

Culvert outfall scoured.

45

Scoured side drain.

Standing water in side drain.

46

Structural damage of bridge superstructure.

Structural damage of bridge abutment.

47

Dirty road sign.

Damaged road sign.

48

Missing road sign.

Damaged road furniture.

49

Deformation of shoulder.

Scour of shoulder and sideslope.

50

High vegetation growth and standing water.

Pot-holes on unpaved road and high roadside vegetation growth.

51

Roughness and lack of camber on unpaved road.

Rutting and lack of camber on unpaved road.

52

Roughness, rutting, pot-holes and lack of camber on unpaved road.

Corrugations on unpaved road.

53

Stripping of aggregate from paved road.

Fatting-up and bleeding on paved road.

54

Pot-hole on paved road.

Edge damage on paved road.

55

Edge step on paved road.

Edge damage and step on paved road.

56

Wheeltrack rutting and cracking on paved road.

Wheeltrack rutting, cracking and pot-hole on paved road.

57

Crocodile cracking on paved road.

Block cracking on paved road.

58

Longitudinal cracking in the wheeltrack on paved road.

Longitudinal cracking on the centre line on paved road.

59

Transverse cracking on paved road.

Longitudinal and transverse cracking on paved road.

60

Shrinkage cracking on paved road.

Cracking (and pumping) at local deformation on paved road.

61

Parabolic cracking and slipping on paved road.

Slipping at edge of paved road.

62

ISSN 0951 8987

OS-E

Maintenance management for district engineers

ORN 1

Transport and Road Research Laboratory Overseas Unit Department of Transport Overseas Development Administration

Overseas Road Note 2

Maintenance techniques for District Engineers (2nd Edition)

Overseas Unit Transport and Road Research Laboratory Crowthorne Berkshire United Kingdom 1985

ACKNOWLEDGEMENTS
This Note was drafted by Mr L S Hitch, Mr T E Jones and Dr R Robinson of the TRRL Overseas Unit. It is based on an original text produced by Scott Wilson Kirkpatrick and Partners, consulting engineers.

First published 1981 Second edition 1985 Reprinted with minor revisions 1987

OVERSEAS ROAD NOTES

Overseas Road Notes are prepared principally for road and road transport authorities in countries receiving technical assistance from the British Government. A limited number of copies is available to other organisa tions and to individuals with an interest in roads over seas, and may be obtained from: Transport and Road Research Laboratory Crowthorne, Berkshire RG11 6AU United Kingdom Crown Copyright 1985 Limited extracts from the text may be produced provided the source is acknowledged. For more extensive reproduction, please write to Head of Overseas Unit, Transport and Road Research Laboratory

CONTENTS Page 1. Introduction 1.1 Scope of this Note 1.2 The Engineers responsibilities Methods of undertaking maintenance 2.1 Labour and equipment 2.2 Direct-labour and contract Safety measures and traffic control 3.1 Principles 3.2 Edge working 3.3 Lane closure 3.4 Centre line working 3.5 Diversions 3.6 Development of safety measures Maintenance of road furniture 4.1 Traffic signs 4.2 Rails and posts 4.3 White lines and reflective studs 4.4 Maintenance gangs Maintenance of drainage, shoulders and slopes 5.1 The drainage system 5.2 Side drains 5.3 Fords and drifts 5.4 Culverts 5.5 Bridges 5.6 Shoulders 5.7 Slopes 5.8 Maintenance gang Maintenance of unpaved roads 6.1 Types of maintenance 6.2 Grading 6.2.1 Types of grading 6.2.2 Crossfall 6.2.3 Steep hills 6.2.4 Grader operation 6.2.5 Grading gang 6.3 Dragging and brushing 6.3.1 Design of drag 6.3.2 Brushing 6.3.3 Method of operation 6.3.4 Dragging gang 6.4 Regravelling 6.4.1 The task 6.4.2 Quality of gravel 6.4.3 Organisation and equipment 6.4.4 Spreading gravel 6.4.5 Compaction 6.4.6 Continuous working 6.5 Filling and patching 6.6 Dust prevention 6.6.1 The dust problem 6.6.2 Remedial treatment 1 1 1 2 2 3 5 5 5 5 7 7 9 9 9 9 10 10 11 11 11 14 14 15 15 16 16 17 17 17 17 18 18 18 20 21 21 21 21 24 24 24 24 24 25 25 26 26 26 26 26

2.

3.

4.

5.

6.

Maintenance of paved roads 7.1 Types of maintenance 7.2 Local sealing 7.3 Crack sealing 7.4 Patching 7.4.1 Procedure 7.4.2 Materials 7.4.3 Patching gang 7.5 Mechanised surface dressing 7.5.1 General principles 7.5.2 Design of surface dressing 7.5.2.1 Type of existing road surface 7.5.2.2 Traffic categories 7.5.2.3 Chippings 7.5.2.4 Binder 7.5.3 Equipment 7.5.4 Application 7.5.4.1 Preparation of the surface 7.5.4.2 Application of binder 7.5.4.3 Spraying temperatures 7.5.4.4 Jointing strips 7.5.4.5 Application of chippings 7.5.4.6 Rolling 7.5.4.7 Aftercare and opening to traffic 7.5.5 Control 7.6 Manual surface dressing 7.6.1 Applicability 7.6.2 Chippings 7.6.3 Binder 7.6.4 Application of binder 7.6.5 Application of chippings 7.6.6 Rolling 7.6.7 Gang size 7.7 Slurry sealing 8. References

7.

Page 27 27 27 27 27 27 28 28 28 28 29 29 29 29 29 30 33 33 33 33 34 34 34 34 34 35 35 35 35 35 36 36 36 36 37 37 37 37 37 39 40 41 41 41 41 42 42 42 42

Appendix A Additional notes on surface dressing A.1 Chippings A.1.1 Size A. 1.2 Average least dimension A.1.3 Flakiness A. 1.4 Adhesion and pre-treatment A.1 .5 Stockpiling A.2 Binder A.2.1 Grade A.2.2 Viscosity A.2.3 Measuring the rate of spread of the binder A.2.3.1 Average rate A.2.3.2 Longitudinal variation A.3 References

1. INTRODUCTION
1.1 SCOPE OF THIS NOTE

This Note describes the principal techniques that will be used to maintain roads within a District in a developing country. Particular attention is paid to the planning and design aspects of the work, as it is in this area that the Engineer will make the largest contribution. Practical execution is not described in detail. The Note first discusses the merits of using labour or equipment-based operations and examines the scope for using local contractors for maintenance work. The importance of safety during maintenance work is then discussed and each of the various maintenance techniques is described in turn.

The Engineer will require adequate trained staff to carry out Ids instructions and he will need to arrange for supervisers, foremen and specialist artisans to attend suitable training courses. Accordingly the Maintenance Engineer must arrange for on-the-job training for labourers and other junior staff. The supply and maintenance of equipment and vehicles is usually organised as a separate Departmental function and will be beyond the control of the Maintenance Engineer. lack of suitable vehicles is always a major factor in maintenance organisations that are inefficient. The Maintenance Engineer must be aware of the importance of proper maintenance of equipment and vehicles and should actively concern himself with the organisation of this, as far as Departmental procedures allow. The work of the District or Maintenance Engineer falls into three categories:(i) Assessment of requirements: establishing what needs to be done in physical terms; Allocation of resources: estimating the men, materials and equipment required for different tasks, determining priorities and allocating resources to ensure the most costeffective results; Monitoring: checking that the work done produces the desired results.

1.2

THE ENGINEER'S RESPONSIBILITIES

One of the key persons in any maintenance organisation is the professional engineer who is responsible for running the organisation at District or operational level Although the District or Maintenance Engineer will usually be constrained to work within whatever management system is operated by his Department, there is invariably scope for improving the cost-effectiveness of the use of the resources available for road maintenance. By adopting the same professional approach towards road maintenance as to any other engineering activity, he can inspire Ids staff to improve their own performance. The Maintenance Engineer should therefore show a personal commitment to maintenance work by regularly inspecting the roads under his control and making Ins staff aware of his interest. By getting out of the office and into the field as much as possible, the Engineer will be able to get to know his road network thoroughly and will readily identify trouble spots and other areas of difficulty. He will be able to assess priorities and will gain first-hand knowledge of what maintenance has actually been carried out, instead of having to rely on reports from others. He will also see the quality of the maintenance work carried out and will be able to use his professional skill and expertise to solve problems on-the -spot as they arise. Seeing him actually on site will give a boost to staff morale and this will result in an improvement in both the quality and quantity of work done. If there is one single factor which influences the standard of road maintenance more than all others, it is the attitude of the Engineer responsible. The Engineer will find it difficult to follow this advice if he is overburdened with routine administrative work and it is essential that he delegates such activities to clerical and administrative staff as much as possible.

(ii)

(iii)

The management aspects of this work are described in more detail in a separate Note (TRRL Overseas Unit, 1981).

2.

METHODS OF UNDERTAKING MAINTENANCE

LABOUR AND EQUIPMENT

obtained by equipment. For example, on low-volume roads maintained by labour-based methods, there is unlikely to be any economic justification for insisting on the close tolerances of level and smoothness that can be achieved by the use of equipment. Equipment and labour cannot be substituted directly for each other as, in addition to the question of standards already mentioned, it will usually be necessary to make changes in work organisation. The following points should be borne in mind when considering the use of labour-intensive maintenance methods: (i) It is necessary to check carefully that labour will be available in the actual place where it is wanted, and at the time it is wanted.

2.1

Most maintenance operations offer considerable scope for the application of labour-based methods and some are only practicable by such methods. No recommendation is made here for either equipment-intensive or labourintensive methods, but Table 1 indicates the potential for equipment and labour-based methods in different maintenance operations. In choosing between equipment-based and labour-based methods, consideration should be given to the standard of work achieved by each method as well as to costs and to the way in which the work is organised. It is not always necessary for labour-based operations to have the same standards of finish that can be

(ii) Adequate organisation and management are critical considerations for large scale labour intensive works.

TABLE 1 POTENTIAL FOR EQUIPMENT AND LABOUR POTENTIAL FOR ACTIVITY Ditch cleaning and cutting Cleaning and minor repair to culverts and bridges Building scour controls Repair of structures Grading unpaved surfaces Dragging and brushing of unpaved surfaces Patching sanding or local sealing of bituminous surfaces Filling on unpaved surfaces and slopes Grass cutting Manufacturing signs Repairing and replacing traffic signs Road line markings Stockpiling gravel Regravelling gravel surfaces Stockpiling chippings Surface dressing EQUIPMENT Good (*) Poor Poor Poor Good (Skilled) () Good Poor Poor Good (**) Fair () Poor Good Good Good Good Good (Skilled) () LABOUR Good (*) Good Good Good Impracticable Poor Good Good Good Good () (Skilled) () Good Fair Fair Fair Poor Fair (Skilled) ()

NOTES:

(*) (**) () ()

The potential in these activities is dependent upon suitable design of the ditch cross-section. 'V'-shaped ditches are suitable for maintenance by grader whereas flat bottomed ditches are suitable for maintenance by hand or by mechanical shovel. The potential in this activity is dependent on the width of the shoulder and presence of obstructions such as road furniture and culvert headwalls. Some methods of manufacture may require the use of specialised plant (eg. vacuum application of reflective sheeting to sign plates). The expression 'skilled' implies that specific training of operatives is essential.

(iii) The health and nutrition of the labour force are very important and improvements can increase productivity. (iv) Standardised tools of good quality are vital in order to achieve high levels of output. (v) It is normally inefficient to have labour and machinery working alongside each other on the same task. (vi) Arrangements may need to be made to transport labour to and from the work sites. For large labour forces, this can be a substantial task.

In addition, the supply or maintenance of equipment and vehicles can be undertaken by specialist contractors. The Maintenance Engineer should satisfy himself that any contractor asked to tender is capable of completing the work satisfactorily, that he has the necessary equipment and staff, and that he has sufficient knowledge of estimating (or of current market rates) to be able to submit realistic prices.

2.2

DIRECT-LABOUR AND CONTRACT

Road maintenance organisations usually carry out much of their work by direct-labour. However, some maintenance activities can be done by small local contractors and, where Departmental policy permits, the Maintenance Engineer should consider whether the use of such contractors is advantageous. The use of contractors can reduce the burden on scarce government staff and also bring lower costs as a result of competitive pressures on efficiency which are unlikely to be obtained within a government department. However, contract work does require the preparation of detailed contract documents and a high degree of supervision, and the staff requirements and extra work involved for this should not be underestimated. There is also a danger that contractors bidding for maintenance work on a regular basis might introduce price fixing' to increase their profitability, and this will result in an increased cost to the maintenance department. The following activities should present no serious problems in the control of quality or quantity if carried out by contract on the basis of competitive tender:Supply of Materials Natural gravel Screened gravel Rock aggregate (for subsequent crushing by Departmental crusher) Crushed rock aggregate (for surface dressing and concrete) Washed sand (for concrete) Precast concrete blocks Precast concrete culvert rings Maintenance Operations Excavation of side drains and turnouts Construction of culverts Regravelling Surface dressing Grass-cutting and bush-clearing

Fig.1 Standard signs

3.
3.1

SAFETY MEASURES AND TRAFFIC CONTROL

PRINCIPLES

When work is being carried out on or close to the carriageway, it is the Maintenance Engineer's responsibility to see that adequate measures are taken to warn and protect road users and maintenance workers. He should instruct all supervisors and foremen in safety measures, including traffic control, and the use of temporary road signs. In selecting and positioning temporary road signs, the following principles should be applied:(i) only standard signs should be used,

or orange safety vests should be worn by the supervisor and all of the workforce. All vehicles and equipment should be painted yellow or orange and should carry red and white striped marker boards front and rear. All vehicles and equipment should work with headlights switched on and, where possible, should carry yellow flashing warning lights. If warning lights are not available, vehicles and equipment should carry a yellow or orange flag.

3.2

EDGEWORKING

Where road works are being carried out which do not affect the carriageway, or where works such as grading or dragging are being carried out, the sign layout shown in Fig. 2 is recommended. Warning signs should be placed before work starts and must be placed in the following order: 1. 'Men working' signs should be placed at the approaches to the work area. 'Road clear' signs should be placed at the ends of the work area.

(ii) the signs must be clean and in good condition, (iii) the standard signs should be displayed in a standard layout, 2. (iv) the layout used must give drivers time to understand and respond to the information which the signs convey. Recommended standard signs are shown in Fig. 1. It may not be possible for the Engineer to purchase or obtain the necessary signs from Departmental stores, but the manufacture of adequate signs should be within the capabilities of a central signs workshop. Each road maintenance gang should be provided with signs appropriate to the work it is carrying out and all foremen and supervisors should be trained in their use and layout. All temporary signs must be removed as soon as the work they relate to is complete. If they are not the value of the signing will be reduced. From the point of view of safety and traffic control, road maintenance work may be divided into four categories: (i) work which does not affect the carriageway, such as cleaning out side drains and cutting grass on verges (Section 3.2);

When the work has been completed, signs should be removed in reverse order. Signs must not be left on the road or at the roadside overnight. They should be removed and returned to the depot.

3.3

LANE CLOSURE

For repairs to the carriageway such as patching which requires closure of one lane, the sign layout shown in Fig. 3 is recommended. Before work starts, warning signs, barriers and cones must be placed around the work area. Work will be carried out on one side of the road at a time allowing traffic to pass on the other. Signs must be placed in the following order: 1. 'Men working' signs should be placed 200 metres in front of the work area. 'Road narrows' signs should be placed 100 metres in front of the work area. 'Keep left/right' arrows should be placed at the start of the work area. Barriers should be placed at each end of the work area. 'Keep left/right' arrows should be placed next to the barriers. Cones should be placed in a taper at the approaches to the work area and at a spacing of 10 metres along the middle of the road next to the work area. 'Road clear' signs should be placed 200 metres beyond the work area.

2.

(ii) work requiring partial closure of the carriageway, such as repair work to surface or base which is restricted to one lane while traffic continues to use the other lane (Section 3.3); (iii) work on the centre line, such as white line painting (Section 3.4); (iv) work requiring total closure of the road, with construction of a temporary diversion, such as the reconstruction of a damaged culvert (Section 3.5). Whenever possible during maintenance work, yellow

3.

4.

5.

6.

7.

Fig.2 Sign layout for edge working (drawn for driving on the left)

Fig.3 Sign layout for lane closure (drawn for driving on left)

Traffic controllers should stand opposite the barriers on the other side of the road holding reversible 'stop/go' signs. One controller should be appointed by the supervisor as the leader. He should decide when to change the direction of the signs and the other controller should follow his lead. In this way, the controllers work closely together to ensure that only traffic from one direction is allowed to pass at a time. When the work area is short, only one traffic controller may be needed. The Maintenance Engineer should train these men in the use of the signs and only the trained men should be allowed to operate them. Police assistance in training may be helpful. On low traffic roads, the Maintenance Engineer may approve the use of a simpler system of traffic control. Patching work should not be left unfinished over-night but, if this sign layout is being used for repairs to a culvert or a washout, then at night the approaches must be adequately lit. Kerosene lanterns may be adequate on lightly-trafficked roads or where speeds are low, but high-intensity lamps should, if possible, be used on heavily trafficked roads. A watchman should always be in attendance to ensure that the lamps are working and are not interfered with or stolen. Lamps should show a yellow light. Flashing lamps are best as they consume less power and are more easily seen. When the work has been completed, signs should be removed in reverse order.

6.

'Road clear' signs should be placed 200 metres beyond the ends of the work area at the side of the road.

When the work has been completed, signs should be removed in reverse order. Signs must not be left on the road or at the roadside overnight. They should be removed and returned to the depot.

3.5

DIVERSIONS

A diversion will enable maintenance work to be carried out more efficiently and more safely. In particular, diversions are needed for regravelling work and major culvert repairs. If traffic is to be diverted for more than a day or two, or the work is being carried out during the wet season, the diversion should be constructed with an adequate base and surfaced with gravel. A small gang should be allocated to keep the diversion in good condition. Diversions should be wide enough to allow two lorries to pass. After the diversion has been completed and before work starts, warning signs, barriers and cones must be placed around the work area. The layout shown in Fig. 5 is recommended. Signs must be placed in the following order: 1. 'Men working' signs should be placed 200 metres in front of the work area. 'Turn left/right ahead' arrows should be placed 100 metres in front of the work area. Cones should be placed diagonally across the road to lead into the diversion. 'Keep left/right' arrows should be placed at both ends of the lines of cones. Barriers should be placed behind the lines of cones. 'Road clear' signs should be placed 200 metres beyond the ends of the diversion.

2.

3.4

CENTRE LINE WORKING

3.

When painting centre line markings on the road, considerable care must be taken and it is most important to pay a great deal of attention to safety measures. The sign layout shown in Fig. 4 is recommended. Before work starts, warning signs, barriers and cones must be placed around the work area. They must be placed in the following order: 1. 'Men working' signs should be placed 200 metres in front of the work area at the side of the road. 'Road narrows' signs should be placed 100 metres in front of the work area at the side of the road. 'Keep left' arrows* should be placed in the centre of the road at the start of the work area. Barriers should be placed behind the 'keep left' signs. Cones should be placed at a spacing of 10 metres on either side of the work area. Assumes driving on the left.

4.

5. 6.

2.

3.

4.

5.

Diversions will usually be in operation at night as well as by day and the approaches must be adequately lit. Kerosene lanterns may be adequate on lightly-trafficked roads or where speeds are low but high-intensity lamps should if possible be used on heavily trafficked roads. A watchman should always be in attendance to ensure that the lamps are working and are not interfered with or stolen. Lamps should show a yellow light. Flashing lamps are best as they consume less power and are more easily seen. When the work has been completed, signs should be removed in reverse order.

Fig.4 Sign layout for centre line working (drawn for driving on the left)

Fig.5 Sign layout for diversion (drawn for driving on the left)

3.6

DEVELOPMENT OF SAFETY MEASURES

4. MAINTENANCE OF ROAD FURNITURE

4.1 TRAFFIC SIGNS

In many developing countries, no safety measures are provided at all during maintenance works. The Maintenance Engineer must understand that the safety of his workers and of the road user during maintenance operations is his responsibility. He must train his staff in methods of signing and traffic control and make them understand the need for safe working. Initially it will probably not be possible to provide all the safety equipment recommended here. However, simple signs can be easy and cheap to make and it is possible to improvise cones and barriers. Turfs of grass cut from the roadside can be used instead of cones and effective road markers can be made from old oil drums painted with black and white bands and filled with sand, stones or water. If signs cannot be provided, a maintenance vehicle should be parked on the road between the on-coming traffic and the maintenance workers with its lights switched on. When traffic volumes are very low, the Engineers may allow his workers to relax these standards. In particular it may not be necessary to use 'stop/go' signs. However, by relaxing standards, road users and maintenance workers must not be placed in a hazardous position.

Traffic signs are the principal means of conveying information about the road to drivers and, as the road network becomes more extensive, the number of traffic signs increases. As traffic flows increase, an increasing effort on their maintenance is needed. The sizes, colours and layouts of signs have, in most developing countries, been standardised in accordance with international protocol and incorporated in Departmental standards. It is important that all signs should comply with these standards. Signs which are clean and in good repair can be easily seen and understood and inspire confidence, that their message is accurate and reliable. Damaged or missing signs should, for the same reasons, be replaced promptly. Signs which are no longer needed (e.g. because of road improvements) should be removed, as should temporary signs upon completion of the maintenance works to which they are related. The provision and care of signs is very cheap and cost effective. Signs should be inspected and cleaned at least twice a year. If there are enough signs, it may be worth setting up a small team which tours the District, covering all roads in rotation, which carries out this work as well as repainting posts and replacing damaged signs. The cleaning and inspection team will need to be provided with equipment such as that listed in Section 4.4. Major repairs, particularly to sign faces, are better carried out in a central workshop which is equipped to do this work under good conditions and supervision. (This is particularly important in relation to signs which have legal force.) It is worthwhile keeping records of traffic signs; they should be included in an inventory and transferred to a signs register (which may be a card-index) in which inspections, repairs and replacements are recorded. It is useful if their location is recorded on a strip-map.

4.2

RAILS AND POSTS

Guard rails and parapet rails are provided to protect road users guard rails to prevent vehicles from running over high embankments and parapet rails on bridges to safeguard pedestrians. They should be repaired promptly if damaged, and kept clean and repainted regularly so as to maintain their visibility and prevent corrosion. Kilometre posts provide both drivers and the maintenance organisation with the basic reference for the position of any point on the road. Kilometre posts

should be kept clean and repainted regularly, and vegetation should be cleared so that they can easily be read from a moving vehicle. Similar requirements apply to bridge and culvert marker posts. It is recommended that each bridge and culvert is provided with a marker post (or posts), serially numbered within the kilometre in which it is located, thus: 79/3 is the third culvert or bridge in km 79. These numbers should be used as references in the culvert and bridge registers. Delineator posts are usually provided only on bends. They should be kept clean and repainted regularly, preferably with reflective paint if reflectors are not fitted. Vegetation should be cleared so that they can be easily seen.

4.4

MAINTENANCE GANGS

The composition of a typical road furniture gang is given below. Such a gang would be reponsible for field repairs to traffic signs, guard rails, parapet rails, marker and delineator posts. They should also be responsible for cleaning signs. The Maintenance Engineer may find that local circumstances mean that a different composition with different tools and equipment are needed. Personnel 1 Overseer/Gang leader 1 Driver 2 Labourers Vehicles and equipment 1 Truck or van, equipped for carrying signs without damage Assorted signs Assorted tools, including spanners, etc 1 Pick-axe 1 Spade or shovel 1 Saw 1 Hand rammer Dry mixed concrete 1 Small ladder Water Detergent Paints Brushes Solvent for brush cleaning Safety equipment as recommended in Section 3. A typical gang for road marking might consist of the following. These would be responsible for painting white lines and replacing reflective studs. Personnel 1 Overseer/Gang leader 1 Driver 2 Painters or labourers Vehicles and equipment 1 Small truck Stencils 30 metre tape String Paints Brushes Solvents for brush cleaning Safety equipment as recommended in Section 3.2. If thermoplastic paint is being used, it must be applied using a specially equipped truck and skilled operators in place of the painters or labourers. The truck should be painted yellow or orange and be equipped with flashing yellow warning lights.

4.3

WHITE LINES AND REFLECTIVE STUDS

White lines are being used increasingly on principal twolane roads in developing countries for centre line, lane and edge markings, and for 'stop' or 'give way' lines. They are particularly useful as an aid to night driving. A note on their maintenance is therefore appropriate. White lines may be of paint or hot-applied plastic. Both materials may be reflectorised with small glass beads (ballotini). Emulsion paint gives a good performance on surface dressed roads with a good surface texture but, on premix roads or roads with less texture depth paint has a short life (perhaps only a few months under heavy traffic) and will require renewing regularly. Hot-applied plastic has a much longer life and should require much less maintenance effort. Its application requires the use of specialised equipment and properly trained operators. Paint, on the other hand, can be applied manually, although this will be a very slow procedure and the use of a portable paint sprayer is preferable. Whichever method is used, road markings should always be applied using templates or a straight edge. Such equipment will usually be provided through the central equipment organisation. Reflective road studs may be used in conjunction with white lines. They should be replaced promptly if they are lost or damaged and protected by masking when surface dressing is being carried out. They should be installed by men who have been suitably trained. Particular attention should be paid to safety measures when men are working in the middle of busy roads (see Section 3.4).

10

MAINTENANCE OF DRAINAGE, SHOULDERS AND SLOPES

THE DRAINAGE SYSTEM

5.1

One of the most important aspects of the design of a road is the provision made for protecting the road from surface water or ground water. If water is allowed to enter the structure of the road, the pavement will be weakened and it will be much more susceptible to damage by traffic. Water can enter the road as a result of rain penetrating the surface or as a result of the infiltration of ground water. The road surface must be constructed with a camber so that it sheds rain-water quickly and the formation of the road must be raised above the level of the local water table to prevent it being soaked by ground water. Water can also have a harmful effect on shoulders, slopes, ditches and other features. High water velocities can cause erosion which, when severe, can lead to the road being cut. Alternatively, low velocities in drainage facilities can lead to silt being deposited which, in turn, can lead to a blockage. Blockages often result in further erosion. A good road drainage system, which is properly maintained, is vital to the successful operation of a road. It has four main functions: (i) to convey rainwater from the surface of the carriageway to outfalls (streams and turnouts); to control the level of the water table in the subgrade beneath the carriageway; to intercept surface water flowing towards the road; to convey water across the line of the road in a controlled fashion.

Even if the drainage system of a new road has been carefully designed, it is likely that for several years after construction it will be necessary to observe its performance closely and to make additions and amendments to it. Reference is therefore made here to design matters where these can be seen as forming part of the maintenance function. Both in the design and in maintenance of drainage, it is important to interfere as little as possible with the natural flow of water. Culverts on natural water-courses should follow the existing alignment as closely as practicable and re-alignment (often resulting in sharp changes in direction) should be avoided. The surface flows in drains and culverts should also be kept to a minimum by the use of frequent turn-outs where side drains cannot be discharged to existing water courses. In side-long ground, where discharge from the side drain on the high side passes to the low side, it is best to use frequent small culverts rather than occasional large ones. In such cases, the spacing will be governed by the maximum flow acceptable in the side drains and the capacity of the culverts will not usually be a constraint as the minimum requirements for access for maintenance (often taken as 600mm diameter or 600mm x 600mm box) will ensure adequate capacity.

5.2

SIDE DRAINS

(ii)

(iii)

(iv)

The level of the water table beneath the carriageway is a major influence on the strength of the subgrade. The bottom of side drains should normally be maintained at a level at least one metre below formation level (the underside of the sub-base). If side drains have been constructed too shallow and they are not performing properly, they should be deepened as part of the maintenance operation. The performance of side drains should be monitored over time by the Maintenance Engineer to determine designs and dimensions appropriate to local topographical, climatic and soil conditions. Side drains are usually built to the same gradient as the road. This may result in high velocities and erosion on steep gradients and silting on flat or reductions in gradient. Volumes in the side drains can usually be reduced by constructing frequent turn-outs. In highly erodible soils, additional measures may be needed to prevent or control erosion. Grass should be encouraged to grow in drainage ditches as this helps bind together the topsoil and inhibits erosion. Where the erosion is only just starting, the most effective control is likely to be to dam the side drain at frequent intervals and to construct additional turn-outs. More severe erosion may need check-dams as shown in Fig. 6.

The first three functions are performed by side drains and the fourth by culverts, drifts and bridges. Common drainage problems include: (i) (ii) blocking of drains by debris or vegetation; silting: the deposition of silt in the bottom of drains and culverts, often reducing the gradient; erosion of the bottom of side drains in erodible soils or on steep gradients, particularly where insufficient turn-outs result in large flows in drains; erosion at culvert outfalls, resulting from high discharge velocities; erosion of shoulders and side slopes.

(iii)

(iv)

(v)

11

Fig.6 Check-dams for side drains

Fig.7 Layout of turn-outs

Dumping rock into an erosion gully is not usually effective unless a filter of finer material is also used. Rocks in steel wire mesh baskets or gabions may be more effective. These are shown in Fig. 6. Gabions should be extended well into the sides of ditches to prevent scour around their edges. Protection should also be provided at the foot of check-dams either by rock riprap or gabion mattresses as found appropriate. In very steep sections on highly erodible soil, it may be necessary to line the ditch with concrete, masonry or riprap. In time, erosion prevention measures may themselves require repair.

In agricultural areas, the construction and maintenance of turn-outs can cause difficulty with farmers. The matter should be discussed at the outset with the farmer as it is often possible to integrate road and farm drainage. In the last resort, it may be necessary to ask for the assistance of the Civil Authorities. Compensation may need to be paid. Turn-outs should receive the same attention as side drains during maintenance. Both can be maintained by hand or with a grader. The cross-section of the ditch or turn-out determines the most appropriate method of maintenance. The use of cut-off drains should be avoided as they sometimes cause small land-slips. Because they are often difficult for maintenance crews to locate, they are often neglected and become ineffective.

Removal of silt should form part of the routine maintenance programme. Silt should be thrown at least two metres clear of the side drains away from the road. It must never be used to repair road surfaces. If silting of side drains persists, it may be advisable to realign the ditches to increase water velocities, if this is possible.

5.3

FORDS AND DRIFTS

The discharge from side drains should, wherever possible, be taken to existing natural watercourses. Where this is not possible, the side drains should be interrupted at intervals by dams, and the flow discharged to turn-outs (see Fig. 7). Here the water will be dispersed by seepage, evaporation, or by surface flow. Turn-outs should be built so that they have a longitudinal gradient of about 1 in 50 to 1 in 100 (ie. nearly parallel to the contours) and of such length that they run out to zero depth. The spoil from turn-outs should be thrown to the low side. The junction of the side drain and the turn-out should have a radius of at least five metres. This avoids the tendency for scour to occur at this point. Spacing of turn-outs depends on several factors including the erodibility of the soil, the width and gradient of road and cross-slope of the ground. On highly erodible soils, the spacing may initially be based on those indicated in Table 2. Spacings may be increased on soils with greater resistance to erosion. Observation of the performance of the side drains and turn-outs in practice will indicate if any changes are necessary.

Fords and drifts are often used on lightly trafficked roads at watercourses with seasonal flows. They may be covered by deep water for short periods when traffic may not be able to pass, or by shallow water for longer periods. For much of the year they may be dry. Maintenance of fords and drifts should be carried out after flooding and much of it consists of maintenance to carriageway, earthworks and culverts as for a normal length of road. In addition, warning signs, marker posts and depth indicators should be checked, and repaired or replaced as necessary. The crossing should be checked to ensure that the carriageway has not scoured or washed out, and any damage should be repaired. Erosion downstream of the ford is a common problem and, if necessary, a masonry or concrete apron should be constructed. An alternative may be to use stone gabions or mattresses. Any silt or debris must be cleared from the carriageway. Some drifts have a culvert passing under them. These can cause the deposition of silt up-stream of the drift which tends to block the entrance to the culvert, making it inoperative. If culverts are used in drifts, they should be large enough and steep enough to be self-cleaning.

TABLE 2 TURN-OUT SPACING Gradient 1 in 100 or less 1 in 100 to 1 in 50 1 in 50 to 1 in 20 1 in 20 to 1 in 10 more than 1 in 10 Spacing (metres) 50 40 25 15 10

5.4

CULVERTS

Culverts are provided to convey water from the upstream side of the road to the downstream side. They may be built on the line of existing watercourses or to carry the build-up of water which results from the presence of the road. In either case, silting, choking by debris or structural collapse will usually result in over-topping and damage to the road. Maintenance comprises keeping the waterway clear, controlling scour and repairing structural damage.

14

Erosion of outlet channels from culverts is a common problem and if not dealt with promptly is likely to result in damage to the culvert and the road. This is caused by high discharge velocities from the outlet, and solutions which do not take account of this are not likely to be permanently effective. If the discharge velocity cannot be reduced by increasing the area of cross-section of the culvert and reducing its gradient, perhaps with a dropinlet, then some type of energy-dissipating outlet should be used. As a first step, a fan discharge constructed with masonry or concrete should be tried. Stone gabions or mattresses downstream of the outlet may be a cheap alternative. Drop outlets are good, but on many sites there is insufficient height to construct these. The checking and removal of debris from culverts can be difficult, particularly if the culverts are small. Longhandled shovels or shovels attached to rods are useful for clearing out culverts which are too small for a man to get inside. Trees or branches blocking culvert entrances should be sawn into convenient sized pieces to help their removal and carting away. Culverts which get regularly blocked by debris should have a grill constructed at their upstream entrance. Culverts made with corrugated metal pipes can be abraded by water carrying silt and sand which can wear away the protective zinc coating. This will result in the culvert pipe rusting away. If the culvert is large enough for a man to enter, the rust can be removed with a wire brush and the pipe coated with a thick layer of hot bitumen or tar. Alternatively, a flat concrete invert can be constructed. Culverts made from concrete rings can be subject to differential settlement. This is a construction fault and major settlement problems can only be corrected by reconstruction. Minor mis-alignments should be repaired by grouting the joints in the pipes with concrete to provide a waterproof seal.

as soon as possible after floods, and eroded and scoured areas should be repaired. These should be filled with compacted gravel and then protected with rip-rap, concrete or gabions. Deposits of silt and sand should be removed where necessary to restore the original channel. Simple repairs can be carried out to the decks of timber bridges. Loose plans can be re-fixed using screws or nails whose length is two or three times the thickness of the plank. Defective planks should be replaced with new planks of the correct thickness, length and width, which should be treated with wood preservative. All nail heads should be driven flush with the surface. Whenever planks are replaced, the condition of the timber underneath should be checked. Small painting jobs can be undertaken such as the repainting of railings or, occasionally, steel beams. All dust, dirt, rust and old paint scale should be removed, where possible with an oxy-acetylene burner, and then with a wire brush. The steelwork should be given coats of primer, undercoat and topcoat. Only good quality paint should be used and it should be brushed thoroughly into the steel. Ample time should be allowed between coats for the paint to dry. Brushes should be washed thoroughly after use with thinners, petrol or kerosene. The colour of the topcoat should be as bright as possible for better visibility and safety.

5.6

SHOULDERS

Shoulders provide lateral support for the carriageway. They also provide accommodation for stopped vehicles and, on narrow roads, provide room for passing vehicles. Besides being used by vehicles, they are also used by pedestrians, cyclists and for animal driving. Shoulders may be paved, gravel, earth or may be grassed. Paved shoulders are subject to the same type of defects as paved carriageways and are repaired using the same methods as described in Section 7 (Maintenance of paved roads). It will usually be convenient to carry out repairs to this type of shoulder at the same time as the repairs to the carriageway with the same maintenance gang. Periodically, it is necessary to reseal shoulders and this should be carried out in the same way as for manual surface dressing (Section 7.6). Similarly, maintenance of earth and gravel shoulders is essentially the same as for unpaved roads. Most defects are corrected by grading, although ruts and pot-holes can be repaired manually (see Section 6.5). One of the objects of grading is to retrieve gravel lost to the shoulder and place it back on to the carriageway. On no account, must sediment from the ditch be graded on to the shoulder. The grader should cut to the bottom of all defects in the shoulder and spread the material so that, when compacted, it is level with the carriageway edge and slopes away to the ditch at a

5.5

BRIDGES

Bridge inspection and maintenance requires specialist knowledge and skills, some of which will only be held by a qualified bridge engineer. The maintenance of large bridges is beyond the scope of this Note. However, some repairs to small bridges can be accomplished by a normal road maintenance unit and notes on these are included here. The maintenance of the waterways under bridges involves the same principles as that for culverts. The main operations involve keeping the waterway clear, controlling scour and repairing structural damage. Maintenance methods are similar to those described under the maintenance of side drains, fords, drifts and culverts (Section 5.2 to 5.4). Debris should be removed

15

steeper crossfall than the carriageway. For unpaved roads, this grading should be carried out at the same time as grading of the carriageway and in the same way (see Section 6.2). If shoulders are worn away by the action of traffic and the level falls below that of the carriageway at the edge, water can collect at the edge of the road and infiltrate the base. This considerably weakens the structure of the road and usually results in severe deformation of the carriageway. In the case of paved roads, the edge starts to disintegrate and break away. This type of edge damage is difficult to repair effectively, but must be patched using methods similar to those described in Section 7.4. Edge repairs must be accompanied by the placing and compacting of new material on the shoulder or the repairs will be ineffective. Periodic regravelling of the shoulders should be carried out in a similar way to that described in Section 6.4. Grassed shoulders can cause problems if the grass traps material washed off the road, leading to the shoulder level building up and trapping water at the edge of the carriageway. If this occurs, the shoulders should be graded-off as described above. Where there are bushes and high grass growing on the shoulder which interfere with lines of sight, these should be cut back as short as possible. This can be done using a tractor-mower or by hand using scythes and cutlasses. Where there is much road side furniture, hand cutting will be easier than using a machine. When there is extensive damage to shoulders as a result of erosion or washouts, repairs are needed urgently and should be carried out as described in Section 5.7.

reinstated. When shoulders have been cleared, their levels should be checked to ensure that water can flow from the carriageway uninterrupted into the drain. When the failed slope is not to be reinstated, the sides of the slip area should be rounded off. Where there has been severe erosion or slips which have removed part of the carriageway or shoulder, or if erosion of slopes threatens the road structure, repairs are needed urgently. Loose material must be removed and the damaged area should be cut back to sound material. New material must be placed and compacted in 50-7 mm layers using small vibrating rollers or hand tampers. It may be necessary to add water to the fill material to assist compaction. When reinstating slopes in these cases, gab ions can be used for all or part of the repair work. In all cases, protection measures should be taken to stop the slip occurring again. Thin will probably involve improvements to the drainage system and may require paving of the slope itself using rip-rap, masonry or concrete. The use of gabions for filling material removes the need for slope protection. Establishing grass or other vegetation on the slope may be a cheaper form of protection. It may also be necessary to surface the shoulder with gravel or with a bituminous surface dressing.

5.8

MAINTENANCE GANG

5.7

SLOPES

For general maintenance work to drainage features, shoulders and slopes, the detailed composition of the maintenance gang will depend on whether the work is to be carried out by labour4ntensive or equipment. intensive methods. A suggested basic gang size is given below, but this will need to change for different operations and situations, and the Maintenance Engineer should develop his own gang compositions to meet his own local circumstances. Personnel 1 Overseer/Gang leader 1 Driver Several labourers Vehicles and equipment 1 Small truck or tractor-and-trailer 1 Hand-held vibrating roller (0.25 Mg) plus a plank to help load onto truck or trailer, or one hand rammer for each labourer used on compaction work 1 Pick-axe for every two labourers 1 Broom for every two labourers 1 Shovel or hoe for every two labourers plus extension rods for culvert clearing 1 Rake for every two labourers 1 200 litre drum for water 1 Bucket or watering can 1 Axe Safety equipment as recommended in Section 3.

For all slips and settlements of cut and fill slopes which affect the road, the maintenance gang should place warning signs, cones and barriers around the site according to the layouts described in Section 3. Severe damage, where the road is cut by a landslide or a washout will usually have to be repaired by special gangs often needing heavy construction equipment. Less severe damage can often be repaired by the maintenance gang. Although slips and settlements can occur as a result of slopes being too steep, they are much more usually associated with the presence of water in the soil. In the case of landslide material blocking the road, all the material should be removed and carted away on a tipper truck or on wheelbarrows and dumped where it cannot affect the road. Roadside ditches should be cleared of all debris and ditch shapes should be

16

Materials Patching materials as indicated in Tables 3 and 4 for unpaved shoulders and slopes and as indicated in Tables 5-7 for paved shoulders. Cement, sand and ballast may be needed for making concrete for repairs to culverts and paved ditches. Water, if not obtainable on site, must be carried. Grass cutting work can be done by this same gang equipped with scythes and cutlasses or by a speciallyequipped tractor mower. Shoulder grading should be carried out with a grader of at least 135 horse power working in conjunction with compaction equipment wherever possible.

6.
6.1

MAINTENANCE OF UNPAVED ROADS

TYPES OF MAINTENANCE

The principal operation in maintaining earth and gravel roads is grading. Dragging or brushing may also be carried out with the objective of controlling the development of corrugations, and light or routine grading is also carried out for this reason. Heavy grading is used to reshape the road surface and to restore it to its correct camber or crossfall and to provide a smooth running surface. Heavy grading can be combined with regravelling to restore the thickness of the gravel surface. Filling or patching are labour-intensive operations to deal with the worst defects on low-volume roads for which the expense of grading or other machine activities cannot be justified.

6.2

GRADING

6.2.1 Types of grading

Light grading is a light trimming of the surface of the road which should be carried out on a routine basis particularly in the dry season to control roughness and corrugations. When undertaking light grading in the dry season, loose material should be bladed towards the edge of the road. If several graders are available, it is more efficient to use them together on the same job. In this case, they should work one behind the other covering the whole width of the road. In the wet season, material should be graded towards the centre of the road. Light grading maybe carried out by motor graders, but a more cost-effective technique is to use tractor-towed graders which are capable of similar outputs and standard of work on properly constructed roads. Heavy grading consists of scarifying and cutting to the bottom of deformations and then reshaping the surface. It usually requires the use of a 135 horse power motor grader but, in some cases, the tractor-towed grader could also be utilised. Heavy grading operations should always be carried out at the beginning of the wet season to ensure that the road has the correct profile for effectively shedding water during the rains. When possible, it should also be carried out at the end of the wet season when the moisture content of the surfacing material is still high enough to help recompaction and prevent loss of fines. This is particularly important when heavy grading is needed to remove ruts and potholes. Scarifying to the depth required to remove these will result in the production of a considerable depth of loose materials and, in the dry season, this cannot be recompacted unless large amounts of water are added. The surface will then be

17

quickly deformed and fines will be scattered by traffic. Heavy grading of gravel is inadvisable without the provision of additional surfacing material if the remaining thickness of gravel is less than 75mm. The frequency at which grading should be carried out will depend upon the traffic, the climate and the nature of the surface material. Gravels of average quality will probably need grading after 12,000-15,000 vehicle passes and good quality gravels may sometimes be left for 25,000 vehicle passes. For roads that are liable to corrugate, grading may be needed after the passage of only 1,500-2,000 vehicles. Some gravels, particularly those which are self-cementing or which contain large size material, are not suitable for grading as this results in the surface being torn up. Patching or regravelling is needed to repair these. The frequency of grading will also depend upon the daily traffic level as, at high traffic levels, a higher level of service may be expected, requiring more frequent grading. More guidance on the choice of grading frequency is given in Overseas Road Note 1.

should be made at approximately 100 metre intervals along the road and if the camber is too steep or too flat, then the road must be graded again.

6.2.3

Steep hills

Steep hills on unpaved roads, where the longitudinal gradient is steeper than the crossfall, are prone to severe erosion in the wheel paths, particularly when these coincide in the centre of the road. Considerable attention must be paid to maintaining adequate cross-fall in these situations as this will minimise the erosion. If severe damage persists, consideration should be given to paving the gradient either by surface dressing or preferably with a concrete pavement.

6.2.3

Grader operation

6.2.2

Crossfall

Earth and gravel roads require steeper crossfalls than bituminous surfaces if rainwater is to be shed satisfactorily. If the grader operator has been trained on construction work for paved roads, it will probably be necessary to ensure that he understands the different requirements for unpaved surfaces. Crossfall on gravel and earth roads should be between 1 in 25 and 1 in 15 (4-6 per cent). It is very important to ensure correct camber on steep alignments. 'Flat' cambers are frequently the cause of the longitudinal gullying commonly found on such alignments. Crossfall should be checked on site using a simple camber board, such as that illustrated in Fig. 8 which can be carried on the grader. Use of the camber board is illustrated in Fig. 9. It should be placed on its edge across the road with its narrower end pointing towards the centre line. If the level bubble is central, then the camber is correct. Checks

The quality of workmanship in maintaining earth and gravel roads depends to a great extent on the skill and judgement of the individual grader operator. Careless operation can cause extensive damage to a road, for example by flattening the crown so that rainwater is not discharged, by cutting too deeply in dry weather or by blading plastic material from side drains on to the carriageway. The Maintenance Engineer must get to know the ability of each grader operator and should find out the capability of each type of grader in use in his District. The manufacturers of graders publish manuals which explain the correct use of their machines in various applications and these should be obtained and studied by the Engineer. A particularly useful document is Grading Illustrated (Aveling-Barford, undated). With this information he will be better able to explain his requirements to grader operators and to ensure that they are carried out. Arrangements must be made for graders to be checked mechanically, greased and fuelled before each day's work. The Maintenance Engineer must make operators understand the need for preventative maintenance of their machines and must try and ensure that equipment is not mishandled, misused or abused by operators in

Fig.8 Five per cent camber board

Fig.9 Use of camber board

Fig.10 Transition of camber on approach to bends

the field. This can only be achieved by proper training and regular site visits by the Engineer. Arrangements may also need to be made to refuel graders on the job if necessary. This may be done either with a refuelling truck visiting the working site, or with an arrangement to refuel at a maintenance camp. For heavy grading, the grader works on one side of the road at a time, if at all possible, and works in passes of about 200 metres in length. Graders must not stop on junctions or on bends where they will be a danger to traffic. On straight stretches of road the operator should aim to develop a crown on the road. The surface should be cambered to fall away from the crown with a slope of 4-6 per cent. The shape of the road must be maintained across culverts but, on sharp bends, the surface must be superelevated and must be flat from shoulder to shoulder with the outer shoulder higher. Any crown on a bend can be very dangerous to traffic. On the transition from bends in the road to straight sections, the camber on the outside lane should be gradually reduced until the normal crosssection shape is obtained again (see Fig. 10). For heavy grading, it is important to cut to the bottom of surface defects and, if the road surface is hard, the grader's tines should be used to loosen the material. The grader should start from the edge of the road and work towards the centre. Gravelled shoulders should be treated as part of the running surface. The first and second passes cut to the bottom of the surface irregularity and deposit a windrow just beyond the centre line. If water is to be added, then the water tankers should spray the road at this point. The windrow is then spread back across the road depositing all the material on the carriageway to give the correct camber. The material may need to be sprayed again with water during this operation. After the camber has been checked, the other side of the carriageway is graded in a similar way to complete the work and leave a smooth even surface. This is illustrated in Fig. 11. It is essential that the grader does not make a final pass down the centre of the road with the blade horizontal. This flattens the centre of the road and causes water to pond. This leads to rapid deterioration of the surface. Windrows must not be left in the middle of the road overnight as this is a danger to traffic. If compaction equipment is available for use, it must follow up closely behind the grader but must only work on sections where grading has been completed to avoid interference with the grading operation. Rolling should start at the edge of the road and work towards the middle. Providing that the work is carried out in the wet season, watering of the road will not be necessary before rolling. Otherwise water should

Fig.11 Grading
be added, as necessary, during the rolling operation to give the correct moisture content for compaction. The rollers should aim to progress from section to section at the same rate as the graders. In the dry season, grading is essentially a dragging operation to remove loose dry material from the surface of the road and to fill in potholes and ruts (see Section 6.3).

6.2.4

Grading gang

For heavy grading, the grader needs to be powerful enough to cut to the bottom of the deformation in the road and a minimum size of 135 horse power is usually recommended. However, if the wearing course does not contain an excessive amount of oversize material, a tractor-towed grader could be used instead. The operator will normally be assisted by a machine attendant who helps direct traffic and grader turning, and removes large stones and other unwanted material from the path of the grader. He should work well ahead of the grader to stop it being delayed. The grader should carry a camber board and traffic signs as indicated in Section 3.2. Rollers should be used if available and water tankers should be used in conjunction with rollers if necessary.

20

For light grading, up to three graders can be used as this is more efficient for supervision, maintenance and refueling. The graders can be less powerful than those needed for heavy grading. Traffic signs should be used as described in Section 3.2. All graders should be fitted with yellow flashing warning lights, and these plus headlights should be switched on when the graders are working.

6.3.3

Method of operation

The basic method of operation for both dragging and brushing is the same. The frequency with which dragging should be carried out depends on the traffic loading, the rate of development of corrugations and the soil type. A road carrying 100 vehicles per day may need to be dragged every two weeks using the metal drags illustrated in Fig. 12. One dragging pass will probably be necessary every 3 to 4 weeks for roads carrying 50 vehicles per day and every four to six weeks for traffic levels of 25 vehicles per day. Simple experiments should be carried out by the Maintenance Engineer to determine the optimum frequencies for different conditions. The drag should be designed where possible so that its width adequately covers half the road. This enables the maintenance to be carried out with a single pass in each direction. For the best results, four main adjustments can be made according to the conditions; namely depth of cut, angle of cutting blades relative to direction of traffic, towing angle of drag and weight of drag. The drags illustrated in Fig. 13 have height adjustments at each end of the cutting blades and the position of the blades on the drag can also be changed to obtain the required volume of windrowed material. Varying the towing angle between tractor and drag can also achieve different volumes of windrows, but this is more difficult to control. The depth of cut can also be varied by weighting the drag. The level of surface roughness and size of the gravel wearing course will largely dictate the optimum weight of drag. However, the designs in Fig. 13 can still be utilised by changing the thickness of the channel sections. The tractor or grader towing the drag should always work in the same direction as traffic and should not stop on junctions or on bends. Drags should be towed at speeds of 5-8 km/h depending on the type of drag and on the condition of the road surface. Care must be taken not to drive too fast or tile drag will skip over the surface irregularities and will also generate a lot of dust. With brushes made of thorn scrub or brushwood, the operator must ensure that pieces which may break off the drag are not left lying on the road surface where they will be a danger to following traffic. Pass lengths should be as long as possible, preferably of the order of several kilometres.

6.3

DRAGGING AND BRUSHING

Regular and frequent dragging can be used, in the dry season, to delay the formation of corrugations on earth and gravel roads by removing loose material from the surface. Dragging will not remove severe corrugations once they have formed, nor will it restore camber or lost material. These defects must be corrected by heavy grading.

6.3.1

Design of drag

Typical drag units are shown in Fig. 12. The first consists of a metal 'A' frame constructed from 100 x 65 x 3mm channel on to which are bolted used grader blades. The leading one of these is angled to the direction of travel. The second unit is constructed from 100 x 75 x 3mm channel, but has additional blades for re-distributing the windrow. The approximate weights of these units are 250 and 375 kg respectively and they are relatively cheap to manufacture. Many other forms of drag have been used in various countries, such as railway rail, roIled steel joists, tolards, timber baulks, etc. However, the two types illustrated in Fig. 12 are recommended because of their proven performance in quantified field trials and their ease of manufacture. They should be towed at speeds from 5-8 kilometres per hour and are capable of maintaining roads carrying up to about 100 vehicles per day. They are also effective for dealing with corrugations.

6.3.2

Brushing

Brushing is generally effective only on very lightly trafficked roads with surfaces containing loose material. Typical tractor drawn brushes constructed from locally available material are shown in Fig. 13. The actual brushes used on the first type illustrated are made from brushwood tied tightly together. These must not be made out of old steel cable which is dangerous if broken off and left lying on the road. The tyre sledge illustrated consists of old tractor or heavy lorry tyres cut in half around their circumference and bolted or chained together. It is important that all the tyres are in contact with each other, as in the illustration, to ensure proper distribution of the loose material. Brushes made out of trees dragged behind a vehicle are not very effective for redistributing loose material on the road.

6.3.4

Dragging gang

Dragging can be carried out by a small crew of a tractor driver and/or grader operator and a machine attendant, depending on the availability of equipment. It is most

21

Fig.12 Metal cutting drags

Fig.13 Types of brush drag

efficient to use several graders working in a team, one behind the other, spreading across the whole width of the road. Machinery should, where possible, be fitted with flashing yellow warning lights, and these plus headlights should be switched on when working. If warning lights are not available, machinery should carry flags. Traffic signs should be used as described in Section 3.2.

TABLE 3 PARTICLE SIZE DISTRIBUTION FOR GRAVEL SURFACINGS Percentage passing (*) Nominal maximum size 37.5mm 19mm 9.5mm 100 80-100 100 55-80 80-100 100 40-60 60-85 80-100 30-50 45-70 50-80 15-30 25-45 25-45 5-15 10-25 10-25

B S sieve size 37.5 19 9.5 4.75 2.36 425m 75m

6.4 6.4.1

REGRAVELLING The task

The surfacing material of gravel roads is worn away by traffic, eroded by rain and blown away as dust. Where this occurs the subgrade will be exposed particularly in ruts and depressions. Before all the material has been lost and the subgrade loses shape and gets damaged, the road requires regravelling. Regravelling is also used to correct loss of shape, ruts, potholes and erosion gullies, when these have become severe. Before regravelling work is carried out, it is important to make any necessary repairs or improvements to the drainage system of the road. If this is not done, the new gravel surface will deteriorate very quickly.

(*) Not less than 10% should be retained between each pair of such successive sieves specified for use, excepting the largest pair.

TABLE 4 PREFERRED PLASTICITY CHARACTERISTICS FOR GRAVEL SURFACINGS Liquid Limit not to Exceed (%) (*) 35

Climate

6.4.2

Quality of gravel
Moist tropical and wet tropical Seasonal wet Tropical Arid and semiarid

Most Roads Departments have standard specifications for gravels for surfacings (and for bases, where these are used). In practice, what is used will depend largely on what is available, and it may be necessary to use lowergrade material than is commonly specified. In dry climates, a fairly high proportion of clay binder is desirable to prevent the surface from ravelling and becoming corrugated. In wet climates, the presence of clay in the material is a disadvantage since it makes the surface slippery and prone to soften and rut under traffic. Suggested specifications in terms of grading and plasticity characteristics are given in Tables 3 and 4.

Plasticity Index range (%) (*) 49

Linear Shrinkage (%) 25

45

6 20

3 10

55

15 30

8 - 15

6.4.3

Organisation and equipment

(*) Higher limits may be acceptable for some laterites or concretionary gravels that have a structure that is not easily broken down by traffic. Lower limits may be appropriate for some other gravels that are easily broken down by traffic. Any variation from these limits should be based on carefully collated local experience.

Regravelling will be the major item of expenditure on the maintenance of gravel roads and its organisation should be carefully planned to ensure maximum efficiency. The following will provide a basis for an estimate of the plant required for regravelling: Gravel production: 1 bulldozer 1 loading shovel 1 grader 8 tipping lorries 6 labourers Production can be estimated as 450-500m3/day on a 5 kilometre average haul. If the haul exceeds 5 kilometres, additional lorries should be provided so that the other plant is fully utilised. Regravelling: 1 grader 1 6/8 tonne steel-wheeled roller and 1 1 6/8 tonne tyred roller roller and pneumatic steel-wheeled 1 pneumatic-tyred roller

24

2 water tankers, if water is available 1 water pump 1 lorry 20 labourers Production can be estimated as 300-350m3/day. To provide 100mm of gravel on one kilometre of road 7.5 metres wide will require 750 m3 of gravel (compacted volume) so that the above team could regravel about half a kilometre of road per day. Well in advance of the work, a start should be made to stockpile gravel at the borrow pit or quarry. In addition, arrangements must be made to obtain water close to the regravelling site. Whenever possible, a bulldozer or grader should open up a diversion track adjacent to the road. If a traffic diversion can be opened adjacent to the work site, it will enable the job to be carried out more efficiently and more safely. If traffic is heavy, the diversion may need to be gravelled and a grader assigned to keep the surface in good condition. After the diversion has been completed and before the work starts, warning signs, barriers and cones must be erected around the work area as described in Section 3.5. If it is not possible to open a diversion, 'lane closed' signing must be used as described in Section 3.3.

grader, but does require a continuous feed of lorries or the method becomes very ineffcient. Ideally, the moisture content of new gravel should be adjusted to optimum for compaction. In practice, this may be impracticable in view of the cost of providing, transporting and applying large amounts of water. However, it is usually possible to take advantage of the climate and to carry out regravelling work at the beginning of the dry season when the natural moisture content of gravel from borrow pits is close to optimum. It is important that the work is planned to obtain maximum benefit in this way. If additional water is required and available, the tankers should spray the road initially, before the new gravel is spread. The new gravel is then spread right across the road using the grader. The new material is alternately spread by the grader and watered by the tanker until its moisture content is correct for compaction. The tankers circulate continuously between the site and the source of water. Once the material has been spread evenly across the road and it is at the correct moisture content, it should be graded to shape as described in Section 6.2.3. Finally, the camber should be checked with the camber board and, if the required standard has not been reached, the grading should be repeated.

6.4.4

Spreading gravel

6.4.5

Compaction

It is advisable to reshape the existing surface before placing additional material. If this is not done it is likely that existing deformation will be reflected in the new surface. A hard surface should be scarified with a grader to a depth of about 50 mm to ensure a good bond between the new and existing material. The edges of the road should be 'boxed-out' to provide support for the new material. The camber of the graded surface should be checked to ensure that it is between 4 and 6 per cent. At the quarry, the tippers should be loaded for transport to site. The supervisor at the quarry should ensure that gravel is taken from the right place and that the trucks are loaded correctly. Tippers circulate continuously between the quarry and the site. Usually the gravel is supplied in advance and tipped in heaps on one side of the road at the correct spacing to give the required thickness of material when spread across the road. If the diversion has not been opened, material must be tipped onto the shoulder and warning signs placed at either end. If the gravel is fairly moist, this will not cause any serious problems, but dry gravel is likely to segregate during the loading, tipping and subsequent spreading operations. A better method of spreading gravel is to use a spreader box towed by a lorry. This is much faster than spreading using a

Compaction should not be left to the action of traffic as this quickly becomes concentrated in the wheeltracks and leads to deformation of the road. Rollers should be used if available as, even though it may not be possible to achieve full compaction, the limited compaction obtained will improve the quality of the surface. Four passes of a roller will give a worthwhile degree of compaction at optimum moisture content, while eight or more will be needed to bring relative compaction up to that required. Compaction should start as soon as the grader has finished a section. The rollers should start at the edge of the road and work towards the centre and continue rolling until full compaction has been achieved. This should be organised to finish at the same time as the grader finishes the next section. It is essential to check the thickness of the compacted layer.

6.4.6

Continuous working

The work should continue along the road in sections. As each section is completed, the traffic signs, cones and barriers are moved along the road to the next section. This opens the road at the completed

25

end for traffic, and closes it at the other end to allow new gravel to be dumped. As the work proceeds, it will be necessary to open new diversions.

6.6 6.6.1

DUST PREVENTION The dust problem

6.5

FILLING AND PATCHING

These are manual operations which can be used for repairs to the surface where defects develop on a small scale and heavy grading or regravelling is not justified. They can also be used when equipment is not available. They can be used to repair pot-holes, ruts, soft-spots and erosion gullies. The operations are sometimes known as spot regravelling. Patching may also be needed on selfcementing gravels or gravels containing large lumps of material as, in these cases, grading will only cause more damage to the surface. Filling and patching are not satisfactory methods of repairing corrugations and dragging or grading is needed to remove these. Where there are large numbers of pot-holes, the road will need heavy grading and possibly regravelling. The quality of the material used should be the same as that used for regravelling (Tables 3 and 4). It should be stockpiled at the nearest maintenance camp or dumped by the side of the road near where it will be used. Before work starts, signs must be set up as described in Section 3.3. Loose material and standing water should then be brushed from the area to be repaired. large or deep pot-holes should have their sides cut back to be vertical and should be deepened to reach sound material. If the material is dry, the area to be repaired should be sprinkled with water and it is then also useful to mix the patching material with water as well. The patch should then be filled in layers of about 50-70mm at a time. Each layer should be compacted with hand rammers or with small vibrating compactors. It is not advisable to roll with the wheels of the truck or tractor as insufficient compaction can be obtained in this way. The layers of the patch should be built up in this way and, finally, the patch is filled with gravel to approximately 30mm above the level of the road surface and is spread and raked to the correct shape. The patch is then compacted to give a surface which is slightly above the level of the surrounding road. Both large and small areas are repaired in the same way. Patching work started must not be left unfinished overnight. At the end of each day, tools and traffic signs should be taken back to the maintenance camp and the site must be left clean and tidy with no stockpiles of material left on the road. Details of the maintenance gang and equipment needed for filling and patching work are given in Section 5.8. The District Engineer will need to modify this basic unit to meet his own local conditions and for different situations.

In the dry areas of many developing countries, unpaved roads are often affected by dust. This may have been generated by the interaction between vehicular tyres and the unbound surface or by the effect of wind. Dust is a maintenance problem because it results in the loss of material from the road surface which has to be replaced. It is a contributory factor to road accidents because of the reduction in visibility and it also pollutes the atmosphere close to the road. It has been shown that well constructed and adequately maintained unpaved roads carrying 100 vehicles per day in dry areas can lose over 25 tonnes of dust per kilometre each year. In addition, it has been found that, when the wearing course contains an excessive amount of material finer than l0mm, the annual dust loss can be greater than 33 tonnes per kilometre. Losses on older roads with lower levels of maintenance will be much higher. This loss of material increases the permeability of the surface layer and results in the early development of pot-holes, all of which accelerate the need for regravelling.

6.6.2

Remedial treatment

The major types of remedial treatment for dust proofing unpaved roads are as follows:(i) (ii) application of a bituminous seal chemical stabilisation and coating with deliquescent salts application of waste local materials such as oiled gravels, sulphur liquors, molasses, palm oil, vegetable oil, bamboo oil, lime, charcoal, etc addition of water.

(iii)

(iv)

With the exception of the bituminous seal, all of these treatments provide only temporary improvements. Many of the additives are soluble in water and will require renewal at the end of the rainy season. Apart from the use of local materials, most treatments are also expensive. If dust treatment is being considered, careful costings should be carried out to ensure that the costs of repeated applications of the additive over several years are cheaper than the more permanent treatment by surface dressing.

26

7.
7.1

MAINTENANCE OF PAVED ROADS

TYPES OF MAINTENANCE

Maintenance operations on paved roads can be divided into those used to repair local failures and those used to improve the whole carriageway. Local surfacing failures such as cracking (without rutting), stripping and fretting may be repaired by local sealing. Fatting-up of bitumen over small areas is not a problem unless the bitumen is 'bleeding' when it should be repaired by sanding. Localised structural failure, where there is rutting and cracking, or pot-holing should be repaired by patching. Whilst surface dressing is essentially a waterproofing operation some minor surfacing defects such as loss of texture, fine cracking and fretting can often be corrected by this process. An alternative to surface dressing is slurry sealing. This is almost always carried out by specialist contractors and only brief notes are given on it here. It is usually more costly than surface dressing. Sand sealing is a form of surface dressing, but it should only be used where there is a shortage of aggregate and an abundance of sand. Its use is not recommended in most cases and no notes are given on it here. Reflection cracking sometimes occurs on roads constructed with cement or lime stabilised bases. It is characterised by a regular rectangular crack pattern and is caused by shrinkage of the base. It is not a sign of failure, but the surface should be sealed to keep out water. If the cracks are widely spaced (greater than 1 metre), they can be sealed with a bituminous binder. If they are closely spaced, it is probably better to surface dress. Where there is extensive structural failure, the road will need to be overlaid or reconstructed. For overlays to be economical, they must be placed when the road is in a 'critical' condition and before complete failure has occurred. Such major repairs are very expensive and use techniques more similar to construction than maintenance. Their use is beyond the scope of this Note and they are not described.

etc. Such considerations automatically limit the range of binders which can be used to bitumen emulsions and low viscosity cut-backs. Control of the bitumen application rate will often be poor and requires careful super-vision by an experienced foreman. Uneven distribution of bitumen is a common fault when it is applied using handlances. Typical productivities for hand-spraying emulsion are 600-1000 litres per day for a unit of three men with one sprayer. Cover aggregate should be of a small nominal size such as 6 mm chippings or coarse sand and should be lightly rolled. This work should be carried out by the patching gang (see Sections 7.4.3 and 5.8).

7.3

CRACK SEALING

This is used where reflection cracking has occurred and the aim is to fill the cracks as completely as possible with bituminous binder to keep out water. The method and materials used will depend on the equipment available for the work. Viscous binders will need a hand-lance with a relatively fine jet capable of getting the binder into the cracks. The use of a watering can for viscous binders is not effective, besides being wasteful and untidy. If sophisticated spraying equipment is available, it should be capable of filling cracks with penetration grade binders such as 80/100 or 60/70. Less effective equipment can only use bitumen emulsion for sealing the cracks. Sealed cracks may be blinded with quarry fines. This work should be carried out by the patching gang (see Sections 7.4.3 and 5.8).

7.4 7.4.1

PATCHING Procedure

7.2

LOCAL SEALING

The need for patching is usually the first sign of major pavement deterioration. It may be carried out on a recurrent basis or prior to the application of a surface dressing or overlay. The cause of the local failure that requires patching will often be related to problems of onroad or off-road drainage, and it is most important that these defects are identified and remedied before the patching operation begins. With all patching work, it is important to remove the failed area entirely and to cut the road back to sound material. The sides and bottom of the patch should be squared-off to provide a firm coherent surface. Where bituminous patching material is to be used, the excavated and trimmed area should be carefully brushed, moistened slightly with water and painted with bitumen emulsion or rapid-curing cut-back so as to provide a good bond with the in-filling material.

This may be used to seal relatively small areas of surfacing failure where the size of the works does not justify using a bulk bitumen distributor (eg. lengths of road less than 100 metres long). The method used will need to take account of the equipment, materials and men available to do the work. Frequently the process relies on the manual application of binder using watering cans, hand-pumps and lances,

27

Recommended materials for patching are described in Section 7.4.2. The material should be compacted into the hole in 50-70mm layers using hand-rammers or a small vibrating roller. The surface of the completed patch should be slightly higher than the road surface so as to permit final compaction by traffic. The patch should be sealed using bitumen emulsion or cut-back and blinded with sand, crushed rock fines or small chippings.

Premix macadam should comply with the requirements given in Table 7. TABLE 7 REQUIREMENTS FOR PREMIXED MACADAM USED FOR PATCHING MATERIAL 14mm nominal size

7.4.2

Materials

The materials used for patching will depend on those present in the existing road. Recommendations are given in Table 5. TABLE 5 RECOMMENDATIONS FOR PATCHING MATERIALS Layer Base Existing material Gravel Repair material Gravel of at least as good quality as the original Graded crushed rock plus fines, soil cement or premix macadam Seal with binder and small chippings Premix macadam plus seal of binder covered with sand, crushed rock fines or small chippings

BS sieve % passing (mm) 20 100 14 90-100 10 55-75 6.3 25-45 3.35 15-25 75m 2-6 Binder content 4 5% The binder to be used for premix patching material will usually be a soft penetration grade bitumen or a fairly viscous medium-curing cut-back, such as MC800 or MC3000, depending upon local conditions and the possible need to stockpile patching material. Premix macadam patching material must be mixed using a pugmill-type mixer. This may be at a central mixing plant where quality can be most easily controlled, although purpose-built mobile mixers are now available to permit cold or hot-mixed material to be manufactured on site. The mixes described above are adequate for small-scale patching work but, for larger-scale work. it is desirable to use higher-grade materials under close quality control.

Surface

Cement stabilised, bitumen stabilised, crushed rock or bitumen macadam Surface dressing (new, thin) Surface dressings which have built up to several centimetres thickness and premix surfaces

Specifications for suitable gravels, which may be used for patching, have been given earlier in Tables 3 and 4. These are also suitable for stabilisation with 5-10 per cent of cement. If crushed rock is used for the patching material, gradings can be again taken from Table 3 or alternatively from Table 6. TABLE 6 ALTERNATIVE GRADING FOR CRUSHED ROCK PATCHING MATERIAL BS sieve 37.5 mm 20 10 5 2.36 600m 75m % passing 100 80-100 55-80 40-60 30-50 15-30 5-15

7.4.3

Patching gang

Details of the maintenance gang and equipment needed for patching work are given in Section 5.8. The District Engineer will need to modify this basic unit to meet his own local conditions and for different situations.

7.5

MECHANISED SURFACE DRESSING General principles

7.5.1

Surface dressing consists of spraying the road surface with a film of binder followed by the application of a layer of stone chippings which is then rolled. Surface dressing has three main purposes: (i) (ii) (iii) to seal the road surface against water, to prevent disintegration of the surface, to provide a non-skid wearing surface.

Surface dressing will not restore the riding quality of mis-shapen roads nor will it significantly strengthen the road structure.

28

The use of mechanised surface dressing methods offers important advantages over manual methods. Mechanical bitumen sprayers allow the close adjustment of rates of application so that these can be accurately controlled and the adverse affects of excessive, insufficient or variable amounts of binder can be avoided. The rate of progress of a mechanised unit is much higher than can be achieved by manual methods. A separate document (TRRL Overseas Unit, 1982) gives detailed guidance for specialist surface dressing operations.

TABLE 9 LANE TRAFFIC CATEGORIES Approximate number of vehicles with unladen weight greater than 1.5 tonnes carried per day in the lane under consideration Over 2000 1000-2000 200-1000 20- 200 Less than 20

Category 1 2 3 4 5

7.5.2

Design of surface dressing

The design should take into account the type of existing road surface, the traffic, the available chippings and the climate.

7.5.2.1 Type of existing road surface Before

deciding on the specification for the surface dressing, a site visit is essential to assess the condition and hardness of the old road surface. For all practical purposes, surfaces may be divided into the following categories:TABLE 8 ROAD SURFACE TYPES Very hard Surfaces such as concrete or exceptionally lean bituminous mixtures with dry stony surfaces into which there will be negligible penetration of chippings under very heavy traffic. Surfaces containing some hard bituminous mortar into which chippings will penetrate only slightly under heavy traffic. Surfaces into which chippings will penetrate moderately under heavy and medium traffic. Surfaces into which chippings will penetrate considerably under heavy and medium traffic. Surfaces into which even the largest chip pings will be submerged under heavy traffic. Such surfaces are usually rich in binder.

7.5.2.3 Chippings The chippings should comply in all respects with the requirements of BS 63: Specification for single sized roadstone and chippings (BSI, 1971), the main points of which are summarised in Appendix A. Samples of the chippings to be used should be tested before the start of the work and subsequently as new deliveries are received. The sampling and testing should be in accordance with the methods described in BS 812 (BSI, 1975).
The size of chippings should be chosen to suit the amount of traffic and hardness of the existing road surface, as given in Table 10. The quantity of chippings needed must be sufficient to cover the entire surface of the binder film after rolling. The most reliable way of ensuring a complete cover is to lay a slight excess of chippings. The rate at which chippings should be spread depends on their size, shape and specific gravity, but rates can be estimated using Fig. 14. The least dimension of at least 200 chip pings should be measured, using the method described in Appendix A, and the 'Average Least Dimension' (ALD) determined. The Average Least Dimension is then entered in Fig. 14 to intersect line AB and the approximate rate of application of chippings can be read off. This rate should only be used as a guide for supply purposes; the actual rate of spread should be adjusted as required on site when the spreading characteristics of the chippings can be observed.

Hard

Normal

Soft

Very soft

7.5.2.2 Traffic categories The volume of traffic

that the road is required to carry is an important factor in the design of the surface dressing. Heavy vehicles embed chippings into the road surface and the amount of this embedment depends on the number and weights of these vehicles. Traffic should be considered in terms of the number of commercial vehicles a day in the lane under consideration. Lane traffic categories are shown in Table 9.

7.5.2.4 Binder Surface dressing binders should have a viscosity of between 104 and 5 x 105 centistokes at the prevailing road temperature. At higher viscosities, stone will not be wetted by the binder and will be lost by whipoff. At lower viscosities, the binder will be too fluid to hold the stone. Figure 15 shows the viscosity/ road temperature relationship for a wide range of binders. Day time road temperatures in the tropics range generally from 300C to 700C, from which it will

29

TABLE 10 RECOMMENDED CHIPPING SIZE (mm) Lane traffic category Type of surface (1) Over 2000 Very hard Hard Normal Soft Very soft * ** 10 mm 14 20** * * Approximate number of commercial vehicles currently carried per day in the lane under consideration (2) 1000-2000 10 mm 14 14 20** * (3) 200-1000 6 mm 10 14 14 20** (4) 20-200 6 mm 6 10 14 14 (5) Less than 20 6 mm 6 6 10 10

Unsuitable for surface dressing. 20 mm chippings may be used for remedial treatment where traffic speeds are low. Very particular care should be taken when using 20 mm chippings to ensure that no loose chippings remain on the surface when the road is opened to unrestricted traffic as there is a high risk of windscreen breakage.

Note: The size of chippings specified is related to the mid-points of lane traffic category ranges 2-5: lighter traffic conditions may make the next smaller size of stone more appropriate. TABLE 11 CONDITION CONSTANTS Traffic Very Light Light Medium Medium Heavy Heavy Very Heavy Existing surface Untreated/primed base Very lean bituminous Lean bituminous Average bituminous Very rich bituminous +6 +4 0 -1 -3 Vehicles/day 0-50 50-250 250-500 500-1500 1500-3000 3000+ Constant +3 +1 0 -1 -3 -5 Types of Chippings Round/dusty Cubical Flaky Pre-coated Constant +2 0 -2 -2

Climatic conditions Wet and cold Tropical (wet and hot) Temperate Semi-arid (dry and hot) Arid (very dry and very hot) +2 +1 0 -1 -2

be seen that the most appropriate binders are MC 3000 or penetration grades up to 80/100. The choice may be restricted in practice either as a result of Departmental purchasing policy or by what is commercially available. In order to design the rate of application of binder, an appropriate constant should be selected from Table 11 for each of the four sets of conditions listed. The four constants are then added together to give the appropriate factor.

The Average Least Dimension of the chippings and the factor obtained from the condition constants in Table 11 are then entered in Fig. 14 to give the rate of application of binder.

7.5.3

Equipment

Few districts will have a length of bitumen-surfaced road large enough to justify a permanent, full-time surface

30

Fig. 14 Surface dressing design chart

*1 2. For slow traffic or climbing grades steeper than 3 per cent, reduce the rate of spread of binder by 10 per cent For fast traffic or down grades steeper than 3 per cent increase the rate of spread of binder by 10 to 20 per cent

Fig.15 Surface temperature/choice of binder for surface dressing

dressing unit. It is more likely that a mobile unit, comprising the specialised machines and operators, will move from district to district in accordance with a centrally-planned schedule, with non-specialised plant and labour provided by the district in which the unit is working. However, a large part of the work involved in organising and carrying out the surface dressing operation will fall to the District Engineer. The operation of a mechanised surface dressing unit will probably represent the largest single item of road maintenance expenditure and should receive corresponding attention. A typical surface dressing unit might comprise: 1 bitumen distributor (capacity 5000-8000 litres) with pressurised spray bar and oilfired heaters bitumen preheaters, oil-fired tipper lorries (supplied by District) chipping spreaders, if lorries are not fitted with spreaders rubber-tyred rollers loading shovel (unless hand loading is used).

surface should be filled. Patching should be done well in advance of the surface dressing. For best results, this should be completed during the previous season to ensure that as uniform a texture as possible is avail-able for the subsequent dressing. It is important that the surface to be treated is clean. Dust and loose aggregate must be swept from the surface so that the binder film can adhere easily.

7.5.4.2 Application of binder The final quality of

the surface dressing will depend largely on the efficiency of the method of application of the binder, which must be applied uniformly over the road surface. The mechanised bitumen distributor must therefore be fitted with means of controlling the rate of spread of binder longitudinally on the road. In addition, the transverse rate of application across the spray bar should be checked at least once a year in the depot. Because the distributor is mounted on springs, the spray bar will tend to rise as the load of bitumen in the tank decreases. The correct spray pattern is dependent on the spray bar being kept at the correct height above the road surface and the spray bar must therefore be adjustable in height, under the control of the operator. The amount of binder actually sprayed for any single run of the distributor must be within plus or minus 10 per cent of the amount calculated by multiplying the area sprayed by the specified rate. In addition to checking the average rate of spread over a substantial area of road by means of readings from the dipstick fitted to the distributor, the actual rate of spread on the road should be determined by means of a tray test as described in Appendix A.

1 or 2 8 or more 1 or 2

2 1

In determining the make-up of the surface dressing unit it is important to keep a balance between the various items of equipment so that each is utilised to the maximum possible extent. A common fault is that there are too few lorries hauling chippings to keep up with the output of the bitumen distributor. Spraying must stop when no chippings are available for spreading. Such a unit is potentially capable of single-coat application at rates of up to two kilometres of 2-lane road per day, using some 20,000 litres of bitumen and 200 tonnes of chippings. In practice progress would fall short of this as a consequence of imbalances in equipment, interference by traffic and non-productive time, but the rate of use of materials will still be very high and it is important to ensure that stockpiles are adequate both in extent and location.

7.5.4.3 Spraying temperatures Thermometers

must be fitted to the bitumen distributor to measure the temperature of the binder delivered to the spray bar. The spraying temperature for any given grade of bitumen will be related to the design of the spray bar nozzles. It is good practice for a thermometer also to be fitted to the spray bar as inaccurate readings are likely to occur when thermometers within the tank become exposed as the level of binder falls. The application temperatures for different grades of bitumen are given in Table 12 for each type of jet. Whirling spray jets atomise the hot binder and expel it as a finely divided spray in the shape of a hollow cone. A hood is required to eliminate wind effects. No special setting is required for these jets and the transverse distribution is fairly insensitive to the height of the bar above the road. An unexpected characteristic of these jets is that the rate of spread is increased as the temperature of the binder falls. Slotted jets expel

7.5.4

Application

7.5.4.1 Preparation of the surface A surface

dressing does not contribute directly to the strength of a road pavement, neither does it improve the riding quality of the existing surface. Therefore it is essential that the underlying material has sufficient strength to withstand the expected traffic loading and is also within the tolerance required for the level of the final running surface. Any area of the road showing signs of structural weakness should be strengthened and any depressions in the

33

the hot binder in a thin fan-shaped film and consequently a hood is not required over the bar to protect it from wind. Slotted jets need to be set with the slots at an TABLE 12 SPRAYING TEMPERATURES Spraying temperature Whirling spray jets Slot jets

a good butt joint as an inaccuracy will lead to chippings not adhering. A good clean transverse joint may be obtained by the use of a strip of building paper spread across the road on which to start and stop spraying.

C
Cutback grades MC30 RC/MC70 RC/MC250 RC/MC800 RC/MC3000 Penetration grades 400/500 280/320 180/200 80/100 50-60 65-80 95-115 115-135 135-155

C
40-50 55-70 80-90 105-115 120-130

7.5.4.5 Application of chippings Chippings should be applied uniformly over the freshly sprayed binder film by a mechanical spreader which should follow closely behind the sprayer. Greater uniformity of spread will be achieved with spreaders fitted with metering devices. Any thinly chipped areas found after the passage of a mechanical spreader should be chipped over by hand to obtain shoulder-to-shoulder cover. Any surplus chippings accidentally spilled on to the road surface should be removed. 7.5.4.6 Rolling Rubber tyred rollers are strongly recommended for rolling surface dressings. Rolling should start immediately after the chippings have been spread. The faces of the chippings in contact with the binders should be pressed into it, so that maximum adhesion occurs and slow-moving rubber tyres are ideal for this purpose. Steel-tyred rollers tend to crush chippings and to bridge local depressions. If steel rollers have to be used, they should be the lightest available and in no case should their weight exceed 8 tonnes. 7.5.4.7 Aftercare and opening to traffic Slow
moving traffic can help to roll and align the chippings immediately after surface dressing, but vehicle speeds must be restricted to 30 km/h or less until there is sufficient adhesion to ensure that the chippings will not be plucked from the surface. This period of time will vary from a few hours to a day or more, depending on the climatic conditions and the type of binder employed. Excess chippings should be swept up and taken away before opening the road to unrestricted traffic. Signs may be used to warn of possible damage to windscreens.

160-170 165-175 170-190 180-200

140-150 150-160 155-165 165-175

angle to the horizontal of 15 to 30 so as to permit overlap of adjacent fan patterns. The slots in these jets are not easily blocked and have a fairly high rate of delivery thus the road speed of the machine when spraying is higher than that of a machine fitted with atomising, whirling spray jets. Because of the inflammable nature of the solvent used in RC-type cutbacks, application temperatures for RC grades should be restricted to the lower parts of the ranges given in Table 12. No smoking or naked flames should be allowed when heating, pumping or spraying any cutbacks. Fire extinguishers should always be readily at hand. 7.5.4.4 Jointing strips It is most important to ensure that the thickness of the binder film is not reduced at longitudinal joints. There must be a clear understanding of the jointing technique appropriate to the type of distributor used. For example, on some machines the intensity of application falls away towards the edges of the strip being sprayed. Adjacent strips should then be overlapped sufficiently to give the required uniformity of spread. This means that the first strip laid must be left uncovered by chippings for a width of several centimetres along the edge to be overlapped. Other types of machine are designed to apply the full thickness of the binder film to the extreme edge of the strip. With these, great care must be taken to make

7.5.4

Control

The success of any surface dressing depends to a large extent on the control of all aspects of the design, application and quality of materials. Table 13 is a checklist of the items that require checking in chronological order. This should be read in conjunction with Appendix A.

7.6 7.6.1

MANUAL SURFACE DRESSING Applicability

Labour-based techniques for surface dressing may be

34

TABLE 13 CONTROL OF SURFACE DRESSING Stage when action should be taken 1. Before start of job

wage rates and is not recommended. Hand-screening of river gravel, however, may be worth considering provided that the product complies with the requirements noted in Appendix A. Notes on hand-screening are given in a separate document (Robinson, 1979).

Action Required (i) Check that any pot-holes or edge-failures have been repaired. (ii) Check grading, shape and (if necessary) the resistance to polishing of aggregate. (iii) Ensure that chippings are not dusty or dirty. (iv) Check that binder is of correct type and viscosity. (v) Check availability and condition of plant. (vi) Ensure that traffic control measures are adequate. (i) Ensure that the surface to be dressed has been adequately swept and cleaned. (ii) Check binder temperature. (iii) Test spray bar jets. (i) Check rates of spread of binder. (ii) Check that chipping spreading follows closely behind binder distributor and that coverage is correct. (iii) Check that rollers follow immediately after the spreading of chippings. (iv) Check binder temperature at intervals. (i) Limit traffic speed when road is first opened to traffic. (ii) Remove surplus chippings.

7.6.3

Binder

For small-scale work it may be more convenient to use bitumen-emulsion, applied at the ambient temperature, rather than cut-back or penetration grade which require pre-heating. Cationic emulsions are now probably more commonly used than anionics as they are claimed to be more effective in depositing a binder film on to acidic aggregates. However, as with anionic emulsions, the principal break mechanism is the evaporation of water. Disadvantages common to both types are: (i) Poor adhesion to dry or dusty surfaces. This may be overcome by slightly dampening the surface before spraying. Low effective bitumen content (usually not more than 60%), which must be allowed for when calculating rates of application. Emulsions containing up to 70 per cent bitumen have been developed, but have to be heated (to 70-800C) before spraying, thus losing the advantage of use with unheated equipment.

2.

Before starting daily work

(ii)

3.

During work

(iii) Low viscosity emulsions tend to drain from the crown of the road before the break occurs; cover aggregate is therefore poorly held at the crown, while excess binder accumulates at the edges. (iv) Some emulsions tend to coagulate if stored or transported in unsuitable conditions. If emulsion has to be stored for long periods, the drums should be turned over occasionally.

3.

After surface dressing

7.6.4
used on small or remote projects, where the use of plant is too expensive or impracticable. The design principles and standards described in the preceding section relating to mechanical methods apply generally to labour-based work. The following paragraphs describe differences where these are justified, as well as outlining some aspects of manual techniques.

Application of binder

Bitumen (cut-back, penetration grade or emulsion) may be applied manually using hand-operated sprayers or by means of watering-cans. The choice of binder type will need to be related at least partly to the application equipment available. In either case, the road surface should be marked out in a rectangular grid such that the area to be covered by each full container (drum or can) is defined. This may be done by placing stones (about fistsize) at the corners of the area to be covered. Provided that the work is set out in this way and the bitumen applied systematically, it should be possible to get acceptably uniform rates of application. It will be necessary to supervise the work carefully and to instruct operators in detail in the methods to be used. Uneven distribution of bitumen is a common fault when bitumen is applied by hand-lances.

7.6.2

Chippings

Chippings for surface dressing will normally be smaller than 20 mm nominal size. Production of these relatively small sizes by hand-breaking is uneconomic even at low

35

Bitumen for labour-based surface dressing work should be brought to the site in drums and distributed along the work in accordance with the requirements for rate of spread. Typical productivities for hand-spraying emulsion are 600-1000 litres per day for a unit of three men with one sprayer.

voids and cracks and hold together surfaces that are starting to fret or ravel. Costs are higher than for surface dressing, but the resulting layer can improve the riding quality of the road which a surface dressing would not. The higher costs may therefore sometimes be justified, particularly for badly cracked or lean bituminous surfacings. However, a slurry seal will not have such good surface texture as a surface dressing and may therefore be slippery in wet weather. Both anionic and cationic emulsions may be used, but cationic emulsion is normally used in slurries containing acidic aggregates. Anionic slurries may be premixed in a static mixing plant but, because emulsions in cationic slurries break relatively quickly, these have to be prepared in a purpose-built mixing and laying machine. A contractor usually supplies this equipment, its operators and the bitumen binder; the maintenance organisation will usually supply the aggregate, water and the labour force. The laying technique can range from simple 'squeegees' to modern mixer-spreader units. The following specification is reproduced by permission of Colas (East Africa) Ltd. Aggregate Gradations (all percent cumulative, passing)

7.6.5

Application of chippings

Chippings should be delivered to the site in dumps at intervals of about 400 metres, from where they are taken to the work in wheelbarrows, with an average haul of 100 metres. The chippings are usually spread by a swinging movement with the shovel and this operation also requires close supervision if full coverage is to be achieved. It is necessary to apply a greater excess when spreading chippings manually than when mechanised spreaders are used, the excess being recovered by brooming after rolling and trafficking. Productivity for spreading chippings, including haulage from dumps by wheelbarrows, is about one m3 /man-day.

7.6.6

Rolling
Sieve Fine Seal General Seal Coarse Surface 100 70-90 45-70 28-50 19-34 15-25 7-18 5-15

There is no practicable alternative to mechanical rolling. The low outputs from labour-based work will result in under-utilisation of self-propelled rollers, and it may be better to consider the use of tractor-drawn rollers. Any roller should preferably be rubber-tyred, as explained previously.

7.6.7

Gang size

The following gang sizes have been found to be practicable: Spraying bitumen 500 m2 /day - 3 men plus 1 sprayer Spreading chippings 500 m2/day - 6 men plus 4 wheelbarrows Loading/unloading drums, cleaning surface and other odd jobs - 7 men Total - 16 men plus 1 foreman. Also required: (i) periodic use of truck or tractor/trailer for delivery of chippings (ii) tractor-drawn rubber-tyred roller.

10mm 6.3 mm 5.0 mm 2.36 mm 1.18 mm 600 m 300 m 150 m 75 m

100 100 100 100 65-90 40-60 25-42 15-30 10-20

100 100 85-100 65-90 45-70 30-50 18-30 10-21 5-15

Slurry Composition Aggregate Cement (catalyst) Terolas A3 bitumen emulsion * Per m3 1m3 15kg Percent by mass 81-83 1

280-330 litres As required

16-18

Water *

Type A3 is a slow setting anionic emulsion, 55 per cent bitumen (minimum)

7.7

SLURRY SEALING

Coverage Old surfaces m2 /m3 130-250

Slurry seals are mixtures of fine aggregates, bitumen emulsions and additional water. When freshly mixed, they have a thick creamy consistency and can be spread to a thickness of 5 to 10 mm. The principal application of slurry seals is as a maintenance treatment for old bituminous surfaces. Because of their low viscosity, they can readily penetrate surface

New surfaces: On 20mm aggregate 130 170 On 14mm aggregate 170 240 On 10mm aggregate 180 250 On primed base 150 180 (in 2 layers)

36

8.

REFERENCES

APPENDIX A
ADDITIONAL NOTES ON SURFACE DRESSING A.1 CHIPPINGS

AVELING-BARFORD, undated. Grading illustrated. Technical Publication TP 549. Grantham: AvelingBarford Ltd. BSI, 1971. Specification for single-sized road-stone and chippings. BS63: 1971 (Part 2). London: British Standards Institution. BSI, 1975. Methods for sampling and testing of mineral aggregates, sands and fillers. BS812: 1975 (Parts 1, 2 and 3). London: British Standards Institution. ROBINSON, R, 1979. Some manual methods of screening aggregates for labour-intensive road construction. TRRL Supplementary Report 503. Crowthorne: Transport and Road Research Laboratory. TRRL OVERSEAS UNIT, 1981. Maintenance management for District Engineers. Overseas Road Note 1. Crowthorne: Transport and Road Research Laboratory. TRRL OVERSEAS UNIT, 1982. A guide to surface dressing in tropical and sub4ropical countries. Overseas Road Note 3. Crowthorne: Transport and Road Research Laboratory.

A.1.1 Size
The chippings for a surface dressing should be of singlesized crushed rock, roughly cubical in shape and clean and free from dust. The nominal sizes used may vary between 6 mm to 20 mm according to the local requirements. The chippings should comply with BS63.Specification for single sized roadstone and chippings (BSI, 1971). It is most important that the chippings should be single sized. Graded chippings make design and construction much more difficult and their use, without the much closer control which is required, will lead to loss of the larger sizes of chipping and to fattingup of the surface dressing. Single sized aggregate may be more expensive to produce than graded aggregates, but their higher cost can be offset by better performance. The remaining sizes of stone can be absorbed into other sectors of the road construction programme. Recommendations for the grading of surface dressing aggregate are given in Table A1. The use of rounded-gravel aggregates should, if possible, be avoided because it is difficult for the binder film to hold them in place and because of their poor surface friction properties. If their use cannot be avoided, adjustments must be made to the rate of spread of bitumen. The rounded aggregates do not interlock and more bitumen is required to hold the particles firmly to the road surface than is required for cubical aggregates.

A.1.2 Average least dimension

Once the surface dressing is laid, the traffic will orientate the chipping to lie as flat as possible with the least dimension vertical. Therefore the Average Least Dimension (ALD) is the measurement which best classifies the chippings and gives a guide to the required rate of spread of bitumen and the rate of spread of chippings. Two methods for determining the Average Least Dimension of the chippings are given below:-

Method 1
The ALD of chippings may be determined by taking a representative sample, say 200 chippings, and measuring the least dimension of each chipping with a pair of calipers. The average of a series of such measurements is then the ALD of the chippings.

Method 2
A sieve analysis is first carried out on a surface-dry sample of the chippings and the results plotted on a

37

TABLE A1 GRADING AND SHAPE OF CHIPPINGS

38

Fig. A1 Average least dimension chart

large scale grading curve. The median size of chipping is then determined as that theoretical sieve size through which 50 per cent of the material will pass. The flakiness index is determined as in Section A.1.3. Then using Figure Al the median size is marked on scale A and flakiness index on scale C. A straight line is drawn between the two marks and the Average Least Dimension read off on scale B. Descriptions of these tests are given in BS812. Methods for sampling and testing mineral aggregates, sands and fillers Part 3: Mechanical properties (BSI, 1975). The flakiness index of an aggregate is the percentage by weight of particles in it whose least dimension (thickness) is less than three fifths of the nominal size. The test is not applicable to material passing a 6.30 mm sieve. Apparatus consists of:(i) A metal thickness gauge of the pattern shown in Figure A2. The width of the slot used in the gauge shall be the dimension specified in the 'thickness gauge' column of Table A2 for the appropriate fractions.

A.l.3 Flakiness
The flakiness index of chippings should not exceed 35. The method for determining flakiness index is given below. The chippings should also satisfy at least one of the following strength criteria:Aggregate crushing value - less than 25 (Values up to 35 may be permitted for lightly trafficked roads) Aggregate abrasion value - less than 12 Ten per cent fines value - not less than 8

(ii) BS test sieves as shown in Table A2 (BSI, 1969). (iii) A balance accurate to 0.5 per cent of the weight of the test sample. 1. Sample quantities are given in Table A2. Allowance should be made for the later rejection of particles retained on 63 mm BS

Aggregate impact value - less than 30

39

Fig.A2 Thickness gauge

TABLE A2 DIMENSIONS OF THICKNESS GAUGES Aggregate size-fraction BS test sieve nominal aperture size 100% passing mm 63.0 50.0 37.5 28.0 20.0 14.0 10.0 100% retained mm 50.0 37.5 28.0 20.0 14.0 10.0 6.30 mm 33.9 + 0.3 26.3 + 0.3 19.7 + 0.3 14.4 + 0.15 10.2 + 0.15 7.2 + 0.1 4.9 + 0.1 kg 50 35 15 5 2 1 0.5 Thickness gauge Width of slot Minimum mass for subdivision

test sieve and passing 6.30 mm BS test sieve. The test sample shall be obtained from the laboratory sample by quartering, and dried. 2. The sample shall be weighed and separated into the appropriate size fractions from Table A2, columns 1 and 2, by sieving, rejecting material retained on a 63 mm sieve or passing a 6.3 mm sieve. Record the sum of all fractions M1, discard those which are 5 per cent or less of M1 and record the remainder as M2. Each appropriate fraction shall be gauged in turn for thickness on the thickness gauge.

4.

The total amount M3 passing the thickness gauge shall be weighed to an accuracy of at least 0.5 per cent of the weight of the test sample.

5.

3.

The flakiness index is the total weight M3 of the material passing the various thickness gauges, expressed as a percentage of the total weight M2 of the sample gauged to the nearest whole number, and shall be reported together with the sieve analysis. M3 x 100 Flakiness Index = M2

40

A.1.4

Adhesion and pre-treatment

(vi)

retain the chippings even at the highest ambient temperatures, be flexible at the lowest ambient temperature, neither cracking and allowing water to enter nor brittle thus allowing the chippings to break free, resist excessive weathering and hardening once the initial hardening has taken place.

The most critical period for a surface dressing occurs immediately after the chippings have been applied. At this stage the chippings have not formed an interlocking mosaic and they are held mainly by adhesion forces in the binder film. This adhesion takes time to develop and is completely stopped if the chippings remain wet. Adhesion is very slow to develop if the chippings are dusty. However in hot climates the chippings may be washed with fresh or salt water just prior to spreading since the chippings dry out in a few minutes after they are spread on the road and the development of adhesion is not significantly delayed. If adhesion problems are experienced it may be useful to treat the chippings with a light spray of diesel oil or kerosene prior to spreading. Enough should be added to render the surface of the chippings slightly damp.

(vii)

A.1.5

Stockpiling

Chippings should be stored in stockpiles which are large enough to supply four to five days work without hauls exceeding 15 kilometres. Stockpiles will typically be about 2000 tonnes and will require an area of about 1000 m2 if formed by tipping from lorries. If suitably located, these areas will be reused at intervals of three to five years, and some permanence in their construction will be justified by a reduction in wastage of chippings. They may be surfaced, either with a two-coat surface dressing, bricks or concrete, with a compacted gravel base in each case. A concrete or brick wall about one metre high on three sides will define the stockpile, avoid contaimination and simplify measurement.

Surface dressing work will normally be carried out with cut-back or penetration-grade bitumens. Cut-back bitumens are likely to be of the high-viscosity, mediumcuring type. The advantage of these binders is that the cutting agent (normally kerosene) lowers the viscosity sufficiently to obtain good wetting of the chippings by the binder. Also the small quantity of cutting agent present results in a reasonably rapid set to hold the chippings after initial adhesion has been achieved. Penetration-grade bitumens may be used in the drier. hotter regions, depending upon actual road temperatures at the time of working. With mechanised methods, the rate of consumption of bitumen is high and it is desirable that it is supplied in bulk rather than in drums, with their associated problems of handling and leakage. It is more economic to transport bitumen in bulk than in drums and the pre-heaters can, if large enough, act as holding tanks and ensure that occasional interruptions to the supply do riot disrupt the whole operation. It is not usually necessary to test the grade of the bitumen on site, reliance being placed on the supplier's certificate.

A.2.2

Viscosity

A.2 A.2.1

BINDER Grade

The performance and qualities required of a surface dressing binder makes the choice of the binder critical. The binder must:(i) (ii) be sprayable at a reasonable temperature, 'wet' the surface of the road and remain in a continuous film waterproofing the road structure, not run off a steep gradient or cambered road or form pools of binder, 'wet' and adhere to the stone chippings at road temperature, be strong enough to resist the traffic force and

(iii)

Surface dressing binders should have a viscosity of between l04 and 5 x l05 centistokes at the prevailing road temperature. At higher viscosities, stone will not be wetted by the binder and will be lost by whip-off; at lower viscosities wetting will occur but the binder will be too fluid to hold the stone. Figure 15 shows the relationship between binder viscosity and road temperature for a wide range of binders and provides a means of selecting a suitable binder for use with a particular road surface temperature. Daytime road surface temperatures in dry weather in the tropics are likely to range from 300C to 700C according to region and season. This suggests that the most suitable binders for surface dressing will be between MC 3000 and 80/100 pen bitumen. The use of a more viscous bitumen than 80/100 pen is not recommended. Occasionally chippings have to be spread manually instead of by mechanical gritters. in such cases it may be advisable to use a slightly less viscous binder than would otherwise have been required. (See also Section 7.6 on manual surface dressing).

(iv)

(v)

41

It is unlikely that the complete range of possible binders will be available to the engineer and therefore for practical reasons it will probably be necessary to select perhaps two binders for general surface dressing use. In this case MC 3000 and 180/200 pen are likely to be the most suitable: MC 3000 for roads with prevailing surface temperatures less than 450C and 180/ 200 pen when prevailing surface temperatures are 450C and above.

trays should be used. When using this test, care must be taken to raise the protective hood on the distributor, if fitted, so that it does not touch the trays, but not so much as to let any wind disturb the spray. The tray test only gives an approximation to the average rate of spread. If it is to be used, it must be remembered that the maximum variation shown by individual trays will be greater than that of groups of trays. The mean result from three or more trays, spaced at intervals of not less than 10 metres, should fall within 10 per cent of specification.

A.2.3

Measuring the rate of spread of the binder

A.2.3.1 Average rate There are two methods of

measuring the average rate of spread of the binder. In the first the volume of binder sprayed is calculated from weighbridge records and this is divided by the measured area covered in a day. Although this is useful in checking quantities for payment, it does not reveal differences in rates of spread along the road and is therefore of limited value. A better method, which should be used on every site, is to measure the average rate over an area of 500 - 800 square metres. If the distributor is fitted with an accurate dipstick, this method can be of considerable value even though it does not give a measure of maximum variation. To make the test, the distributor is placed on a level stretch of road and the position of the wheels are marked. Dipstick readings are taken, repeating the dips until consistent readings are obtained. Binder is sprayed by the normal technique until about 10 per cent of the total volume in the tank has been used. Dipstick readings are taken with the distributor in the same position as before. The area covered is then measured accurately and the average rate of spread calculated after allowance has been made for any variation in the intensity of application at the longitudinal joint (see jointing of strips, Section 7.5.4.4).

A.3

REFERENCES

BSI, 1969. Specification for test sieves. BS4IO: 1969. London: British Standards Institution. BSI, 1971. Specification for single sized road-stone and chippings. BS63: 1971 (Part 2). London: British Standards Institution. BSI, 1975. Methods for sampling and testing mineral aggregates, sands and fillers. BS812: 1975 (Parts 1, 2 and 3). London: British Standards Institution.

A.2.3.2 Longitudinal variation Longitudinal

variation in the rate of spread of the binder can be checked with a tray test. In this test, light metal trays about 200 mm square by 5 mm deep, and numbered, are placed at intervals along the road in the path of the binder distributor. After the distributor has passed, the trays are removed, wrapped in weighed sheets of paper and taken away to be weighed; the rate of spread can then be calculated for each point on the road. A balance accurate to 0.1 gramme installed in a car, van or mobile laboratory permits the results from five trays to be obtained within 10 minutes. The spacing and number of trays used can be varied to suit the circumstances of the particular investigation, but at least five

42

NOTES

ISSN 0951 - 8987

OS-E

Maintenance techniques for district engineers

ORN 2

Overseas Road Note 3 (2nd Edition)

A guide to surface dressing in tropical and sub-tropical countries

ORN 3 (2 nd Edition)

First Published 2000 ISSN 0951-8797 Copyright TRL Limited 2000.

Subsector: Theme: T2

Transport

Project title: A guide to surface dressing in tropical and sub-tropical countries Project reference: D251

This document is an output from a project funded by the UK Department for International Development (DFID) for the benefit of developing countries. The views expressed are not necessarily those of the DFID.

IRL is committed to optimising energy efficiency, reducing waste and promoting recycling and re-use. In support of these environmental goals, this report has been printed on recycled paper, comprising 100% post-consumer waste, manufactured using a TCF (totally chlorine free) process.

ii

ACKNOWLEDGEMENTS This Overseas Road Note was produced in the International Division of TRL (Director Mr S W Colwill) on behalf of the Department for International Development. The research has been carried out with the active collaboration of highway authorities in many countries and their help and cooperation has been essential to the success of the project. The TRL project team responsible for this Road Note were Mr H R Smith (Project Officer), Dr J Rolt and Mr W G Ford. The helpful comments of the World Bank are gratefully acknowledged. Where necessary, use has been made of work published by other research and road authorities.

OVERSEAS ROAD NOTES Overseas Road Notes are prepared principally for road and transport authorities in countries receiving technical assistance from the British Government. A limited number of copies is available to other organisations and to individuals with an interest in roads overseas, and may be obtained from:

International Development Advisory and Information Unit TRL Limited Crowthome, Berkshire, RG45 6AU United Kingdom

Limited extracts from the text may be reproduced provided the source is acknowledged. For more extensive reproduction, please write to the address given above.

iii

iv

CONTENTS
Page

Preface 1 Introduction
2 Types of surface dressing Single surface dressing Double surface dressing Triple surface dressing Racked-in surface dressing Other types of surface dressing Chippings for surface dressings 5 Bitumens Prime coats Bitumens for surface dressings Penetration grade bitumens Bitumen emulsion Cutback bitumens Polymer modified bitumens Adhesion agents Design Existing site conditions Selecting the binder Choice of binder and timing of construction work

1 3
3 3 3 4 4 5

3 4

6 6 7 7 7 8 9 10 10 10 12 13 13 13 14 15 15 15 17 17 17

Designing the surface dressing

Basis for the design method Determining the average least dimension of chippings Determining the overall weighting factor Determining the basic bitumen spray Spray rate adjustment factors Adjusting rates of spray for maximum durability Surface dressing design for low volume roads Spread rate of chippings

6 Plant and equipment

Methods of distributing binder Constant volume distributors Constant pressure distributors Principal components of binder distributors Spray bars and spray jets Binder pumps and air pumps Tanks and burners 18 18 19 19 19 20 20

Page Distributor speed control and calibration Chip spreaders Rollers and other equipment 7 The surface dressing process Planning The surface dressing operation After-care 8 Other surface treatments Slurry seals Otta seal Sand seals Synthetic aggregate and resin treatments Applications of light bitument sprays Fog sprays Enrichment sprays 9 References 20 21 21 22 22 23 23 24 24 25 25 25 25 26 26 26

Appendix A: Requirements for grading and particle shape (Reproduced from BS 63: Part 2: 1987) Appendix B: Recommended polished stone values of chippings for roads in Britain Appendix C: The immersion tray test for determining the concentration of adhesion agent required Appendix D: The probe penetration for test for measuring road surface hardness Appendix E: Example of a surface dressing design Appendix F: Tests for uniformity of transverse distribution of binder (depot tray tests) (reproduced from British Standard 1707:1989) Appendix G: Bitumen distributor: rate of spread/speed calibration Appendix H: Miscellaneous equipment required for a surface dressing unit

28

29

30

31 32

33

35

35

vi

Preface
This Road Note is the 2nd edition of Overseas Road Note 3 'A guide to surface dressing in tropical and sub-tropical countries' The revisions include descriptions of a wider range of types of surface dressing, current materials specifications and more detail on the use of bitumen emulsions. An equation has been introduced to allow the design spray rate to be calculated as an alternative to reading from a nomograph. A simplified presentation of spray rate adjustment factors, related to different site conditions, has also been provided. The durability of surface dressings, particularly for low volume roads, is discussed and additional spray rate adjustment factors have been suggested for these roads. The use of Otta seals has also been suggested for low volume roads. It is important that this Road Note, is not regarded as a specification. However, the advice contained in this Road Note together with local experience of relevant materials and surface dressing performance should be of value to those drawing up specifications.

Introduction

1.1 Surface dressing is a simple, highly effective and inexpensive road surface treatment if adequate care is taken in the planning and execution of the work. The process is used throughout the world for surfacing both medium and lightlytrafficked roads, and also as a maintenance treatment for roads of all kinds. 1.2 Surface dressing comprises a thin film of binder, generally bitumen or tar, which is sprayed onto the road surface and then covered with a layer of stone chippings. The thin film of binder acts as a waterproofing seal preventing the entry of surface water into the road structure. The stone chippings protect this film of binder from damage by vehicle tyres, and form a durable, skid-resistant and dust-free wearing surface. In some circumstances the process may be repeated to provide double or triple layers of chippings. 1.3 Surface dressing is a very effective maintenance technique which is capable of greatly extending the life of a structurally sound road pavement if the process is undertaken at the optimum time. Under certain circumstances surface dressing may also retard the rate of failure of a structurally inadequate road pavement by preventing the ingress of water and thus preserving the inherent strength of the pavement layers and the subgrade. 1.4 In addition to its maintenance role. surface dressing can provide an effective and economical running surface for newly constructed road pavements. Existing roads with bituminous surfacings, carrying in excess of 1000 vehicles/lane/day, have been successfully surfaced with multiple surface dressings. For sealing new roadbases traffic flows of up to 500 vehicles/lane/day are more appropriate, although this can be higher if the roadbase is very stable or if a triple seal is used. A correctly designed and constructed surface dressing should last at least 5 years before resealing with another surface dressing becomes necessary. If traffic growth over a period of several years necessitates a more substantial surfacing or increased pavement thickness, a bituminous overlay can be laid over the original surface dressing when the need arises. 1.5 The success of a surface dressing depends primarily on the adhesion of the chippings to the road surface, hence both the chippings and the road surface must be clean and free from dust during the surface dressing process. Inappropriate specifications, poor materials, and bad workmanship, can also drastically reduce the service life of a surface dressing. 1.6 This Road Note is a general guide to the design and construction of surface dressings in tropical and sub-tropical environments and draws attention to some of the more common mistakes that are made. It

provides a framework on which the engineer can base more specific decisions made to suit particular local conditions thereby producing cost effective results. It also contains brief descriptions of certain other types of surface treatment.

Types of surface dressing

2.1 Surface dressings can be constructed in a number of ways to suit site conditions. The common types of dressing are illustrated in Figure 1. Single surface dressing 2.2 When applied as a maintenance operation to an existing bituminous road surface a single surface dressing can fulfil the functions required of a maintenance re-seal, namely waterproofing the road surface, arresting deterioration, and restoring skid resistance. A single surface dressing would not normally be used on a new roadbase because of the risk that the film of bitumen will not give complete coverage. It is also particularly important to minimise the need for future maintenance and a double dressing should be considerably more durable than a single dressing. However, a 'racked-in' dressing (see paragraph 2.7) may be suitable for use on a new roadbase which has a tightly knit surface because of the heavier applications of binder which is used with this type of single dressing.

Double surface dressing 2.3 Double surface dressings are robust and should be used when: A new roadbase is surface dressed. Extra 'cover' is required on an existing bituminous road surface because of its condition (e.g. when the surface is slightly cracked or patched). There is a requirement to maximise durability and minimise the frequency of maintenance and resealing operations.

2.4 The quality of a double surface dressing will be greatly enhanced if traffic is allowed to run on the first dressing for a minimum period of 2-3 weeks (and preferably longer) before the second dressing is applied. This allows the chippings of the first dressing to adopt a stable interlocking mosaic which provides a firm foundation for the second dressing. However, traffic and animals may cause contamination of the surface with mud or soil during this period and this must be thoroughly swept off before the second dressing is applied. Such cleaning is sometimes difficult to achieve and the early application of the second seal to prevent such contamination may give a better result.

Figure 1 Type of surface dressings

2.5 Sand may sometimes be used as an alternative to chippings for the second dressing. Although it cannot contribute to the overall thickness of the surfacing, the combination of binder and sand provides a useful grouting medium for the chippings of the first seal and helps to hold them in place more firmly when they are poorly shaped. A slurry seal may also be used for the same purpose (see paragraph 8.2). Triple surface dressings 2.6 A triple surface dressing (not illustrated in Figure 1) may be used to advantage where a new road is expected to carry high traffic volumes from the outset. The application of a small chipping in the

third seal will reduce noise generated by traffic and the additional binder will ensure a longer maintenance-free service life. Racked-in surface dressing 2.7 This system is recommended for use where traffic is particularly heavy or fast (TRL, 1996). A heavy single application of binder is made and a layer of large chippings is spread to give approximately 90 per cent coverage. This is followed immediately by the application of smaller chippings which should lock-in' the larger aggregate and form a stable mosaic. The amount of bitumen used is more than would be used with a single seal but less than for

a double seal. The main advantages of the racked-in surface dressing are: Less risk of dislodged large chippings. Early stability through good mechanical interlock. Good surface texture.

Other types of surface dressing

2.8 'Sandwich' surface dressings are principally used on existing binder rich surfaces and sometimes on gradients to reduce the tendency for the binder to flow down the slope. 2.9 'Pad coats' are used where the hardness of the existing road surface allows very little embedment of the first layer of chippings, such as on a newly constructed cement stabilised roadbase or a dense crushed rock base. A first layer of nominal 6mm chippings will adhere well to the hard surface and will provide a 'key' for larger l0mm or l4mm chippings in the second layer of the dressing.

surface dressing is to restore or enhance the skid resistance of the road surface. The PSV required in a particular situation is related to the nature of the road site and the speed and intensity of the traffic (Salt and Szatkowski, 1973). The resistance to skidding is also dependent upon the macro texture of the surface which, in turn, is affected by the durability of the exposed aggregate. This property is measured by the aggregate abrasion value (AAV). Appendix B gives recommended values of PSV and AAV for various road and traffic conditions in Britain and provides an indication of the required aggregate properties. 3.5 The nominal sizes of chippings normally used for surface dressing are 6, 10, 14 and 20 mm. Flaky chippings are those with a thickness (smallest dimension) less than 0.6 of their nominal size. The proportion of flaky chippings clearly affects the average thickness of a single layer of the chippings, and it is for this reason that Jackson (1963) introduced the concept of the 'average least dimension' (ALD) of chippings. 3.6 In effect, the ALD is the average thickness of a single layer of chippings when they have bedded down into their final interlocked positions. The amount of binder required to retain a layer of chippings is thus related to the ALD of the chippings rather than to their nominal size. This is discussed further in Section 5 where guidance is given on the selection of the appropriate nominal size of chipping and the effect of flakiness on surface dressing design. 3.7 The most critical period for a surface dressing occurs immediately after the chippings have been spread on the binder film. At this stage the chippings have yet to become an interlocking mosaic and are held in place solely by the adhesion of the binder film. Dusty chippings can seriously impede adhesion and can cause immediate failure of the dressing. 3.8 The effect of dust can sometimes be mitigated by dampening them prior to spreading them on the road. The chippings dry out quickly in contact with the binder and, when a cutback bitumen or emulsion is used, good adhesion develops more rapidly than when the coating of dust is dry. 3.9 Most aggregates have a preferential attraction for water rather than for bitumen. Hence if heavy rain occurs within the first few hours when adhesion has not fully developed, loss of chippings under the action of traffic is possible. where wet weather damage is considered to be a severe risk, or the immersion tray test, described in Appendix C, shows that the chippings have poor affinity with bitumen, an adhesion agent should be used. An adhesion agent can be added to the binder or, used in a dilute solution to pre-coat the chippings. However, the additional cost of the adhesion agent will be wasted if proper care and attention is not given to all other aspects of the surface dressing process.

Chippings for surface dressings

3.1 The selection of chipping sizes is based on the volume of commercial vehicles having unladen weights of more than 1.5 tonnes and the hardness of the existing pavement. Ideally, chippings used for surface dressing should be single sized, cubical in shape, clean and free from dust, strong, durable, and not susceptible to polishing under the action of traffic. In practice the chippings available usually fall short of this ideal but it is recommended that chippings used for surface dressing should comply with the requirements of BS 63: Part 2 (1987) for the nominal size of chipping selected by the engineer. In this standard, some control of shape is ensured by the limits set for the flakiness index for each nominal size (except 6mm). Part of BS 63 is reproduced in Appendix A. 3.2 Samples of the chippings should be tested for grading, flakiness index, aggregate crushing value and, when appropriate, the polished stone value and aggregate abrasion value. Sampling and testing should be in accordance with the methods described in British Standard BS 812 (1985,1989a, 1989b 1990a, 1990b). 3.3 Specifications for maximum aggregate crushing value (ACV) for surface dressing chippings typically lie in the range 20 to 35. For lightly trafficked roads the higher value is likely to be adequate but on more heavily trafficked roads a maximum ACV of 20 is recommended. 3.4 The polished stone value (PSV) of the chippings is important if the primary purpose of the

3.10 Improved adhesion of chippings to the binder film can also be obtained by pre-treating the chippings before spreading. This is likely to be most beneficial if the available chippings are very dusty or poorly shaped, or if traffic conditions are severe. There are basically two ways of pretreating chippings: Spraying the chippings with a light application of creosote, diesel oil, or kerosene at ambient temperature (NAASRA, 1986). This can be conveniently done as the chippings are transferred from stockpile to gritting lorries by a belt conveyor or. alternatively, they can be mixed in a simple concrete mixer. Pre-coating the chippings with a thin coating of hard bitumen such that the chippings do not stick together and can flow freely.

before the use of an adhesion agent or pre-treated chippings is considered.

Bitumens

4.1 It is essential that good bonding is achieved between the surface dressing and the existing road surface. This means that non-bituminous materials must be primed before surface dressing is carried out.

Prime coats
4.2 Where a surface dressing is to be applied to a previously untreated road surface it is essential that the surface should be dry, clean and as dust-free as possible. On granular, cement or lime-stabilised surfaces a prime coat of bitumen ensures that these conditions are met. The functions of a prime coat can be summarised as follows. It assists in promoting and maintaining adhesion between the roadbase and a surface dressing by pre-coating the roadbase and penetrating surface voids. It helps to seal the surface pores in the roadbase thus reducing the absorption of the first spray of binder of the surface dressing. It helps to strengthen the roadbase near its surface by binding the finer particles of aggregate together. If the application of the surface dressing is delayed for some reason it provides the roadbase with a temporary protection against rainfall and light traffic until the surfacing can be laid.

3.11 Chippings which are pre-coated with bitumen enable the use of a harder grade of binder for construction which can provide early strong adhesion and thus help to obtain high quality dressings. The binder used for pre-coating need not necessarily be the same kind as that used for the surface dressing; for example, tar-coated chippings adhere well to a sprayed bitumen film. Pre-coating is usually undertaken in a hot-mix plant and the hardness of the coating, and thus the tendency for the chippings to adhere to each other, can be controlled by the mixing temperature and/or the duration of mixing; typical coating temperature are about 1400C for bitumen binders and 1200C for tar binders. Table 1 indicates the amount of binder recommended for lightly coating chippings. Table 1 Binder contents for lightly-coated chippings Target binder content (per cent by mass) Bitumen Tar (TRL, 1996) (TRL, 1992) 1.0 0.8 0.6 0.5 1.2 1.0 0.8 0.7

Nominal size of chippings (mm) 6 10 14 20

Reproduced from Road Note 39

3.12 Pre-coated chippings should not be used with emulsions because the breaking of the emulsion will be adversely affected. 3.13 In some countries adhesion agents or pre-treated chippings are often used in an attempt to counteract the adverse effect of some fundamental fault in the surface dressing operation. If loss of chippings has occurred, it is advisable to check whether the viscosity of the binder was appropriate for the ambient road temperature at the time of spraying. The effectiveness of the chipping and traffic control operations should also be reviewed

4.3 The depth of penetration of the prime should be between 3-l0mm and the quantity sprayed should be such that the surface is dry within a few hours. The correct viscosity and application rate are dependent primarily on the texture and density of the surface being primed. The application rate is. however, likely to lie within the range 0.3-1.1 kg/m2 . Low viscosity cutbacks are necessary for dense cement or limestabilised surfaces, and higher viscosity cutbacks for untreated coarse-textured surfaces. It is usually beneficial to spray the surface lightly with water before applying the prime coat as this helps to suppress dust and allows the primer to spread more easily over the surface and to penetrate. Bitumen emulsions are not suitable for priming as they tend to form a skin on the surface. 4.4 Low viscosity, medium curing cutback bitumens such as MC-30, MC-70, or in rare circumstances MC-250, can be used for prime coats (Asphalt Institute, 1983). The relationship between grade and viscosity for cutback primes is shown in Table 2.

Table 2 Kinematic viscosities of current cutback binders

Grade of cutback binder MC 250 MC 70 MC 30 ASTM D2027, 1998 Permitted viscosity range (centistokes at 60C) 250-500 70-140 30-60

the water evaporates. and particles of high viscosity bitumen adhere to the chippings and the road surface. 4.9 Depending upon availability and local conditions at the time of construction, the following types of bitumen are either commonly used in the tropics or are becoming so: Penetration grade. Cutback. Emulsion. Modified bitumens.

Bitumens for surface dressings

4.5 The correct choice of bitumen for surface dressing work is critical. The bitumen must fulfil a number of important requirements. They must: be capable of being sprayed; 'wet' the surface of the road in a continuous film; not run off a cambered road or form pools of binder in local depressions; 'wet' and adhere to the chippings at road temperature; be strong enough to resist traffic forces and hold the chippings at the highest prevailing ambient temperatures; remain flexible at the lowest ambient temperature, neither cracking nor becoming brittle enough to allow traffic to 'pick-off' the chippings; and resist premature weathering and hardening.

Penetration grade bitumens 4.10 Penetration grade bitumens vary between 80/ 100 to approximately 700 penetration. The softer penetration grade binders are usually produced at the refinery but can be made in the field by blending appropriate amounts of kerosene, diesel, or a blend of kerosene and diesel. With higher solvent contents the binder has too low a viscosity to be classed as being of penetration grade and is then referred to as a cutback bitumen which, for surface dressing work, is usually an MC or RC 3000 grade. In very rare circumstances a less viscous grade such as MC or RC 800 may be used if the pavement temperature is below 150C for long periods of the year. Bitumen emulsion 4.11 Cationic bitumen emulsion with a bitumen content of 70 to 75 per cent is recommended for most surface dressing work. This type of binder can be applied through whirling spray jets at a temperature between 70 and 850C and, once applied, it will break rapidly on contact with chippings of most mineral types. The cationic emulsifier is normally an antistripping agent and this ensures good initial bonding between chippings and the bitumen. 4.12 When high rates of spray are required, the road is on a gradient, or has considerable camber, the emulsion is likely to drain from the road or from high parts of the road surface before 'break' occurs. In these cases it may be possible to obtain a satisfactory result if the bitumen application is 'split', with a reduced initial rate of spray and a heavier application after the chippings have been applied. If the intention was to construct a single seal then the second application of binder will have to be covered with sand or quarry fines to prevent the binder adhering to roller and vehicle wheels. If a double dressing is being constructed then it should be possible to apply sufficient binder in the second spray to give the required total rate of spray for the finished dressing. 4.13 If split application of the binder is used care

4.6 Some of these requirements conflict. hence the optimum choice of binder involves a careful compromise. For example, the binder must be sufficiently fluid at road temperature to 'wet' the chippings whilst being sufficiently viscous to retain the chippings against the dislodging effect of vehicle tyres when traffic is first allowed to run on the new dressing. 4.7 Figure 2 shows the permissible range of binder viscosity for successful surface dressing at various road surface temperatures. In the tropics, daytime road temperatures typically lie between about 250C and 500C, normally being in the upper half of this range unless heavy rain is falling. For these temperatures the viscosity of the binder should lie between approximately l04 and 7 x l05 centistokes. At the lower road temperatures cutback grades of bitumen are most appropriate. whilst at higher road temperatures penetration grade bitumens can be used. 4.8 The temperature/viscosity relationships shown in Figure 2 do not apply to bitumen emulsions. These have a relatively low viscosity and 'wet' the chippings readily, after which the emulsion 'breaks',

Figure 2 Surface temperature/choice of binder for surface dressings

must be taken with the following: The rate of application of chippings must be correct so that there is a minimum of excess chippings. The second application of binder must be applied before traffic is allowed onto the dressing. For a single seal it will be necessary to apply grit or sand after the second application of binder. 4.15 In some areas of the world the range of binders available to the engineer is restricted. In this situation it may then be necessary to blend two grades together or to 'cut-back' a supplied grade with diesel oil or kerosene in order to obtain a binder with the required viscosity characteristics. Diesel oil, which is less volatile than kerosene and is generally more easily available, is preferable to kerosene for blending purposes. Only relatively small amounts of diesel oil or kerosene are required to modify a penetration grade bitumen such that its viscosity is suitable for surface dressing at road temperatures in the tropics. For example Figure 3 shows that, for the road temperatures prevailing during trials in Kenya, between 2 and 10 per cent of diesel oil was required to modify 80/100 pen bitumen to produce binders with viscosities within the recommended range for use (Figure 2). Figure 4 shows the temperature/viscosity relationships for five of the blends made for these trials.

Cutback bitumens
4.14 Except for very cold conditions, MC or RC 3000 grade cutback is normally the most fluid binder used for surface dressings. This grade of cutback is basically an 80/100 penetration grade bitumen blended with approximately 12 to 17 per cent of cutter.

Figure 3 Blending characteristics of 80/100 pen bitumen with diesel fuel

Figure 4 Viscosity/temperature relationships for blends of 80/100 pen bitumen with diesel fuel

4.16 The blending process is not difficult but it must be undertaken with great care by staff who are properly trained. A convenient method is to pump the required amount of cutter (e.g. diesel oil) into the distributor whilst simultaneously pumping in hot bitumen. Before pumping in the cutter, sufficient bitumen should be pumped into the distributor to enable the cutter to discharge below the surface of the bitumen. Because of the fire risk, all the burners

must be extinguished and naked lights and smoking prohibited during this operation. Polymer modified bitumens 4.17 Polymers can be used in surface dressing to modify penetration grade, cutback bitumens and emulsions. Usually these modified binders are used at locations where the road geometry, traffic

characteristics or the environment, dictate that the road surface experiences high stresses. Generally the purpose of the polymers is to reduce binder temperature susceptibility so that variation in viscosity over the ambient temperature range is as small as possible. Polymers can also improve the cohesive strength of the binder so that it is more able to retain chippings when under stress from the action of traffic. They also improve the early adhesive qualities of the binder allowing the road to be reopened to traffic earlier than may be the case with conventional unmodified binders. Other advantages claimed for modified binders are improved elasticity in bridging hairline cracks and overall improved durability. 4.18 Examples of polymers that may be used to modify bitumens are proprietary thermoplastic rubbers such as Styrene-Butadiene-Styrene (SBS), crumb rubber derived from waste car tyres and also glove rubber from domestic gloves. Latex rubber may also be used to modify emulsions. Binders of this type are best applied by distributors fitted with slotted jets of a suitable size. 4.19 Rubber modified bitumen may consist, typically, of a blend of 80/100 penetration grade bitumen and three per cent powdered rubber. Blending and digestion of the rubber with the penetration grade bitumen should be carried out by experienced personnel prior to loading into a distributor. This must be done in static tanks which incorporate integral motor driven paddles. The blending temperature is approximately 2000C. 4.20 Cationic emulsion can be modified in purpose made plant by the addition of three per cent latex rubber. One of the advantages of using emulsions is that they can be sprayed at much lower temperatures than penetration grade bitumens, which reduces the risk of partial degradation of the rubber which can occur at high spraying temperatures. 4.21 Bitumen modified with SBS exhibits thermoplastic qualities at high temperatures while having a rubbery nature at lower ambient temperatures. With three per cent of SBS, noticeable changes in binder viscosity and temperature susceptibility occur and good early adhesion of the chippings is achieved. SBS can be obtained in a carrier bitumen in blocks of approximately 20kg mass. The blocks can be blended, at a concentration recommended by the manufacturer, with 80/100 penetration binder in a distributor. In this procedure it is best to place half of the required polymer into the empty distributor, add hot bitumen from a main storage tank and then circulate the binder in the distributor tank. The remaining blocks are added after about 30 minutes and then about 2 hours is likely to be required to complete blending and heating of the modified binder. Every effort should be made to use the modified bitumen on the day it is blended.

Adhesion agents
4.22 Proprietary additives, known as adhesion agents. are available for adding to binders to help to minimise the damage to surface dressings that may occur in wet weather with some types of stone. When correctly used in the right proportions. these agents can enhance adhesion between the binder film and the chippings even though they may be wet. The effectiveness and the amount of an additive needed to provide satisfactory adhesion of the binder to the chippings in the presence of free water must be determined by tests such as the Immersion Tray Test which is described in Appendix C. 4.23 Fresh hydrated lime can also be used to enhance adhesion. It can be mixed with the binder in the distributor before spraying (slotted jets are probably best suited for this) or the chippings can be pre-coated with the lime just before use. by spraying with a lime slurry. The amount of lime to be blended with the bitumen should be determined in laboratory trials bitt approximately 12 per cent by mass of the bitumen will improve bitumen-aggregate adhesion and it should also improve the resistance of the bitumen to oxidative hardening (Dickinson,1984). 4.24 Cationic emulsions inherently contain an adhesion agent and lime should not be used with this type of binder.

Design

5.1 The key stages in the surface dressing design procedure are illustrated in Figure 5.

Existing site conditions

5.2 Selection of a suitable surface dressing system for a road and the nominal size of chippings to be used is based on the daily volume of commercial vehicles using each lane of the road and the hardness of the existing pavement surface. 5.3 With time, the action of traffic on a surface dressing gradually forces the chippings into the underlying surface, thus diminishing the surface texture. When the loss of surface texture reaches an unacceptable level a reseal will be required to restore skid resistance. The embedment process occurs more rapidly when the underlying road surface is softer, or when the volume of traffic, particularly of commercial vehicles, is high. Accordingly, larger chippings are required on soft surfaces or where traffic is heavy whilst small chippings are best for hard surfaces. For example, on a very soft surface carrying 1000 commercial vehicles per lane per day, 20mm chippings are appropriate, whilst on a very hard surface such as concrete, 6mm chippings would be the best choice.

10

Figure 5 Outline procedure for design of surface dressings

5.4 Guidance on the selection of chipping size for single surface dressings, relating the nominal size of chipping to the hardness of the underlying road surface and the weight of traffic expressed in terms of the number of commercial vehicles carried per lane per day. These recommendations are shown in Table 3. 5.5 Road surface hardness may be assessed by a simple penetration probe test (TRL. 1996). This test utilises a modified soil assessment cone penetrometer and is described briefly in Appendix D. Alternatively the hardness of the existing road surface may be made on the basis of judgement with the help of the definitions given in Table 4. 5.6 Although the recommendations for the selection of chipping size were developed for conditions in the United Kingdom they have been found to be applicable to roads in tropical and sub-tropical countries. 5.7 If larger sized chippings are used than is recommended in Table 3 then the necessary bitumen spray rate, required to hold the chippings in place, is likely to be underestimated by the design procedure

11

Table 3 Recommended nominal size of chippings (mm) Approximate number of commercial vehicles with an unladen weight greater than 1.5 tonnes currently carried per day in the design lane Type of surface Veryhard Hard Normal Soft Very soft 2000-4000 10 14 20 * * 1000-2000 10 14 14 20 * 200-1000 6 10 10 14 20 20-200 6 6 10 14 14 Less than 20 6 6 6 10 10

The size of chipping specified is related to the mid point of each lane traffic category. Lighter traffic conditions may, make the next smaller size of stone more appropriate.

Very particular care should be taken when using 20mm chippings to ensure that no loose chippings remain on the surface when the road is opened to unrestricted traffic as there is a high risk of windscreen breakage. * Unsuitable for surface dressing.

Table 4 Categories of road surface hardness Category Penetration1 of surface at 300C (mm) Very hard Hard Normal 0-2 2-5 5-8

Definition Concrete or very lean bituminous structures with dry stony surfaces. There would be negligible penetration of chippings under the heaviest traffic. Likely to be an asphalt surfacing which has aged for several years and is showing some cracking. Chippings will penetrate only slightly under heavy traffic. Typically, an existing surface dressing which has aged but retains a dark and slightly bitumen-rich appearance. Chippings will penetrate moderately under medium and heavy traffic. New asphalt surfacings or surface dressings which look bitumen-rich and have only slight surface texture. Surfaces into which chippings will penetrate considerably under medium and heavy traffic. Surfaces, usually a surface dressing which is very rich in binder and has virtually no surface texture. Even large chippings will be submerged under heavy traffic.

Soft

8-12

Very soft

>12

See Appendix D

described in Section 5. This is likely to result in the 'whip-off' of chippings by traffic early in the life of the dressing and also to have a significant effect on the long term durability of low volume roads. 5.8 In selecting the nominal size of chippings for double surface dressings, the size of chipping for the first layer should be selected on the basis of the hardness of the existing surface and the traffic category as indicated in Table 3. The nominal size of chipping selected for the second layer should preferably have an ALD of not more than half that of the chippings used in the first layer. This will promote good interlock between the layers. 5.9 In the case of a hard existing surface, where very little embedment of the first layer of chippings is possible, such as a newly constructed cement stabilised road base or a dense crushed rock base, a 'pad coat' of 6mm chippings should be applied first followed by l0mm or 14 mm chippings in the second

layer. The first layer of small chippings will adhere well to the hard surface and will provide a key' for the larger stone of the second dressing.

Selecting the binder

5.10 The selection of the appropriate binder for a surface dressing is usually constrained by the range of binders available from suppliers, although it is possible for the user to modify the viscosity of penetration grade and cutback binders to suit local conditions as described in paragraphs 4.14 to 4.16. 5.11 The factors to be taken into account in selecting an appropriate binder are: The road surface temperature at the time the surface dressing is undertaken. For penetration grade and cutback binders the viscosity of the binder should be between 1 04 and 7x 105 centistokes at the road surface temperature (see paragraphs 4.5 to 4.9).

12

The nature of the chippings. If dusty chippings are anticipated and no pre-treatment is planned, the viscosity of the binder used should be towards the lower end of the permissible range. if the binder selected is an emulsion it should be borne in mind that anionic emulsions may not adhere well to certain acidic aggregates such as granite and quartzite. The characteristics of the road site. Fluid binders such as emulsions are not suited to steep cross falls or gradients since they may drain off the road before 'breaking'. However, it may be possible to use a 'split application' of binder. The type of binder handling and spraying equipment available. The equipment must be capable of maintaining an adequate quantity of the selected binder at its appropriate spraying temperature and spraying it evenly at the required rate of spread. The available binders. There may be limited choice of binders but a balanced choice should be made where possible. Factors which may influence the final selection of a binder include cost, ease of use, flexibility with regard to adjusting binder viscosity on site and any influence on the quality of the finished dressing.

binder as conditions will permit. For instance, if the road temperature is 400C then for heavy traffic the chart in Figure 2 would suggest that MC 3000 would be only just viscous enough. 400/500 penetration grade bitumen would be on the limit of being too viscous, however, it would be preferable to cut-back the bitumen to a 500/600 penetration grade rather than use a MC3000 grade. If pre-coated chippings could be used then the use of a 400 penetration grade bitumen would be acceptable. 5.15 Penetration grade bitumens as hard as 80/100 are often used for surface dressing work when road temperatures are high. With such a high viscosity bitumen it is very important that the chippings are applied immediately after spraying and, to achieve this, the chipping spreader must follow closely behind the distributor. The construction team must be well organised and skilful. This type of binder will not be tolerant of delays in the application of the chippings nor of the use of dusty chippings. In either situation, early trafficking is very likely to dislodge chippings and seriously damage the seal. 5.16 The use of penetration grade binders in the range 80/100 to 400 is preferred to MC3000 wherever circumstances allow this. For high volume fast traffic, where very early adhesion of the chippings is essential, consideration should be given to the use of pre-coated chippings. This will allow the use of a more viscous binder for a given road temperature and will ensure that a strong early bonding of the chippings is obtained. A polymer modified or rubberised binder can also provide immediate strong adhesion. Alternatively, emulsions will provide good 'wetting' and early adhesion provided rainfall does not interfere with curing. 5.17 The most difficult situations occur when it is required to start work early in the day and temperatures are considerably lower than they will be in the afternoon. It may appear to be appropriate to use a cutback binder, such as MC3000, for the low road temperature but, by the afternoon, the seal is likely to be too 'soft'. In these situations it better to use a more viscous binder and keep the traffic off of the new seal until it has been rolled in the afternoon.

5.12 Consideration of these factors will usually narrow the choice of binder to one or two options. The final selection will be determined by other factors such as the past experience of the surface dressing team.

Choice of binder and timing of construction work

5.13 The choice of cutback grade or penetration grade bitumen for surface dressing work is largely controlled by road temperatures at and shortly after the time of construction. However, there are relative advantages and disadvantages associated with the use of penetration grade binders or cutback bitumen. 5.14 MC 3000 cutback binder typically contains 12 to 17 per cent of cutter. Under warm road conditions this makes the binder very tolerant of short delays in the application of chippings and of the use of moderately dusty chippings. It is therefore a good material to use for training new surface dressing teams and for use in areas where water for cleaning chippings is scarce. However, a substantial percentage of the cutter, especially if it is diesel, can remain in the seal for many months. If road temperatures increase soon after construction, it is likely that MC3000 will be found to be 'tender' and that the seal can be easily damaged. This should not be a problem for lightly trafficked roads and for new roads that are not opened to general traffic for several days after the surface dressing is constructed. If a road must be opened to fast high volume traffic within a few hours of construction then there will be considerable advantage in using as high a viscosity

Designing the surface dressing Basis for the design method

5.18 Having selected the nominal size of chipping and the type of binder to be used, the next step in the design of a surface dressing is to determine the rate of spread of the binder. In this respect the recommendations given in Road Note 39 (TRL, 1996) for conditions in the United Kingdom are not appropriate for most tropical or sub-tropical countries. Differences in climate, uniformity of road surfaces, the quality of aggregates, traffic characteristics and construction practice, necessitate a more general

13

approach to the determination of the rate of spread of the binder for application in tropical countries. 5.19 The method of surface dressing design put forward by Jackson (1963) is suitable for general application and trials undertaken by the TRL in Kenya (Hitch, 1981) indicate that with some minor modifications, it works well under a range of tropical and sub-tropical conditions. Accordingly this method is recommended as a good basis on which to develop national or regional standards for surface dressing design in tropical countries. 5.20 The Jackson method of design incorporates concepts first put forward by Hanson (1934) which relate the voids in a layer of chippings to the amount of binder necessary to hold the chippings in place. Hanson calculated that in a loose single layer of chippings, such as is spread for a surface dressing, the voids are initially about 50 per cent decreasing to about 30 per cent after rolling and subsequently to 20 per cent by the action of traffic. For best results, between 50 and 70 per cent of the voids in the compacted aggregate should be filled with binder.

Hence it is possible to calculate the amount of binder required to retain a layer of regular, cubical chippings of any size. However, in practice chippings are rarely the ideal cubical shape (especially when unsuitable crushing plant has been used) and this is why the ALD concept was originally introduced.

Determining the average least dimension of chippings

5.21 The ALD of chippings is a function of both the average size of the chippings, as determined by normal square mesh sieves, and the degree of flakiness. The ALD may be determined in two ways. Method A. A grading analysis is performed on a representative sample of the chippings in accordance with British Standard 812:1985. The sieve size through which 50 per cent of the chippings pass is determined (i.e. the median size'). The flakiness index is then also determined in accordance with British Standard 812:1985. The ALD of the chippings is then derived from the nomograph shown in Figure 6.

Figure 6 Determination of average least dimension

14

Method B. A representative sample of the chippings is carefully subdivided (in accordance with British Standard 812:1985) to give approximately 200 chippings. The least dimension of each chipping is measured manually and the mean value, or ALD, is calculated.

Determining the overall weighting factor

5.22 The ALD of the chippings is used with an overall weighting factor to determine the basic rate of spray of bitumen. The overall weighting factor F' is determined by adding together four factors that represent: the level of traffic, the condition of the existing road surface, the climate and the type of chippings that will be used. Factors appropriate to the site to be surface dressed are selected from Table 5. Table 5 Weighting factors for surface dressing design

5.24 The rating for the existing surface allows for the amount of binder which is required to fill the surface voids and which is therefore not available to contribute to the binder film that retains the chippings. If the existing surface of the road is rough, it should be rated as 'very lean bituminous' even if its overall colour is dark with bitumen. Similarly, when determining the rate of spread of binder for the second layer of a double surface dressing, the first layer should also be rated 'very lean bituminous'. 5.25 The Jackson method of determining the rate of spread of binder requires the estimation of traffic in terms of numbers of vehicles only. However, if the proportion of commercial vehicles in the traffic stream is high (say more than 20 per cent) the traffic factor selected should be for the next higher category of traffic than is indicated by the simple volume count.

Determining the basic bitumen spray rate

Description Factor 5.26 Using the ALD and 'F' values in equation 1 will give the required basic rate of spread of binder. R = 0.625+(F*0.023)+[0.0375+(F*0.0011)]ALD (1) +3 +1 0 -l -3 -5 Where F = Overall weighting factor the average least dimension of the chippings (mm) Basic rate of spread of bitumen (kg/m2)

Total traffic (all classes)

Very light Light Medium Medium-heavy Heavy Very heavy Vehicles/lane/day 0 - 50 50 - 250 250 - 500 500 - 1500 1500 - 3000 3000+

ALD = R =

Existing surface
Untreated or primed base Very lean bituminous Lean bituminous Average bituminous Very rich bituminous +6 +4 0 -1 -3 5.27 Alternatively, the two values can be used in the design chart given in Figure 7. The intercept between the appropriate factor line and the ALD line is located and the rate of spread of the binder is then read off directly at the bottom of the chart. The basic rate of spread of bitumen (R) is the mass of MC3000 binder per unit area on the road surface immediately after spraying. The relative density of MC3000 can be assumed to be 1.0 and the spread rate can therefore also be expressed in 1/m2, however, calibration of a distributor is easier to do by measuring spray rates in terms of mass.

Climatic conditions
Wet and cold Tropical (wet and hot) Temperate Semi-arid (hot and dry) Arid (very dry and very hot) +2 +1 0 -l -2

Type of chippings
Round/dusty Cubical Flaky (see Appendix A) Pre-coated +2 0 -2 -2

Spray rate adjustment factors

5.28 Research in Kenya (Hitch, 1981) and elsewhere, has indicated that best results will be obtained if the basic rate of spread of binder is adjusted to take account of traffic speed and road gradient as follows. For slow traffic or climbing grades with gradients steeper than 3 per cent, the basic rate of spread of binder should be reduced by approximately 10 per cent. For fast traffic or downgrades steeper than 3 per cent the basic rate of spread of binder should be increased by approximately 10 per cent.

5.23 For example, if flaky chippings (factor -2) are to be used at a road site carrying medium to heavy traffic (factor - 1) and which has a very rich bituminous surface (factor -3) in a wet tropical climate (factor +1) the overall weighting factor 'F' is: -2-1-3 + 1 = -5

15

Figure 7 Surface dressing design chart

5.29 The definition of traffic speed is not precise but is meant to differentiate between roads with a high proportion of heavy vehicles and those carrying mainly cars travelling at 80km/h or more. 5.30 The basic rate of spread of binder must also be modified to allow for the type of binder used. The following modifications are appropriate:

5.31 Suggested adjustment factors for different binders and different site conditions are given in Table 6. The adjustment factors reflect the amount of cutter used in the base 80/100 penetration grade bitumen but must be regarded as approximate values. 5.32 The amount of cutter required for 'on-site' blending should be determined in the laboratory by making viscosity tests on a range of blends of bitumen and cutter. Work at TRL (Hitch and Stewart, 1987) has shown that MC3000 can be made in the field by blending 90 penetration bitumen with 12 to 14 per cent by volume of a 3:1 mixture of kerosene and diesel. It is suggested that if there is significantly more than 14 per cent of cutter by volume then the spray rate should be adjusted to compensate for this. For binders which have been cutback at the refinery, the cutter content should be obtained from the manufacturer. 5.33 If a different grade of binder is required then the adjustment factor should reflect the different amount of cutter used. For instance, a 200 penetration

Penetration grade binders: decrease the rate of spread by 10 per cent. Cutback binders: for MC/RC 3000 no modification is required. (In the rare cases when cutbacks with lower viscosity are used the rate of spread should be increased to allow for the additional percentage of cutter used). Emulsion binders: multiply the rate of spread given in the chart by 90/bitumen content of the emulsion (per cent). This calculation includes a reduction of ten per cent for the residual penetration grade binder.

16

Table 6 Typical bitumen spray rate adjustment factors Basic spray rate from Figure 7 or equation 1 Flat terrain, moderate traffic speeds High speed traffic, down-hill grades >3% Low speed traffic, up-hill grades >3%

Binder grade

MC3000 300 pen 80/100 pen Emulsion1

1

R R R R

R R*0.95 R*0.9 R*(90/%binder)

R*l.1 R*1.05 R*0.99 R*(99/%binder)

R*0.9 R*0.86 R*0.81 R*(8l/%binder)

% binder is the percentage of bitumen in the emulsion. Table 7 Nominal size of chippings for different hardness of road surface No. of commercial 1 Vehicles/lane/day Category of road surface hardness Very hard Hard Normal Soft
1

binder may have 3 per cent cutter in it and therefore the spray rate is 103 per cent of the rate for a 80/100 penetration bitumen. Appendix E gives an example of the use of the design chart and adjustment factors.

20-100

<20

Adjusting rates of spray for maximum durability

5.34 The spray rate which will be arrived at after applying the adjustment factors in Table 6 will provide very good surface texture and use an 'economic' quantity of binder. However, because of the difficulties experienced in many countries in carrying out effective maintenance, there is considerable merit in sacrificing some surface texture for increased durability of the seal. For roads on flat terrain and carrying moderate to high speed traffic it is possible to increase the spray rates obtained by applying the factors given in Table 6 by approximately 8 per cent. The heavier spray rate may result in the surface having a 'bitumen-rich' appearance in the wheel paths of roads carrying appreciable volumes of traffic. However, the additional binder should not result in bleeding and it can still be expected that more surface texture will be retained than is usual in an asphalt concrete wearing course.

Nominal chipping size (mm) 6 6 10 14 6 6 6 10

Vehicles with an unladen weight greater than 1.5 tonnes

bitumen determined in the normal way will be too low to obtain good durability. Low volumes of traffic are also unlikely to cause the chippings to be 'rotated' into a tight matrix and this will result in the layer being of greater depth than the ALD of the chippings, which is assumed in the design process. It should therefore be safe to increase bitumen spray rates on low volume roads to compensate for the reduced embedment of 'oversize' chippings and the increased texture depth that results from less reorientation of the chippings under light traffic. 5.39 Ideally the ALD of the two aggregate sizes used in a double surface dressing should differ by at least a factor of two. If the ALD of the chippings in the second seal is more than half the ALD of the chippings in the first seal then the texture depth will be further increased and the capacity of the aggregate structure for bitumen will be increased. 5.40 It is suggested that on low volume roads the bitumen spray rates should be increased above the basic rate of spread of bitumen (see paragraphs 5.26 to 5.27) by up to the percentages given in Table 8. It is important that these increased spray rates are adjusted on the basis of trial sections and local experience.

Surface dressing design for low volume roads

5.35 If a low volume road, carrying less than about 100 vehicles per day, is surface dressed it is very important that the seal is designed to be as durable as possible to minimise the need for subsequent maintenance. 5.36 A double surface dressing should be used on new roadbases and the maximum durability of the seal can be obtained by using the heaviest application of bitumen which does not result in bleeding. 5.37 Where crushing facilities are put in place solely to produce chippings for a project, it will be important to maximise use of the crusher output. This will require the use of different combinations of chipping sizes and correspondingly different bitumen spray rates. The normally recommended sizes of chippings for different road hardness and low commercial traffic volumes are reproduced in Table 7. 5.38 It may be desirable to use chippings of a larger size than those recommended in Table 7 for reasons of economy. It is likely that the rate of application of

Spread rate of chippings

5.41 An estimate of the rate of application of the chippings assuming that the chippings have a loose density of 1.35Mg/m3, can be obtained from the following equation: Chipping application rate (kg/m2) = 1 .364*ALD

17

Table 8 Suggested maximum increases in bitumen spray rate for low volume roads
ALD of

chippings (mm) <20 15

3 20-100 10 <20 20

6 20-100 15 <20 30

>6 20-100 20

All traffic (vehicles/lane/day) Increase in bitumen spray rate (per cent)

5.42 The chipping application rate should be regarded as a rough guide only. It is useful in estimating the quantity of chippings that is required for a surface dressing project before crushing and stockpiling of the chippings is carried out. A better method of estimating the approximate application rate of the chippings is to spread a single layer of chippings taken from the stockpile on a tray of known area. The chippings are then weighed, the process repeated ten times with fresh chippings, and the mean value calculated. An additional ten per cent is allowed for whip off. Storage and handling losses must also be allowed for when stockpiling chippings. 5.43 The precise chipping application rate must be determined by observing on site whether any exposed binder remains after spreading the chippings, indicating too low a rate of application of chippings, or whether chippings are resting on top of each other, indicating too high an application rate. Best results are obtained when the chippings are tightly packed together, one layer thick. To achieve this. a slight excess of chippings must be applied. Some will be moved by the traffic and will tend to fill small areas where there are insufficient chippings. Too great an excess of chippings will increase the risk of whip-off and windscreen damage.

application is to use hand lances. if skilfully used, they can produce an acceptably uniform rate of spread but it is very difficult to achieve a specified rate of spread with them. They cannot therefore be recommended for other than small-scale work and limited maintenance operations. The use of either of these hand methods of binder application for larger scale work invariably results in waste of valuable binder and a poor quality surface dressing which will have a short 'life'. 6.3 The spreading of binder on a larger scale requires the use of a bulk binder distributor, which may be either a self propelled or a towed unit (British Standards BS 1707:1989, and BS 3136:Part 2:1972) 6.4 There are two basic types of bulk binder distributors, the pressurised tank, constant rate of spread, constant volume, and constant pressure machines.

Constant volume distributors

6.5 These distributors are fitted with positive displacement pumps, the output of which can be pre-set. All the binder delivered by the pump is fed to the spray-bar when spraying is in progress and there is no by-pass arrangement for recirculating binder to the tank. For a spray bar of given length and output, the rate of spread of binder on the road is inversely proportional to the forward road speed of the distributor. On most constant volume machines it is possible to preheat the spray bar by circulating hot binder to it before spraying commences but this facility is not available on all machines. 6.6 Constant volume distributors can spray a wide range of types of binder and they are quite common in tropical developing countries. Disadvantages of constant volume distributors are; Calibration involves three inter-related variables, i.e. the pump output, the road speed and the spray bar width; hence the calibration procedures need to be extensive if, for example, it is required to vary spray bar width to allow for different lane widths. However, some constant volume machines have a limited but useful degree of automatic control of bitumen pump speed to compensate for variation in road speed. The relative mechanical complexity of the machines means that they are not suitable for operation by partly skilled operators.

Plant and equipment

Methods of distributing binder 6.1 The success of a surface dressing is very dependent on the binder being applied uniformly at the correct rate of spread. The method adopted for distributing binder must therefore; be capable of spreading the binder uniformly and at the predetermined rate of spread; and be able to spray a large enough area in a working day to match the required surface dressing programme.

6.2 The use of hand-held containers such as watering cans, perforated buckets etc, has a place for minor works. Any type of binder from penetration grades to emulsion can be applied in this way but uniform spreading of predetermined amounts cannot be achieved by this method and hence it is not recommended for anything other than small-scale work. A rather more controllable method of hand

18

6.7 Most distributors manufactured in the USA are constant volume machines.

Table 9 Spraying temperatures for binders Whirling spray jets Slotted jets Min C Max C 40 55 80 105 120 50 70 90 115 130

Constant pressure distributors

6.8 In these machines a pump of adequate capacity delivers binder to the spray bar at a pre-set pressure. A relief valve regulates the pressure and permits binder to bypass the spray bar and return to the tank. The pressure in the spray bar is not affected by the number of jets in use, and hence re-calibration is not required when spray bar extensions are fitted or the number of jets are reduced. As with constant volume machines, the rate of spread of binder varies inversely with the road speed of the distributor. 6.9 Most distributors made in the UK are of the constant pressure type.

Cutback grades MC30 RC/MC70 RC/MC250 RC/MC800 RC/MC3000

Min C Max C 50 65 95 115 135 60 80 115 135 150

Penetration grades 400/500 160 280/320 165 180/200 170 80/100 180

170 175 190 200

140 150 155 165

150 160 165 175

Because of the flammable nature of the solvent used in RC-type cutbacks, application temperatures should be restricted to the lower parts of the ranges given above. It is essential to extinguish flames and prohibit smoking when heating, pumping or spraying all cutbacks. Fire extinguishers should always be readily at hand.

Principal components of binder distributors 2 6.10 Distributors spray the bitumen through a spray bar to which the binder is delivered by a pump. or under pressure, from a heated insulated storage tank. Brief descriptions of these principal components and general guidelines on their operation are given below. Manufacturers instruction manuals give detailed operating instructions for each model of distributor. These should be carefully followed and used to train operators so that they fully understand the principles and the correct method of operation of their distributor.

6.13 On constant pressure distributors a pressure gauge fitted to the spray bar registers the spraying pressure during spraying (though not when re-circulating only), and on some machines a temperature gauge is also fitted to the spray bar. 6.14 The uniformity of transverse distribution of a spray bar should be checked by the 'Depot tray test' at least once a year. This test is specified fully in BS 1707:1989 (1989) and is described in Appendix F. 6.15 Attention should also be paid to maintaining the correct height of the spray bar above the road. Whilst jets are positioned on the spray bar so that their sprays overlap to minimise the effect of variations in spray bar height on the uniformity of transverse distribution of binder. some adverse effects are likely if the spray bar is operated at an incorrect height. Slotted jets are more critical than whirling spray jets in this respect. 6.16 Since the spray of the last jet at each end of a spray bar is not overlapped by an adjacent spray the rate of spread of binder is less at the ends of a spray bar than along its length. For this reason, adjacent spraying runs of a distributor are normally overlapped. Some distributors are fitted with a larger jet at the end of the spray bar to compensate for this effect. The alternative practice of turning the last jet of a slotted jet spray bar at right angles is not recommended, nor is the practice of attempting to spray butt joints. This invariably results in narrow unsprayed strips between adjacent paths of the distributor. 6.17 To ensure satisfactory performance of the spray bar, strainers and in-line filters in the binder feed system must be cleaned regularly otherwise blocked jets will result. Before commencing

Spray bars and spray jets

6.11 There are basically two types of spray jets, slotted jets and whirling spray jets. Slotted jets are usually high output jets and are particularly suitable for spraying polymer modified binders or for grouting. However, some manufacturers can supply jets with a range of different slot widths or whirling spray jets for the same spray bar. Whirling spray jets are of lower output and have the advantage for normal surface dressing in that the forward speed of the distributor can be slower than when slotted jets are used. This can enable the speed to be controlled more easily and for the chipping operation to keep pace with the spraying. Higher bitumen temperatures are necessary when spraying with whirling jets and suitable spraying temperatures for both types of jets are given in Table 9. 6.12 The swirl chamber of whirling spray jets is enclosed in the spray bar so that the jets can be pre-heated effectively by circulating hot binder through the spray bar prior to spraying. The fine spray produced by whirling spray jets necessitates protecting the spray bar with a hood and canvas curtains to prevent wind from deflecting the spray. This is not required with slotted jets.

19

spraying, the spray bar and jets should be preheated by circulating hot binder and then the jets should be operated for a few seconds, discharging on to waste ground, to ensure they are operating freely. 6.18 If spraying is interrupted briefly. for example, to allow the chipping operation to catch up, the spray bar should be kept hot by circulating binder, preferably with the distributor standing off the road. When spraying is stopped for a longer period, such as at the end of the day or when the tank is being re-filled, the binder pump should be opened to air and the feed line to the spray bar, the spray bar itself, and the return line back to the tank emptied. The return valve should then be closed and the jets blown out with air. If the machine is being allowed to cool completely the binder pump should be flushed out with diesel fuel. Most spray bars are fitted with a drain cock so that binder or flushing oil can be drained off when required.

kerosene or diesel fuel which is usually drawn from the main fuel tank of the vehicle. Vaporising burners require the vaporising coil to be heated before they can operate, whilst atomising burners, which are preferable. can start up from cold. Fire extinguishers, suitable for fighting fires fuelled by bitumen or solvents, should be located in convenient positions. It is important that professional advice is obtained on fire-fighting matters well before work commences. 6.24 The burners in a distributor should be used to make only relatively small adjustments to the binder temperature. Wherever possible the main operation of heating the binder should be done in pre-heaters and the binder transferred to the distributor at or above the spraying temperature. 6.25 When heating binder in the tank it is necessary to ensure that the burner flues are fully covered by the binder, preferably with a depth of at least 150mm of binder over the top of the flues. On some distributors a danger level is indicated on the contents gauge. If this precaution is not observed the burner flues may burn out, causing a fire or explosion. 6.26 Burners must not be operated when the distributor is spraying or moving or if any blending is in progress. To prevent 'coking' of binder in the vicinity of the flues it is recommended that the binder is circulated when the burners are lit. This will also speed up the transfer of heat throughout the binder.

Binder pumps and air pumps

6.19 On most distributors the binder pump is driven by a separate engine, usually mounted either at the rear of the tank or between the tank and the driving cab. The pump itself is normally located inside the binder tank so that it is kept hot by the surrounding binder. The engine drive to the pump is usually through a clutch and the same engine usually drives a small air compressor which supplies air and fuel under pressure to the burners. 6.20 The binder tank should be emptied at the end of a day's work so that when the tank is next filled with hot binder there is no cold binder around the pump to prevent it from warming up quickly. If the binder system is not cleaned out as described above the pump will not work until it has been cleared of cold bitumen. This should be done by turning the engine crank manually as the bitumen in the distributor is heated and not by using the engine. 6.21 On some distributors the pump drives are taken either from the main power transmission of the vehicle or are driven by the main engine through a hydraulic system.

Distributor speed control and calibration 6.27 Most binder distributors are equipped with a 'fifth wheel' which operates a low range speedometer. The speedometer is located in the driver's cab in a prominent position so that a steady forward speed can be maintained relatively easily. 6.28 To spray binder at a specified rate of spread all that is necessary with constant pressure machines is to read off the corresponding road speed from the 'Driver's chart' or calibration chart which should be carried by every distributor. With constant volume machines it is necessary to select from the chart both the pump output and the road speed necessary to give the required rate of spread for the width of spray bar being used. 6.29 If the distributor has not previously been calibrated or if the calibration chart has been lost either of the following two methods can be used to calibrate the machine. Method A. This method is preferred for initial calibration. The distributor is loaded with binder which is raised to the correct spraying temperature and circulated around the spray bar to heat it. Static spraying is done into suitable containers to check the evenness of the appearance of the binder spray.

Tanks and burners

6.22 Most binder distributors have tanks with a capacity of between 500 and 16000 litres. The tanks are invariably made of steel and are lagged to reduce heat loss. Baffles are fitted internally to minimise surge. An inspection hatch fitted with a strainer basket provides access at the top of the tank, and a dipstick or contents gauge indicates the level of the binder. 6.23 Flues fitted with burners run through the tank to heat the binder and a thermometer is fitted to indicate the temperature. The burners use either

20

Binder is then sprayed into weighed containers of suitable dimensions for an accurately measured period of time and the mass of sprayed binder determined by weighing. The mass of binder delivered per unit time is calculated and the rate of spread/speed of distributor relationship is determined as described in Appendix G. Method B. Four or five weighed metal trays of known area (0.lm square is a suitable size) are placed in the path of the distributor as it makes a spraying run at a constant speed. The trays are then picked up and weighed and the rate of spread of binder is calculated. The process is repeated with different distributor speeds until the required rate of spread/speed chart can be drawn up. This tray test should be repeated periodically during surface dressing operations to check the consistency of the rate of spread of bitumen. It will, of course, be necessary to complete the dressing by hand on the areas where the trays were located. 6.30 Tar and bitumen binders have been found to have different outflow characteristics when sprayed from whirling spray jets; hence, if both kinds of binder are likely to be used, it is advisable to draw up a calibration chart for both binders for machines fitted with this type of jet.

rate of spread of the chippings is dependent on gravity and the speed of the tipper lorry acting independently, the skill of the lorry driver is crucial in ensuring an even distribution of the chippings. Nevertheless good results can be obtained with these simple machines. 6.34 However, to reduce dependence on the skill of the tipper driver, metering devices are available for tailboard chip spreaders that control the rate of discharge of the chippings by delivering them over a roller which is driven from the road wheels of the lorry or from a fifth wheel attached to the chip spreader. In this way variations in road speed of the tipper produce corresponding variations in the rate of discharge of the chippings. 6.35 Pushed metering chip spreaders operate on a similar principle but the metering roll is located at the base of a wheeled hopper which is pushed along the road by a reversing tipper lorry. The roll is driven by the road wheels of the hopper and the chippings in the hopper are replenished from the raised body of the tipper. 6.36 Self-propelled metering chip spreaders are the most effective machines available for applying chippings. They have a hopper at the rear into which chippings are discharged from the delivering tipper lorry which, during the transfer of the chippings, is towed along in reverse by the chip spreader through a quick release mechanism. Conveyor belts transfer the chippings to a transverse hopper at the front of the machine at the bottom of which is the metering roll that delivers the chippings to the road. However, there are selfpropelled models which do not meter the chippings but rely on gravity feed and these machines require careful operation to ensure that a constant road speed is maintained. 6.37 It should be noted that none of these chip spreaders can deliver chippings at a pre-determined rate of spread; they simply facilitate an even distribution of the chippings and the operator must ensure that an adequate, but not excessive, rate of application is maintained. 6.38 The number of tipper lorries must be sufficient to provide a steady supply of chippings at a rate that allows the planned daily output of the surface dressing unit to be achieved. Depending on the distance of the stockpile of chippings from the surface dressing site, a minimum of four or five tippers is usually required plus one spare tipper for applying chippings by hand to awkward shaped corners and other areas that may not have been covered by the chip spreader.

Chip spreaders 6.31 Chippings can be spread on the sprayed binder by hand and good results can be obtained by this method with a well-trained and plentiful labour force. In general, however, better results will be obtained when chippings are spread mechanically since this facilitates a more even distribution and rapid application of the chippings after the binder has been sprayed. 6.32 There are three main types of chip spreader; Metering or non-metering 'tail-board' types. Pushed metering chip spreaders. Self-propelled metering or non metering chip spreaders.

6.33 Non-metering tail board chip spreaders are bolted in place of the tailgate of a normal tipping lorry. They are the cheapest and simplest kind of mechanical chip spreader, having very few moving parts. A serrated steel comb controls the flow of chippings and a rotary gate with a helical edge controls the width of spread and the starting and stopping of the flow. The 'Hornsey gritter' is a popular example of this type. The flow of chippings is controlled by an operator who walks beside the tipper lorry, whilst it is driven in reverse at walking speed with the tipper body partly raised. Since the

Rollers and other equipment 6.39 The rolling of a surface dressing plays an important part in ensuring the retention of the chippings by assisting in the initial orientation and

21

bedding down of the chippings in the binder. Traditionally, steel-wheeled rollers have been used but these tend to crush weaker aggregates and to crack poorly shaped chippings. Accordingly, if steel-wheeled rollers are used they should not exceed 8 tonnes in weight and should only be used on chippings which are strong enough. Some steel-wheeled rollers are fitted with rubber sleeves which makes them more suitable for surface dressing work but, as for any roller of this type. they will bridge' depressions in the existing road surface. In general. pneumatic tyred rollers are preferred because the tyres have a kneading action which tends to manoeuvre the chippings into a tight mosaic without splitting them and they do not bridge' depressions. 6.40 In favourable conditions, adhesion should be well established within 30 minutes of rolling after which considerable benefit can be obtained by allowing slowmoving traffic, particularly heavy lorries, to traverse the dressing provided that traffic speed is kept below 20 to 30 km/hr. This is very important and the use of a lead vehicle to 'convoy' traffic at slow speed is recommended. 6.41 Other important items of equipment required for surface dressing are mechanical brooms, binder heaters, decanters and transporters, and front-end loaders. Mechanical brooms, either towed or powered, are invaluable for obtaining a clean road surface prior to spraying the binder. Whilst hand brooming is an alternative, it is difficult to obtain as good results by this method, particularly when sweeping the surface of a newly constructed roadbase from which all loose particles should be removed. 6.42 Binder heaters are required to raise bulk stocks of binder to the spraying temperature. They should have sufficient capacity to supply. at the correct temperature, all the binder required for the planned output of the surface dressing unit. Binder decanters fulfil the same function when the binder is supplied in drums. Their capacity tends to be small hence it is usually necessary to provide several decanters to supply the required quantity of hot binder. When binder is supplied in bulk it is desirable to transport it from the bulk supply point in binder transporters. It is not desirable to use distributors for this, nor should the burners in a distributor be used for raising the binder from pumping temperature, at which it is usually discharged at the supply point, to the spraying temperature. 6.43 Front-end loaders are required primarily for handling chippings. They are essential for loading tipping lorries quickly with chippings at the stockpile to ensure that a continuous supply of chippings is delivered to the surface dressing site. Front-end loaders are, of course, also useful for many general lifting duties such as lifting drums of binder onto decanters. In addition to these major items of

equipment a surface dressing unit requires a variety of small equipment such as road signs, hand tools etc. These are listed in Appendix H. 6.44 The routine maintenance and servicing of the mechanical equipment of a surface dressing unit usually has to be undertaken in the field, often remote from a base maintenance workshop. Machine operators do not normally have the skills to undertake running repairs or adjustments to their machines, nor are they usually authorised to do such work. There is thus a need to provide for a certain level of running repairs on site if long delays are to be avoided whilst plant fitters and spares are supplied from a distant maintenance workshop. The employment in the unit of a plant operator who is also a trained filler is therefore an advantage. together with the provision on site of a basic set of spares and the necessary tools.

The surface dressing process

Planning 7.1 A typical sequence of events in the planning of a surface dressing operation is as follows: 1 Select lengths of road requiring surface dressing and detail the preliminary work required on each road before the surface dressing can be carried out. Implement the necessary preliminary work such as patching. heating and planing, shoulder and edge repairs, drainage works. reinstatement of service trenches etc. Allow as much time as possible for trafficking before commencement of surface dressing operations. Decide on the type of surface dressing, the binder to be used and nominal chipping sizes, i.e. use the road surface hardness probe and make a preliminary design, taking into account constraints on the supplies of binder and chippings and limitations of plant and labour. Ensure that the chippings will be of adequate quality with suitable ALD value(s). Stockpile chippings at convenient points along the road to be surface dressed so as to minimise haul distances during construction. Sample the chippings and confirm their suitability, modify the surface dressing design if necessary. Order the appropriate binder for the anticipated weather conditions, or make provision for blending and adding adhesion agents as required. Ensure that all the plant and equipment of the unit is in good working order.

22

10 7 Instruct the construction team of the details of the work programme. Inform the police and other organisations likely to be affected by the surface dressing operation. Inform the Materials Laboratory and arrange for the testing of aggregates and binder and rate of spread checks during spraying. 11

The burners on the distributor are extinguished and the distributor is positioned at the beginning of the spray run. The driver adjusts the guide chain, the fifth wheel is lowered to the ground, and the height of the spray bar is adjusted. The distributor then commences the spraying run, the cut-off sheets being removed immediately the distributor passes to avoid contamination of the wheels of the chip spreader or tipper lorries. The chip spreader, tippers and the rollers should follow closely behind the distributor. Spraying should he stopped if the chipping operation is delayed for any reason. A strip of binder 150mm wide is left un-chipped at the edge of the lane to allow for the overlap of the adjacent run of the distributor. A tipper and crew should move slowly over the new dressing, spreading chippings by hand shovels on areas where there is a deficiency of chippings. The operation is then repeated on the adjacent pass (if any) and traffic is allowed to move slowly over the new dressing. The distributor then returns to the original level parking site and the volume of binder remaining in the tank is checked with the 'dip-stick'. The supervisor records the amount of binder used and, knowing the total area sprayed, calculates the average rate-of-spread. Speed control and other traffic warning signs are left in position along the length of the new surface dressing. At the completion of the day's work the distributor spray bar is cleaned, all vehicles and plant refuelled and lubricated and the supervisor checks that the bitumen heaters are loaded ready to supply the binder required for the next day.

12

The surface dressing operation 7.2 The following sequence of events normally comprises the complete surface dressing operation: 1 Raising the temperature of the binder in the depot tanks is started early in the morning so that the distributor can be loaded with bitumen. The temperature should, preferably, be just above the ideal spraying temperature.

13

14 2 The supervisor arrives on site with traffic control equipment and supervises the placing of warning signs, control barriers, traffic cones, etc. 15 3 The surface dressing unit arrives on site and the distributor is parked off the road, preferably on a level site where the tank can be 'dipped' before and after spraying. If the parking area is not level and alternative site must be located. Whilst the binder temperature is adjusted using the burners; binder is circulated through the spray bar, and the jets are checked for correct operation. 17 5 The calibrated 'dip-stick' supplied with the distributor is used to measure the volume of binder in the tank at the start of the day's work. The supervisor instructs the distributor crew on the spray rate required, the corresponding road speed and the pump output, where this is necessary. The chipping crew load the tipper lorries with chippings and the lorries line up ready to follow the distributor at the location specified by the supervisor. The rollers also prepare to follow the distributor after the chipping lorries. The road is thoroughly swept and road furniture such as manhole covers, reflective studs etc, is masked so as to prevent contamination with binder. Cut-off sheets of paper or other material are placed at the beginning and end of the spray run. The supervisor checks that the road is in fit condition for spraying and that laboratory staff, if present, are ready to do tray tests.

16

18

After-care 7.3 After-care is an essential part of the surfacing process and consists of removing excess chippings within 24 to 48 hours of the construction of a dressing. Some of the excess chippings will have been thrown clear by passing vehicles but some loose chippings will remain on the surface and these are a hazard to windscreens and, hence, a source of public complaint. They can be removed by brooming or by purposed-made suction cleaners. Care must be taken with brooming to avoid damage to the new dressing and it is usually best to do this work in the early

23

morning when the surface dressing binder is still relatively stiff. It is Important to stress that over-chipping can reduce the quality of a dressing, make after-care a more time consuming process and also unnecessarily increase costs.

early breaking characteristics are also advantageous when rainfall is likely to occur. Suitable specifications for slurry seals and for a Cape-seal are given in Tables 10 and 11. Table 10 Aggregate particle size distribution for slurry seals

Other surface treatments

Percentage by mass of total aggregate passing test sieve BS test sieve (mm) 10 5.0 2.36 1.18 0.6 0.3 0.15 0.075 Bitumen content (per cent by mass of dry aggregate) Fine 100 90-100 65-90 40-60 25-42 15-30 10-20 General 100 90-100 65-90 45-70 30-50 18-30 10-21 5-15 Coarse 100 70-90 45-70 28-50 19-34 12-25 7-18 5-15

8.1 Apart from surface dressing there are several other kinds of surface treatment that complement surface dressings, five of which are described briefly below. Slurry seals 8.2 A slurry seal is a mixture of fine aggregates, Portland cement filler, bitumen emulsion and additional water (ASTM, D 3910, 1996; BS 434, Parts 1 and 2, 1984). When freshly mixed they have a thick creamy consistency and can be spread to a thickness of 5 to 10 mm. This method of surfacing is not normally used for new construction because it is more expensive than surface dressing, does not provide as good a surface texture, and is not as durable as a properly designed and constructed surface dressing. Slurry mixes are best made and spread by purpose made machines as shown in Figure 8. 8.3 Slurry seals are often used in combination with a surface dressing to make a 'Cape-seal'. In this technique the slurry seal is applied on top of a single surface dressing to produce a surface texture which is less harsh than a surface dressing alone and a surface which is flexible and durable. However, the combination is more expensive than a double surface dressing and requires careful control during construction. 8.4 Both anionic and cationic emulsions may be used in slurry seals but cationic emulsion is normally used in slurries containing acidic aggregates, and its

10-16

7.5-13.5

6.5-12.0

The optimum mix design for the aggregate, filler, water and emulsion mixture should be determined using ASTM D 3910-84 (1996).

Table 11 Typical coverage for a new 'Cape seal' Size of chipping in surface dressing (mm) 20 14 10 Coverage (m2 /m3) 130-170 170-240 180-250

Figure 8 Slurry seal machine (diagrammatic)

24

Otta seal 8.5 An Otta seal is different to surface dressing in that a graded gravel or crushed aggregate containing all sizes, including filler, is used instead of single sized-chippings. There is no formal design procedure but recommendations based on case studies have been published (Norwegian Public Roads Administration, 1999). An Otta seal may be applied in a single or double layer. Evidence on the performance of these types of seal has shown them to have been satisfactory for over 12 years on roads carrying up to 300 vehicles per day (Overby, 1998). 8.6 The grading of the material is based on the level of traffic expected. Recommended grading envelopes are given in Table 12. Generally for roads carrying light traffic (<100 vehicles per day), a 'coarse' grading should be chosen while a 'dense' grading should be applied to one carrying greater than 100 vehicles per day.

8.9 An important aspect of Otta seal construction is the need for extensive rolling by pneumatic rollers for two or three days after construction. The action of rolling ensures the binder is forced upwards, coating the aggregate, and thereby initiating the process, continued by subsequent trafficking, of forming a premix like appearance to the surface. 8.10 After care can take as long as twelve days and involves sweeping dislodged aggregate back into the wheelpaths for further compaction by traffic.

Sand seals 8.11 Where chippings for a surface dressing are unobtainable or are very costly to provide, sand can be used as 'cover material' for a seal. Sand seals are less durable than surface dressings; the surface tends to abrade away under traffic. Nevertheless a sand seal can provide a satisfactory surfacing for lightly trafficked roads carrying less than 100 vehicles per lane per day. 8.12 It is not possible to design a sand seal in the same sense that a surface dressing can be designed. The particles of sand become submerged in the binder film, and the net result is a thin layer of sand-binder mixture adhering to the road surface. 8.13 The sand should be a clean coarse sand, with a maximum size of 6mm, containing no more than 15 per cent of material finer than 0.3 mm and a maximum of 2 per cent of material finer than 0.1 5mm. The sand should be applied at a rate of 6 to 7 x 10- m3 /m2 (CSRA, 1986). The binder, which may be a cutback or an emulsion, should be spread at a rate of approximately 1.0 to 1.2 kg/m2 depending on the type of surface being sealed. Synthetic aggregate and resin treatments 8.14 These treatments are costly and are used only on relatively small areas, usually in urban situations. where high skidding resistance is required. The aggregate is normally a small. single-sized, calcined bauxite which has a high resistance to polishing under traffic. The aggregate is held by a film of epoxy-resin binder (Denning, 1978). The process requires special mixing and laying equipment and is normally undertaken by specialist contractors.

Table 12 Otta seal aggregate grading requirements Percentage passing Sieve (mm) 19.0 16.0 12.0 9.5 4.750 2.360 1.180 0.600 0.300 0.150 0.075 Dense 100 79-100 61-100 42-100 19-68 8-51 6-40 3-30 2-21 1-16 0-10 Coarse 100 77-100 59-100 40-85 17-46 1-20 0-10 0-3 0-2 0-1 0-1

Aggregate should be screened to remove stone greater than 19mm

8.7 The viscosities of binders used in construction should reflect the quality of aggregate employed but normally cut back bitumen MC 800, MC 3000 or 150/200 penetration grade bitumen is used depending upon the traffic volumes and type of aggregate cover. Spray rates can not be calculated by design and must be chosen empirically. Typically, spray rates (hot) for single seals are between 1.6 and 2 1/m2 but reference must be made to the Design Guide (NPRA, 1999) so that necessary detailed adjustments can be made. 8.8 It is because of the broad range of materials that may be used and the empirical nature of the design of this type of seal that it is imperative that pre-construction trials be carried out. This strategy will identify any special local conditions concerning the available aggregates and binders to become apparent to enable the engineer to adjust the nominal design.

Applications of light bitumen sprays 8.15 There are two main uses for light sprays of bitumen: A light film of binder which can be applied as the final spray on a new surface dressing. The advantage of this procedure is that the risk of whip-off of chippings under fast traffic is reduced. This is particularly useful where management of traffic speed is difficult.

25

A light spray of binder can be used to extend the life of a bituminous surfacing. This is particularly useful where a surfacing is showing signs of bitumen ageing by fretting or cracking.

American Society for Testing and Materials (1996). D3910, Annual book of ASTM standards, Vol.4.03. Philadelphia. British Standards Institution (1972). Cold emulsion spraying machines for roads. Part 2, Metric Units. British Standard, BS 3136:Part 2:1972. (British Standards Institution, London). British Standards Institution (1984). Specification for bitumen road emulsions (anionic and cationic). British Standard 434:1984 (Parts 1 and 2). (British Standards Institution, London). British Standards Institution (1985). Testing aggregates; sieve tests. British Standard, BS 812: Part 103.1:1985. (British Standards Institution, London). British Standards Institution (1986). Specification for test sieves. British Standard 410:1976. (British Standards Institution, London). British Standards Institution (1987). Specification for single sized aggregates for surface dressing. British Standard, BS 63: Part 2:1987. (British Standards Institution, London). British Standards Institution (1989). Specification for hot binder distributors for road surface dressing. British Standard 1707:1989. (British Standards Institution, London). British Standards Institution (1989a). Testing aggregates: method for sampling. British Standard, BS 812: Part 102:1989. (British Standards Institution, London). British Standards Institution (1989b). Testing aggregates: method for determination of the polished-stone value. British Standard, BS 812: Part 114:1989. (British Standards Institution, London). British Standards Institution (1990a). Testing aggregates: method for determination of aggregate crushing value (ACV). British Standard, BS 812: Part 110:1990. (British Standards Institution, London). British Standards Institution (1990b). Testing aggregates: method for determination of aggregate abrasion value (AAV). British Standard, BS 812: Part 113:1990. (British Standards Institution, London). Committee for State Road Authorities (1972). TRH 7. The use of bitumen emulsions in the constriction and maintenance of roads. Technical Recommendations for Highways. Department of Transport, Pretoria. Committee for State Road Authorities (1986). TRH 3. Surfacing seals for rural and urban roads and compendium of design methods for surfacing seals used in the Republic of South Africa. Technical Recommendations for Highways. Department of Transport, Pretoria.

8.16 These applications may be referred to by different authorities as Fog Sprays or Enrichment sprays.

Fog sprays
8.17 A light spray of bitumen emulsion is ideal for improving early retention of chippings in a new dressing (CSRA, 1972). The road surface is usually dampened before spraying or, if a low bitumen content emulsion (45 per cent) is available, this dampening can be omitted. Complete breaking of the emulsion must occur before traffic is allowed onto the dressing and it may be necessary to dust the surface with sand or crusher fines to prevent pick-up by traffic. If emulsion is diluted with water, to obtain a 45 per cent bitumen content to ensure the bitumen will flow around the chippings, then the suitability of the water must be established by mixing small trial batches. 8.18 The spray rate for the diluted emulsion will depend upon the surface texture of the new dressing but the best results will be achieved if the residual bitumen in the fog spray is treated as part of the design spray rate for the surface dressing. The spray rate is likely to be between 0.4 and 0.8 litre/m2. It is important to avoid over application of bitumen which could result in poor skid resistance.

Enrichment sprays
8.19 Surfaces which are showing obvious signs of disintegration through bitumen ageing can be enriched by applying stable grade anionic bitumen emulsion which has been diluted at a rate of 1: 1 with water (CSRA, 1972). The rate of application will depend upon the texture of the surfacing and this must be determined by trial sprays, however, it is likely to be between 0.2 and 0.5 litres/m2 of residual bitumen. Great care must be taken to avoid leaving a slippery surface and a light application of sand sized fines may be require in some cases.

9 References
Asphalt Institute (1983). Specifications for paving and industrial asphalts. Specification Series No.2 (SS-2). College Park, Maryland (The Asphalt Institute). American Society for Testing and Materials (1998). D2097, Standard specification for cutback asphalt (Medium Curing Type). Annual Book of ASTM Standards, Vol.4.03. Philadelphia.

26

Denning J H (1978). Epoxy- resin/calcined bauxite surface dressing on A1, Sandy, Bedfordshire: skid resistance measurements 1968 to 1977. Laboratory Report LR867. TRL Limited, Crowthorne. Dickinson E J (1984). Bituminous roads in Australia. Australian Road Research Board, Vermont South, Victoria. Hanson F M (1934). Bituminous surface treatment of rural highways. Proc. New Zealand Soc. Civ. Engrs., Vol 21.1934/35 (New Zealand Society of Civil Engineers). Hitch L S (1981). Surface dressing in developing countries: research in Kenya. Laboratory Report LR 1019. TRL Limited, Crowthorne. Hitch L S and Stewart M (1987). The preparation of cutback bitumens to ASTM specifications by blending readily available constituents. Research Report RR 104. TRL Limited, Crowthorne. Jackson G P (1963). Surface dressing. Shell International Petroleum Co., London. National Association of Australian State Road Authorities (1986). Principles and practice of bituminous surfacing Vol 1: Sprayed work. (National Association of Australian State Road Authorities, Sydney). Norwegian Public Roads Administration (1999). A guide to the use of Otta Seals. Publication No.93. Directorate of Public Roads, Road Technology Department, International Division. Oslo. Overby C (1998). Otta seal - a durable and cost effective global solution for low volume sealed roads. 9th REAAA Conference, 'An International Focus on Roads: Strategies for the Future'. Wellington. Salt G F and Szatkowski W S (1973). A guide to levels of skidding resistance for roads. Laboratory Report LR510. TRL Limited, Crowthorne. Transport Research Laboratory (1992). Design guide for road surface dressings. Road Note 39 3rd edition. (TRL Limited, Crowthorne). Transport Research Laboratory (1996). Design guide for road surface dressings. Road Note 39 4th edition. (TRL Limited, Crowthorne).

27

Appendix A: Requirements for grading and particle shape (Reproduced from BS 63: Part 2: 1987)
Specified sizes for given nominal sizes are given in Table Al and the grading limits for each nominal size of aggregate are given in Table A2. Specified values which can be applied to roads carrying up to 250 vehicles per day are given in Table A3.

Table Al Specified sizes for given nominal sizes (mm) Specified size Nominal size 20 14 10 6
1

Table A3

Grading limits, specified size and maximum flakiness index for surface dressing aggregates for lightly trafficked roads Nominal size of aggregates (mm)

Passing BS test sieve 1 20 14 10 6.3

Retained on BS test sieve 14 10 6.3 3.35

Grading limits BS test sieve' 28 20 14 10 6.3 5.0 3.35 2.36 0.600 0.075

20 100 85-100 0-40 0-7 0-3 0-2 -

14 100 85-100 0-40 0-7 0-3 0-2 -

10 100 85-100 0-35 0-10 0-3 0-2 -

6.3 100 85-100 0-35 0-10 0-2 -

In accordance with BS410 (1986), specification for test sieves

Table A2 Grading limits, specified size and maximum flakiness index for surface dressing aggregates Nominal size of aggregates (mm) Grading limits BS test sieve1 28 20 14 10 6.3 5.0 3.35 2.36 0.600 0.075 20 100 85-100 0-35 0-7 0-2 0-1 14 10 6.3 100 85-100 0-35 0-10 0-2 0-1

Minimum percentage by mass retained on BS test sieve Specified size Specified size 60 60 65 65

100 85-100 100 0-35 85-100 0-7 0-35 0-10 0-2 0-2 0-1 0-1

Maximum flakiness index Maximum flakiness index 35

1

35

35

In accordance with BS 410 (1986), specification for test sieves

Minimum percentage by mass retained on BS test sieve Specified size Specified size 65 65 65 65

Maximum flakiness index Maximum flakiness index 25

1

25

25

In accordance with BS 410 (1986), Specification for test sieves

28

Appendix B:

Recommended polished stone values of chippings for roads in Britain

Table B1 Minimum PSV of chippings for roads in Britain

29

Appendix C:

The immersion tray test for determining the concentration of adhesion agent required
A tin lid approximately 135mm diameter, or other suitable tray, is covered with binder to give a film some l.5mm thick. Place at least 10 chippings which are damp, but not with shiny wet surfaces, in the film of binder at the 'design road temperature' and leave for 10 minutes. Then withdraw some of the chips to confirm coating. Add water to about half the depth of the remaining chippings at the chosen test temperature and leave for 10 minutes before withdrawing them and noting the degree of coating. If the coating is less than 90 per cent on any chipping then an adhesion agents should be tried. In this case different percentages of the adhesion agent are added to samples of the binder until 90100 per cent coverage is obtained, after soaking, on all chippings. If limestone chippings are available they will provide a good comparison of adhesion properties with the chippings to be used on site because limestone has good affinity with bitumen

The following test procedure has been has been included in Road Note 14 (1964) and editions of Road Note 39, Design Guide for Road Surface Dressing (in the UK) since at least 1964 and it has in the past been published unchanged in ORN3. The method is reproduced below and then suggestions are made which may help to make it more appropriate for tropical conditions. In this test a tin lid approximately 135mm diameter is covered with 15 to 20g of binder giving a film some l.5mm thick. When this has cooled to the test temperature* it is immersed in water also at test temperature to a depth of about 25mm. Nominal l4mm chippings are then applied by hand and lightly pressed in. At least six pieces of the aggregate are used. The chippings are left for 10 minutes and are then carefully removed from the binder film: the percentage of binder retained on the chippings is assessed visually. When testing an adhesion agent, a known quantity of agent is added to the binder and thoroughly stirred to ensure good dispersion. The procedure is then as outlined above. The test is repeated with varying concentrations of agent in the binder until the minimum concentration required to give satisfactory results has been found. The concentration normally falls in the range 0.5 to 2.5 per cent by mass of agent. The agent may be considered satisfactory for use on the road if, in the test, when the chippings are lifted from the binder film the faces which have been in contact with the film are all 90-100 per cent coated with binder. * The temperature of water and tray of binder in the above test should be the expected temperature of the road surface during the treatment. Where it is desired to compare the behaviour of different agents with a given stone and binder it is suggested that 200C should be used as the test temperature.

Suggested new procedure For tropical conditions the test bitumen should be of the grade to be used on site and it should be tested at appropriate site temperatures. Testing different adhesion agents at 200C is not practical if, for instance, hot conditions warrant the use of an 80/100 penetration grade bitumen. It is considered appropriate to test the adhesion agents at a temperature which relates to the design road temperature on which binder selection was based.

30

Appendix D:

The probe penetration for test for measuring road surface hardness
cross. The probe tip should not be centred on any large stones present in the road surface. Before each measurement the collar is slid down the probe rod until it is flush with the end of the probe. The probe is then centred on the measurement mark and a pressure of 35 kgf is applied for 10 seconds care being taken to keep the probe vertical. The probe is then lifted clear and the distance the collar has slid up the probe is recorded in millimetres. It sometimes occurs that the point selected for test is below the general level of the surrounding road surface. It is then necessary to deduct the measurement of the initial projection of the probe tip from the final figure. The road surface temperature should be measured at the same time that the probe is used and the tests should not be made when the surface temperature exceeds 350C. This will limit probe testing to the early morning in many tropical countries. The probe readings are corrected to a standard temperature of 300C using Figure D 1, and the mean of ten probe measurements is calculated and reported as the mean penetration at 300C. Categories of road surface hardness and the corresponding ranges of surface penetration values are shown in Table 4 (paragraph 5.5).

General description This test utilises a modified soil assessment cone penetrometer, originally designed by the UK Military Engineering Experimental Establishment for the assessment of in-situ soil strength. The standard cone normally used with this penetrometer is replaced by a 4mm diameter probe rod with a hemispherical tip made of hardened steel. The probe is forced into the road surface under a load of 35 kgf (343N) applied for 10 seconds and the depth of penetration is measured by a spring loaded collar that slides up the probe rod. The distance the collar has moved is measured with a modified dial gauge. The temperature of the road surface is recorded and a graphical method is used to correct the probe measurements to an equivalent value at a standard temperature of 300C.

Method of operation All measurements are made in the nearside wheel track of each traffic lane where maximum embedment of chippings can be expected. A minimum of ten measurements are required at each location. These should be evenly spaced along the road at intervals of 0.5m. any recently repaired or patched areas being ignored. For convenience the measurement points can be marked with a chalk

Figure D1 Graphical method for correcting measurements of road surface hardness to the standard test temperature of 30oC

31

Appendix E:

Example of a surface dressing design

Site description A two-lane trunk road at an altitude of approximately 1500m. Vehicle count averaged 3370 per day/lane (i.e. 'Heavy' rating). Bitumen to be used is 400 penetration grade (made by cutting back 80/100 pen bitumen with 6.7 per cent by mass (or approximately 7.5 per cent by volume) of a 3:1 mixture of kerosene and diesel. Design Traffic (Heavy) Existing surface (average bituminous) Chippings (cubical) Climate (hot/dry) Overall weighing factor (F) Factor -3 -1 0 -1 -5

Aggregate (nominal 19 mm) Median size (ie 50 per cent passing) Flakiness Index Average Least Dimension (from nomograph, Figure 6) l6mm 16 12

The determination of spread rates of 80/100 and 400 pen bitumen for an F factor of -5 and an ALD of 12 on a site where maximum durability is required are summarised in Table El.

Table El Determination of spread rates for 400 penetration grade bitumen Spread rates for penetration grade binders (kg/m2) 80/100 pen (RD *0.9) 0.87 0.78 0.95 400 pen (RD *0.9*1.067) 0.92 0.84 1.02

Type Terrain Flat Uphill grade>3% Downhill grade>3%

Basic spread rate R for MC3000 (from Fig. 7 or equation 1) (kg/m2) 0.89 0.89*0.9=0.80 0.89*1.1=0.98

For increased durability RD = (R*1.08) (kg/m2) 0.96 0.87 1.06

32

Appendix F:

Tests for uniformity of transverse distribution of binder (depot tray tests) (reproduced from British Standard 1707:1989)
Before each test the containers are examined for damage and replacements made if such damage is likely to affect the test. The trolley runs on steel rails fastened to the top of a 1500 litres catch tank, the rails being horizontal and parallel to the sides of the tank and sufficiently long to allow the trolley to lie clear of the spray before the test. The top rim of each container, when fitted on the trolley is parallel to the rails and the same distance below the nozzles or distributing gear as the road surface under normal working conditions. The distributor is backed into position with the spray bar over the catch tank, precautions being taken to see that the spraybar is horizontal and at right angles to the rails. The trolley and containers rest on the rails clear of the spray hood. A short preliminary spray is made to ensure that all nozzles are functioning and that the machine is otherwise in normal working condition. The trolley and containers are then pushed underneath the spray hood; spraying is commenced, and maintained for a period of time sufficient almost to fill the containers. The trolley is then withdrawn to the previous position. The depth of binder in each container is measured by dipping with a steel rule graduated in millimetres. Each container is dipped in the same position, a convenient place being some 300 mm from one end. Dipping takes place when the froth has settled.

General This standard includes requirements and tolerances for uniformity of distribution of binder across the surface being sprayed. A standard method for determining the transverse uniformity of distribution has been developed, the requirements of which are: The conditions prevailing during the test are comparable with those occurring during normal operations as regards: a b c d e Temperature of binder. Viscosity of binder. Height of distributor gear above the test surface. Pressure in the distribution system. Speed of operation of mechanical distributing gear, when applicable.

The test surface is divided into strips of equal width, usually 50 mm, the length of the strips being parallel to the direction of travel of the distributor. The test is so arranged that the distributor can operate for a sufficient period to obtain the normal working conditions, and when this has been achieved the test surface is exposed to the discharge for a suitable period. The amount of binder delivered on each 50 mm strip is then measured and the results expressed as a percentage deviation from the mean for all the 50mm units over the effective width. (The effective width is defined as the sprayed width less the 150 mm margin at each side.) The rests of the test are recorded in the form indicated in Figure Fl.

Depot tray test apparatus and procedure The apparatus consists of a wheeled trolley carrying a set of removable containers. Each container is 50 mm wide x 1000 mm long x 150 mm deep, made of 0.9 mm mild steel sheet, and of approximately 7 litres capacity. The containers extend to a width of 150mm greater than the full spray width of the distributor there being six containers per 300 mm of spray width. The rim of each container is lipped on one side in order that the containers will overlap and prevent binder escaping.

33

Figure F1 Typical results for uniformity of transverse distribution of binder. Reproduced from British Standard 1707:1970

34

Appendix G:

Bitumen distributor: rate of spread/speed calibration

The output of the spray bar will have been determined for the given bitumen, operating temperature and distributor settings. In the case of constant pressure distributors, the operating pressure must be fixed. Constant volume distributors require separate calibrations for different combinations of bitumen delivery pump speed and number of jets in operation. By varying one of these variable at a time calibration charts can be developed. Having completed the static calibration tests, so that the rate of delivery of bitumen is known, and determined the design spray rate the required road speed of the distributor can be calculated from equation G1. Speed of distributor = S/(R*W) Where S R W (G1)

is the rate of delivery (mass) from the spray bar in kg/s is the design spray rate (mass) in kg/m2 is the sprayed width in metres

The required speed will then be in m/s. The calibration should then be checked dynamically as construction work progresses as described in Method B in paragraph 6.29.

Appendix H:
1

Miscellaneous equipment required for a surface dressing unit

6 7 Sample cans and bags. Cut-off sheets of building paper or flattened drums for clean starting and finishing of spraying. Rotatherm' type thermometers. Cleaning materials for plant and personnel. A walking type distance measuring device.

Fuel and lubricant for servicing plant. (A purpose-built vehicle is often used and this is very effective). Temporary traffic warning signs.

2 3 4

8 Stop/go signs and warning flags. 9 Hand brooms, shovels, wheelbarrows (for chipping small areas) watering cans, hammers and chisels (for opening drums etc.). Masking tape for protecting road furniture and covering road markings. 10

11 First aid kit, including burn treatment supplies and supply of water 12 Fire extinguishers.

35

36

37

38

ISSN 0951-8797

OS-E

A Guide to surface dressing in tropical And sub-tropical countries

ORN 3 (2nd Edition)

Transport and Road Research Laboratory Overseas Unit Department of Transport Overseas Development Administration

Overseas Road Note 4

Field survey techniques and analysis for urban bus operators

Overseas Unit Transport and Road Research Laboratory Crowthorne Berkshire United Kingdom 1987

ACKNOWLEDGEMENTS
This note was drafted by Mr P R Fouracre, Dr A C Maunder, and Dr G D Jacobs, of the TRRL Overseas Unit (Unit Head: Mr J S Yerrell). It is largely based on research work undertaken with the support and co-operation of the Association of State Road Transport Undertakings, N. Delhi, India. The authors gratefully acknowledge the Advice and contributions of Dr P S Rana Traffic Development Advisor at Delhi Transport Cooperation, And Mr P R White, Senior Lecturer, Polytechnic of Central London.

OVERSEAS ROAD NOTES

Overseas Road Notes are prepared principally for road and Transport authorities in countries receiving technical Assistance from the British Government. A limited number Of copies are available to other organisations and to Individuals with an interest in roads overseas. Enquiries Should be made to: Transport and Road Research Laboratory Crowthorne, Berkshire RG11 6AU England Crown Copyright 1987 Limited extracts from the text may be reproduced, provided The source is acknowledged. For more extensive Reproduction please write to: Head of Overseas Unit, Transport & Road Research Laboratory.

ISSN 0951-8987

CONTENTS
1. Introduction Objectives The need for surveys Content and structure Information sources and performance indicators Data sources Maintenance records Traffic supervision Financial accounting Ticketing Performance measures Field surveys Purpose of surveys Bus loading surveys Journey time surveys Waiting times and frequencies Passenger interviews Household surveys Survey logistics, sampling and other considerations Practical applications Introduction Monitoring route performance - profitability Monitoring route performance - indicators Allocating buses between routes Fare levels and subsidies Appraising the development of new services Journey times and bus priority Concluding remarks References Appendix A: Demand elasticities References Appendix B: A simple cost model Appendix C: Examples of survey output Basic route characteristics Journey time components Passenger waiting times and bus frequencies Passenger characteristics Reference Appendix D: Passenger interview questionnaire Appendix E: Example of calculating a sample size for large populations Appendix F: Example of route screening in Delhi Appendix G: Standard pro-formas Page 1 1 1 1 2 2 2 2 2 3 3 5 5 6 8 12 13 15 15 17 17 17 17 17 17 20 21 21 21 21 23 24 24 24 27 28 28 30 32 32 34

2.

3.

4. 5.

6.

7.

8. 9.

10. 11. 12. 13.

1.

INTRODUCTION

CONTENT AND STRUCTURE

1.7 This guide is structured in two parts: the initial section examines in some detail the inadequacies of existing data sources and the need for appropriate performance and planning indicators; the following sections explain the task of collecting appropriate material, its analysis and presentation. The techniques and analysis employed should find wide application with bus operators throughout the Third World. 1.8 While the emphasis of this guide is on survey data for planning purposes, the role of other information sources is explained, and briefly commented on. This gives some context to the survey data, as well as drawing attention to the overall management information system required for monitoring service levels and long term planning.

OBJECTIVES
1.1 This guide explains how the quality of management information in the bus industry can be improved by means of field surveys. It further explains how the efficiency of public transport operations in towns and cities in Third World countries can be improved by the use of information collected from these surveys. 1.2 The guide is aimed primarily at the middle management of public transport operators and at those who have been delegated the responsibility of collecting relevant data.

THE NEED FOR SURVEYS

1.3 It is often argued that because demand for conventional stage-carnage services in Third World cities is presumed to be captive, operators need only concern themselves with getting the maximum use out of their vehicles while the demand side will take care of itself. Such an attitude ignores factors such as the need or pressure to introduce new modes of public transport (para transit types, for example) which may compete for custom; the growth in ownership of cheap personal transport (like cycles and motorcycles); the need for operators to present to Government (or other finance sources) requests for investment based on sound analysis of market prospects; the pressures which build up amongst users faced with consistently poor services. 1.4 It is important for an operator to be aware of the market structure and how users are likely to respond to fare changes, service changes and the like. The opinions and attitudes of users towards the service are rarely sought and neither is investigation made of how their demand is generated and how they choose their mode of conveyance. 1.5 Most urban bus operators in the developing world collect statistics for purposes of management accounting and control but these data sources are seldom adequate to throw light on the effectiveness of bus services in meeting demand. 1.6 Field surveys of bus operations and the use made of buses should be used to provide information for operators on; better use of existing resources in providing the bus service; more effective long term planning to meet future travel needs. 1

2. INFORMATION SOURCES AND PERFORMANCE INDICATORS

DATA SOURCES
2.1 Data concerning the performance of bus fleets usually comes from three main functions: engineering, traffic and accounts. Table 2.1 presents typical data sources and the information which is readily available from each.

gearbox, etc) and general 'cannibalisation' (making one serviceable vehicle out of two or more unserviceable vehicles) that it is difficult to say whether any individual vehicle maintains a unique identity which can be recorded in a log-book.

Traffic supervision
2.4 Traffic supervisory staff monitor the service to ensure that schedules are being maintained. There may be time keepers at terminals, roving inspectors (who, amongst other things, check on fare evasion), as well as depot staff who ensure that drivers and conductors report for duty and are allocated an appropriate vehicle which leaves the bus depot according to schedule.

Maintenance records
2.2 Information is usually kept in depots and/or central workshops which records maintenance, servicing and daily preparation performed on vehicles. Sometimes vehicle log-books are used to monitor the service record of a vehicle, recording maintenance and servicing together with vehicle kilometrage operated. 2.3 Vehicle log-books are often poorly completed and the information available is thus of dubious quality. Furthermore, vehicles can he so transformed through their lives by the replacement of major assemblies (engine, axles,

Financial accounting
2.5 The financial side of the organisation collects together all cost and revenue information in order to present both the profitability of the company and budget estimates for following years. This information source will contain information of both operating costs and capital costs (including capital structuring or sources of capital).

TABLE 2.1
CURRENT DATA SOURCES AND INFORMATION

Data source

Information Available

1.

Maintenance records

-rate of consumption of spare parts, fuel and tyres -vehicle availability - vehicle breakdowns and accidents. - crew availability - vehicle outshedding - schedules and trips operated - lost mileage - journey speeds of vehicles - daily vehicle utilisation - breakdowns and accidents. - total revenues and sources - total costs and cost components - trends in costs and revenues - unit prices of resources - rates of expenditure - staffing structure and norms. - number of fare-paying passengers carried - average passenger journey distance (lead) - average fare per passenger carried total earnings from fare paying passengers.

2.

Traffic supervision

3.

Financial accounts

4.

Ticketing

Ticketing
2.6 Table 2.1 indicates the information that can be obtained depending on the type of ticketing system used. Systems which provide hand cancelled tickets for each denomination will provide most of this information. Other systems can be used only to record passengers carried (tickets sold) and total revenue per conductorshift. There is unlikely to be a one to one correspondence between tickets sold and numbers of passengers carried. The sale and use of bus passes, through ticketing (where one ticket covers several bus journeys), free travel concessions and fare evasion all lead to ticket information underestimating total patronage. 2.7 The estimation of passenger lead (ie journey distance) is similarly affected by the structure of the fare scale. A flat fare yields no information about passenger journey distance, since everyone pays the same whatever the length of the journey. Clearly the more fare stages there are in a given route network, the easier it is to make a good estimate of passenger lead.

the effectiveness of the system and, in general, the data is too coarse because of the amount of averaging that occurs over the total network. 2.9 Clearly, to satisfy the needs of monitoring performance and of planning, there is a requirement to disaggregate information as much as possible and to seek relationships which explain the way in which output and costs respond to changes in inputs and operating environment. Disaggregating information to, say, the route level, obviously calls for the collection of more data. It would therefore seem reasonable to he selective in what is collected. 2.10 Screening indicators can be used to isolate poorly performing routes and, subsequently, more detailed surveys can be carried out to show how to improve the situation. 2.11 Table 2.3 shows the data needed for both monitoring operational performance at different levels of the organisation and for deriving relationships that can be used for use in policy formulation and planning procedures. Screening indicators for route performance might include, for example, the ratio of EPK:CPK, average route waiting times and average load factors (see para 5.4). 2.12 Information is needed not only in overall performance, but in how policy might be changed to induce better performance. Activities like staffing practices and arrangements, maintenance and scheduling procedures, fare setting and investment will be of primary concern. Profitability (or meeting predetermined financial targets which allow for subsidies) is an indicator of fleet performance. However for long term planning purposes the most critical aspects concern demand and how demand responds to changes in service and fare levels and how unit costs change, as the quality of service changes or as the network expands, (see Appendices A and B for details).

Performance measures
2.8 The most common performance measures used by bus operators are shown in Table 2.2. These ratios are, for the most part, easily determined from data which must necessarily be recorded in running the company. The values presented are often averaged over the fleet as a whole but may sometimes be disaggregated to depot level. They are usually produced daily. The ratios are most usefully employed if some critical values or benchmarks are set for each indicator which are targets set by management. However performance measures give little guidance on the root cause of poor performance and they give little help for long term planning purposes in that they have little information on the way in which output and costs respond to input and fare level changes. Few of the indicators measure

TABLE 2.2
OPERATIONAL RATIOS COMMONLY USED FOR MEASURING PERFORMANCE OF A BUS COMPANY

Fleet availability Vehicle utilisation Schedule out-turn Staff productivity EPK CPK Passengers carried Load factof Breakdown rate Accident rate

- number of vehicles outshed as a proportion of total Fleet stock - average daily km per vehicle operated - proportion of schedules operated - number of staff per schedule or per bus - earnings per km - operating cost per km - either absolute or per bus or per bus km - total passenger km/total seat km - per million vehicle km - per million vehicle km

TABLE 2.3 DATA NEEDS Monitoring/Planning Activity Data needs

1. Route performance

- load factor - passenger lead - passenger throughput - fare revenues - use of bus passes - route costs - waiting times - journey speeds/times - causes for delay. - vehicle availability/utilisation - vehicle breakdowns/accidents - rate of consumption of resources. - profitability - load factor - passenger throughput - passenger lead - use of bus passes - fare leakage - vehicle availability - vehicle output - proportion of schedules missed - regularity and punctuality - vehicle breakdowns/accidents. - all the above items - vehicle handling characteristics (passenger throughput, penalty times, boarding and alighting times) - passenger travel patterns and use of buses - demand elasticities - opinions of service - staff time use - vehicle time use - staff knowledge and training needs - productivity trends - scale economies.

2. Depot performance

3. Fleet performance

4. Policy planning and service changes

3.

FIELD SURVEYS

PURPOSE OF SURVEYS
3.1 Existing data sources are unlikely to contain sufficient information on items 1 and 4 of Table 2.3 ie route performance, policy planning and service changes. In these cases field surveys will be needed to provide additional information. Table 3.1 lists some of the types of survey that are recommended and the information that they can be expected to produce. 3.2 The type of information yielded refers to broad area of interest such as system effectiveness and the demand for bus
TABLE 3.1

services. Some surveys wilt yield information in more than one area of interest. For example, a loading survey will provide data on both system effectiveness as well as passenger handling capabilities. 3.3 Surveys involving interviews are primarily concerned with users obtaining information on their travel habits and opinions of the service. The type of information required determines whether surveys should be conducted on or off the bus. Continuous surveys involve monitoring on a regular basis and will normally be carried out on the basis of a sampling framework. 3.4 In the following sections, these surveys are described in more detail in terms of manpower requirement,

SURVEY TYPES

Information 1. Loading surveys System effectiveness Vehicle performance

Method In-vehicle, continuous, by observation

Output Vehicle load patterns Av. load factors Av. passenger lead Passenger throughput Vehicle handling capability Fare revenues/leakage Use of bus passes Journey speeds/time Boarding/alighting Times Journey speeds Causes of delay Penalty times. User waiting times Passenger arrival patterns Bus arrival patterns Boarding/alighting times

2. Journey time and penalty time surveys

Vehicle performance

In and off-vehicle, continuous or ad-hoc, by observation Off-vehicle, continuous or ad-hoc, by observation

3. Waiting times and bus frequencies

System effectiveness Demand Vehicle performance

4. Passenger interviews

Demand System effectiveness

On or off-vehicle Ad-hoc interviews

Travel patterns and use of buses Estimates wait times and travel Times Opinions of service Demand for transport Modal choice criteria. Staff time use Vehicle use. Knowledge and training needs.

5. Household surveys

Demand System effectiveness Staff/vehicle performance

Off-vehicle Ad-hoc by interview On or off-vehicle Ad-hoc by observation Off-vehicle Ad-hoc interviews On/off-vehicle by observation

6. Time and motion surveys 7. Staff interviews

Staff performance

8. Boarding/alighting

Vehicle effectiveness

Time boarding/alighting.

questionnaire or pro-forma design and output. These descriptions are based on manual recording and data storage techniques. More sophisticated techniques for directly recording information in machine readable format are rapidly becoming available and should be considered where resources permit. 3.5 The overall logistics of the field surveys are considered in Section 4 and practical applications of the survey results

are contained in Section 5. In order to illustrate the output expected, an example is presented in Appendix C.

BUS LOADING SURVEYS

3.6 The main purpose of the bus loading survey is to determine passenger load patterns on vehicles and routes. Other information on revenue generation, the use of bus

Fig 3.1 Pro-forma for bus loading survey

passes, journey speeds and boarding/alighting times are also determined from this type of survey. 3.7 A bus loading survey requires two survey assistants per monitored bus trip. They sit in the vehicle, one at the entrance and the other at the exit. Where the vehicle has only one exit/entry door, two survey assistants are still required to cope with data collection. The pro-forma for data collection is shown in Fig 3.1, completed for a typical bus trip. Times are recorded with stop watches (if available) or wrist-watches, which have been synchronised. 3.8 At the start of each bus trip the following are recorded: day and date of survey; route number; journey direction; vehicle type; start time.

recorded from the conductor's waybill. The conductor should also be able to provide information on the number of bus passes produced by travellers, particularly if he has been prompted by the survey assistants to make a note of this. 3.10 During the terminal turn-round period, the survey assistants can undertake preliminary analysis such as calculating bus stand times and bus loadings along the route. The latter is determined by comparison of individual observations of number of passengers boarding and alighting along the route. 3.11 Apart from information on trip times and speeds, trip revenues and total number of passengers carried. it is also possible to determine the average passenger trip length (or lead) and the vehicle load factor for the trip. The lead is determined from a histogram of passenger loadings along the route. Fig 3.2 presents the loading pattern for the trip information contained in Fig 3.1, where each bar represents the number of passengers on the bus at any point along the route. The height of each bar is proportional to the number of passengers on each link (ie section of the route), and the width proportional to the link distance. The shaded area of the histogram is equal to the total number of passenger-km undertaken on the bus trip, each bar representing a certain number of passengers who have travelled the link distance; the summation of all these bars is the total passenger-km for the bus trip. The average passenger lead is the total passenger-km divided by the total number of passengers carried.

3.9 Each survey assistant records the times of arrival and departure from each bus stand at which the vehicle stops. The man at the entry door records the number of passengers boarding, while the man at the exit door notes the number of passengers alighting. During the trip, any undue delay (caused by congestion, accident, etc) can be noted in the remarks column, together with any action taken by traffic staff as a result. At the trip destination the finish time is noted so that total trip time can be determined. Information on trip revenue (from the number of tickets sold by denomination) can also be

Fig. 3.2 Histogram of number of passengers on the bus at any point along the route 7

3.12 For accuracy, the distances between bus stands should be known. Where this is not known, one approximation is to assume that bus stands are spaced equally along the route, with the inter-bus stand distance equal to the route length divided by the number of stands less one. Thus the passenger lead for a trip is given by the expression:
sum of (passengers on the bus on each link x link distance) total number of passengers using the bus.

(see below) are required to assess the importance of different journey time components, and the factors which affect them. The analysis of journey time data which is available from loading surveys is included in the next section.

JOURNEY TIME SURVEYS

3.16 There are three broad components of bus journey time:

3.13 From Fig 3.2 the total passenger kilometrage was 559 while from Fig 3.1 the number of passengers carried (ie who boarded the vehicle) was 103. Thus the average passenger lead on this trip was 5.4 km. 3.14 The load factor relates passenger km to seat km, the latter being the route length times the number of seats (or rated bus capacity) on offer. For the trip illustrated the load factor was 1.2 or 120 per cent. 3.15 The timing information contained on the data sheets (shown in Fig 3.1) can be used in a number of ways. For time and motion studies it is possible to assess the amount of staff time which is spent actively, ie steering time (when the vehicle is being productively driven) as a proportion of total time. It is also possible to extract information on passenger boarding and alighting times, together with journey times and speeds. The variability in journey times can be determined, though relating this to specific causes may not be possible from the loading survey. Journey time surveys

free running time; bus stand delay; other delays which can be subdivided into a) stop or stationary delays; b) general delays.

3.17 Stop or stationary delays include delay at traffic signals whereas general delays do not involve stops but take account of periods when speeds are below those which might reasonably be achieved on open roads. General delay is due to such factors as prevailing traffic or weather conditions. 3.18 Each bus stand or stop delay involves two separate elements: the time when the vehicle is stopped and the so called penalty time which is incurred due to deceleration from and acceleration to the optimum travel speed. 3.19 The objective of journey time surveys is to identify the importance of the component parts of travel time and to identify possible causal relationships. Fig 3.3 illustrates how

Fig. 3.3 Examples of a breakdown of bus running times

components of bus running time can be analysed and presented. Detailed analysis of this sort is useful when considering such things as new bus interior designs or other changes which might affect vehicle passenger handling characteristics. The data would also be useful to support a case for traffic management measures to improve bus journey times. 3.20 Fig 3.3 shows that in this example, bus stand

delays represent about 25 per cent of journey time, whilst other delays (stop and general) account for between 2030 per cent of journey time. Traffic management techniques for improving bus running speeds can therefore be expected to reduce journey times by up to 20-30 per cent. As a proportion of total journey times total delay ranges from 38-54 per cent with some evidence suggesting it is higher in the peak period.

Fig. 3.4 Pro-forma used to carry out bus journey time surveys 9

3.21 In its simplest form, the journey time survey is undertaken by one survey assistant per bus trip. He is located near the driver where he can observe traffic conditions. A pro-forma like that shown in Fig 3.4 is used, together with a stop-watch. The survey assistant records the

day, date, route number, direction of trip and vehicle type before the trip starts. He then records the start time, followed by all further relevant timings. when the vehicle slows to a speed slower than walking pace (as estimated by the survey assistant) but does not actually stop; when the

Fig. 3.5a,b Proformas used for carrying out penalty time surveys

10

vehicle picks up a speed faster than walking pace again, when a vehicle actually stops and starts; trip finish time. The causes of each delay are denoted by one of the codes 1 to 8. It will be seen that stop delays are recorded separately from general delays (which do not involve stops). Delay locations are noted so that congestion points can be identified. Stop or stationary delay is the delay to vehicles caused by stops other than at bus stands. It is defined as the extra time taken by delayed buses to travel between bus stands over and above that taken by undelayed buses. The measurement of time is that from when a bus first stops in a queue to when it clears the area of congestion (taken to be when the vehicle is moving faster than walking pace). Stop delays are the addition of stop times and the penalty time incurred as a result of slowing down. 3.22 Penalty times are monitored off the vehicle at bus stands. The location of the survey site for penalty times should offer a flat gradient, good line of sight for the survey assistants and traffic conditions which are free of congestion. Three survey assistants are required, one located 200 metres before the stand, one at the stand and one 200 metres beyond the stand. The person at the bus stand has a pro-forma like that shown in Fig 3.5(a), on which he records the arrival and departure times of each bus which stops at the stand. A description of each bus is also kept: vehicle type, registration number, number of exit! entry doors. Timings are made using a stop-watch, if available. The other two assistants use stop-watches which are synchronised with that of the assistant located at the bus stand. They use the pro-forma shown in Fig 3.5(b) to monitor the exact time that every bus passes them, the bus being identified by its registration number. Vehicles that do not stop at the stand do not have to decelerate or accelerate and therefore travel the distance of 400 metres (between the two assistants located 200 metres either side of the stand) at a much faster speed than vehicles which do stop (ignoring the actual stop time). The penalty time associated with stopping vehicles can be determined by comparing the travel times of non-stop and stopping vehicles over the 400 metre section, allowing for the actual stop time of vehicles which do stop. These times are determined by comparing the information from the three observers: vehicle numbers are matched, stopping vehicles appearing on all three proformas, non-stopping vehicles only on the pro-formas of the two outer observers.

3.24 The penalty time information is used to provide additional information to that obtained during the journey time survey. It provides additional material on the time losses caused by stops and starts which the survey assistant sitting on a bus would be unable to monitor on his own. 3.25 Bus stand delay (which includes both penalty time and the time when the vehicle is not in motion) is dependent on a number of factors which include vehicle design, driver and passenger behaviour and numbers waiting to board or alight. The latter information will have been collected as part of the loading survey (see above) or the waiting time frequency survey (see below). For predictive purposes a statistical test known as linear regression* can be used to determine relationships between times of boarding and alighting, and the numbers of passengers involved, for given vehicle types. The bus stand stop time has two elements which are the dead time (time between bus coming to a standstill and persons boarding and/or alighting) and the actual boarding and/or alighting time (total time taken by those boarding and/or alighting).

3.26 The dead time should be independent of the number boarding and alighting. For single entrance vehicles the boarding and alighting times are additive, ie those boarding must wait till those alighting have finished. The total stop time will thus be dependent on the total numbers boarding and alighting. The marginal alighting time (the time each passenger takes to alight) is likely to be smaller than the marginal boarding time (the time that each passenger takes to board), and therefore total stop time will vary with the proportion of alighting to boarding passengers. Total stop time for single entrance buses can be represented by: Y = C + aA + bB where Y is the total time, C is the dead time, A and B are the numbers alighting and boarding and a and b the marginal alighting and boarding times.

3.27 For double door buses boarding and alighting take place independently of one another. At any stand, the stop time will be the result of dead time plus the boarding or alighting time, whichever is greater. Total boarding and alighting times can be represented separately by:

3.23 An analysis pro-forma is shown in Fig3.5(c). Average values of penalty time should be determined for different vehicle types, if relevant. From a series of surveys in Delhi the weighted average penalty time value was 13 seconds which is high in comparison to observations in the UK where values of 9 seconds have been monitored. The high penalty time may be connected with extreme overloading experienced in single deck operations or the lower performance of some technologies in use in India when the surveys were implemented.

YA = C1 + a1A YB = C2 + b1B where YA and YB are the total alighting and boarding times respectively.

*(Linear regression is a statistical technique which seeks to derive an equation which best explains any relationship between two or more variables)

11

(c)
Fig. 3.5c Pro-forma for analysis of penalty time survey

SURVEYS OF PASSENGER WAITING TIMES AND BUS FREQUENCIES

3.28 The purpose of these surveys is to measure user waiting times and the reliability of the bus service, both key components of the overall level of service effectiveness. The same surveys can also be used to monitor boarding and alighting times, as well as passenger arrival patterns. 3.29 Waiting times can be monitored using a 'Q' Enquiry card as shown in Fig 3.6. Two (or three) survey assistants are

required who are located at a bus stand. One (or two) is required to issue the cards to passengers arriving at the bus stand, while the other assistant monitors the arrival and departure times of buses at the stand. A 'Q' Enquiry card is handed to each passenger as he arrives at the bus stand. (Where demand is heavy only a sample of passengers may be selected, say every other one, or every third one to arrive.) The survey assistant completes a few of the details before handing out the card: day, date, intended route number, direction of the bus and, most important, arrival time (of the passenger) at the bus stand. Additional

12

Fig. 3.6 'Q' Enquiry card information on fare to be paid, use of bus pass and destination can also be sought, though this is not essential to the main purpose of the survey. The passenger is instructed to return the card to the survey assistant when his (the passengers') bus arrives. At this point the survey assistant notes the arrival time of the bus on the card, and the waiting time is the difference between passenger arrival and bus arrival time. Where possible a note is made of the number of buses that arrived which the passenger could not board (either the bus did not stop or the vehicle was too full to board). This can be obtained from the assistant who is monitoring bus arrivals, or by asking the passenger. 3.30 Problems sometimes arise in the retrieval of Q Enquiry cards, especially when large numbers of passengers try to board a vehicle. It is advisable in these circumstances to have one assistant stationed at the entrance to the bus checking passengers one by one. 3.31 Data collected from the waiting time surveys tends to be more accurate than using passengers' own estimates which can be greatly exaggerated and hence unreliable as an indicator of service reliability. 3.32 The form used for monitoring bus arrival times at bus stands is shown in Fig 3.7. A record is kept of the route number, vehicle registration number, and whether the vehicle stopped or not. The numbers of passengers boarding should also be noted, since this will help verify the sampling rate, if adopted. Stop watches should be used if available, synchronised between all three assistants. 3.33 record Both survey types enable the operator to keep a of the effectiveness of the system overall (and specifically certain routes) and the level of service reliability. Clearly such surveys need to be undertaken frequently throughout the network so that any deterioration can be spotted quickly and investigated. For the existing (and potential) passenger a high service frequency and hence minimal waiting time are key elements as to whether frequent use is made of the service.

PASSENGER INTERVIEWS
3.34 These are undertaken in order to obtain several Sorts of information concerning patterns of travel and use of public transport, as well as attitudes to and opinions of the service. The nature of the survey will strongly influence the way in which it is carried out. Where the information to be obtained is fairly simple, a single card may be sufficient for recording purposes. Information like origin and destination, route number, time of day and journey purpose could be recorded quickly by a survey assistant for all passengers using a bus. (It would be helpful to hand such cards out to passengers for completion, but it cannot be assumed that either all passengers have a pen or that all passengers are literate.) Such a survey might be useful where, for example, the bus company wanted to find out more about the use of bus passes, or the extent to which passengers have to make interchanges in their trips, or passengers' estimates of waiting times. 3.35 Where several sets of information of this type are being sought then the questionnaire becomes more lengthy and complex. The interview may take some time to

13

Fig. 3.7 Pro-forma for bus arrival time survey complete, and it becomes necessary to sample from those using the bus. Processing of the data gathered also becomes more complex, and sorting and tabulating data may best be undertaken using computers, if available. 3.36 The advantage of the in-depth passenger interview is that it is possible to establish something of the travellers'

social and economic background, his travel characteristics (when using a bus) and the market structure which is currently being met. It is also possible to seek users' views of the service and their opinions as to how improvements could be made, often a useful public relations exercise in itself. Socio-economic data of passengers, in relation to data on their level and frequency of trip making enable the

14

operator to build up considerable information on his actual (and potential) market. This is often crucial when holding discussions with the relevant authorities on the setting of fare levels, amount of subsidy required etc. In addition the data also allows the operator to locate and plan services for certain sectors of the population (or potential markets) specifically catering to their needs eg standing only, express buses, limited stop. seating only, air conditioned services etc. 3.37 Interviews can be undertaken both in or off the vehicle. However, for convenience, the more complex questionnaires are probably best completed off the bus, at bus stands. In this case interviews can take place at a sample of bus stands (perhaps the busiest), and a sample of passengers (say every fifth one) interviewed. 3.38 Appendix D shows an example of an in-depth questionnaire used by the Overseas Unit TRRL to determine the travel and socio-economic characteristics of users of public transport in a number of Third World Cities. Questions were framed to determine the extent to which bus passes were in use, the number of bus interchanges being made, and other details of the trip including waiting and walking times.

4.

SURVEY LOGISTICS, SAMPLING AND OTHER CONSIDERATIONS

4.1 The manpower requirement for surveys depends on the type of survey being undertaken, its duration and extent, and the work-rate of survey assistants. The latter can normally be expected to work an 8 hour shift, the same as the bus operating staff. A bus may typically be operated for two shifts (morning and evening) and, consequently, if it is proposed to monitor a complete days operations, two shifts or survey assistants will be required per bus-day. Table 4.1 gives the manpower requirements for the main surveys in terms of the number of man-days per survey working day.
TABLE 4.1 MANPOWER REQUIREMENTS

Man-days expended per survey working day* Loading surveys Journey times Waiting times/bus frequencies Passenger interviews *per bus or per observation point 4 2 4-6 4

HOUSEHOLD SURVEYS
3.39 A disadvantage of the passenger interview survey is that it provides information only on those who use the busservice, ie the current market. Nothing is learned about potential users and/or users of competing modes. A more comprehensive understanding of the total demand for transport and how this varies with city structure and affluence will help the operator in planning future investments. A household survey should go a long way to meet this requirement. However, it is unlikely that bus operators would become involved in undertaking household surveys on a regular basis. These surveys are usually carried out on a large scale at some considerable cost by urban authorities to assess the potential for city development and/or transport in general. What should be of interest to operators, however, is some of the output from household surveys regarding modal choice and travel patterns associated with the different modes of transport.

4.2 In addition to the actual survey manpower requirement there is also the effort required for pre-analysis of the data eg coding (if necessary), sorting, tabulating and presenting material. For every five man days of data collection, one to two man days of manual, analytical effort is probably required. 4.3 From the above figures it is clear that surveys involve considerable manpower requirements. Some form of sampling is required to keep the surveys within manageable limits. It is also wise to have a programme of surveys mapped out well in advance, with the aim being: to keep the work load fairly constant; to move survey assistants around to relieve boredom; to be prepared (in terms of preparation of pro-formas and location of survey points); to provide for a mixture of continuous and temporary survey work.

4.44 The sampling rate may well be dictated by the manpower available for survey work. The danger is that . the4 sample might be so small as to produce unacceptable levels of accuracy. Appendix Fwell be dictated by the of The sampling rate may illustrates an example manpower available for thed of the standard deviation of the distribution of use sample means (standard error) in calculating a sample e size for large populations. As the example shows, some g degree of r e e o 15

15 The sampling

accuracy may have to be lost to keep manpower requirements to an acceptable level. The next problem concerns the sample population and the need to select representative or random samples. The sample population could be the complete network or organisation, a regional area (associated, say, with one depot) or a route. One bus operator in the UK for example samples from the whole network in order to estimate information on passenger loadings, the use of bus passes, etc (CIPFA, 1979). Samples can be drawn from crew duties, having further subdivided these by day of the week, type of duty, garage and type of operation (whether one-man-operated or not). If the survey is continuous, and over a long period of time (several months) information can be built up on individual routes.

route performance and output. Whatever technique is used for sampling some account of these patterns is necessary when trying to understand data recorded at different times of the year. Obvious distortions due to festivals and other similar events must be avoided by undertaking surveys outside such periods.

4.5 An operator might prefer to rotate his survey team around each route in turn, ensuring that each route is surveyed for a complete days' operations. Where there are a large number of buses employed on one route it may not be possible to survey that route in one day using available manpower; it might take as much as four days to cover all the duties being operated. Although this provides a great deal of detailed information route by route, it may take some time to get a total picture of the network as a whole. For example, if each route occupies the whole of the survey team for one working week then clearly only 50 routes could be surveyed in a year. For large networks routes would be covered only once every two or three years. In this case it may be necessary to sample from all the bus duties associated with each route, thus completing each route survey in only one or at most two days. 4.6 Seasonal variations are likely to be influential in
TABLE 4.2

4.7 The organisation of surveys is clearly quite complex and forward programming is required, taking account of data needs, priorities and resources available. It is suggested that it would not be unreasonable for a bus operator to spend one per cent of total revenues on planning activities, including both short term monitoring and the development of long-term policies. Not all information need be collected frequently and Table 4.2 sets out a possible timetable for the main data requirements. Some information is specific to a route, and some is of a more general character related to the total network. Some surveys, specific to the monitoring of a particular service change, might be carried out infrequently, but on a 'before and after' basis.

4.8 Apart from the programming of surveys, management will also be concerned with the control of staff. Surveys undertaken off the vehicle are more easily controlled because staff are not constantly on the move. Spot checks are necessary, not only to ensure the work is being undertaken in a professional manner, but also to answer queries and to give some moral support in what can be a tiring job. Needless to say, survey assistants should be thoroughly familiar with the work to be done prior to the start; trial runs can provide useful experience for both staff and management.

FREQUENCY OF SURVEYS

16

5. PRACTICAL APPLICATIONS
INTRODUCTION
5.1 As noted earlier, it is in the areas of route performance, policy planning and service changes that current data resources are usually inadequate, and for which special field surveys are necessary. This section is addressed specifically to these topics to show how the information gathered from the surveys, described in Section 3. can be used for better, more informed decision making.

levels are most important. Route profitability can be measured by comparing earnings per Km (EPK) with operating cost per Km (CPK), (ensuring that the data is as near as possible relevant to the route in question). Service level to passengers has many aspects. Perhaps the most easily measured are bus frequency, headway, regularity and punctuality, but waiting times and load factors are also indicators of service levels. It is suggested that the three indicators, ratio EPK to CPK, average route waiting time and average load factor, could provide a useful screening process for route performance. 5.5 Table 5.2 illustrates a possible screening procedure using three ratios and shows how possible improvements could be made to bus operations on the different routes. In all cases the average wait times and load factors on the route maybe high because of poor regularity. Regularity might be assessed by relating average wait times to scheduled headways or expected wait times. Some low-demand routes may also inevitably have high wait times because of low frequency of buses. This must also be taken into account where necessary. To make the screening process as realistic as possible the peak and off-peak operating performance should be separately assessed. Appendix F illustrates the route screening procedure using values of specified performance criteria for five routes operated by the DTC in Delhi.

Monitoring route performance profitability

5.2 Using some of the basic measurements derived from a loading survey it is possible to estimate route revenues. Using a simple cost model (described in Appendix B) it is possible to estimate route costs. Hence, from a knowledge of route costs and revenue it is possible to estimate route profitability. 5.3 Table 5.1 presents the estimated costs and revenues associated with five routes operated in a major Third World city (see Appendix C). The extra buses used on routes 80 and 720 during peak hour operations incur higher costs for the reasons described in Appendix B. Any positive difference in revenues and costs is profit for that route. (More precisely the profit is in fact the contribution to the fixed overheads of the total network, which are not accounted for in the above costs). The average costs and revenues show each route to have been loss-making, given the particular design of each route and the numbers of buses being operated.
TABLE 5.1 EXAMPLE OF ROUTE COSTS AND REVENUES, DELHI

Allocating buses between routes

5.6 To maximise profits (or minimise losses) for a given level of operations, an operator would ideally like to switch buses between routes, such that if there is a net gain in demand (or revenue) through switching a bus from one route to another, then, assuming no change in costs, the move would increase profits. (There could well be changes in costs associated with moving a bus from one route to another, and these would have to be off-set against the change in profits to assess whether the move is worthwhile.) As an example, the estimated revenues and costs of Routes 155 and 430 (described in Appendix C) are compared in Table 5.3. 5.7 Overall, Table 5.3 shows that the ten buses on Route 430 are more profitable than the ten buses on Route 155. However, if a choice had to be made between operating a tenth bus on either route, the bus operator should logically choose to put it on Route 155. On this route the tenth bus looses only Rs 196 per day as against Rs 316 per day on Route 430. Expressed in a different way, for the tenth bus each rupee of net cost on Route 155 generates 44 passenger km, while each rupee of net cost on Route 430 generates only 15 passenger km.

Route number 80
Cost per km (Rs)

89 2.86 2.86 2.31 1.61 1.89

155 1.88 1.88 1.77 1.46 1.60

430 1.80 1.80 2.07 1.41 1.61

720 1.85 2.42 2.03 1.35 1.19 1.29

Normal duties Extra buses All buses

Revenue per km (Rs)

1.97 2.94 2.09 2.23 1.74 1.94

Peak time Off-peak All buses

Monitoring route performance - indicators

5.4 From the operators' view, profitability is most important, while from the users' view adequate service

Fare levels and subsidies

5.8 Comparison of the effects of a fares increase with improvements to the service highlights some important

17

TABLE 5.2 ROUTE PERFORMANCE SCREENING PROCEDURE

issues. Small improvements in service levels often provide large increases in demand while large increases in fares cause little loss in demand. This suggests that fares could be raised quite substantially with the expected loss in demand being more than easily compensated by increases in service levels, ie there would often appear to be great scope for increasing fares and using additional revenues generated to expand the service and reduce the need for operating loans. At the same time the probability is that there would be no net loss in demand. 5.9 This type of analysis can also be used to assess the effects of subsidies. For example, an operator who is receiving subsidy may be meeting a demand of 10 passenger-km for every rupee of net cost (or subsidy). If a change in the service gives a higher level of passenger handling per rupee spent than this then it is worth undertaking, because for the same financial loss more demand can be met. Changing the level of subsidy however 18

alters the comparison. If more subsidy is given, it could be used to either expand the service and/or reduce fares. In both cases the amount of extra demand carried per rupee spent is likely to be lower than previously. Which course of action to follow may be pre-determined by the political process of giving subsidy, but given the choice the bus company would ideally use the extra subsidy on the scheme which goes nearest to meeting company objectives (say that which brings in most additional demand per rupee spent). This would set the level against which to compare all other possible schemes. This could be called the 'norm'. 5.10 Apart from changes in subsidy level (or financial target), changes in productivity which affect costs will also have an effect on the value of the norm. Improved productivity will reduce the net loss (or increase net gain, if appropriate) which is equivalent to a reduction in subsidy, thus increasing the value of the norm. There is a very real danger for bus operators that worsening productivity will

TABLE 5.3 COMPARABLE PROFITABILITY OF TWO ROUTES, (COSTS GIVEN IN RUPEES)

Route 155 Buses Passenger km(00) 157 155 150 148 145 125 119 102 98 86 Revenue Rs 452 446 432 426 418 360 343 294 282 248 3701 Cost Rs 444 444 444 444 444 444 444 444 444 444 4440 Contribution Rs 8 2 -12 -18 -26 -84 -101 -150 -162 -196 -739 Passenger km(00) 187 186 185 181 170 164 122 91 59 47 Revenue Rs 589 586 583 570 536 517 384 287 186 148 4386

Route 430 Cost Rs 464 464 464 464 464 464 464 464 464 464 4640 Contribution Rs 125 122 119 106 72 53 -80 -177 -278 -316 -252

1 2 3 4 5 6 7 8 9 10 Total

attract subsidy which is not used for either service improvements or fare reductions. 5.11 A thorough analysis of an operator's market will indicate differences in demand on different types of route, as well as between different times of day and between different journey lengths that passengers make. The latter would be of particular relevance when examining fare structures and the expected revenues that alternatives would yield.

should be monitored for this purpose. These routes should be representative of all other routes where this service is being provided so that actual service performance rather than individual route performance is being assessed. 5.13 One example of a specialised service is the railway special operated by the DTC in Delhi (Maunder and Fouracre, 1983). The railway special services were introduced so that rail passengers could be provided with direct routes from the main railway stations to various residential areas of Delhi at a reasonable fare (by comparison to taxi services). The charge imposed in 1980 was a flat rate of Re 1(50 paise for children) as compared to the average fare on ordinary DTC services of about 40 paise and a typical taxi fare of Rs 15. 5.14 During February, 1980 two railway special routes were monitored over a four day period to obtain the operational data presented in Table 5.4. Data for the whole DTC network for the year 1980/81 is also shown.
TABLE 5.4

Appraising the development of new services

5.12 It may be the bus companies policy to treat sections of the travelling public differently, perhaps, with a view to providing specialised services. Market surveys should be undertaken before introducing such services, in order to estimate their usefulness. Where these services are already in operation, the operator should check their performance to see if they are meeting their objectives and whether any modifications are required. Individual routes on which the special service is being provided

COMPARATIVE EARNINGS ON RAILWAY SPECIAL AND ORDINARY SERVICES, DELHI

Route

Av. bus load per trip

Av. fare (paise)

Revenue per trip (Rs)

Revenue per bus per day (Rs) 396 480

EPK (paise)

Est. load factor

Railway Sp. 2 Railway Sp. 3

33 41

99 99

33 40

143 153

0.43 0.50

Total DTC network

91

42

38

460

203

0.80

19

5.15 Despite lower load factors on the railway services, the earnings per bus/day are of the same order as those for the total DTC network. This is because of the higher fare levels and also better vehicle utilisation of the railway special service. Unfortunately, for the operator, this higher output involves additional (variable) costs and the railway special service was not attracting sufficient additional revenue to cover this extra cost. 5.16 Surveys of users are very appropriate in assessing how well a specialised service is meeting its objectives. An extensive survey by the Overseas Unit TRRL was undertaken to see whether the service was being used for the purposes for which it had been designed, ie to provide a special service for those carrying luggage to or from the railway station. On the two routes monitored only about 20 per cent of passengers were using the railway special in this way; over half the passengers were travelling to or from work. Furthermore, these passengers were undertaking the journey by the rail special on a frequent basis, often daily, although not necessarily in both directions. 5.17 When asked why they used the railway special few users referred to the specific purpose of the service; they seemed to value things like comfort (seating only) and convenience. (Perhaps this explains the fact that the service was clearly being used by commuters, to and from work, rather than the intended market, those travelling to and from the railway station.) 5.18 Faced with evidence like this the operator might well question the value of railway specials. However, it

would appear to demonstrate a demand for more specialised commuter services such as a seating only high-fare service on high demand corridors.

Journey times and bus priority

5.19 Monitoring the causes of bus journey delays can indicate specific bottlenecks, places where priority for buses could improve journey time and/or service reliability. For example bus lanes were introduced in Bangkok in 1980. Surveys carried out on six different sections, before and after the event, showed that in almost all cases, either bus travel times, or car travel times, or both, were improved significantly (Marler, 1982). The most successful section showed improvements to both bus and car mean travel times of 25-30 per cent Figure 5.1 shows the change in travel time distribution for a particular bus lane introduced in Bangkok. It can be seen that average bus travel times were reduced by 27 per cent and journey time variability improved considerably. 5.20 It is unlikely in any city that the introduction of a bus priority system will be the responsibility of the operator. This is usually carried out by the City Traffic Engineers Department. However, the bus operator by means of journey time surveys can indicate to the Traffic Engineers Department places where bus lanes etc could most sensibly be located.

Fig. 5.1 Bus travel time changes, Paholyothin Road

20

6.

CONCLUDING REMARKS

7.

APPENDIX A DEMAND ELASTICITIES

6.1 This note has examined the purpose, logistics and implementation of field surveys designed to improve management information on bus service performance. Practical examples of the analysis of survey findings have also been presented to demonstrate how this information can be positively used. 6.2 Prevailing operating conditions, available resources, size of operations and company objectives vary considerably between operators. As a result, management information systems and requirements are likely to differ considerably. This note has presented a range of practical options which can be developed by an operator to meet his specific needs. 6.3 The control and development planning of bus operations should be based on sound quantitative data of both the efficiency and effectiveness of the service. The use of some, or all, of the techniques described in this note will greatly contribute to this management process, to the general benefit of the urban transport sector of the Third World.

7.1 The demand for services is usually measured in terms of passengers or passenger kms per unit of time (eg per peak hour, per day, per annum). Total demand will be affected by such factors as city size and land use, per capita incomes, vehicle ownership levels, fares on competing modes, and service levels. From the planning point of view it is important to know how demand varies with these (and other) factors. The measure of response in demand to any one of these factors is called the demand elasticity with respect to that particular factor. If demand elasticities can be established with any confidence they are then extremely useful in the planning process. 7.2 The elasticity is the ratio of the percentage change in demand to the corresponding percentage change in the factor being considered: e = (y/y)/(x/x) where x represents a factor which influences demand (an independent variable) x is a small increase in that factor, y is the demand level associated with x and y is the change in demand resulting from x. 7.3 There is little documented evidence on the way in which demand for public transport in Third World cities responds to changes in fare and service levels. What little data there is tends to correspond with the findings of the more voluminous research undertaken in the highly industrialised nations. Until more studies are undertaken in the developing world it would seem appropriate to make use of this material. Table Al contains such estimates of elasticity values, together with values for two Third World cities. 7.4 Fare elasticities are likely to be high in situations where choice of other modes is readily available: for example, in small compact cities the possibility of using a cycle or walking exists as an alternative to using a bus; in larger cities when two or more modes (say bus and trains) are running in parallel, then an increase in fares on one mode is likely to make the other mode(s) more attractive, financially. 7.5 The calculation of elasticity values is usually undertaken on the basis of a statistical analysis of 'cross section' data (ie for say a number of bus companies in a single time-period) or 'time-series' data (for one bus company over an extended period of time).

REFERENCES
CIPFA (1979). Passenger transport operations supplement: peak/off-peak costing and revenue allocation. Passenger Transport Finance Executive, Chartered Institute of Public Finance and Accounts, London. MARLER N W (1982). The performance of high-flow bus lanes in Bangkok. Department of the Environment, Department of Transport. TRRL Supplementary Report SR 723, Crowthorne. (Transport and Road Research Laboratory). MAUNDER D A C and FOURACRE P R (1983). Specialised bus services in three Third World cities. Department of the Environment Department of Transport. TRRL Supplementary Report SR 811, Crowthorne.(Transport and Road Research Laboratory).

REFERENCES
TRANSPORT AND ROAD RESEARCH LABORATORY (1980). The demand for public transport - Report of the International Collaborative Study of the

21

factors affecting public transport patronage. Crowthorne. (Transport and Road Research Laboratory). BUCHANAN, M (1980). The Bombay bus management study. PTRC Summary Annual Meeting, University of Warwick. FOURACRE, P R, D AC MAUNDER, M G PATHAK and C H RAO (1981). Studies of bus operations in Delhi,

India. Department of Transport. TRRL Supplementary Report SR 710, Crowthorne. (Transport and Road Research Laboratory). MODAK, S K and BHANUSHALI V G (1985). Demand elasticities for public bus transport in Bombay. Transportation Research Forum. 26th Annual Meeting, Jacksonville, Florida, Nov.1985.

TABLE Al FARES AND SERVICE ELASTICITY VALUES FOR BUS OPERATIONS

Demand elasticity with respect to Fares

Location

Value or likely range

Comments

Developed Countries (I)

-0.1 to 0.6 Av. 0.3

For: large towns 0.1 to 0.5 small towns 0.2 to 0.7 During: peak 0.1 to 0.35 off-peak 0.25 to 0.7 Higher values in poorer suburbs and where rail competes with bus. Little or no competition for mass

Bombay (ii) (iii)

-0.28 to 0.75 Mean value -0.4 to 0.48 -0.11 transit. 0.4 to 0.5

Delhi (iv)

Service level

Developed Countries (i) Bombay (ii) Delhi (iv)

0.3 to 0.45 0.6 Probably an overestimate.

Sources:

(i) TRRL, 1980 (ii) Buchanan, 1980 (iii) Modak and Bhanushali, 1985 (iv) Fouracre et al, 1981

22

8.

APPENDIX B A SIMPLE COST MODEL

TABLE BI OPERATING COST STRUCTURE (WORKING EXPENDITURE) 1977-78

8.1 A change in the public transport system will usually result in a change in operating costs. The structure of the operating costs of a company can given an indication of how changes in costs of the different items affect total costs. For example, Table B 1 shows the per cent distribution of costs incurred by a major Third World bus operator in the financial year 1977-78. 8.2 If, due to external circumstances, the cost of the diesel fuel is increased by 10 per cent, relative to all other components of cost, then the total costs will rise by 1.8 per cent (ie 10 per cent of 18.3 per cent). Usually a change in output creates changes in more than one cost component. 8.3 The way in which the different parts of total costs change as output changes is important. Establishing each cost components' relationship with a particular measure of output is the basis of a cost model which can be used to analyse any planned system change. The measures of output commonly used in the bus industry are bus Ion, bus hours and number of vehicles in use (or peak hour requirement). A system change may affect one or more of these output measures: for example, rescheduling or rerouting of buses may affect only bus km run, while an increase in fleet size will affect all three measures, with additional bus km, bus hours and buses in use. 8.4 The rate of response of the change in the cost component to the change in output varies greatly. Additional kilometres will immediately affect consumption of fuel and hence the cost of this item. On the other hand, costs such as rent and rates of buildings and administration costs are unlikely to be affected by small changes in fleet size. Only large fleet additions requiring the acquisition of new buildings and administrative staff would affect costs. Three cost categories are usually specified which are variable costs, semi-variable costs and fixed costs. Variable costs are taken as those which respond almost instantaneously to changes in output. They are particularly important when considering the more productive use of the existing stock of vehicles. Semi-variable costs are the costs incurred when there is a marginal increase in stock of vehicles, or the costs which result from the longer term (several weeks or months) effects of the more productive use of existing stock. Fixed costs are those costs which, though dependent on output, are not particularly responsive to changes in level of output except when the changes are very large. 8.5 Typically, then, in a cost model the costs are broken down as far as possible and allocated in a way which is based on their degree of variability over time and the measure of output to which they most directly respond. There is no

Per cent of total Variable costs Diesel Oil Tyres Spares Tickets Sub-total Semi-variable Drivers and conductors Traffic Supervisory staff Central workshop staff Depot staff (maintenance) Uniforms Tax on vehicles Insurance Welfare and superannuation Sub-total Fixed costs HQ: officers, clerical and cash staff Central workshops: officers and clerical Depots:officers and clerical Other admn. staff and expenses Rent and rates Sundries Sub-total Total 18.3 4.3 8.7 12.8 0.4 44.5 26.0 2.6 1.7 7.1 1.0 2.2 0.3 3.9 44.8 3.0 0.3 0.8 3.0 1.1 2.5 10.7 100.0

standard format for such a model since different operators will undoubtedly have different views on which output and time factors have most effect on each component. 8.6 In Table B 1, all costs itemised as variable could be taken as dependent on bus km; all the semi-variable coald be taken as dependent on number of buses held; all fixed costs depend on aset number of vehicles (for example, these costs may increase in units of 100 vehicles, this being the equivalent of one new depot). A simple cost model could be expressed in the form: TC = FC + b1K + b2V where TC is the total daily operating cost, FC is the fixed cost per day (for the given output level), K is the daily kilometrage of the fleet, V is the number of vehicles in use per day, b1 is the cost per km and b2 is the cost per vehicle

23

employed. Thus b1K is the variable cost and b2V is the semi-variable cost. The daily cost of an individual vehicle is given by: C = b1K1 + b2 (V = 1) where k1 is the daily kilometrage output of a vehicle.

9. APPENDIX C EXAMPLES OF SURVEY OUTPUT

BASIC ROUTE CHARACTERISTICS
9.1 In order to iflustrate the output expected from the surveys described in Section 3, material from six routes studied in Delhi in 1978 (Fouracre et al, 1981) is presented. As background to these surveys, Table Cl presents the physical characteristics of the routes which were monitored. 9.2 Table ~ shows typical output from a loading survey carried out on five of the routes studied in Delhi during a complete days operations in 1978. The presentation of survey material in this basic form (averaged for the whole route, with some distinction made between peak and offpeak operations) provides useful background information for further analysis. For example, on all routes passenger lead tended to be slightly higher in the peak than in the offpeak. Further, as would be expected, passenger lead was higher for the longer routes, except in the case of Route 89 which had a lead of only about 8 km for a route length of 18.5km. Route 89 is a cross-city route, and many travellers would be using the service for up to half its trip length (to and from the city centre). 9.3 As expected load factors were significantly higher on all routes in the peak direction during the peak period, than in the off-peak. Route 430 has a fairly even load factor in the peak for both directions of travel whereas route 720 was extremely 'unbalanced' in this respect. 9.4 The high revenues earned by buses on route 89 reflect the fact that high capacity double-deck vehicles were used on this route. 9.5 Average journey speeds are shown for both peak and off-peak periods; the figures include both direction speeds. Average speeds were marginally higher than scheduled, which were themselves high for urban traffic conditions. Only route 89 with its use of double-deck buses had average trip speeds of less than 20 km/h.

JOURNEY TIME COMPONENTS

9.6 In a journey time survey in Delhi the average stop time at bus stands was measured at between 0.31 and 0.42 minutes on a selection of routes. Penalty time was found to be 0.22 minutes per bus stand, giving a total delay time at bus stands of between 0.53 and 0.64 minutes per stand. Table C3 provides more detailed information for the routes surveyed, of how bus stand delay is incurred. 9.7 Penalty time will also include stationary delays (ie when the vehicle comes to a halt for reasons other than setting down or picking up passengers). Table C4 shows the

24

TABLE Cl PHYSICAL CHARACTERISTICS OF SIX BUS ROUTES IN DELHI

Route number 80 Route length (km) Number of bus stands Number of vehicles used: peak off-peak Scheduled daily km Scheduled daily km per bus: normal services peakextra Scheduled journey time (mm) Scheduled joumey speed (km/h) 8.0 14/15* 6 2 1488 219 88 25 19.2 89 18/19* 36/39* 7 1554 222 55/60* 19.3 155 21.1 33 10 2532 235 60 21.1 430 16.4 33 10 2624 262 40 24.6 521 14.9 27/28* 7 3 1937 226 119 43 20.8 720 20/21* 36/39* 6 5 2132 248 114 50 25.1

*Some routes have differences in route layout and bus stand locafion depending on direction.
TABLE C2 BASIC OUTPUT DATA FROM A LOADING SURVEY

Route number 80 Km. operated Operated: Schedule km No. of passengers: peak off-peak Av. passenger lead (km): peak off-peak Load factor: peak (peak direction) peak (both directions) off-peak Av. revenue per bus (Rs): peak off-peak Av. revenue per pass. (paise) Earning per km (paise): peak off-peak Av. journey speed (km/h): peak off-peak 1472 0.99 4620 5199 5.3 5.0 0.94 0.70 0.49 17.9 13.9 29 223 174 20.8 23.3 89 1332 0.86 3998 4177 8.3 7.3 0.65 0.53 0.33 42.8 29.8 31 231 161 19.5 19.8 155 2363 0.93 5933 5732 11.8 10.4 1.53 1.02 0.71 37.4 30.9 32 177 146 23.1 24.5 430 2575 0.98 5320 8408 10.2 9.1 1.16 1.10 0.71 34.0 23.2 30 207 141 24.6 26.6 720 2050 0.96 4178 2609 11.3 9.7 1.23 0.64 0.50 27.7 24.4 38 135 119 24.3 26.2

importance of penalty time in these delays for the same selection of routes in Delhi. 9.8 For these routes the average stationary delay time per km ranged from 0.10 to 0.54 minutes per km. (These differences broadly reflected known operating conditions.)

This measure provides an important indicator of the relative congestion faced on different routes and points to those routes which have particular problems. Where aper km ranged from 0.10 For these routes the average stationary delay tirne particular congestion black-spot affects a number of routes there may weH be a case for remedial action involving traffic management techniques. Yet another part of total delay is 25

TABLE C3 BUS STAND DELAYS SURVEYS IN DELHI, 1978

Route/time

Average total stop time at bus stands per trip (min) 3.4 3.1 12.0 10.6 5.1 5.6 8.0 8.2

Average Average stop number of time per bus stops at stands stand (min) per trip 10 9 29 29 19 21 19 26 0.34 0.35 0.41 0.37 0.27 0.27 0.42 0.31

Average Total average penalty time bus stand per trip (mm) delay per trip (min) 2.2 2.0 6.3 6.3 4.1 4.6 2.2 5.6 5.6 5.1 18.3 16.9 9.2 10.2 10.2 13.8

Bus stand delay as percentage of total trip time 26 26 32 31 25 32 24 30

80 am peak am off-peak 89 pm peak pm off-peak 430ampeak am off-peak 521 pm off-peak 720pm off-peak

TABLE C4 STOP DELAYS - SURVEYS IN DELHI, 1978 Average number stops per trip Average stop time per trip (min) Average penalty time per trip (min) Total average delay time per trip (min) Average delay time per km (min) Stop delay as a percentage of total trip time

Route/time

80 am peak am off-peak 89 pm peak pm off-peak 430 am peak am off-peak 521 pm off-peak 720 pm off-peak

5 4 13 12 7 4 12 6

1.7 1.3 7.1 4.7 1.8 0.8 3.2 3.6

1.1 0.9 2.8 2.6 1.5 0.9 2.6 1.3

2.8 2.2 9.9 7.3 3.3 1.7 5.8 4.9

0.35 0.28 0.54 0.39 0.20 0.10 0.39 0.24

13 11 17 13 9 5 14 10

the effect on vehicle speeds of general traffic conditions and the like, ie those factors which cause slow running (below some optimum for the type of road) rather than actual stoppages. This can only be estimated by assuming some free-running speed and comparing the time it would take to cover the route distance at that speed with actual observed speeds (allowing for the stops and starts due to serving bus stands and stop delay). Taking route 80 of Table C4 as an example: a bus would cover the 8 km route distance in 9.6 minutes, plus 0.2 minutes penalty time, if travelling at 50km per hour an assumed optimum speed. Assuming the same pattern of stops and starts (at bus stands and because of stationary delay) then the total estimated travel time, operating at 50 km per hour between stops, is 18.2 minutes in the am peak and 17.1 minutes in the am off-peak. Actual observed average journey times for these same periods were

21.2 and 19.3 minutes respectively. The difference may be ascribed to general delay. 9.9 In Table CS all the journey time components for the same routes in Delhi are combined. 9.10 In this particular example bus stand delays accounted for about 25 per cent of journey time. Other delays (both stationary and general) accounted for between 20 to 30 per cent of journey time. Traffic management techniques for improving bus running speeds could therefore be expected to reduce journey times, in these examples, up to a maximum of 20 to 30 per cent. More efficient passenger handling techniques (through better interior design or off-bus ticket sales) could possibly reduce bus stand delays especially in the peak period, though to assess the value of

26

TABLE C5 COMPONENTS OF BUS JOURNEY TIME - SURVEYS IN DELHI, 1978 Average journey time (mm) Stationary delays and penalty Bus stand delays and penalty General delay (min) Total delay as a percentage of total journey time 54 49 51 48 46 38 53 53

Route/time

80ampeak amoff-peak 89pmpeak pm off-peak 430 am peak amoff-peak 521 pm off-peak 720pm off-peak

21.2 19.3 57.3 54.1 36.7 32.0 42.3 46.7

2.8 2.2 9.9 7.3 3.3 1.7 5.8 4.9 TABLE C6

5.6 5.1 18.3 16.9 9.2 10.2 10.2 138

3.0 2.2 1.1 1.9 4.3 0.2 6.3 6.3

TYPICAL BUS BOARDING AND ALIGHTING TIMES - SURVEYS IN DELHI, 1978 Boarding time per passenger (secs) 1.3 1.4 2.1 1.9 Alighting time per passenger (secs) 0.7 1.1 1.5 1.3

Bus type

Peak/off-peak

Dead time (secs)

Double-decker (single entrance/ exit) Single-decker (separate entrance and exit)

peak off-peak peak (boarding) off-peak (boarding) peak (alighting) off-peak (alighting)

7.2 5.0 3.0 3.1 3.0 3.4

this requires an understanding of boarding and alighting rates. 9.11 Some typical values for boarding and alighting rates are contained in Table C6. The base data for these estimates was a loading survey undertaken in Delhi. With information in this form it is possible to compare different vehicle designs and their impact on passenger handling and overall journey times.

during the survey period. Three or four bus stands had been selected for each direction of each route, the stands having been identified as having high passenger activity from a previous loading survey. 9.13 The coefficient of variation is a uselul measure of service regularity. If passengers arrive at a bus stand in a random way and can board the first arriving bus, then thefr average waiting time (AWT) can be expressed as: _ AWT = H/2 (1 + V2 ) where H is the mean headway and V is the coefficient of variation of the distribution of headways. This expression is minimum when V = 0, ie when the service is regular and every bus arrives at a bus stand exactly H minutes behind the last bus. Irregular services will have a high of V, as shown for most routes in Table C7.

PASSENGER WAITING TIMES AND BUS FREQUENCIES

9.12 Typical data obtained in a waiting time survey in Delhi are shown in Table C7. Weighted averages of observations at stands along each route are shown. The weightings were the number of passengers boarding at each stand monitored

27

TABLE C7 BUS HEADWAYS AND WAITING TIMES - SURVEYS IN DELHI, 1978 Peak hours W min. 7.0 18.8 11.5 9.1 10.0 15.3 H min. 6.9 24.9 14.9 8.4 11.9 17.9 Key W min. 8.2 16.1 13.3 9.2 15.4 21.8 Off-peak hours H min. 9.8 24.2 20.5 9.8 19.6 31.9

Route 80 89 155 430 521 720

V 0.63 0.51 0.34 0.44 0.46 0.63

P 0.24 0.13 0.18 0.36 0.26 0.17

V 0.55 0.49 0.41 0.57 0.45 0.60

P 0.13 0.09 0.07 0.15 0.15 0.13

W - observed average waiting time H - mean headway V - coefficient of variation (ratio of standard deviation of distribution of headway times to the mean) P - probability of not being able to board first arriving bus

TABLE C8 PASSENGER SURVEY OUTPUT - DELHI, 1978 Distance travelled (km) Total journey time (mm) Travel for work purposes (%) Passholder (%)

Time period

Single journeys Peak morning Peak evening Off-peak morning Off-peak evening Multiple journeys* Peak morning Peak evening Off-peak morning Off-peak evening

11.5 10.9 9.9 10.5

47 48 43 47

80 76 49 60

7 6 6 8

18.1 19.1 18.7 21.5

75 82 78 85

80 71 55 92

10 11 6 11

*Journeys involving at least one interchange

9.14 The probability of a passenger being unable to board a bus, shown in Table C7, is derived directly from the 'Q' Enquiry cards. For the routes shown this probabilitv is higher in the peak than the off-peak.

9.16 In this example the information was aggregated for all respondents, to represent a picture for the network as a whole. Something like 70 per cent of respondents were making single journeys, ie involving no interchange. Information presented by route is shown in Table C9.

PASSENGER CHARACTERISTICS
9.15 Data derived from the passenger interviews can be by route or for the network as a whole, depending on the survey objectives. A distinction can also be made between peak and off-peak travellers. Table C8 shows some output from the survey which used a questionnaire similar to that contained in Appendix D.

REFERENCE
FOURACRE, PR, MAUNDER D AC, PATHAK MG and C H RAO (1981). Studies of bus operations in Delhi, India, Department of Transport. TRRL Supplementary Report SR 710, Crowthorne (Transport and Road Research Laboratory)

28

TABLE C9 PASSENGER SURVEY RESULTS, PRESENTED BY ROUTE - DELHI, 1978

Route 80 Single journeys Distance travelled (km) Total journey time (mm) Travel for work purposes (%) Passholder (%) Monthly income (Rs) Multiple journeys* Distance travelled (km) Total journey time (mm) Travel for work purposes (%) Passholder (%) Monthly income (Rs) 89 155 430 521 720

5.5 36 60 4 750

8.5 41 63 4 830

13.4 52 67 9 290

10 40 77 10 690

9.7 50 69 5 410

13.7 48 82 7 790

16.5 62 59 11 520

16.1 67 55 9 960

18.9 66 65 9 380

22.8 72 74 7 610

17.9 73 75 10 450

19.7 66 67 12 740

*Journeys involving at least one interchange

29

10.

APPENDIX D PASSENGER INTERVEW QUESTIONNAIRE

30

31

11. APPENDIX E EXAMPLE OF CALCULATING A SAMPLE SIZE FOR LARGE POPULATIONS

11.1 For large populations the standard deviation of the distribution of sample means (known as the standard error of the mean) is approximately equivalent to the standard deviation of the population divided by the square root of the sample size. ie standard error mean = n when n is the sample size and is the standard deviation of the population, represented by the sample deviation. From sampling theory it can be demonstrated that the population mean will lie within two standard errors on either side of the sample mean, with 95 per cent confidence or certainty. For example, if the average daily bus load is to be monitored and it is known to be of the order of 1000, with a population standard deviation of 200, and an accuracy of 10 percent is required from the sample estimate of the mean, then the following reasoning can be used to determine sample size: Required accuracy = 1000 + 2se with 95% confidence where 2se = 100 (ie 10% of mean) and se = / n with se = 50 and =200 n = /se = 4 n = 16 11.2 In this example, sixteen buses should be monitored to give the required degree of accuracy. If each bus is engaged on two shifts or duties. then 32 duties would have to be covered, requiring 64 survey assistants. Clearly some degree of accuracy may have to be lost in order to keep the manpower requirement at an acceptable level. Reducing both the accuracy to 20 per cent and the confidence of acceptance to 90 per cent gives a sample size of about three vehicles, requiring 12 survey assistants. 11.3 There wlll necessarily be some trial and error involved in selecting the sample size, since population parameters are unlikely to be known prior to sampling.

12. APPENDIX F EXAMPLE OF ROUTE SCREENING USING SPECIFIED PERFORMANCE VALUE CRITERIA FOR FIVE ROUTES IN DELHI
12.1 In Table Fl the values of the performance criteria are specified for five of the routes described in Appendix C. 12.2 In order to see how the screening procedure works it is necessary to specify some cut-off points to distinguish between good and poor performance for each indicator. Ideally, these values should be based on operating experience and an appreciation of what service levels passengers should be able to expect. The process of establishing these cut-off points will involve some trial and error; setting them too low will produce too many poor routes and too high will produce too few. Furthermore if the service is improving (or degenerating) over time, it will be necessary to adjust the target values accordingly. In Delhi, for example, the following seemed appropriate: the EPK to CPK ratio is high if greater than 1.0 in the peak and 0.7 in the off-peak; the average route waiting time is high if greater than 15 minutes in the peak and 20 minutes in the off-peak; the average route load factoris high if greater than 1.0 (measured in the peak direction) in the peak and 0.7 in the off-peak. Using these norms Table P' shows how well the five routes performed. 12.3 Route 80 has a similar profile in both the peak and off-peak with high profitability (relative to other routes), low wait times and low load factors. The route would appear to be working well from both operator's and user's view. However, the low load factor suggests that there may be a case for reducing the frequency of operations. 12.4 Route 89 is characterised by low profitability and correspondingly low load factors. Waiting times are high in the peak periods. There is some suggestion of poor route or possibly the use of the wrong vehicle type. (Route 89 is a cross-city route using double-deck buses.) 12.5 Route 155 has high load factors in both the peak and off-peak, coupled with low waiting times. Profitability is poor in the peak, but good in the off-peak. There might be a case for using larger vehicles on this route, or re-assessing the route layout. Route 430 is in many ways sirailar to Route 155, though it has good profitability throughout the operational periods. Extra buses might be usefully deployed on Route 430. (But see Table 5.3

32

12.6 Route 720 has a bleak profile in both the peak and off-peak. The high waiting times are associated with low frequency of operations, although further analysis indicates poor reliability. The route suffers badly from poor return loads and a high peak to off-peak imbalance. Route layout may be at fault, or possibly smaller buses (minibuses) might usefully be deployed on this route. 12.7 The screening process may thus indicate particularly poor routes such as 89 and 720 which merit further attention. This procedure does nothing more than this and it would be wrong to take decisions purely on the basis of the indicators.

TABLE Fl EXAMPLE OF ROUTE PERFORMANCE INDICATORS FOR FIVE ROUTES, DELHI

Route number 80 Peak EPK:CPK 1.01** Waiting time (mm) 7.0 Loadfactor* 0.94 Off-peak EPK:CPK 0.88 Waiting time (mm) 8.2 Load factor 0.49 89 155 430 720

0.81 0.94 18.8 11.5 0.65 1.53

1.15 9.1 1.16

0.64** 15.3 1.23

0.56 0.78 16.1 13.3 0.33 0.71

0.78 9.2 0.71

0.64 21.8 0.50

*measured in the peak direction **extra buses are used in the peak on these routes: the CPK for peak-time operations has been calculated as the weighted average of costs for normal duty and extra buses, the weights being the number of buses used.
TABLE F2 EXAMPLE OF ROUTE SCREENING ANALYSIS, DELHI

Route number 80 Peak EPK:CPK Waiting time Load factor Off-peak EPK:CPK Waiting time Load factor 89 155 430 720

high low low high low low

low high low low low low

low low high high low high

high low high high low high

low high high low high low

33

13

APPENDIX G STANDARD PRO-FORMAS

Pro-forma for bus loading survey

13.1 Day: Date: Route no:

No of tickets 15 paise: 30 paise: No of bus passes:

Revenue: Revenue: 2

Route length:

Trip start time: Trip finish time: Total trip time:

Total fare revenue: Rs

Total no of passengers:

Vehicle type Single deck Double deck Minibus Other (specify)

Time Name of bus stand Arrival Departure Boarding

Passengers Alighting Remaining

Journey Remarks

Total

34

13.2 Pro-forma used to carry out bus journey time surveys Day: Date: Route no: Route length: Trip start time: Trip finish time: Total trip time: Vehicle type: Single deck Double deck Minibus Artic double deck Other (spectfy) Causes 1. 2. 3. 4. 5. 6. 7 8. Roundabout Traffic signals Bus stands Pedestrians Animals Uncontrolled junction Controlled junction Accident Direction of trip:

Time slower than walking speed

Stop time

Time faster than walking speed

Causes of delay

Delay time in seconds

Delay location

35

13.3 Pro-formas used to carry out penalty time surveys Day: Date: Location:

Vehicle type: single deck SD double deck DD artic double deck ADD minibus MB

Direction:

Vehicle type

Bus registration number

Passing time

Vehicle type

Bus registration number

Passing time

(b)

Vehicle type

Bus registration number

Number of entry exit doors

Depart time

Stop time (secs)

(a)

36

13~4 Pro-forma for analysis of penalty time survey Day: Date: Direction: to: Location:

Vehicle type

Bus number

Time at first point

Time at arrival

Bus stand departure

Time at second point

Total journey time (mins)

Stop time (mins)

Actual journey (mins)

13.5 Q' Enquirey card ROUTE NO: DAY: DIRECTION OF BUS: DATE: TO:

Name of the bus stop Time of passenger arrival at the bus stop Time of boarding bus Waiting time No. of buses arriving during this time which did not stop or were too full to board FARE PAID

hours hours hours

mins mins mins

37

Pro forma for bus arrival time survey

Day: Date: Direction of trip:

Route No: Route Length:

Bus Stand:

Bus number

No of passengers boarding

Bus did not stop Arrival time Headway time Overcrowded No reason

38

NOTES

39

NOTES

Printed in the United Kingdom for HMSO DdT0670N 4/95 C5 51-00 10170

40

Transport and Road Research Laboratory Overseas Unit Department of Transport Overseas Development Administration

Overseas Road Note 5

A guide to road project appraisal

Overseas Unit Transport and Road Research Laboratory Crowthorne Berkshire United Kingdom 1988

ACKNOWLEDGEMENTS
This Note was produced in the Overseas Unit of the Transport and Road Research Laboratory Head: Mr J S Yerrell) on behalf of the Overseas Development Administration. The following members of the Unit contributed to the draft: Dr R Robinson (Project Leader and Editor), Mr P J Beaven, Mr D M Brooks, Dr M A Cundill, Mr A J Downing, Dr B L Hills, Mr J L Hine, Dr G D Jacobs, Mr C J Lawrance, Mr R S Mansfield, Mr D Newill, Mr P W D H Roberts, Dr J Rolt and Mrs J M White. The section on cost estimating is based on the ODA draft cost estimating manual prepared by UMIST. The Note was reviewed for ODA by Mr J B Wilmshurst.

First published 1988

OVERSEAS ROAD NOTES

Overseas Road Notes are prepared principally for road and transport authorities in countries receiving technical assistance from the British Government. A limited number of copies is available to other organisations and to individuals with an interest in roads overseas, and may be obtained from: Transport and Road Research Laboratory Crowthorne, Berkshire, RG11 6AU United Kingdom Crown copyright 1988 Limited extracts from the text may be produced provided the source is acknowledged. For more extensive reproduction, please write to Head of Overseas Unit, Transport and Road Research Laboratory ISSN 0951-8987

ii

CONTENTS
Page 1. IntroductionI Purpose of this Note The project cycle Identification Feasibility Design Commitment and negotiation Implementation Operation Evaluation Structure of this Note Project types Factors to be considered Upgrading and new construction Reconstruction and rehabilitation Stage construction Maintenance projects Network considerations Preliminary project considerations Setting objectives Analysis period and design life Underlying issues Uncertainty and risk Choice of technology Institutional issues The institutional framework Improving institutional development Road maintenance organisations Assessment of maintenance capability Socio-economic considerations Environmental considerations Factors to be considered Environmental assessment Assessing traffic demand Types of traffic Baseline traffic flows Traffic counts Traffic counters Moving observer counters Traffic forecasting Normal traffic Diverted traffic Generated traffic Uncertainty of traffic estimates Cost estimation Purpose of the estimate Cost estimating stages The estimator Information in the estimate Estimating techniques 1 1 1 1 1 1 1 2 2 2 3 3 3 4 5 5 5 6 6 6 7 7 7 8 8 8 9 9 10 11 11 12 13 13 13 13 14 14 15 15 15 15 16 17 17 17 18 19 19

2.

3.

4.

iii

The techniques available Global Man hours Unit rates Operational (resources cost) Suitability of individual techniques Sources of data Principal sources Project specific data Data banks Published data Contract strategy The design process and project costs 5. Geotechnics Object of geotechnical surveys Information needed Route location Subgrade strength Materials Earthworks Drainage Structures Sources of information Steps in the survey process Project identification Feasibility Design for project implementation Costs and accuracies of geotechnical surveys Pavement design Structural classification Earth roads Gravel roads Choice of paved road construction type Rigid pavements Flexible pavements materials Surfacings Road bases Sub-bases and other pavement layers Use of marginal materials Factors affecting flexible pavement design Subgrade strength Traffic loading Materials Variability and uncertainty Shoulders Drainage of pavement layers Preparation and checking of flexible pavement designs Collection of information Choice of design method Strengthening flexible pavements Costing Geometric design Purpose of geometric design Elements of geometric design Rational basis for geometric design

19 19 20 21 21 22 22 22 22 23 23 24 25 26 26 26 26 26 27 27 28 28 28 29 29 29 29 30 31 31 31 31 32 32 32 32 33 34 34 34 34 34 35 35 36 36 37 37 37 37 38 38 38 38 40

6.

7.

iv

Design related to terrain Horizontal alignment Vertical alignment Cross-section Road reserve Junction design Signs and road markings Costing earthworks 8. Drainage and structures The drainage system Side drains Requirements for cross drainage Fords and drifts Culverts Bridges Bridge design Site conditions Abutments and piers Timber decks Concrete decks Steel decks Arch bridges Replacement of existing bridges Costing bridges Maintenance Ferries Introduction of the assessment of benefits

40 41 42 42 43 43 43 44 44 44 44 45 45 46 46 46 46 47 47 47 48 48 48 49 49 49 50 51 51 51 51 52 52 52 52 53 56 57 57 57 57 58 58 58 59 59 59 59 60 61 61

9.

10. Vehicle operating cost savings Factors affecting vehicle operating costs Road investment models Vehicle operating cost tables Data requirements Measured values Relative importance of data items Physical parameters of the road Vehicle descriptions Determining costs 11. Road maintenance benefits Paving gravel roads Strengthening and reconstruction Concrete roads Diverted traffic Traffic delays during maintenance works Determining costs 12. Time savings General considerations Vehicle fleet Vehicle occupants Freight 13. Reduction in road accidents Forecasting accident reductions

Factors leading to reductions Effect of highway design Low cost remedial measures Effects of traffic Road accident costs Material and subjective factors Methods available for costing road accidents The relevance of alternative methods for developing countries Recommended method of costing 14. Economic development benefits The impact of different forms of road investment Consumer surplus Producer surplus 15. Cost-benefit analysis Principles Prices Inflation Discounting Shadow prices Comparison of alternatives Net present value Internal rate of return Project timing Recommended approach 16. Rural access roads The need for special considerations Non economic considerations Screening and simplification of data collection Upgrading and maintaining existing access New road access 17. Analysis of uncertainty Scenario and risk analysis Expected values Contingency Sensitivity analysis Traffic Project costs Delay Generated traffic Time and accident savings Shadow prices Maintenance Special factors Investment models Risk analysis 18. The feasibility study report Preparation Presentation 19. Checklist of key points Objectives Background Institutional and managerial aspects

61 61 61 63 63 63 63 64 64 65 65 65 66 67 67 68 68 68 68 69 69 69 70 70 71 71 71 72 72 72 73 73 73 74 74 74 74 74 74 75 75 75 75 75 75 76 76 77 78 78 78 78

vi

Traffic Design Costs Benefits References Appendix A: Road upgrading project Components of the study Traffic surveys Traffic growth Engineering design Construction costs Maintenance costs Road user costs Cost-benefit analysis Sensitivity tests Discount rate Variations in ADT Traffic growth rates Construction cost Maintenance capability Other benefits Time savings for freight Diverted traffic Socio-economic integration Improved consumer choice Increased export earnings Road safety Appendix B: Road shortening project Background to the study Road user cost savings Engineering costs Benefits Sensitivity testing Discount rate, developmental benefits and time Traffic Construction costs Relative sensitivities

79 79 79 79 79 81 81 81 82 83 84 85 85 86 87 87 87 88 88 88 89 89 90 90 90 90 90 91 91 91 93 93 93 96 96 97 97

vii

1.

INTRODUCTION

PURPOSE OF THIS NOTE

1.1 This Note gives guidance on carrying out feasibility studies for road projects in developing countries. It is intended for administrators, economists, transport planners and engineers in road and transport ministries in developing countries who are responsible for preparing or appraising project submissions. It will also be of interest to personnel in aid agencies and consultancies who are responsible for road projects. 1.2 The Note deals with rural (non urban) road projects for new construction, upgrading, rehabilitation, stage construction and maintenance. A background description of the engineering and transport issues involved is provided, and guidance is given as to which aspects of feasibility studies should be undertaken by a transport planner and which require the advice of a road engineer. The phases involved in executing a road project are outlined and attention is drawn to the need to collect good data and to identify which data and decisions are the most important. 1.3 The document is consistent in its approach with ODA's 'Appraisal of projects in developing countries: a guide for economists' (Overseas Development Administration 1988) and its book on planning development projects (Bridger and Winpenny 1983). It is the document to which staff at ODA will refer when appraising road projects, and should be used by consultants when preparing project submissions for ODA.

1.6 It must be remembered that the decision to proceed to the feasibility stage arouses expectations for the project which can create their own momentum. Dubious projects should therefore be rejected at the identification stage, particularly if there is no lack of more promising ones. It becomes increasingly difficult to stop a project at the later stages in the cycle when minor changes of detail are often all that is possible.

Feasibility
1.7 The feasibility study will provide enough information for deciding whether to proceed to a more advanced stage of planning. The level of detail of this study will depend on the complexity of the project and how much is known already about the proposal. Sometimes, a succession of increasingly detailed studies will be needed. The feasibility study should define the objectives of the project. It should consider alternative ways of achieving these and eliminating poor alternatives. The study provides the opportunity to mould the project to fit its physical and social environment in such a way to maximise the return on the investment. 1.9 Once the feasibility study has indicated which project is likely to be the most worthwhile, detailed planning and analysis can begin. Even though the less promising projects will have been eliminated by this time, the selected project may be redefined and modified as more detailed information becomes available. At this stage, studies of traffic, geotechnics and design will be carried out in order to refine the prediction of costs and benefits and to enable an economic analysis to be carried out.

THE PROJECT CYCLE

1.4 Projects are planned and carried out using a sequence of activities known as the 'project cycle'. There are many ways of defining the steps in this sequence but, in this Note, the following terminology is used: identification, feasibility, design, commitment and negotiation, implementation, operation, and evaluation.

Design
1.10 Preliminary design and feasibility are often simultaneous, but detailed design, which can be very costly (up to 15 per cent of project costs), usually follows provisional commitment to the project. Numerous decisions which will affect economic performance are taken throughout design; and economic appraisal often results in redesign.

Identification
1.5 The first stage of the cycle is to find potential projects. This is sometimes known as the 'pre-feasibility' stage. There are many sources from which suggestions may come including well-informed technical specialists and local leaders. Ideas for new projects will also come from proposals to extend existing programmes or projects. In the process of preparing an economic development plan, specific suggestions for projects may come from operating agencies responsible for project implementation. Sometimes a sector survey will identify the need for specific projects.

Commitment and negotiation

1.11 Commitment of funds often takes place in a series of stages. This is followed by invitations to tender and negotiations with contractors, potential financiers and suppliers. At this stage, there are still considerable uncertainties.

Implementation
1.12 Detailed recommendations on project implementa tion are beyond the scope of this Note. However,

several aspects of the earlier stages in the project cycle will affect the success of the implementation. 1.13 The better and more realistic plan, the more likely it is that the plan can actually be carried out and the full benefits be realised. A flexible implementation plan should also be sought. It is almost inevitable that some circumstances will change during the implementation. Technical changes may be required as more detailed soils information becomes available or as the relative prices of construction materials change. Project managers may need to change and replan parts of the project to take account of such variations. The more innovative and novel the project is, the greater is the likelihood that changes will have to be made during implementation.

will also be of use at the identification, design and evaluation stages. 1.20 Technical, financial, social and environmental appraisals should also be undertaken for a project, and these are discussed in this Note where they affect the economic appraisal. Political and strategic considerations are beyond the scope of this Note. 1.21 When carrying out feasibility studies, it is recommended that the following steps are undertaken. References are made to the section of this Note where the various items are covered. 1. 2. Define objectives (para 1.53-54)

Operation
1.14 This refers to the actual use of the road by traffic; it is during this phase that benefits are realised and maintenance is undertaken.

Determine alternative ways (para 1.26-51) of meeting objectives: modal choice choice between new road or upgrading existing road, etc Preliminary considerations Assess traffic demand Design and cost different options: Route location Pavement design Geometric design Design of structures (para 1.52-57 and Section 2)

3.

Evaluation
1.15 The final phase of the project cycle is evaluation. This consists of looking back systematically at the successful and unsuccessful elements of the project experience to learn how planning can be improved in the future. 1.16 For evaluation to be successful, it is important that data about the project is collected and recorded in a systematic way throughout all stages of the project cycle. Without this, it is usually impossible to determine details of events and information that were available during periods leading up to the taking of important decisions. 1.17 Evaluation may be carried out by many different people. The sponsoring organisation or external agency may undertake evaluation. In large and innovative projects, a separate unit may be needed to monitor each stage of the project by collecting data for identifying problems that need to be brought to the attention of the project's management. In some cases, outside staff will be used to provide an independent audit and specialist university staff may well be suited to undertake such a task. 1.18 The evaluation should result in specific recommendations about improving aspects of the project design which can be used to improve ongoing and future planning.

4. 5.

(Section 3) (Sections 4-8)

6.

Determine benefits (Sections 9-14) for each alternative: Road user cost savings Road maintenance benefits Time savings Reduction in road accidents Developmental and social benefits Economic analysis and comparison of alternative Recommendations (Sections 15 and 16)

7.

8.

(Section 18)

These steps are not necessarily sequential and iteration between them is usually needed. The analysis undertaken should be used as a project design tool. 1.22 As indicated above, the layout of this Note follows a similar structure. Sections 4-8, on design and costing, are written for a non-engineering audience, but should still be of interest to engineers. Sections 9-14, on the assessment of benefits, are wntten primarily for administrators, economists and engineers, but should still be of interest to transport planners who are familiar with benefit assessment methods. Section 15, on cost-benefit analysis, will be of interest to economists, although written primarily for an audience who are not familiar with economic analysis techniques. All other chapters are written for a broad audience and should be consulted by all these engaged in feasibility studies.

STRUCTURE OF THIS NOTE

1.19 This Note is concerned primarily with the feasibility stage of the project cycle, but guidance in the Note

1.23 Throughout the feasibility study phase, there are a number of underlying issues that need to te borne in mind. These include considerations of: uncertainty and risk choice of technology institutional issues socio-economic issues environmental aspects.

show a spectacular increase in traffic over the present situation, then they should be treated with scepticism and the basis of these forecasts should be re-examined carefully. 1.30 Will there be enough inputs at the right time and of the right quality to meet the needs for the project? It is particularly important for road and transport projects to consider not only the inputs required during the design and construction phases, but also the recurrent inputs required during operation. The maintenance requirements for road projects are particularly important and will be discussed in some detail later. 1.31 How appropriate and cost-effective is the project for converting inputs into Outputs? For example, an alternative to building an all-weather rural access road may be to invest in crop storage facilities so that farmers' produce can be held until travelling conditions improve. 1.32 Does the management expertise exist to enable the project to operate efficiently? It is necessary to assess the existing numbers of staff, their skills and their availability. There may be some scope for upgrading skills and increasing the numbers of skilled staff by training as part of the project. However, it would be unwise to assume that additional staff will become available to operate the project, or that staff, once trained, will necessarily remain on the project. 1.33 If a satisfactory answer can be given to these four questions, it is likely that the project is worth analysing. The larger and more complex the project, the greater should be the effort put into the feasibility study. For particularly large oneoff projects, specialist outside expertise may be required. 1.34 Before any analysis is carried out, decisions must be made about the type of project and the proposed method of construction, etc. All these factors will have an important bearing on the engineering decisions that must be made and on the benefits to be expected.

1.24 Section 19 contains a checklist of key points to be considered during the feasibility study and is designed to assist those submitting and appraising project reports to check quickly whether all of the key issues have been included. 1.25 Case studies have been included as appendices to illustrate different aspects of the feasibility study stage. Further case studies which should be consulted are given by Adler (1987).

PROJECT TYPES Factors to be considered

1.26 It is rare for a project under consideration to be entirely new. Most projects have a history and have been awaiting the most appropriate time for them to be considered for funding. The decision to carry out a study of a project is not a neutral action. The process of examining a project arouses expectations and creates momentum. Once a decision is taken to carry out a feasibility study, it is sometimes difficult to halt the progress of a project. Rejecting a proposed study on a doubtful-looking project at an early stage is rnore important than adding sophistication to the appraisal itself to try and obtain a more objective result. 1.27 Occasionally, projects exist which are manifestly viable, such as when a bridge on a major route collapses. Detailed economic feasibility studies are necessary in such cases, although a design study might be appropriate. 1.28 For practical national planning, it is necessary to study a project in the context of the particular needs and problems of the country before proposing solutions. This will help to determine what type of project will be most useful and where the effort should best be applied. A project can normally be considered as a means of converting inputs, such as the cost of a new or improved road, into outputs, such as benefits to vehicle operators or generated traffic. In this context, the following questions should be considered prior to the start of the study to help determine priorities. 1.29 Is there an effective demand for the project? It is vital to differentiate between what people would like, and what they are prepared to pay to use. If forecasts

Upgrading and new construction

1.35 Upgrading projects aim specifically at providing additional capacity when a road is nearing the end of its design life or because there has been an unforeseen change in use of the road. Typical examples of upgrading projects are the paving of gravel roads, the provision of strengthening overlays for paved roads and the widening of roads. Upgrading projects must not be confused with maintenance, and it is important that they are not undertaken without a proper assessment of maintenance capability as described in para 2.27-34. The appraisal of upgrading projects is similar to that of new projects. In fact most new projects are essentially upgrading projects

because some form of track or lower standard road is usually in existence. Only in the case of bypasses or new roads provided to support other investment are roads likely to be needed where none existed before. 1.36 The most important point about upgrading is that the timing of the project is crucial. For example, in the case of a paved road which requires strengthening to enable it to carry an increased traffic loading, the increased strength can normally be achieved by placing an asphalt overlay on the road. If the placing of this overlay is delayed for any reason, the road may start to deteriorate very rapidly. The cost of appropriate strengthening measures will then escalate and eventually complete rehabilitation will be the only sensible engineering solution costing many times that of a timely overlay and causing greater interruption to traffic. 1.37 Upgrading from a gravel surface to a paved road will be justified principally by savings in vehicle operating costs arising from the smoother running surface, but time savings may also be important. The level of traffic at which this kind of upgrading becomes justified can be determined by manual calculations, but it is easier to use road investment models as described in para 10.5. The availability of such models makes it inappropriate to quote levels of traffic at which upgrading from gravel to paved is justified; each case can and should be studied individually to determine the optimum timing.

responsible for the need for rehabilitation, then it is unrealistic to assume that the maintenance of the reconstructed road will be carried Out any better than was the case with the original road. This will result in very rapid deterioration and this should be taken into account in the appraisal. Unless the project includes a special component which has a realistic chance of increasing maintenance performance, the feasibility of the new project will be in doubt. The comments made in para 2.27-34 about assessing the capacity of the maintenance organisation are particularly relevant to this situation. 1.41 In order to assess whether a paved road has reached the end of its design life or whether the design specification was actually met at the time of construction, it is necessary to have access to the original design and construction reports and to have reliable traffic and axle load data. 1.42 For paved roads, unfortunately, it is common for design specifications not to be met at the time of construction, particularly with regard to the quality of materials used and the thickness of pavement layers. When assessments are made of roads requiring rehabilitation, it is important that trial pits are dug and cores taken from the road pavement to enable tests to be carried out. A sufficient number of test holes must be dug to enable statistically meaningful results to be obtained and it is recommended that there should be a minimum of one hole every half kilometre on long lengths of road and that a minimum of six holes should be dug in any apparently homogeneous short section of road. More holes than this are preferable, and they should be spread across the road to cover the various wheel tracks and other features. The cores and test pits will provide information on the actual layer thicknesses achieved during construction. Material from the test pits can be tested in the laboratory to compare its characteristics with those specified for the construction. However, it is possible that some of the material properties may have changed with time. If design specifications have not been met, the same construction and supervision teams should not be used again on a project. 1.43 Although design life for economic purposes is expressed in years, for the design of the pavements of bituminous roads, it must be related to the traffic 'loading' that the road is designed to carry. This is explained in para 6.2830. If the number of vehicles using the road or the axle loads of these vehicles has been higher than was originally forecast at the time of the design, then the road can 'fail' prematurely. Traffic and axle load levels should be checked by examining records for the road and related roads where these exist but, in most cases, new classified traffic counts and axle load surveys should be carried out at the time of the appraisal of the project to provide new and up-to-date information.

Reconstruction and rehabilitation

1.38 Rehabilitation is needed if the road has deteriorated beyond the condition at which overlaying is a satisfactory engineering alternative. This may often be because the road has received insufficient maintenance over its design life. Additionally, rehabilitation may be needed because the original road was not built to the standards of quality required by the original design. Reconstruction to provide a new alignment should be considered as an upgrading project as discussed in the previous sub-section. 1.39 The rehabilitation of gravel roads is usually relatively straightforward. The existing surface will need to be reshaped and the road can then be regravelled. The rehabilitation of bituminous surfaced roads is usually more complicated and much more expensive. Usually, the surface of the road will have deteriorated to such an extent that it must be scarified and a new pavement provided. It is important that rehabilitation projects include remedial works to both the shoulders and the drainage of the road. Without these, the condition of the new pavement, is likely to deteriorate rapidly. 1.40 When rehabilitation work is being proposed, it is important to identify the reasons why the work is needed. If lack of effective maintenance is wholly or partly

Stage construction
1.44 Stage construction consists of planned improvements to the pavement standards of a road at fixed stages through the project life. Normally, the road alignment needed at the final stages of the project is provided from the outset. A typical policy will be to construct a gravel road initially which will be paved when traffic flows have reached a given level. Stage construction differs conceptually from upgrading in that any later improvements are planned from the outset, whereas upgrading projects aim specifically at providing additional capacity only when a road is nearing the end of its design life or because there has been an unforeseen change in use of the road. 1.45 When considered purely from the point of view of optimal economic balance, stage construction policies often have much to commend them. However, difficulties can arise in practice, particularly with regard to the future funding of such projects. If a stage construction policy is proposed, its viability will depend crucially on the successful implementation of the upgrading at the correct time in the project life. Experience has shown that budget constraints often prevent the later upgrading phase of stage construction projects from being funded, with the result that anticipated benefits from the project have not materialised. This has resulted in lower rates of return than were originally expected. If stage construction projects are considered, it is important to ensure that full account has been taken of the setting-up and overhead costs at the time of the upgrading as these can represent a substantial part of the contract.

1 .48 Most maintenance projects will be attempting to rectify fundamental institutional problems of a maintenance organisation with the objective of increasing the general capability or, perhaps, the capability in some specific area. Lack of capability will often result from some combination of the particular institutional situation, such as lack of management, technical skills, insufficient resources and lack of foreign exchange. Many problems are likely to be due to basic institutional factors existing outside the roads authority (see para 2.18-21) and which are not susceptible to change, even over a long period. It is vital to identify these factors clearly, and to look for means of carrying out maintenance which will avoid the effects of these factors, or at least minimise them. 1.49 If lack of maintenance has resulted in badly deteriorated roads then, because the cost of construction is already sunk and the possible savings in vehicle operating costs and deferred reconstruction are high, the potential return from road maintenance projects can be extremely large. However, the difficulty of realising these benefits, in practice, is considerable because of the difficulty of overcoming institutional factors. 1.50 The second type of project is usually easier to define and the work is relatively straightforward to execute and monitor. Benefits, and the probability of achieving the benefits, should be relatively easy to predict. However, there is a danger that this type of project will only remedy the symptoms of a problem and will do very little to rectify the cause, which will often be institutional constraints on the general capability to carry out maintenance. Short-term maintenance projects will only be appropriate in a very few cases and proposals in which they are recommended should be vetted very carefully to ensure that a deep-rooted and longterm problem is not being tackled with short-term measures. 1.51 It is important that the two types of maintenance project that have been described are not confused with projects for upgrading, rehabilitation and stage construction as described earlier.

Maintenance projects
1.46 The purpose of maintenance is to ensure that the road does not 'fail' before the end of its design life. In doing this, maintenance reduces the rate of deterioration of the road, it lowers the cost of operating vehicles on the road by providing a smooth running surface, and it keeps the road open on a continuous basis by preventing it from becoming impassable. It is a relatively low cost activity and specifically excludes those works designed to increase the strength or improve the alignment of the road. 1.47 The form of maintenance projects can be very varied, but there are two principal types of project. The first has the objective of building up the institutional capability of the maintenance organisation to carry out maintenance in an efficient manner. The second form of project has the objective of overcoming a short-term problem and includes such items as the supply of maintenance equipment or spare parts, surface dressing or regravelling a particular length of road by contract, or provision of specialist courses or technical assistance. The second type of project may be one component of the first type.

Network considerations
1.52 In general, when constructing or improving a road network where economic constraints apply, the most economical solution for one road link may not necessarily be the best solution for the network as a whole. The cost of implementing one project to high standards may consume resources that would be better spent over the whole network, or in filling other gaps in the network with lower standard roads. In those countries where the basic road network is incomplete, it will usually be appropriate to adopt a relatively low level of geometric standards in order to release resources to provide more basic road links. This policy will generally do more to foster economic development than building a smaller number of road links to a higher standard.

PRELIMINARY PROJECT CONSIDERATIONS

Setting objectives
1.53 A road project should, wherever possible, be set against the background of a national or regional transport plan or, at least, a road plan. The feasibility study process involves several steps. Firstly, the project's objectives must be defined to provide the basic framework for the analysis. Within this framework, a range of alternatives should be considered. For each, it is necessary to assess the demand, determine physical resources and costs, and to predict benefits in order to compare alternatives. This should be done in conjunction with the analysis of uncertainty in order to provide a robust recommendation. For appraisal purposes, all costs and prices are expressed in 'economic' as opposed to 'market' terms to reflect the real value of the resource to the economy.
1.54 The project objectives need to be clearly defined from the outset. The objective of providing a new road may be to support some other developmental activity, to provide fundamental links in the national or a district road network, or perhaps to meet a strategic need. Alternatively, a road improvement may be proposed to increase the structural or volumetric capacity of an existing road to cope with higher traffic flows. Often, there will be multiple objectives. Depending on the objectives of the investment, the project will be appraised against different sets of criteria.

future. For most road projects, an analysis period of 15 years from the date of opening is appropriate, but this time period should be tested by the appraisal. Such a period of analysis reduces both the problems of making forecasts for long periods into the future and the impact on the project of the size of the residual value. Choosing the same value for pavement design life in years, and the analysis period, simplifies the calculation of residual values. 1.57 When choosing design standards for a road, a fundamental decision must be made as to whether those design standards should hold only for the analysis period for which a project is being analysed or whether standards should be chosen for a shorter or longer period than this. In the past, geometric standards have effectively been chosen for a life far in excess of the economic analysis period, whereas pavement design standards have been chosen based on the actual analysis period itself, or even for a shorter period when coupled with stage construction. However, there is rarely any economic justification for providing a higher standard of geometric design than is required by the most optimistic traffic forecast for the latter years of the economic analysis period.

Analysis period and design life

1.55 The viability of a road project should not depend on the length of the economic analysis period. Long analysis periods are useful when comparing mutually exclusive projects. Short analysis periods may be appropriate for small projects such as regravelling of rural access roads, where the life of the investment is expected to be limited to a few years. 1.56 Whatever time period is chosen for the economic analysis, the project will usually have some residual value at the end of this period. The residual value can be approximated as the difference in cost between rebuilding the road at the end of the analysis period to satisfy further demand, using the structure remaining from the initial project, and the building cost if the first project were not to take place. The size of this residual value can sometimes be large, especially for structures and concrete roads, and therefore have considerable impact on the choice of project or the feasibility of a project. The problem of determining the residual value can be reduced by extending the period for which the project is appraised, but this causes additional problems of making forecasts, particularly of traffic, for long periods into the

2.

UNDERLYING ISSUES

et al. (1986). Risk analysis and the role of sensitivity analysis

are discussed in Section 17.

2.1 Several issues underly all steps of the feasibility stage of the project cycle and are discussed here.

CHOICE OF TECHNOLOGY
2.8 When building engineering projects, there is a choice between using technology dominated by mechanical equipment or dominated by labour. Most engineers are familiar with equipment based technology, so emphasis here is on the use of labour based methods. 2.9 Although the use of relatively sophisticated equipment for construction and maintenance can be very effective, it usually has two serious drawbacks for developing countries. Firstly, it has to be paid for in foreign currency and, secondly, its use tends to reduce the employment of local labour. In countries where there are high levels of unemployment or underemployment, there are often good economic, social and political reasons for using labour based methods. However, such methods are very dependent on competent management and organisation at various levels. 2.10 Labour based methods should not be used unless they are competitive from an economic point of view, which may not be identical to the financial viewpoint. Comparison of the economics of labour and equipment should ensure that foreign exchange is properly priced (see para 15.14) and that realistic estimates of equipment availability and output are used (see para 2.29). The introduction of extra cash into the local economy through the utilisation of labour based techniques may well give rise to other effects if large numbers of wage earners are involved and these effects should also be considered. 2.11 The following points should be borne in mind when considering the use of labour based construction methods: With certain qualifications and assuming certain improvements over traditional methods, labour based construction can be made competitive with other methods when the average unskilled wage rate is equivalent to about US$4.00 per day (1985) or less; in some circumstances, the competitive wage threshold can be higher than this figure It is necessary to check carefully that labour will be available in the actual place where it is wanted, and at the time it is wanted Adequate organisation and management are critical considerations for large-scale labour based works Design can be modified to suit labour based methods, but any change in benefits from the project that result must be accounted for explicitly The health and nutrition of the labour force are very important. Improvement can increase productivity, besides bringing other benefits Standardised tools of good quality are vital in order to achieve high levels of outputs

UNCERTAINTY AND RISK

2.2 All stages of the project cycle involve uncertainty and risk. Projects in developing countries are always set against a background of economic, social and political uncertainty to some degree. 2.3 The appraisal of a project involves the collection of a large amount of data and the forecasting of trends into the future. All data collected in the field are subject to errors and some can be particularly inaccurate. By the time these data have been used to make future projections, any error can be magnified significantly. When this is coupled with the uncertainties which exist in the projection process itself, the appraisal can be subject to substantial errors. 2.4 It is important to recognise that uncertainties exist and to take steps to minimse them. It is also necessary to determine the effect of uncertainty on the robustness of the conclusions reached as this may affect the final recommendations.
2.5 Projects should not only be appraised with a recognition of uncertainty, but they should be designed with it in mind in order to minimise risk. The approach that is necessary to deal with uncertainty should depend on the level of project development. If the project is well defined, 'risk analysis', is likely to be appropriate. This involves formal probability analysis of the likely range of outcomes. If the project is exploratory, with project identification as a component, then 'scenario analysis' is more appropriate.

2.6 Scenario analysis requires the examination of a range of future possibilities that might reasonably be expected to occur. Normally 3 to 5 scenarios would be examined, each reflecting an internally consistent combination of possibilities for the major socio-economic uncertainties relevant to the project. The intention of the set of scenarios is not to act as a forecast of what will occur, but to span a wide but plausible range of possibilities. Projects should be chosen on their ability to deliver a satisfactory level of service across a range of scenarios. In this way, the economic return of a project need not be the sole criterion since social and political realities can also be taken into account. 2.7 Scenario analysis is discussed in the sections on assessing traffic demand and on project design and costing. The general principles are described by Allport

There are limited prospects for a half-way stage of socalled 'intermediate technology' between the labour based and equipment based methods Supervision and management requirements are very different for the two methods of construction.

are considered with regard to the specific wage levels, labour availabilities, etc, that are applicable in the project area. 2.15 Where appropriate, feasibility studies should include the World Bank type of screening process and demonstrate that the design has been neutralised to ensure that use of the most appropriate technology is being recommended.

2.12 Before making direct cost comparisons between the two methods of working, consideration must be given to each of the above factors. It is also particularly important when carrying out feasibility studies to consider the design of the scheme because current construction procurement practices are strongly biased towards the use of equipment based methods. This is largely the result of the influence of the industrial countries where the use of advanced technology, availability of capital and high cost of labour have led to designs, specifications, conditions of contract and methods of finance which take account of these factors. These equipmentbiased practices have usually been transferred with little or no modification to labour-abundant countries and have resulted in a failure to consider utilising the natural labour resource which is widely available. A further difficulty is that engineering and training institutions in the industrialised countries expose students from development countries almost exclusively to equipment based construction concepts. 2.13 It is important when projects are being formulated that the aspects of design, specifications and contractual procedures are defined in such a way that any bias is removed and alternative construction methods can be considered on a competitive basis. This process is known as 'neutralisation' and is described in a systematic way by the World Bank in their publication: 'Guide to competitive bidding on construction projects in labour-abundant economies' (International Bank for Reconstruction and Development and Scott Wilson Kirkpatrick & Partners 1978). 2.14 In this World Bank document, it is recommended that a screening process is applied to projects that are being formulated. The object of this is to identify, in two stages, those projects which are likely to benefit from neutralisation and therefore minimise unnecessary effort on projects which for one reason or another must be constructed by mainly labour based or equipment based methods. Even where there are overriding circumstances which determine the method of working, such as government policy towards employment creation, screening should still be carried out to determine the cost of this policy on the economics of the project being considered. Screening must be carried out as early as possible in the project cycle since, once a decision has been made on the broad technology to be followed, it is difficult to change later on in the project preparation process. Most of the factors considered in the two parts of the screening process are the same but, whereas in the first stage they are considered from a country-wide point of view with a long-term perspective, in the second, they

INSTITUTIONAL ISSUES The institutional framework

2.16 The success of many projects will depend upon the institutional framework in which they are set. Aspects that need to be considered are the organisation, staffing, training, procedures, planning, maintenance, funding and controls within the agency responsible for the project, and also within other agencies who may need to make an input to the project. The activities of all agencies responsible for development projects need to be coordinated to ensure that the best use is made of the country's resources and that a project being undertaken by one agency does not undermine the likelihood of success of a project being carried out elsewhere. 2.17 For a project to be a success, it must have full local support and not be undertaken solely because of pressure from a donor or other agency. Such issues should be clarified at the project identification stage. There is also a need for the involvement of the people living in the area of a project in its formulation, and this is particularly important in the case of rural access roads.

Improving institutional development

2.18 There is often a need to strengthen the institutions associated with the implementation of projects. Particular attention should normally be paid to: Whether the agency responsible for the project is the most suitable, not only for the implementation of the project, but also for its operation after the project is complete The need to develop permanent local training courses which will assist in long term institutional expansion; training should be introduced as soon as possible in the project, rather than at the closing stages and should be aimed at policy makers, managers and technicians; it is essential that counterparts are available to be trained and that there are sufficient incentives for them to wish to be trained The agency's ability to hire and fire staff and to provide incentives to reward good performance Whether there is effective coordination between agencies involved in the road subsector and the

transport sector as a whole to ensure the optimum use of resources in these areas. 2.19 It is normally to be recommended that institutional development should be carried out as a component of a project rather than as a complementary measure. Technical assistance and tralning are specific aspects of this and should be treated as subcomponents of the institutional development component. 2.20 The time horizon for institution building projects needs to be long. It is unrealistic to expect to make fundamental changes to the workings of an organisation which are sustained in less than about 10 years. The fundamental problem behind lack of resources may be that the macro economic performance of the country is insufficient to generate the resources that are needed. The lack of trained staff may be due to an insufficient number of students graduating from schools, colleges and universities, or may be due to the lack of a work situation in which training can develop. It may also be due to an inadequate wage structure. Often a combination of several problems will apply simultaneously. 2.21 Proposals for tackling institutional problems should be exarnined to ensure that they are tackling the root cause. Well-founded projects may be in operation for several years before any measurable improvement can be observed. This must be clearly understood, not only by the people promoting and funding the project, but also by the organisation responsible for its execution. It can be expected that the final benefits from institution building projects may be substantially lower than those predicted, and this should be taken into account when the project is appraised by examining the effects of different scenarios (see para 2.6-7).

investment funds away from other priority areas and undermine the creation of proper maintenance departments which are essential in the long run. 2.23 On a well maintained road, vehicle operating costs over the project analysis period will typically be four times the size of the construction costs. Maintenance costs over the same period will be only a few per cent of the total. Lack of maintenance, or lack of effective maintenance, which leads to a road pot-holding can lead to a 15 per cent increase in vehicle operating costs. If this situation continues, and the road disintegrates, vehicle operating costs can increase by up to 50 per cent. Similarly, lack of maintenance can lead to the need for premature rehabilitation of a road. Thus, the adequacy of maintenance can have a significant effect on the feasibility of a project. 2.24 Many construction and improvement project appraisals in the past have apparently overlooked the fact that insufficient funds, personnel, equipment, materials and appropriate skills exist to carry out the necessary level of maintenance on the new project, or indeed on any other road. If maintenance fails to reach the planned standard, anticipated benefits will not be achieved which, in most cases, negates the case for the project. 2.25 It is unrealistic to assume that subsequent maintenance of a new project will be carried out any better than maintenance is carried out on the remainder of the road network. Appraisals should therefore be based on measurements of maintenance capability in the existing maintenance organisation. Unless the project includes a special component aimed at improving maintenance performance which has a realistic chance of success, the appraisal and design of a new project should be based firmly on the assumption that it will receive the same level of maintenance as is being applied to comparable roads in the country at the time the appraisal is being carried out. 2.26 For any road development project, it is therefore necessary to assess the institutional capability to carry out effective maintenance in order to appraise the project properly.

Road maintenance organisations

2.22 Although maintenance costs normally only account for a small fraction of the initial investment costs of the road, they are often an important item in the budget. This can lead to particular problems where investment costs are covered by foreign aid donors and maintenance must be financed locally. In addition, because periodic maintenance can sometimes be delayed from one year to another without immediate disastrous consequence, it is tempting for hard-pressed roads ministers to economise on this item rather than on some other element of the budget. The cumulative effect of such skimping is that road maintenance departments are frequently poorly staffed, badly trained and demoralised compared to other departments dealing with new investment. These factors often result in pressures on the financiers and consultants for projects to provide higher standards than are justified in an attempt to reduce the future maintenance burden. The adoption of higher standards than are necessary draw

Assessment of maintenance capability

2.27 There are several ways of assessing the existing maintenance capacity of a roads organisation. A competent road engineer should be able to make a subjective assessment by simply inspecting a sample of roads. Lack of effective maintenance will be shown up by the level of deterioration of the road network taken as a whole. Watching maintenance gangs working in the field will also give a good indication of the likely productivity and durability of maintenance operations.

2.28 However, appraisals should normally be expected to contain objective assessments of maintenance performance. Field measurements can be taken of several functions and these should be used in conjunction with available records from the road maintenance organisation to enable the assessment of capability to be made. Appropriate maintenance functions that can be included here are the availability and utilisation of specific items of maintenance equipment, the productivity achieved in certain maintenance activities and the frequency of carrying out periodic maintenance. In addition, the record of maintenance funding compared with the estimated amount necessary, with an assessment of work efficiency and the effectiveness of expenditure control will all give indications of maintenance capability. Feasibility studies should include assessments of maintenance capability based on field measurements in these areas. 2.29 Poor availability of maintenance equipment is almost always a sign of an inefficient maintenance organisation. Hence measurements of availability and utilisation, particularly of key times of equipment such as graders and bitumen distributors, gives a good indication as to whether a particular organisation has the capability to carry out maintenance work efficiently. Availability is measured simply as the number of working days that the equipment is in a suitable condition to work. Utilisation is the actual hours that equipment is working as a proportion of the total machine hours in a given period of time. Clearly, the actual utilisation of equipment is constrained by its availability. An efficient organisation would normally expect to have average equipment utilisations of at least 50 per cent. Availabilities and utilisations for the previous years should be obtained from records, if available. Non-availability of records may in itself be a good indicator that equipment availability is low. It is vital that field checks of equipment use are made over as long a period of time as possible. The results from the field checks should be compared with the organisation's own record as this will highlight systematic errors of recording that might invalidate these records. 2.30 In a similar way, records should be collected and compared with field measurements of productivity rates for selected maintenance activities. These should be compared with the performance standards in Table 8.2 of Overseas Road Note 1 (TRRL Overseas Unit 1987). It is important that field measurements made over a reasonable length of time are used for this assessment, and that the measured productivities take account of any unproductive time, irrespective of the reasons why this arises. It is not sufficient for the appraisal to quote standard productivity rates used by the maintenance organisation as these will normally represent ideal or target values which are unlikely to be achieved in practice. The actual productivities achieved by a reasonably efficient maintenance organisation should lie within the

range of outputs quoted in Table 8.2 of Overseas Road Note 1. 2.31 A third check can be made to assess the maintenance capability. On average, paved roads carrying average traffic levels should have been surface dressed every 5-8 years, and all gravel roads carrying about one hundred vehicles per day should have been regravelled every 5-8 years. The surface dressing and regravelling achievements for the last three years should have been checked to see if, on average, between 1/8th and 1/5th of the network has been surface dressed or regravelled in each of those years, and also whether the proportion achieved each year is remaining constant, increasing or declining. 2.32 Inspections should be made of typical lengths of paved roads using the methods and criteria recommended in Overseas Road Note 1. If maintenance has been effective, very few lengths of road should be in need of 'further investigation' as defined in Table 7.2 of Overseas Road Note 1. The existence of some corrugated gravel and earth roads is not a good indication of maintenance capability as, even with a high level of maintenance, corrugations can appear within a few days of the maintenance having been carried out. 2.33 The capability to carry out maintenance, as estimated from the above, must be compared with the requirements for the maintenance organisation. Their work programme and budget allocations should be studied, and information obtained from these should be compared with the results of field inspections in order to assess the maintenance requirement. The requirement and the assessment of capability can then be compared. 2.34 Appraisals should use the results of these observations of maintenance capability to estimate a realistic level of maintenance that can be expected to be carried out on the project, and this is the level that should be used in the assessment of feasibility of the project. The sensitivity of the costs and benefits of the project to changes in this predicted level of maintenance should also be examined.

SOCIO-ECONOMIC CONSIDERATIONS
2.35 Several socio-economic factors may influence the way that a project should be executed. Projects may produce sudden effects on economic and social activity in the surrounding area of the road, or changing demands on health services and education. Construction camps can introduce populations from outside the project area. Disease problems may arise both from those brought in by the new population, and from their potential exposure to diseases to which they do not have immunity. Depending on how the project is designed, it may increase or decrease the rate of road accidents with the possible

10

consequences of pain, grief and suffering, in addition to direct costs to the community. A check-list of socioeconomic factors follows. 2.36 Social changes. What will be the social consequences of the project and what steps can be taken to deal with these? Socio-economic advice should be sought if the project is likely to introduce sudden changes in the social and economic activity of the surrounding area. If, for instance, the project will result in the enforced movement of people's homes, the advice of a socio-economist and an anthropologist should be sought at an early stage. Effects on health services and education may also be important. 2.37 Construction consequences. Has the impact of construction on the indigenous community been considered? What are the implications for the physical environment of the local population and their settlements? What are the implications of the introduction of relatively high earning construction workers into the community? Are there any disease implications? What steps can be taken to mitigate adverse effects? 2.38 Road accidents. Has the project been formulated considering the effects on road safety? The recommendations outlined in Section 13 should be followed in order to provide a safer environment surrounding the project. 2.39 Severance. Have problems of severance been taken into account? The severance of communities by road projects may not only result in social inconvenience, but may also give rise to an increase in road accidents. 2.40 Minorities. Have any special needs of women or minority groups been taken into account in the project formulation? 2.41 Expertise and resources. Does the local design organisation and contracting industry have the in-house expertise and resources to mount a project of the nature and scale involved? If not, how can the project strengthen these by institution building or provide additional resources. 2.42 Data. Is the information on the local social environment, the site conditions and climate likely to be reliable? If not, what data should be collected under the project to increase this reliability to an acceptable level, or how can the project be formulated to reduce the level of risk?

undisturbed country, this can be profound. The impact of road improvement projects will be less, but should still be considered. 2.44 Particularly in the case of new roads, there will be a direct environmental impact along the line of construction which will encourage the spread of new settlements and agriculture into previously uncultivated areas. These indirect effects are often more potentially damaging than the project itself. Such access may encourage the depletion of forestry resources, the loss of fertile soil through desertification, the opening of mineral extraction facilities, or the extinction or depletion of species of plants and wildlife. It could be argued that environmental damage of this type resulting from improved access is not the concern of the transport planner, and should be considered at the broader planning level before individual road projects are apraised. In reality, environmental problems are likely to be ignored unless they are raised in connection with specific projects and, if there is likely to be any significant environmental impact, it is recommended that specialist advice from environmentalists and conservationists should be sought. 2.45 Geotechnical damage can often result from projects in the road subsector and steps should be taken to minimise this at an early stage in the project cycle. In hilly environments, considerable attention needs to be given to existing and relic landslide areas. While the former create serious but obvious problems for the road engineer, there are locations where former landslides have become stabilised. Cutting through these sites can reactivate the slides with serious effects for both the road and the slopes above it. Cuttings may also increase the likelihood of soil erosion from slopes above the cutting. Geomorphological mapping of the road line may identify potential risks and allow redesign to reduce the longterm maintenance costs. It is not uncommon for mountain communities to face substantial annual maintenance requirements and these should not be underestimated. Careful and appropriate planning and design, such as that outlined in Section 5, can do much to avoid or minimise problems of erosion and landslides. 2.46 The future depletion of local roadbuilding and maintenance materials should also be taken into account during the feasibility study and design stages. Most road projects will make heavy demands on gravel, sand, rock, cement and timber resources which will have to be extracted and transported to the construction site. The effects of this extraction can be considerable especially in wetlands and coastal zones. 2.47 Design of bridges and culverts normally takes account of flood discharges, but these can be difficult to estimate when data on rainfall, soil and vegetation are absent or scarce. Furthermore, the effects of land

ENVIRONMENTAL CONDITIONS Factors to be considered

2.43 Any road project will have an impact on the environment. In the case of new roads penetrating

11

clearance may increase run-off and sedimentation in stream channels, so flood peaks become higher and arrive more rapidly after rainfall. It may be necessary in some hilly areas to consider the sensitivity of the design of drainage structures to changes in land use. Roads can impede drainage and may provide suitable habitats for disease vectors. Local water supplies and washing places may be disrupted at bridging points. 2.48 Along the line of construction and around borrow-pits and quarries, there will be direct disturbance of agricultural or natural land. Consideration should be given to the likelihood of damage to wetlands and woodlands in particular, avoiding them where possible, especially when they are isolated elements of the landscape. During feasibility study surveys, it is recommended that an ecological reconnaissance is carried out to provide an initial rapid environmental appraisal which can be accommodated by the initial road design. Particular attention should be given to migration routes, not only for higher animals but for lower animals and amphibians as well. There is often a disturbance corridor along construction routes. Depending upon the terrain and the species, this may extend for about one kilometre on either side. Roads may cut off slower moving species from breeding or feeding grounds. In the latter case, this may cause local overgrazing and land degradation. Roads through existing reserves and national parks should be avoided not only because of the disturbance but also because, unless there is a strong warden system, this can encourage new settlements and more permanent disturbance. 2.49 Borrow-pits often cover large areas and are comparatively shallow. If they receive surface run-off, they may become shallow ponds and habitats for waterborne disease vectors. They may also become fish ponds or provide a new aquatic habitat, in which cases the advice of a fisheries adviser could be sought. Dry pits may also be adopted for farm land, industrial uses or develop as new habitats. 2.50 In coastal areas, roads are often routed across estuaries and mangrove forests upon embankments. These will concentrate tidal flows at bridges and changes will take place in tidal patterns and velocities. This may have 'knock-on' effects on salt-water intrusion into estuaries. pollutant dispersion and concentration, fisheries, sedimentation and disease vector habitats. 2.51 In industrialised countries, environmental problems of air pollution, noise and vibration, and visual intrusion are often considered to be significant, particularly in urban areas. Knowledge of the costs and benefits of reductions of such environmental damage in developing countries is, at present, inadequate to enable satisfactory policy guidance to be given. Nevertheless, it is always advisable to consider possible environmental effects in

qualitative terms during the feasibility studies phase and when considering any subsequent design modifications. Often, low cost changes at the feasibility stage can avoid or minimise problems later.

Environmental assessment
2.52 The general environmental pollution and damage caused by roads is closely associated with the level of economic activity. An increase in GNP is likely to lead to an increase in the environmental cost of transport, although the increased national wealth may also make it possible to raise the level of investment in measures designed to combat pollution. If a nation wishes to reduce the environmental impact of schemes, then it must be realised that there is a definite cost associated with protective measures. 2.53 Environmental benefits do not have infinite value, so it is essential to try and compare the costs and benefits of any change which will reduce the environmental impact, but increase other costs. Maximising the 'general good' or social welfare is commendable, but an impossible quantifiable task because the quality of life as a result of economic changes cannot be measured. 2.54 In view of the high degree of uncertainty about the value of environmental measures, value judgements usually have to be made to determine which measures will yield net benefits to the community. Although it is unrealistic that the 'man-in-the-street' can be involved in the details of environmental decisions, particularly in developing countries, it is important that the impact on the public at large is taken into account when adopting investment policies and plans that will impact on the environment. It is also important that the views of all socio-economic groups are considered and not just those of the affluent or vocal minority. 2.55 In the past, project appraisal in developing countries has either ignored environmental issues or has taken them into acount in a very imperfect manner. It is recommended that, when project reports are prepared, they include an environmental impact statement based on an appropriate study of the issues involved. The statement can be presented in the form of a balance sheet where all project costs and benefits are set out using monetary or physical units, where possible, or a points rating where not. The statement should also indicate who are the bearers of costs and the recipients of benefits. For unquantifiable environmental effects, the following general principles for presentation can be considered: The population exposed to the effect should be enumerated and described in terms of its level of sensitivity to the effect; the population may need to be classified where levels of sensitivities are different

12

The existing level of the effect should be identified and measured, where possible, to show the current degree of exposure of the population The anticipated level of the effect resulting from the project should then be predicted, in the same units as for the existing level, to enable comparisons to be made in terms of population affected.

ASSESSING TRAFFIC DEMAND

TYPES OF TRAFFIC
3.1 For the purposes of geometric design and the evaluation of economic benefits, the volume and composition of current and future traffic needs to be known in terms of cars, light goods vehicles, trucks, buses, non-motorised vehicles, etc. For the structural design purposes of paved roads, because the lighter vehicles contribute so little to pavement damage, they can be ignored and only the number and axle loading of the heavier vehicles need be considered. 3.2 In order to assess benefits, it is also necessary to separate traffic into the following three categories: Normal traffic. Traffic which would pass along the existing road being considered by the project if no investment took place, including normal growth. Diverted traffic. Traffic that changes from another route (or mode) to the project road, but still travels between the same origin and destination (this is termed 'reassigned' traffic in transport modelling). Generated traffic. Additional traffic which occurs in response to the provision or improvement of a road (this includes 'redistributed' traffic as defined in transport models).

2.56 The following checklist includes those issues that should be considered in an environmental assessment. 2.57 Consequential developments. Will the project stimulate land clearance for agriculture, the development of industry or mineral extraction? What steps can be taken to mitigate long term adverse effects? 2.58 Geotechnical damage. Has the project been designed to minimise the possibility of landslides and other geotechnical problems? Have long term maintenance consequences been taken into account? 2.59 Material resources. Will the project result in the unacceptable depletion of material resources that may be needed for subsequent maintenance or other construction projects? Will borrow pits be restored and can their effect on the landscape be minimised? 2.60 Drainage. Will the project result in increased risks from flooding or landslides as a result of disturbing natural drainage patterns? Will consequential development of agricultural land and other settlements affect hydrological conditions sufficiently to require modification to drainage works and bridges? What will be the effects on coastal habitats and landforms, especially swamps, lagoons and mangroves? Both reclamation and the effects of embankments on estuarine hydrology need consideration, as to disturbance and increased sedimentation during construction. Will any water impoundments create health hazards? 2.61 Ecology. Have the effects on animals and plants been considered? Has an ecological reconnaissance been carried out to assess effects? 2.62 Other factors. Are the problems of air pollution, noise and vibration, and visual intrusion matters for concern? If so, what can be done to mitigate their effects?

These categories are each treated separately in an economic appraisal.

BASELINE TRAFFIC FLOWS Traffic counts

3.3 The first step in assessing demand is to estimate baseline traffic flows. The estimate used should be the Annual Average Daily Traffic (ADT) of traffic currently using the route, classified into vehicle categories such as those described above. This is the total annual traffic in both directions divided by 365, typically obtained by recording actual traffic flows over a specific shorter period from which the ADT is estimated. 3.4 Traffic counts carried out over a short period as a basis for estimating the traffic flow can produce estimates which are subject to large errors because traffic flows can have large daily, weekly, monthly and seasonal variations. The daily variability in traffic flow depends on the volume of traffic, increasing as traffic levels fall, and with high variability on roads carrying less than

13

1000 vehicles per day. Traffic flows vary more from day-today than week-to-week over the year, so that there are large errors associated with estimating annual traffic flows (and subsequently annual average daily traffic) from traffic counts of a few days duration, or excluding the weekend. For the same reason, there is a rapid fall in the likely size of error as the duration of counting period increases up to one week, but there is a marked decrease in the reduction of error for counts of longer duration. Traffic flows also vary from month-tomonth, so that a weekly traffic count repeated at intervals during the year provides a better base for estimating the annual volume of traffic than a continuous traffic count of the same length. Traffic also varies considerably through the day, but this is unlikely to affect the estimate of ADT providing sufficient hours are covered by the daily counts.

Traffic counters
3.8 Classified traffic counts are normally obtained by counting manually. However, these counts can be supplemented by the use of automatic traffic counters. The commonest types of counter use either a pneumatic tube laid across the carriageway or a loop of wire buried beneath the road surface. Buried loops have a longer life than pneumatic tubes, but can be difficult and expensive to install, especially on a paved road. Pneumatic tubes are easy to install, but require regular maintenance and are subject to vandalism. For short duration counting on a paved road, a loop can be fixed to the carriageway surface (Kember-Smith 1984). 3.9 In their basic form, automatic counters do not distinguish between different types of vehicle, so they cannot provide a classified count. Modern detector systems are now becoming available which can perform classified vehicle counting, but such systems are expensive and not yet considered to be robust enough for most developing country applications.

3.5 In order to reduce the magnitude of errors, it is recommended that traffic counts to establish ADT at a specific site conform to the following practice:

the counts are for seven consecutive days. the counts on some of the days are for a full 24 hours, with preferably one 24 hour count on a weekday and one during a weekend. On the other days, 16 hour counts should be sufficient. These should be grossed up to 24 hour values in the same proportion as the 16 hour/24 hours split on those days when full 24 hour counts have been undertaken. counts are avoided on roads at times when travel activity increases abnormally due to the payment of wages and salaries, or at harvest time, public holidays, etc, or on any occasion when traffic is abnormally high or low. if possible, the seven day counts should be repeated several times throughout the year.

Moving observer counts

3.10 In order to obtain a broad impression of the changes in vehicle flows along a length of road, moving observer counts can be carried out. These can be used to guide the number and choice of sites for carrying out the detailed traffic counts.

3.11 A hand-tally can be used to record the number of vehicles. The flow can be estimated from the expression: (x + y) q= t where q x y t = total flow in both directions in time t = number of vehicles met (ie travelling in the opposite direction) = number of vehicles that overtake the observer minus the number he overtakes = journey time

3.6 Country-wide traffic data must be collected on a systematic basis to enable seasonal trends in traffic flows to be quantified. Unfortunately, many of the counts that are available are unreliable. Therefore, where seasonal adjustment factors are applied to raw traffic data from a traffic survey in order to improve the accuracy of baseline traffic figures, the quality of the statistics on which they are based should be rechecked in the field.

3.7 When determining design flows for bridge or other projects close to urban communities, it is usually appropriate to base the design on peak hourly flows in order to take account of commuter traffic. However, delays must be considered when costs and benefits are assessed, as allowing traffic to queue may be the most economic solution.

3.12 This expression assumes that flows in each direction are equal. If the observer can drive so that he passes as many vehicles as pass him in the direction of travel (this can usually be achieved without risk at low traffic volumes) then the expression becomes: x q= t 3.13 Counts of this type are also useful as a cross-check on static counts and also to assess the extent of variations in flow from day to day or between one season and another.

14

TRAFFIC FORECASTING
3.14 Even with a developed economy and stable economic conditions, traffic forecasting is an uncertain process. In a developing economy, the problem becomes more intractable. The economies of developing countries are often very sensitive to world prices of just one or two particular commodities, and fluctuations in world oil price and supply over the last decade have added a new dimension to the difficulties. Despite the uncertainty, the economic viability of a road project is often very sensitive to the forecast traffic levels.

component of fuel costs and vehicle import restrictions could tend to depress future growth rates.

Diverted traffic
3.19 Where parallel routes exist, traffic will usually travel on the quickest and cheapest route, although this may not necessarily be the shortest. Thus, surfacing an existing road may divert traffic from a parallel and shorter route because higher speeds are possible on the surfaced road. Origin and destination surveys should be carried out to provide data which can be used to estimate likely traffic diversions. Assignment of diverted traffic is normally done by an 'all-ornothing' method in which it is assumed that all vehicles that would save time or money by diverting would do so, and that vehicles that would lose time or increase costs would not transfer. With such a method, it is important that all perceived costs are included. In some of the more developed countries, there may be scope for modelling different scenarios using standard assignment computer programs.

Normal traffic
3.15 The commonest method of forecasting the growth of normal traffic is to extrapolate time series data on traffic levels and assume that growth will either remain constant in absolute terms (a linear extrapolation) or constant in relative terms (a constant elasticity extrapolation) ie traffic growth will be a fixed number of vehicles per year or a fixed percentage increase. Data on fuel sales can often be used as a guide to country-wide growth in traffic levels although improvements in fuel economy over time should be taken into account. As a general rule, it is only safe to extrapolate forward for as many years as reliable traffic data exist from the past, and for as many years forward that the same general economic conditions are expected to continue.

3.16 As an alternative to time, growth can be related linearly to GDP. This is normally preferable, since it explicitly takes into account changes in overall economic activity, but it has the disadvantage that, in order to use the relationship for forecasting, a forecast of GDP is needed. The use of additional variables such as population or fuel price brings with it the same problem. If GDP forecasts are not available, then future traffic growth should be based on time series data.

3.20 Diversion frorh other transport modes, such as rail or water, is not so easy to forecast or deal with. Transport of bulk commodities will normally be by the cheapest mode, though this may not be the quickest. However, quality of service, speed and convenience are valued by intending consignors and, for general goods, diversion from other modes should not be estimated solely on the basis of door-todoor transport charges. Similarly, the choice of mode for passenger transport should not be judged purely on the basis of travel charges. The importance attached to quality of service by users has been a major contributory factor to the worldwide decline in rail transport over recent years.

3.21 Diverted traffic is normaily forecast to grow at the same rate as traffic on the road or mode from which it diverted.

3.17 If it is thought that a particular component of the traffic will grow at a different rate to the rest, then it should be specifically identified and dealt with separately. For example, there may be a plan to expand a local township or open a local factory during the life of the project, either of which could lead to different growth rates for different types of vehicle, or there may be a plan to allow heavier freight vehicles on the road, in which case the growth rate of truck numbers may be relatively low because each truck is heavier.

Generated traffic
3.22 Generated traffic arises either because a journey becomes more attractive because of a cost or time reduction, or because of the increased development that is brought about by a road investment. It is difficult to forecast accurately and can be easily overestimated. It is only likely to be significant in those cases where the road investment brings about large reductions in transport costs. For example, in the case of a small improvement within an already developed highway system, generated traffic will be small and can normally be ignored. Similarly, for projects involving the improvement of short lengths of rural roads and tracks, there will usually be little generated traffic. However, in the case of a new

3.18 Whatever the forecasting procedure used, it is essential to consider the realism of forecast future levels. Few developing countries are likely to sustain the high rates of growth experienced in the past, even in the short term, and factors such as the high foreign exchange

15

road allowing access to a hitherto undeveloped area, there could be large reductions in transport costs as a result of changing mode from head4oading to motor vehicle transport and, in this case, generated traffic could be the main component of future traffic flow 3.23 'Producer surplus' models exist for forecasting generated traffic based on the anticipated response of farmers to road investment. However, the predictive accuracy of these models is poor and a major limitation to their use is that they consider only agricultural freight, which typically accounts for less than ten per cent of road traffic. Road traffic in rural areas is usually dominated by personal travel. 3.24 The recommended approach to forecasting generated traffic is to use demand relationships. The price elasticity of demand for transport measures the responsiveness of traffic to a change in transport costs following a road investment. On inter-urban roads, a distinction is normally drawn between passenger and freight traffic and, on roads providing access to rural areas, a further distinction is usually made between agricultural and non-agricultural freight traffic. 3.25 Evidence from several evaluation studies carried out in developing countries give a range of between - 0.6 to - 2.0 for the price elasticity of demand for transport, with an average of about 1.0. This means that a one per cent decrease in transport costs leads to a one per cent increase in traffic. Calculations should be based on door-to-door travel costs estimated as a result of origin and destination surveys, and not just on that part of the trip incurred on the road under study. Generally, this implies that the reduction in travel cost and increase in traffic will be smaller than measurements on the road alone suggest. 3.26 The available evidence suggests that the elasticity of demand for passenger travel is usually slightly greater than unity. In general, the elasticity of demand for goods is much lower and depends on the proportion of transport costs in the commodity price. However, the ability to market or process some crops is very dependent on the availability of good road access.

traffic growth Will increase the uncertainty in the project's economic return even further. 3.28 Particularly bearing in mind the large errors that can be associated with both traffic counting and forecasting, it is vital that considerable attention is paid to the quality and duration of the data collection in this area. In addition, scenario analysis should always be carried out to determine the effects of errors in traffic counts and forecasts on the final recommendations. Projects should normally be analysed using both 'optimistic' (high) and 'pessimistic' (low) levels of future traffic in addition to the scenario of the best estimate.

UNCERTAINTY OF TRAFFIC ESTIMATES

3.27 Estimates of baseline traffic flows and traffic growth rates will inevitably be subject to degrees of error. Errors in both of these parameters will have a great impact on the estimated economic rate of return of a road project. It is difficult to estimate baseline traffic flow to within about 20 per cent and, typically, an error this size will give an error of a similar order of magnitude in the net present value (see para 15.19-23). Errors in estimation of

16

4.

COST ESTIMATION

PURPOSE OF THE ESTIMATE

4.1 The objective of the estimate is to provide the most realistic prediction possible of the total cash expenditure and time that will be necessary to complete the project ready for operation. Cost estimates provided for road projects in developing countries in the past have generally proved to be wildly inaccurate, with two main results: Tendered contract bids have often proved to be considerably higher than the engineer's feasibility study estimate Considerable cost overruns have occurred during projection execution.

deficient in several important areas. This section recommends the use of analytical techniques and rigorous procedures of risk management to produce realistic estimates of cost at all stages of the project cycle. Expected values of project costs and construction periods should reflect past experience that actual values achieved have normally been far in excess of those estimated originally, particularly those estimates produced at the early stages of project preparation. It is worth expending considerable effort to produce realistic estimates of cost at all stages of the project.

COST ESTIMATING STAGES

4.3 The stages of a project in which estimates of cost and time may be needed are illustrated in Table 4.1. The result required at each stage and the information normally available are set out. It is appreciated that, in particular projects, some stages may be omitted or be indistinguishable from adjacent stages.

4.2 Traditionally, unit pricing techniques have been used for project costing, but these have been shown to be

TABLE 4.1 PROJECT STAGES

Stage Identification Activities Identification of project Feasibility Appraisal of the identified project, including basic requirements, alternative schemes, and recommendation of preferred scheme. Normally requires a costbenefit analysis. Recommendation of preferred scheme Design Definition of preferred scheme including basic design data, conceptual design, technical specifications, construction appraisal, contract strategies and estimate of final cash cost. Project definition report for use in a submission for funding Commitment Consideration of submission for funding Implementation Implementation of approved project including - detailed design - issue of tender enquiries - assessment of tenders - placing of contracts - construction - completion - commissioning Basis for assessment of tenders and ongoing monitoring of costs and progress against approved estimate Tender documents Operation/ evaluation Operation of new asset by client. Evaluation of project.

Result

Inclusion in forward programme

Funding approved for defined project. Basis for cost control

Historical cost and productivity data bank

Available information for estimate

No design capacity/size only

Preliminary designs of alternatives

Conceptual design

Conceptual design

Completed contract

17

4.4 It is important to strive for the ideal of evolving a cost history of the project from start to finish with an estimated cash total at each stage near to the eventual achieved cost. This ideal can only be approached throughout the life of the project if the rising level of definition is balanced by reducing tolerances and contingency allowances which are effectively the measure of uncertainty. Each estimate should be directly comparable with its predecessor in a form suitable for cost monitoring during implementation. 4.5 The sequence of estimates throughout the life of a typical project is given below.

This estimate therefore would use the same information as is available to the tendering contractors and should be a good basis for the assessment of bids.

6.

Post contract award

A further refinement of the approved estimate in the light of the contract(s) awarded. It includes redistribution of the monies within the approved total to allow more effective cost monitoring of the project to completion.

7. 1. Preliminary
The quick estimate needed at the project identification stage, with no design available, and only the barest statement of capacity or size.

Achieved cost

A record of the actual costs achieved in order to review the cost performance of the project and for project evaluation. It should include a reconciliation of the actual use of contingencies and of the use of tolerance for dealing with major risks.

2.

Feasibility THE ESTIMATOR

4.6 The estimator must have relevant experience in the type of project envisaged and, wherever possible, in the costs and productivities of construction work at the proposed construction and main supply locations. 4.7 The estimator must have a close working relationship with the project design organisation and will normally be part of that organisation. It is essential that he is able to appreciate the conception and purpose of the project and the intentions of the design, and can easily investigate and clarify any uncertainties with the designers as they arise during the compilation of the estimate. The estimator must also be able to contact the client, the funding agency, visit the site and have access to any local information relevant to the estimate. 4.8 It is highly desirable that the same estimator is used on all the estimates required during the life of the project and is responsible for the subsequent cost monitoring and control. This clearly depends on the continuity achieved in the design organisation. When a new estimator has to be appointed, for whatever reason, or when a check estimate is required from a separate estimator, then it is important to ensure an orderly transfer to him of all relevant information so that the new estimator is able to become accountable for his estimate and the subsequent cost control against it. 4.9 The estimator should be accountable for his estimate and should be involved in the subsequent monitoring of project costs against it. He should be responsible for employing the estimating technique most appropriate for the type of project and the stage of the project. In reaching this decision, he should note the advice on the various techniques, their strengths and weaknesses and their sources of data.

Estimates or alternative schemes under consideration in the feasibility study stage of the project. The essential property of these estimates is that they are directly comparable with each other and therefore base estimates could suffice so long as the same estimating technique and price base data are used. The differences between alternatives will not necessarily be absolute and the danger of their use for forward budgeting must be avoided.

3.

Design

The cost estimate for the selected scheme using the design (usually conceptual) and specifications resulting from the design study and forming part of the project definition report. This estimate would provide the figures for capital cost, cash flow and currency requirements which would then be used in viability calculations for the project and in the submission for donor aid, where appropriate. It must be a cash estimate.

4.

Commitment

The proposal estimate as modified and approved for financing, together with the associated modifications to the project definition and/or the programme. This must be a cash estimate, and will provide the basis for the cost control of the project.

5.

Pre-tender

A refinement of the approved estimate in the light of further design work done during the tender period and using the information given in the enquiry documents.

18

INFORMATION IN THE ESTIMATE

4.10 The information required by the estimator should include the following: The latest description of the intended project including all available drawings, specifications, job descriptions and the site location The intended/required start and completion dates and latest programmes Latest ideas on method of construction Sources of project funding with dates of availability Latest ideas on contract strategy and availability of resources together with any prescribed restrictions of choice Any papers or reports describing performance and problems encountered on similar projects in similar locations Any cost/productivity data relating to the project or current construction projects in the' host country.

4.15 Sample size The data must be from a sufficiently large sample of similar work in a similar location and constructed in similar circumstances. 4.16 Base date Cost data needs to be related to a specific historical date, chosen with care. In the case of construction work carried out over a period of time, an appropriate 'mean' date has to be chosen eg two-thirds through the period. 4.17 Price indices Having selected the relevant price base date, there remains the problem of updating the price to the base date for the estimate. The only practical method is to use an inflation index, but there may not be a sufficiently specific index for the work in question. If there is not, recourse to general indices is usually made. In any event, there is a limited length of time, which probably does not exceed five years, over which such updating has any credibility, particularly in times of high inflation. 4.18 Market effects Overlying the general effect of inflation, is the influence of the 'Market Place'. This will vary with the type of project being undertaken, with the host country and also with the supplying countries. The state of the world economy at the price base date will require careful consideration before historical data can be credibly applied to a later or future date.

4.11 The essential documents to be submitted by the estimator will be: Summary of estimate, together with any further documents necessary for explanation A list of documents and drawings used in compiling the estimate A programme for the project showing key dates.

Global
4.19 This term describes the 'broadest brush' category of technique which relies on libraries of achieved costs of similar projects related to the overall size or capacity of the asset provided. This technique may also be known as 'rule of thumb' or 'ballpark' estimating. Examples are: cost per metre or km of roads cost per square metre of bridge deck areas or per cubic metre of mass concrete.

In all cases the estimator should also be required to submit: A method of construction A contract strategy report.

4.12 Each section of the estimate should be compiled in the working currency envisaged for that section at prices current at a stated price base date. The consequent base estimates will be converted to cash estimates by the use of inflation indices, selected by the estimator, in conjunction with the project programme. Where a funding agency is involved, all cash estimates should be converted to the currency used by that agency using a stated exchange rate.

4.20 The technique relies entirely on historical data and therefore must be used in conjunction with inflation indices and a judgement of the market place influence to allow for the envisaged location and timing of the project. 4.21 The use of this type of rolled up historical data for estimating is beset with dangers, especially inflation, as outlined generally above. The following specific dangers apply. 4.22 Varying definitions of what costs are included engineering fees and expenses by consultants/ contractors/client, including design, construction supervision, procurement and commissioning

ESTIMATING TECHNIQUES The techniques available

4.13 The four basic estimating techniques available to meet the project needs outlined above are summarised, together with the data required for their application, in Table 4.2. 4.14 The global and unit rate techniques rely on historical data of various kinds. Comments on this aspect of each technique are given under the respective headings, but the associated dangers are so critical that it is worth making the following general warning points about the use of historical data in estimating.

19

TABLE 4.2 ESTIMATING COSTS Global Project Data required - Size/capacity - Location - Completion date Man hours - Quantities - Location - Key dates - Simple method statement - Completion date Unit rate - Bill of quantities (at least main items) - Location - Completion date Operation - Materials quantities - Method statement - Programme - Key dates - Completion date

Basic estimating data required

- Achieved overall costs of similar projects (adequately defined)

- Hourly rates - Productivities - Overheads - Materials costs

- Historical unit rates for similar work items - Preliminaries

- Inflation indices - Market trends - General inflation forecasts

- Hourly rate forecasts - Materials costs forecasts

- Inflation indices - Market trends - General inflation forecasts - Plant data

- Labour rates and productivities - Plant costs and productivities - Material costs - Overhead costs - Labour rate forecasts - Materials costs forecasts - Plant capital and operating costs forecast

List of potential problems, risks, uncertainties and peculiarities of the project 4.23

final accounts of all contracts including settlements of claims and any other payments land acquisition costs transport costs of materials financing costs taxes, duties etc. Varying definitions of measurement of the unit of capacity is a metre/kilometre of road an overall average including pro rata costs of bridges or should these be estimated separately? square metre of bridge deck area: including or excluding the cost of abutments?; cubic metre of mass concrete in bridges: height measured from top of ground or top of foundations?

4.26 A scrutiny of all these dangers, especially the effects of inflation, must be made before any reliance can be placed on a collection of data of this type. It follows that the most reliable data banks are those maintained for a specific organisation where there is confidence in the management of the data. The wider the source of the data, the greater is the risk of differences in definition. 4.27 However, so long as care is taken in the choice of data, the global technique is probably as reliable as an over-hasty estimate assembled from more detailed unit rates drawn from separate unrelated sources and applied to 'guesstimates' of quantities.

Man hours
4.28 This is most suitable for labour based construction and erection work where there exist reliable records of productivity of different trades per man hour. The total man hours estimated for a given operation are then costed at the current labour rates and added to the costs of materials and equipment. The advantages of working in current costs are obtained. 4.29 The technique is often used without a detailed programme on the assumption that the methods of construction will not vary from project to project. Experience has shown, however, that where they do vary (eg due to the capacity of heavy lifting equipment available), labour productivities and consequently the total

4.24 Not comparing like with like differing levels of quality such as different pavement thicknesses for different levels of traffic different terrain and ground conditions such as roads across flat plains compared with mountainous regions different logistics depending on site location item prices taken out of total contract prices may be distorted by front end loading eg fob prices for hard currency items. 4.25 Inflation different cost base dates - it is essential to record the 'mean' base date for the achieved cost and use appropriate indices to adjust to the forecast date required.

20

cost can be affected significantly. It is recommended that a detailed programme is prepared when using this technique. The prediction of cash flow requires such a programme.

carried out by a tendering contracter, for any but the simplest of jobs. Neither does it encourage consideration of the particular peculiarities, requirements, constraints and risks affecting the project. 4.35 There is a real danger that the precision and detail of the individual rates can generate a misplaced level of confidence in the figures. It must not be assumed that the previous work was of the same nature, carried out in identical conditions and with the same duration. The duration of the work will have a significant effect on the cost. Many construction costs are time related, as are the fees of supervisory staff and all are affected by inflation. 4.36 It is therefore recommended that a programme embracing mobilisation and construction is prepared. This should be used to produce a check estimate in simplified operational form where there is any doubt about the realism of the unit rates available. 4.37 Nevertheless, unit rate estimating can result in reliable estimates when practised by experienced estimators with good, intuitive judgement and the ability to assess the realistic programme and circumstances of the work.

Unit rates
4.30 This technique is based on the traditional bill of quantity approach to pricing construction work. In its most detailed form, a bill will be available containing the quantities of work to be constructed, measured in accordance with an appropriate method of measurement. The estimator selects historical rates or prices for each item in the bill using either information from recent similar contracts or published information (eg price books for civil engineering), or 'built-up' rates from his own analysis of the operations, plant and materials required for the measured item. As the technique relies on historical data, it is subject to the general dangers outlined earlier. 4.31 When a detailed bill is not available, quantities will be required for the main items of work and these will be priced using 'rolled up' rates which take account of the associated minor items. Taken to an extreme, the cruder unit rate estimates come into the area of global estimates as described above (eg unit rate per metre of road). 4.32 The technique is most appropriate to repetitive work where the allocation of costs to specific operations is reasonably well defined and operational risks are easily manageable. It is less appropriate for civil engineering works where the method of construction is variable and where the uncertainties of ground conditions are significant. It is also likely to be less than successful for engineering projects in locations where few similar schemes have been completed in the past. In these cases, success depends much more on the experience of the estimator and his access to a well understood data bank of relevant 'rolled up' rates. 4.33 Unit rates quoted by contractors in their tenders are not necessarily related directly to the items of work they are pricing. It is common practice for a tenderer to distribute the monies included in his tender across the items in the bill to meet objectives such as cash flow and anticipated changes in volume of work. It is likely that similar weighting is carried out by all tenderers in an enquiry and therefore it is not easily detected. It follows that tendered bill unit rates are not necessarily reliable guides to prices for the work described. 4.34 The technique does not demand an examination of the programme and method of construction and the estimate is compiled by the direct application of historical 'prices'. It therefore does not provide an analysis of the real costs of work, of the kind that would need to be

Operational (resource-cost)
4.38 This is the fundamental estimating technique since the total cost of the work is compiled from consideration of the constituent operations or activities revealed by the method statement and programme, and from the accumulated demand for resources. The advantages of working in current costs are obtained because labour, plant and materials are costed at current rates. 4.39 The most difficult data to obtain are the productivities of labour and construction plant in the geographical location of the project and especially in the circumstances of the specific activity under consideration. Claimed outputs of plant are obtainable from suppliers, but these need to be reviewed in the light of actual experience. Labour productivities will vary from site to site depending on management, organisation, industrial relations, site conditions, etc and also from country to country. Productivity information is a significant part of the 'know how' of a contractor and will naturally be jealously guarded. 4.40 The operational technique is particularly valuable where there are significant uncertainties and risks. Because the technique exposes the basic sources of costs, the sensitivities of the estimate to alternative assumptions/methods can be investigated and the reasons for variations in cost appreciated. It also provides a detailed current cost/time basis for the application of inflation forecasts and hence the compilation of a project cash flow.

21

4.41 In particular, the operational technique for estimating holds the best chance of identifying risks of delay as it involves the preparation of a method of construction and a sequential programme including an appreciation of productivities. Sensitivity analyses can be carried out to determine the most vulnerable operations and appropriate allowances included. Action to reduce the effect of risks should be taken where possible.

Suitability of individual techniques

4.42 At the identification stage, the absence of all but the simplest definition of the project means that only the global' technique can be applied. However, estimating organisations which regularly use operational techniques state that even their crudest overall data are recorded in such a way that the effects of the more obvious uncertainties can be allowed for at this early stage. Clearly the availability of a reliable, well managed, global cost data bank together with associated 'broad brush' analyses is an essential requirement for any organisation involved in the early identification of projects for inclusion in a forward financial programme. 4.43 The essential activity in the feasibility stage of a project is the consideration of many alternatives. The most important characteristic of the estimating technique employed is, therefore, reliable comparability between the alternative schemes, which may be numerous. The technique must also be usable with only preliminary data for the schemes, as the conceptual design will normally be in its very early stages. The most appropriate techniques would be: Global Unit rate, using 'rolled up' unit rates for the main items of work.

For all 'one-off' jobs, there is no credible alternative to operational estimating Accuracy in all estimates depends heavily on a clear definition of scope, the extent of use of local information and on the definition of uncertainties and potential problems There is considerable merit in using an alternative approach to prepare a 'validation' estimate; any differences between the main and validation estimates must be satisfactorily reconciled. An estimate submitted at any stage of a project should be subject to review It is recommended that all submitted estimates should include a carefully considered programme for the work; if this is omitted, there is a reduced likelihood that the effects of risk, delay and inflation will be properly considered It is vital that any modifications to the estimate are backed up by a depth of study not less than the depth of the original estimator's own investigations.

SOURCES OF DATA Principal sources

4.47 Estimating data are normally obtained from three principal sources: Project-specific data collected for a particular project and therefore related to a specific location and time Data banks of previous or current projects collected by an individual estimator or estimating organisation Published data.

4.44 However, if the state of information is good enough and time and funds are available, then the operational technique should be considered even at this stage, especially for plant intensive construction, when the key requirement is to identify the major resources and cost them over their period of required availability. 4.45 As soon as sufficient design detail is available at the appraisal stage, the first preference should be to use the operational technique. Exceptions may be labour based projects, where the man hours technique would be appropriate. In cases where, for whatever reason, there is insufficient time, funds or data available for the operational method, the unit rate technique may have to be resorted to. 4.46 When considering estimating techniques, the following factors should be kept firmly in the foreground:

4.48 It is important that the estimator is directly involved in the collection of these data. This involvement should include visits to the project location and any other appropriate sites. It should also include a search for any significant risks, peculiarities or constraints to which the project may be subject and any factors which might affect the method of construction.

Project specific data

4.49 These data are in two forms: current costs of basic inputs to the work productivity outputs relevant to the type of work, its location and the particular circumstances surrounding it.

4.50 The basic inputs for which the estimator must collect current cost data include: hour wage rates other labour costs and overheads construction plant purchase prices and/or hire rates

22

management, supervisory and administrative salary rates prices of materials prices of services and utilities transport, shipping and freight charges import duties taxes insurances interest rates.

The applicability of data to the specific circumstances, including location and duration, of the project being estimated.

4.55 Wherever it is necessary to access data banks collected by others, the credibility of the information obtained should be assessed using the same criteria. 4.56 The data recorded should be considered in two basic categories: cost data and productivity data. 4.57 Cost data The majority of these data will be historical, derived from compiled prices, and will include, in increasing order of detail: global unit costs related to size of project; eg cost per kilometre of roads 'rolled up' unit rates for main items of work rates for standard bill of quantity items.

4.51 Current cost data should be obtained from the sources closest to the initiation of potential price changes. The estimator must have accurate information on the prices and costs ruling in the market place at the base date assumed for the estimate. The sources must be local to the activity in question and will include: Government institutions Public works departments Contracting organisations (both local and experienced offshore) Consultants (both local and experienced offshore) Aid or development agencies Trade missions Shipping agencies Construction plant and materials manufacturers or importers Transport companies, etc.

4.58 All types of historical cost data must be related to a date from which the subsequent inflation can be estimated, normally using published indices. In addition, historical data must be assessed against changes in the market place over time. It follows that greater weight should be given to the most recent data available, such as that from current projects. 4.59 These data are for use in the global and unit rate estimating techniques. 4.60 Productivity data These data cover outputs and possibly utilisation figures for labour and construction plant. They will be related to specific operations in a known location and in defined circumstances. It is recommended that data are collected in the form of: output of work achieved in unit paid time by production units utilisation figures for the production units.

All these sources are subject to error and the estimator must continually and critically assess their relevance to the specific project. 4.52 Credible productivity data and current cost data of basic inputs are essential information for the compilation of operational and man-hours estimates. They may also be used, with care, to revalidate data for global and unit rate techniques. 4.53 The translation of the base estimate to a cash estimate requires information on inflation and exchange rates which will be normally available from government sources, financial institutions and publications.

4.61 Such data can be collected at several levels, examples of which are given below in increasing order of detail: a. b. a histogram of major resources available, coupled with the main quantities of work achieved information of the form 'x number of machines of y capacity were employed for z weeks to remove V m3 of clay material', together with utilisation figures for the period the output of work achieved on a specific task by a labour gang or item of equipment, together with overall utilisation figures for the production unit.

Data banks
4.54 Each estimating organisation can be expected to maintain a record of the costs and times achieved in the projects in which it has been involved. In-house data banks are more reliable than data banks collected by others as the management and interpretation of the information is within the control of the organisation and therefore consistency in its application should be assured. However, the reliability of any data bank depends on: The size of the sample available The acceptance throughout the organisation of standard methods of measurement and definitions of terms The recording of any special factors and circumstances which affected the performance of the recorded project

c.

4.62 As the level of detail increases, so does the importance of clear definition and consistent use of the terms used for productivity measurement. In particular, the distinction and relation between output and utilisation must be recognised.

23

4.63 It is recommended that levels (a) and (b) are the most appropriate for the initial collection of productivity data. They are the most readily usable for client estimates, since they allow for downtime over a significant period of time. As data are collected from an increasing number of projects, it should become possible to reconcile differences between them. This will be facilitated if a record is kept of the major factors affecting output and utilisation. 4.64 Such productivity data are unaffected by inflation and therefore can be applied in the operational and man-hours estimating techniques, wherever similar circumstances for the operation are foreseen.

Lump sum Admeasure

Cost-reimbursable

Target cost

payment based on a single price for the total work payment for quantities of completed work, valued at tendered rates in a bill of quantities payment for actual cost (requires 'open book' accounting) plus fee for overheads and profit payment based on actual cost plus fee plus incentive.

Choice of one or other type will be largely dictated by the perception of financial risk. 4.69 Price based contracts In lump sum and admeasure contracts, the contractor bears much of the risk and has to price his tender accordingly. When, as may frequently be the case in developing countries, the risks are high, this results in extremely high prices or in offshore contractors being reluctant to tender at all. 4.70 Cost based contracts In a cost-reimbursable contract, the client and donor will bear the main risks, whereas the intention of a target cost contract is to price the work excluding risk, the cost of which is borne by the client. Target cost contracts introduce an incentive for the contractor to work efficiently, aligning his objectives with that of the client to achieve the construction cost-effectively. Flexibility is desirable under the uncertain conditions found in many developing countries. Where the contract is managed with 'open book' accounting, this provides the opportunity, in theory, for client, consultant and contractor to discuss design modifications when these may be desirable. However, such methods of contract are very demanding of senior site staff for both consultant and contractor. 4.71 Where there is an active construction industry, tendering mobilises competition to good advantage. However, to be truly successful, the tender procedure also depends on there being a precise, comprehensive specification and a thorough design. Where the above conditions are not met, it is likely that negotiation of a cost-plus form of contract is more likely to be practical. A mixed approach is often adoped with initial tendering followed by negotiation on some aspects of the contract. 4.72 There are strong developmental arguments for using local construction capacity where this is available. Not only will this capacity usually be much less costly to mobilise, but the experience gained will strengthen the industry and help the country to be self-sufficient. There will also often be multiplier effects on other sectors of the economy. Some donors treat local contractors preferentially. For instance, the World Bank will consider local bids acceptable even if they are 15 per cent more expensive than those from offshore contractors.

Published data
4.65 Wherever required data are not available from inhouse resources, or from specific resources related to the project, the estimator may have to resort to published data. Such data must be used with caution and thorough inquiry made into its basis and the circumstances of its achievement. For instance, the unit rates quoted for a common building activity in the United Kingdom in a range of established publications have been found to vary by - 50% to + 150% from the mean. Other studies have shown that equipment outputs as low as 20 per cent to 30 per cent of the manufacturer's published data might be expected, particularly in developing countries. The estimator is responsible for judging the credibility of any published data he may decide to use. 4.66 Some relevant sources of published data are given below, although there are many other useful publications: 'Caterpillar Performance Handbook'. Caterpillar Tractor Co. Annual editions. The most comprehensive catalogue of plant and its probable performance. 'Contractors Plant Association Handbook'. Contractors Plant Association, London. Current UK hire rates and details of selected plant items are given in the current annual edition. 'Civil Engineering and Public Works Review' publish a quarterly selection of unit rates for items of civil engineering construction in the UK.

CONTRACT STRATEGY
4.67 The choice of construction contract strategy could have a major bearing on the cost and therefore the feasibility of the project. It is therefore appropriate to consider this choice at an early stage. 4.68 The following four types of contract can be used:

24

THE DESIGN PROCESS AND PROJECT COSTS

4.73 For anything other than 'global' cost estimating, it will be necessary to base costs on a design which has been carried out to an appropriate level of detail. The road design process consists of four distinct phases: route location, the structural design of the pavement, the geometric design, and the design of structures including bridges, culverts and earth retaining works.

4.74 Route location applies only to new roads and is discussed in Section 5. It is that stage of the design process where the general positioning of the route or its 'corridor' is determined. The corridor should be chosen to meet the objectives of the project which normally implies a balance between maintaining user benefits by providing the shortest route consistent with engineering constraints of topography, ground conditions, drainage, structures and the location of road building materials.

4.79 The relative contribution that each aspect of design makes to the total costs of the project depends mainly on the various design standards adopted and on the type of terrain through which the road passes. To provide a particular geometric standard in hilly terrain requires the construction of cuts and fills involving the movement of earth or rock, much of which would be unnecessary in flat terrain. On the other hand, the quantity of high quality road making material required to provide the necessary structural strength in the pavement will not usually depend on the terrain. The choice of design standards should be based on an economic analysis, but can only be made within practical limits of design, construction and consistency. It is an iterative process and it is not possible to provide general guidelines or rules of thumb on the relative costs associated with each aspect of design. Each case will need to be assessed on its merits with the objective of optimising the relationship between costs and benefits. 4.80 It is expected that road engineers will already have detailed knowledge of the methods used to design and cost projects at the various stages of their development and implementation. The following four sections on geotechnics, pavement design, geometric design, and drainage and structures are therefore written primarily for transport planners, economists and administrators, and others who may not have experience of practical road engineering.

4.75 The structural design of the road is the process in which the various layers of the pavement are selected so that they are capable of supporting the traffic for as long as required. The principal elements in this process are the choice of materials and their thickness for each pavement layer. This is dealt with in Section 6, where the basic principles of structural design are described and key features pertinent to the appraisal process are highlighted.

4.76 Section 7 covers the topic of geometric design which is the process whereby the detailed layout of the road in the terrain is designed to meet the needs of the road user. The principal elements of this process are the selection of suitable horizontal and vertical alignments and road widths.

4.77 Finally, the design of bridges and structures is concerned with a variety of topics, which include structures which allow the road to cross or be crossed by rivers, water courses, railways and other roads: it is concerned with earth retaining systems, drainage systems and special provisions for erosion control. These issues are discussed in Section 8.

4.78 In general, the four elements of design can be considered separately although, in some areas, there are overlaps and particular subjects need to be considered under more than one heading. For example, the width of road shoulders has implications for both geometric design and structural design. Other important examples are the consideration of road maintenance and the provision of adequate drainage which affect all three elements of design. Where overlap exists, the requirements can sometimes conflict and compromises are necessary. Topics of this kind are specifically mentioned in the text.

25

26
ground conditions as they influence the strength of the soil beneath the road (the subgrade) or present instability problems or natural hazards materials used in construction, including quarried rock earthworks (the volume and stability of cuttings and embankments) surface and sub-surface drainage, including erosion the need for structures.

GEOTECHNICS

OBJECT OF GEOTECHNICAL SURVEYS

5.1 Geotechnical surveys are required principally for projects for new roads, although they still have some application for projects concerned with existing roads. They are usually carried out at three stages in the project preparation process: identification (also called 'reconnaissance' by geotechnical engineers), feasibility, and design. As the project proceeds through these stages, geotechnical information needs to be collected at greater levels of detail. 5.2 The amount of geotechnical work that needs to be carried out will depend on whether the road is new and the alignment has to be selected, or whether the project is only concerned with upgrading or reconstruction of an existing road. In projects for new roads, the geotechnical surveys are usually carried out to select and compare alternative routes for the road and general 'corridor' studies will be required. Information on the supporting ground for the road, earthworks, bridge sites, drainage and materials (including water) for construction are some of the important features which have to be considered. For upgrading and reconstruction projects, geotechnical information is needed to determine the choice and properties of materials that are available for use in pavement construction. Where existing roads have been damaged or are threatened by ground instability, geotechnical information is needed to effect the necessary repairs or provide suitable protection. 5.3 For all projects, information from geotechnical surveys provides the basis for much of the costing of the engineering works. The level of information should be appropriate to the level of accuracy required to estimate costs at the different project stages. It is important that any factors which could have a major impact on engineering costs during or after construction should be identified at an early stage. The actual costs of geotechnical surveys varies depending on the standard of road to be built, the complexity of the conditions encountered, and many other factors, but sums between 0.5 and 3 per cent of the total project cost are typical.

5.5 The emphasis placed on these different factors will vary with the stage of the survey. The choice of route is normally associated with the identification and feasibility stages. One of the major objectives of these stages is to identify critical factors which could have a major impact on engineering costs and therefore deserve extra study at the early stages of the project. Changes to the design that must be made during the course of a construction contract often involve considerable disruption, delay and expense, and are normally the result of insufficient preliminary survey work.

Subgrade strength
5.6 Predictions have to be made about the strength of the subgrade after construction. Areas of very weak soil such as swamps should be identified at early stages of survey and avoided wherever possible. Having rejected the obvious areas, the next decision is to assess the relative subgrade strengths of alternative alignments and their implication on the pavement design. Details of the pavement design process are described in Section 6. 5.7 For route alignment, a method of construction should be suggested that provides an appropriate structure for the design life of the project in relation to subgrade strength and available construction materials. At the early stages of survey, any factors which might have a major influence on route alignment should be identified to guide the more detailed design surveys that will be carried out later. In some areas, construction materials may be abundant, but there may be considerable variation in the strength of the soil. In other areas, soil conditions may be fairly uniform, but known sources of materials are scarce, and so the emphasis should be on finding alternative sources to minimise the cost of haulage. 5.8 The subgrade strength depends on the soil type and moisture condition and its importance depends to some extent on the standard of the road. For a low volume road, it may be economic to lengthen the route to avoid difficult soils and thus to minimise construction costs. On a heavily trafficked road, where the shortest route will produce the highest user benefits, it may be economic to import material to the road line to blanket areas of weak soil. The surveys at the feasibility stage should identify the alternatives and the design stage should quantify the costs. The types of feature that need to be investigated

INFORMATION NEEDED Route location

5.4 Route location consists of selecting the best compromise between 'demand' factors and 'terrain' factors. Demand factors determine the areas to be served and the road standard, and terrain factors influence the engineering cost. The principal terrain factors are:

are areas of unstable Soil, such as expansive black clays, or areas liable to flooding where the road embankment may need to be raised. Problems such as these may give rise to maintenance problems if they are not catered for at the design stage. Other problems that need to be identified are erodible soils which may require expensive anti-scour structures, spring lines and perched water tables that may cause local recurrent failures, or features associated with particular terrain, such as the need for protection against sand dunes covering the road, or corrosive water which attacks concrete.

construction materials are scarce, soil stabilisation may sometimes be used to improve the quality of local materials. 5.11 Soils and aggregates are the most important materials in construction, but water can also be vital. Many construction jobs have been delayed because of an underestimate of the supply of water conveniently available for construction. Construction can be phased to make best use of natural moisture in the materials. In arid areas near the coast, sea water can sometimes be used for compaction, but care must be taken to ensure that levels of salt contamination do not rise above acceptable limits. Dry compaction methods should always be considered in arid areas. 5.12 Regravelling of gravel roads is normally needed about every five to eight years to replace material lost from the surface. The effect of this could be to exhaust sources of material or to increase haul distances if the better materials are used first. Good quality material may be required at a later stage in the road's life if the standard is improved to meet increased traffic demands. Increased costs for regravelling may result in it becoming economic to pave roads at lower traffic levels than would otherwise be the case.

Materials
5.9 Having decided the strength of the foundation, the pavement design process defines the thickness and properties of the separate layers. Before this can be done, information must be obtained on the nature and engineering properties of the road building materials available in the area. Previous experience in the area may assist with this, but often a survey will be necessary. The materials required include rocks suitable for crushing and natural aggregates such as gravels and sands. The specifications for these materials depend on the type of road being constructed and are discussed more fully in Section 6. The purpose of the geotechnical survey is to identify sources of the materials within an economic haulage distance and to ensure that they exist in sufficient quantity and are of sufficient quality for the purposes intended. This process often requires an extensive programme of site and laboratory testing, especially if materials are of marginal quality or occur in small quantities. However, it is important that the answers to these two questions should be very reliable before the contract for the road construction is awarded. Two of the most common reasons for construction costs to escalate is that, once construction has started and material sources fully explored, the material is found to be deficient in quality or quantity leading to expensive delays whilst new sources are investigated, or the road is redesigned to take account of the actual materials available. It is not possible to give general guidelines on the level of detail required for such surveys and specialist advice should be obtained. 5.10 A reconnaissance survey at the project identification stage should indicate the main types of construction materials and estimate if they are likely to be in short supply. A feasibility study can then identify the sources of material with estimates of quantities. It should be borne in mind that different parts of the road structure require materials of different quality. Sufficient testing must be carried out to identify substandard or variable materials. If problems do exist, then extra effort will be needed in the final design when all sources should be identified and proved by digging pits to the full depth of the gravel layer or drilling rock to the full depth of the proposed working. The siting and frequency of such excavations will be decided by an engineer according to site conditions and existing knowledge of material variability. Where good

Earthworks
5.13 Earthworks always form a significant part in the cost of road construction, as even a simple road in flat terrain involves the excavation of ditches, and the formation of a small embankment. When the terrain is not flat, cuttings are required, and their design can greatly affect the cost of earthworks. In tropical areas, shallow cuttings of less than about 3 metres with a steep slope are normally preferred to minimise earthworks and to reduce the area exposed to erosion by rain water. In certain areas, such cuttings are unstable and will need to be set back to a flatter slope. The increase in land acquisition costs is not normally a significant factor. 5.14 In areas of steeper terrain, the expense of earthworks increases and the design engineer has to select the best combination of vertical alignment, horizontal alignment and earthworks costs. Shortening of the road and increasing earthworks can be offset by shorter construction time, and taken in conjunction with reduced operating costs, this can sometimes provide a better solution. The transport planner should be able to provide the engineer with an estimate of the likely benefits in road user cost savings of shortening the road on a per metre basis (Sections 10, 12 and 13) to help guide the design. However, increased earthworks can lead to increased risk of landsliding and the higher standard of road leads not only to a higher cost of repairing such damage, but also a higher cost of delay to traffic if the road is cut completely. 5.15 In deep cuttings, it is likely that rock will be encountered, although in tropical environments a much deeper zone of weathered material may occur in comparison

27

with temperate climates. Information from the desk study or site investigation at identification or feasibility stage should indicate the general geology of the area and the type of rock that Will be present. Rock usually requires blasting and this will be more costly than soil excavation. On the other hand, rock can be cut to steeper slopes than soil, thus reducing the volume of excavation needed and often providing aggregate for use in building the road. When exposed in steep cuttings, metamorphic and sedimentary rocks, are often more prone to instability and failure along joint planes and faults than igneous rocks. The presence of discontinuities or faults should be noted at an early stage as this could be a key feature affecting route alignment. Certain types of rocks may be very hard and apparently durable when first opened up in a cutting but, after being exposed to tropical weathering for a few months, become extremely soft and unstable. The angle of this cut face may need to be reduced to take account of this. Fine-grained sedimentary rocks are very prone to rapid weathering, as can be igneous and metamorphic rocks. Local enquiry and observation of existing road cuttings should be made to identify such rocks. 5.16 An important factor in any earthworks operation, especially in steep terrain or areas where there is evidence of former landslide activity, is that the construction work could upset a delicate equilibrium in terms of ground stability. Rainfall can trigger landslips either by draining into the slope or by causing erosion and it may be necessary at the design stage to take preventative measures to avoid the occurrence of failures. The heterogeneous nature of soils and rocks, as well as a shortage of time available to monitor slope conditions, often makes it difficult to apply theoretical design measures for cut slopes and full advantage should be taken of the observed performance of other slopes in similar conditions in the region.

Structures
5.19 One of the first factors to consider in route alignment is the need for major structures such as bridges or large culverts. The very high cost of such structures is reduced by siting the structure at the most favourable point in the terrain. The cost saving thereby produced may be sufficient to justify shifting the alignment away from the most desirable course for many kilometers on each side of the structure. These 'nodal' points on the route will be located at the identification stage and may affect costings of the whole project. 5.20 A road diverts considerable quantities of water from one part of a slope to another. Care must be taken not to overload a stream with excess water. Stream courses in mountainous areas are often only marginally stable, and the addition of water, poured in from a road drainage system during a tropical thunderstorm, can cause grave erosion from which the watercourse may never recover.

SOURCES OF INFORMATION
5.21 Much of the geotechnical information needed will be obtained from field studies and site investigations, but existing information may be available locally in the country concerned or from international documentation centres. Information that should be sought includes topography, geology, soils, hydrology, vegetation, land-use, earthquake activity and climate in the region. Any constraints which may affect the engineering appraisal such as political, economic, environmental or timing restrictions should also be noted. More specific engineering information may be available from other reports on road projects in similar areas. 5.22 Often in developing countries, there is a scarcity of published information and the desk study should reveal where deficiencies occur. Information from maps can be considerably enhanced by the use of aerial photography and satellite imagery. The use of such remote sensing techniques should never be overlooked in road planning studies as they provide the facility to study the project from a broad to a detailed level. They may also show developments that have taken place after the publication of a map. The main benefit from using aerial photographs and satellite imagery is that unmapped and remote areas can be studied in detail and considerable savings in time in the field can be made. Expensive oversights of advantageous terrain conditions or unsuspected natural hazards can be avoided. A skilled interpreter with geological training should be a member of the survey team, but engineers without formal training in photo interpretation are still able to obtain valuable information.
5.23 Landsat satellite imagery is available for the whole world apart from persistently cloudy regions. The images,

Drainage
5.17 Drainage is vital to the successful performance of a road. The choice of a ridge route will minimise the number of culverts needed and also lead to better foundation conditions. Allied with drainage is the problem of erosion and, depending on soil type, climate and site conditions, anti-erosion measures may be needed on embankment faces, cuttings, culverts, side drains and stream crossings. Some of these may only be temporary measures after construction until vegetation grows and establishes stability, or it may be necessary to install permanent solutions to long term problems. 5.18 A particular hazard in steep terrain is the indiscriminate dumping of spoil on the hillslope below the road or into a convenient stream. The spoil can erode, or may wet up and slide in a mass. Material is carried downslope and may cause scour of watercourses or bury stable vegetated or agricultural land. Material choking stream beds causes the stream to meander from side to side, undercutting the banks and creating further instability.

28

in colour, depict terrain and drainage Systems over very large areas (185 x 185 km per image). They are commonly studied at scales of 1:100 000 to 1:1 000 000 and, as such, are most useful at desk study and project identification stages of investigation. They also show changes in surface features dating back to 1972, by repeated coverage of the same area. Change in major river flow patterns, retreating coastlines, or deforestation can be observed in this way. Since early 1986, the French satellite 'SPOT' has also been collecting images of the earth's surface. These are similar to Landsat images but have the advantage of higher resolution, nominally 25 m in colour or 10 m in black and white, depending on availability. 5.24 Aerial photographs are generally flown at scales ranging from 1:20 000 to 1:60 000, in black and white. Their chief advantage is in giving a highly detailed view of the terrain. When studied with the aid of a stereoscope, the ground surface is seen in full three-dimensional relief. Even sub-surface features, such as tilted or folded rock strata, solution hollows in limestone, or buried gravel deposits, can he interpreted from aerial photographs by the effect they have on surface features. 5.25 As an extension of air photo and satellite image interpretation, terrain evaluation methods have been developed which enable all types of engineering information to be incorporated into a systematic mapping scheme describing the terrain and its attributes (TRRL 1978).

should be used to interpret boundaries between terrain types, where changes in topography, geology, drainage pattern or vegetation (land use) occur. A change in any of these will give rise to different engineering conditions, which could affect the design of the road. Such items as the following should be considered: changing course of major rivers catchment areas of major river systems extent of flooding of low-lying areas possible sources of water for construction possible sources of construction materials pattern of regional instability extent of erosion spread of deforestation assessment of land acquisition/site clearance problems location of all possible bridge sites.

Feasibility
5.30 At this stage, the corridors are appraised to select the best route. This should be carried out mainly using air photos for all detailed interpretations, ideally at a scale of 1:20 0001:60 000, as available. These can be supplemented by colour information from Landsat images. 5.31 Detailed interpretations should be made of conditions on all routes and, if necessary, a more detailed terrain classification of the area should be made. The following items should be investigated: foundation conditions catchment areas and the location of culverts location of spoil areas and possible borrow areas possible sources of construction materials identification of most favourable bridge sites possible major hazard areas such as poorly drained soils, spring lines, unstable areas, erosion in river courses.

STEPS IN THE SURVEY PROCESS

5.26 The steps required to provide geotechnical information at each stage of a road project are described in Appendix 2 of 'Terrain evaluation for highway engineering and transport planning' (TRRL 1978). The way that these steps can be augmented by the use of widely-available remote sensing techniques is summarised in Table 9.2 of 'Remote sensing for highway engineering projects in developing countries' (Lawrance and Beaven 1985). The steps in the survey process are summarised in the following sub-sections.

Project identification
5.27 The purpose of this reconnaissance stage of the survey process is to identify possible alternative routes in terms of the 'corridors' within which they lie. 5.28 Possible routes should be examined on maps, satellite images and air photo mosaics, where available, and a broad terrain classification should be made for collation of the regional information, possibly on a data storage system. Visits should be made to site to check interpretations and findings should be summarised to assist in planning the next stage. 5.29 Air photo mosaics at a scale of approximately 1:100 000 and Landsat images at 1:500 000-1:250 000

5.32 Site investigations should be carried out of alternative routes, guided by the terrain evaluation. These should note key physical and geotechnical features. Selected laboratory and field testing should also be carried out, again guided by the terrain evaluation. If more detailed information is required, specialised air photography may need to be commissioned at a scale appropriate to the size of the task and degree of ground complexity (approximately 1:10 000-1:30 000). 5.33 Finally, cost comparisons should be made of alternative alignments to assist in the recommendation of the best route.

Design for project implementation

5.34 The final stage of the geotechnical survey process is to make detailed field studies of the selected route to enable a design to be carried out to engineering standards.

29

5.35 A further land classification should be carried out at a more detailed level on the selected route corridor, commissioning special photography, if required. Detailed air photo interpretation should be continued in support of all field activities to help plan a comprehensive site investigation of the selected route, which should be carried out with a full sampling and testing programme. This should examine:
5.36

TABLE 5.1 Approximate man-days of effort required to accomplish a geotechnical survey Stage of survey Identification Easy terrain Difficult terrain

construction materials subgrade conditions cuttings and embankinents areas of instability erosion and soft ground requirements for frequency and size of culverts bridge sites

Area (per 100 km2)

1 interpretation of remotely sensed images 1 field work 1 interpretation 1 field work

1 interpretation

3 field work 1 interpretation 2 field work

Line (per 25 km) Feasibility study Area

The geotechnical survey phase of the appraisal process concludes by preparing detailed designs and cost estimates.

COSTS AND ACCURACIES OF GEOTECHNICAL SURVEYS

5.37 Table 5.1 indicates the approximate amount of effort in man-days required to carry out geotechnical investigations at each of the principal stages of road survey. Because of the extreme variety of conditions under which geotechnical surveys are carried out, it is impossible to specify the costs of such surveys. The table represents a range of 'typical' conditions, which have been simplified into two types of project: one covering an area such as might be demarcated for a network of rural access roads, and one linear which is more appropriate for surveys along a single road alignment. The table also shows the difference between effort required in easy terrain and in difficult terrain as defined below. 5.38 Time is allocated both to field work, and to interpretation of aerial photographs and satellite images in advance of field work. Time spent on interpretation is considered essential for economic use of field time. The following assumptions are made about the time allocations in the table: Time taken to collect together maps and photographs, including photographic processing and computer processing, is excluded. Remote sensing imagery is used at all stages. Thus, time spent on interpreting images at the second and third stages is assumed to be minimal because of experience during the earlier stages of the surveys.

1 interpretation 3 field work 1 interpretation 3 field work

1 interpretation 5 field work 1 interpretation 4 field work

Line

Design Area 2 interpretation 10 field work 2 interpretation 10 field work 3 interpretation 20 field work 4 interpretation 25 field work

Line

Terrain EASY. Easy to moderate access by vehicle; good visibility on the ground; flat to rolling topography; simple geology and geomorphology (arrangement of surface land forms such as slopes and drainage patterns). DIFFICULT. Difficult or no access by car; poor visibility on the ground; hilly topography; complex geology and geomorphology. In mountainous areas, or areas covered by jungle, conditions tend to be extremely arduous and field work is very slow. Remote sensing is of value only at a coarse level in jungle terrain.

Project type AREA. A unit area of 100 km2 has been taken. The time taken to cover larger areas increases in linear proportion. LINE. A unit length of 25 km has been taken. 'Line' applies to projects such as single road projects that are essentially linear in nature. The principles of survey, as far as a geotechnical engineer is concerned, do not differ from area surveys, but a linear unit of measurement is more appropriate. Note that an effective geotechnical survey is never restricted to the centreline of the road, but always includes excursions of possibly several kilometres into the country on either side to check on sources of construction material, alternative alignments, off-road hazards etc.

Definitions
The foflowing terms are defined to represent the extremes of a range, with continuous intergrading between. Allowance must be made for combinations of factors when interpreting the table.

30

PAVEMENT DESIGN

EARTH ROADS
6.4 Earth roads have no added pavement and layers are therefore not structurally designed. Their performance depends very strongly on their cross-sectional shape (see Section 7), material properties, location in the terrain (see Section 5) and the drainage facilities incorporated in the design (see Section 8). With very low traffic roads, the most important consideration is whether or not the road is passable, since very high costs may be associated with the road being closed. Consideration should be given to the provision of simple drainage structures and local gravelling and improvements to provide all weather access where appropriate. The engineering aspects of earth roads will not be discussed further.

STRUCTURAL CLASSIFICATION
6.1 The structural or pavement design of the road is the process in which the various layers of the pavement are selected so that they are capable of supporting the traffic for as long as required. The principal elements in this process are the choice of materials and their thickness for each pavement layer. For structural design, roads can be classified as follows: Unimproved earth roads and tracks Gravel surfaced roads Roads incorporating pavement quality concrete or 'rigid' pavements Roads incorporating bituminous materials or 'flexible' pavements.

GRAVEL ROADS
6.5 Roads may be surfaced with gravel to provide traction for vehicles in wet weather at relatively low cost. Surfacing with gravel also retards the increase in deformation of the surface, but regular reshaping is needed as part of recurrent maintenance activities. Even when badly deformed, gravel roads can normally carry traffic successfully as drivers try to avoid deformed areas by choosing different wheelpaths, but vehicle operating costs will be increased considerably as gravel roads deteriorate. Gravel roads normally have properly designed and built drainage structures compared with earth roads and provide all-weather instead of seasonal access. 6.6 Gravel roads are rarely designed in the structural sense. Within the normal range of conditions, differences in performance which can be attributed to gravel thickness are not pronounced except on very weak subgrades. Usually a fixed thickness of gravel (150 mm or

6.2 The aims in designing a pavement are to protect the natural ground, or subgrade, from the high and concentrated load stresses which would be applied to the subgrade by the wheels of vehicles, whilst at the same time ensuring that the pavement layers are strong enough to support the traffic loads. Since the imposed load stresses are higher nearer to the wheel and the road surface, the traditional type of construction consists of various layers of material with the weakest layer at the bottom and the strongest layer at the top. 6.3 There are various ways of describing the pavement layers and this has often led to confusion. Figure 6.1 illustrates the most common method. The most important layers are the surface layers and the road base since these need to be the strongest.

Fig.6.1 Pavement layer description

31

200 mm) is used irrespective of climate, subgrade strength or traffic loading conditions, and this is replenished periodically as it is worn away. Rates of gravel loss are of the order of 2030 mm thickness a year per 100 vehicles per day, but this will vary depending on local materials and conditions. The gravel itself should be selected on the basis of its material properties and its expected behaviour under the climatic conditions prevailing. Recommended specifications are given in Tables 3 and 4 of Overseas Road Note 2 (TRRL Overseas Unit 1985). 6.7 If traffic volumes are high, total vehicle operating costs will rise rapidly as the road deteriorates and rates of gravel loss will be correspondingly large. Under these circumstances, there may be some justification for increasing the gravel thickness, but it is often cheaper to provide a surface dressed road. Considerable information has been gained in recent years on the performance of gravel roads under a variety of conditions. It is now possible to estimate the total transport costs associated with a gravel road including vehicle operating costs, maintenance costs and regravelling costs under a variety of traffic, climatic and maintenance conditions using road investment models. Their use is described in Section 10. The traffic level at which a bituminous surface is justified will depend on many factors including the expected rate of gravel loss and the cost of hauling gravel, and can range from 50-800 vehicles per day. It is not possible to give general guidelines for this and each case must be studied individually on its merits using an investment model. 6.8 As with earth roads, the performance of gravel roads depends very strongly on their positioning in the terrain (see Section 5), their cross-sectional shape (see Section 7) and the adequacy of drainage facilities (see Section 8). The engineering aspects of gravel roads will not be discussed further.

RIGID PAVEMENTS
6.11 Feasibility studies for paved roads should normally consider the potential for building in concrete. Particularly in those countries which manufacture their own cement, but import bitumen, the initial costs of construction are likely to be comparable. Even where the initial cost of construction is higher than for a comparable bituminous surfaced road, the reduced maintenance requirement over the design life may make this type of construction more economic in the long term. This should be considered particularly in those countries experiencing difficulties maintaining their road network to an econotnic standard (see para 2.22-26). It is also probable that the riding quality of concrete, although initially rougher than on bituminous roads, will deteriorate much less, so that future vehicle operating costs will not increase so rapidly. 6.12 A further advantage of concrete roads is that they can be built by labour based methods using skills and technology learned in the building trade. The introduction of concrete technology in the road building sector can also do much to develop local skills and offers scope for the fostering of local contracting industries. 6.13 However, the benefits associated with concrete roads will only be obtained if they are well constructed; if not, remedial works are much more costly than for bituminous roads and vehicle operating costs on a very deteriorated concrete road are likely to be high. Attempts should be made to quantify these longer term effects when comparing the lfietime costs of bituminous and concrete roads. These issues are discussed in Sections 10 and 11. 6.14 No design methods have been produced specifically for concrete roads in developing countries in the tropics. Until such time as a specific method is available, designs should be based on either the ASSHTO (1974), CPCA (1984) or the TRRL (Mayhew and Harding 1987) methods.

CHOICE OF PAVED ROAD CONSTRUCTION TYPE

6.9 Where a paved road is necessary, there are two basic types of construction that can be used. In the past, flexible pavements with an asphalt surfacing have normally been used in most tropical countries because they have provided a more economic solution. However, with variations in oil prices affecting the cost of using bitumen, the cost of using rigid pavements constructed with portland cement concrete has become more competitive, particularly in those countries having their own cement manufacturing capability. 6.10 The choice between flexible and rigid pavements should be made on considerations of the likely cost of both construction and maintenance, the pavement life and effect on road user costs.

FLEXIBLE PAVEMENT MATERIALS Surfacings

6.15 The essential requirement of all bituminous surfacings is that they should be waterproof. They should also provide a skid resistant surface. Surfacings do not necessarily have to perform a load spreading function because this can often be done by the underlying structural layers. 6.16 The surfacing is the most expensive of all the layers and therefore needs to be kept as thin as possible commensurate with the stresses that it can withstand and the tolerances on thickness which can be achieved with

32

the construction methods and materials chosen. The following list of surfacing materials is not complete, but it includes the principal types that are met in practice. 6.17 Surface treatments. The simplest type of surfacing is a surface dressing consisting of a thin layer of bitumen into which single sized stone chippings are rolled. This type of surfacing is very flexible and provides a reas9nably waterproof seal. Depending on traffic and climatic conditions a single, double or even triple surface dressing may be used. A surface dressing is too thin to provide any structural strength. Other similar surfacings whose main function is to waterproof the road surface are sand seals (sand plus bitumen), slurry seals (graded fine aggregate or sand plus bitumen emulsion) and a combination of slurry seal and surface dressing, often called a cape seal. 6.18 Structural surfacings. There are many types of surfacings which provide substantial structural strength to a pavement. These consist of precisely defined mixtures of bitumen, coarse and fine aggregate, sand and fine material called filler. In order to make them properly, it is usually necessary to mix the constituents in specialised plant and hence the materials are generally known as premix or plant mix. However, in some countries, lower quality materials are often made by rnlxing on the road itself or by the side of the road, usually by a more labour intensive method. Such methods can be useful for producing patching material, but are rarely practicable for surfacing or resurfacing. The principal types of premixed structural surfacings are as follows: Hot rolled asphalt (HRA). This type of mix has been used extensively in the United Kingdom. It derives its strength from the properties of a mortar of bitumen, sand and filler. Larger stones are added to the mix mainly to act as an extender. HBA is easier to make successfully than some of the other mixes but has not been used extensively in hot countries because of fears that under hot conditions and heavy traffic it will deform more easily than other mixes. However the deformation properties of HRA can be controlled in the mix design process and can be verified by simple laboratory tests at elevated temperatures. Provided that suitable sand is available, the use of HItA should be encouraged since it is resistant to cracking and therefore provides a more resilient water proof surfacing than other mix types. Asphnltlc concrete (AC). Asphaltic concrete is the most common surfacing material in use on heavily trafficked roads in developing countries. Asphaltic concrete was developed in the USA and derives much of its strength from the interlocking of angular particles within the particle/bitumen matrix. All sizes of particle need to be present in precisely the right proportions to ensure a satisfactory mix. It is more

difficult to make than HRA because the proportions of each sized particle need to be more accurately controlled. It can be made very stiff or strong to reduce the risk of deformation occurring at high temperatures, but it is intrinsically rather brittle and thus liable to crack under heavy traffic loads, allowing water to penetrate the road base. Bitumen macadams. These mixes are similar to AC, deriving much of their strength from the interlocking of angular particles. Dense bitumen macadam (DBM) is suitable as a wearing course. Open textured mixes are suitable as the base course of a surfacing or as road bases (see Figure 6.1) and in other situations where their permeability is of no consequence, such as regulating courses under strengthening overlays on roads which have deformed excessively.

6.19 Mix-in-place surfacings. In some countries, mix-inplace and hand mixed surfacings are constructed for use both on trunk roads as well as more minor roads. The results are not easy to control and the methods are often wasteful in their use of bitumen. Their use is not recommended.

Road bases
6.20 The road base is generally the main structural element of a road. Base materials are conveniently divided into three categories. Unbound bases. Unbound materials are the most common in developing countries. The materials should be a mechanically stable mixture of angular particles of different sizes ranging from about 50 mm in diameter down to dust. Usually rock or gravel needs to be crushed for this purpose although some natural gravels are suitable. It is important that the fine particles should not cause too much weakening of the base when wet, hence they should have little or no clay present. The most common type of unbound base is graded crushed stone, or 'wet mix', but other types, such as dry bound macadam and water bound macadam which, despite their names, are two types of unbound aggregate base are frequently encountered. Cement or lime stabilised bases. If unbound material of suitable strength is not available, use can be made of material which is inadequate in some way. To do this, the material is strengthened and improved by the addition of cement or lime. Not all materials are suitable for lime stabilisation as clay nunerals are necessary in the soil for the stabilisation reaction to occur. For both cement and lime stabilisation to be effective, the material to be stabilised should not be too uniform in size and should be free from organic matter.

33

Bitumen stabilised bnses. Bitumen stabilisation is rarely used for lower grade aggregates in road bases because other alternatives are usually cheaper and more reliable. If bitumen is used in bases at all, it is usually because a high strength, high quality pavement is justified and, in such a situation, good quality aggregates will be used to make a premix. One exception to this general rule occurs in areas where there are no aggregates available. Here bitumen stabilised sand is an alternative which can be used successfully for moderate traffic.

FACTORS AFFECTING FLEXIBLE PAVEMENT DESIGN

6.25 The structural design of road pavements depends primarily on the following factors: strength of the subgrade traffic loading materials variability and uncertainty in the above three items and in the quality control of the construction process.

Sub-bases and other pavement layers

6.21 The quality of material used for sub-base does not need to be as high as for bases. Usually the material is required to meet few selection criteria. The most common materials for use as sub-bases are naturally occurring (unmodified) gravels and gravel-sand-clay mixtures. Sometimes cement or lime stabilised soils are used. Selected fill material and 'capping' layers are of still lower quality and are usually selected on the basis of a simple strength test to ensure a platform of minimum guaranteed strength on which to build the pavement proper.

6.26 In addition, the structural performance of the road will depend on the adequacy of drainage measures within the road structure, the design of the shoulders and the level of maintenance.

Subgrade strength
6.27 The most important factor which controls the pavement thickness is the strength of the subgrade soil. This, in turn, depends on the type of soil, its moisture content and the level of compaction (density) achieved during construction. The thickness of pavement required to carry a particular traffic level is very sensitive to subgrade strength when the subgrade is weak, but insensitive to subgrade strength when the subgrade is very strong. The strength of the subgrade can change with time as a result of moisture changes in the soil. Such changes are often associated with poor maintenance and are therefore unpredictable. Designers often include substantial safety factors at this stage of the design process. It is important to estimate the strength of the subgrade under the most likely adverse conditions and guidance on how this can be done is given in Road Note 31 (Transport and Road Research Laboratory 1977 -currently under revision).

Use of marginal materials

6.22 Specifications for pavement materials used in developing countries have normally been copied from those used in the more industrialised countries. These original specifications have usually been evolved to overcome different climatic and loading conditions to those found in developing countries, such as the need to reduce frost damage. Local experience sometimes suggests that standard specifications can be relaxed to make use of materials that are marginal in quality, but are abundant and relatively cheap to use. The need to do this will be dictated by a lack of conventional materials or a need to build a lower cost road. 6.23 Consulting engineers are often reluctant to allow the use of marginal materials and, in many countries, they are discouraged from trying new techniques. There is often little incentive to propose the use of non-standard techniques under normal contractual arrangements since any benefits are accepted with little acknowledgement, but the results of failure are remembered for a long time. The result is that unnecessarily expensive designs are sometimes recommended. 6.24 The use of marginal materials needs a greater degree of control during construction and may lead to an increased rate of deterioration. However, these materials should ahvays be considered when carrying out pavement design in situations where their use is economic.

Traffic loading
6.28 The second important factor to influence pavement thickness is traffic loading. The damage that vehicles do to a road depends very strongly on the axle loads of the vehicles. The exact relationship is influenced by the type of road structure and the way the road deteriorates but a 'fourth power' damage law gives a good approximation for most practical applications. All axle loads are converted to an equivalent number of 80 kN (8.157 tonne) axles, referred to as standard axles, using Figure 6.2. Multiple axles are treated as separate axles for this purpose. 6.29 Figure 6.2 illustrates the importance of axle load surveys for structural design. An increase in axle load of 60 per cent increases the number of standard axles by 700 per cent and the passage of one 13 tonne axle causes as much damage as the passage of eight 8.2 tonne axles.

34

Fig.6.2 Pavement damage relationship

One of the most common causes of premature pavement failure in third world countries is incorrect estimates of traffic loading. In most developing countries, overloading is common and it is also unwise to assume that axle loads on all roads in a country are similar. It is essential to carry out independent axle load surveys when planning paved road projects. Guidance on how to carry out such surveys is given in Road Note 40 (Transport and Road Research Laboratory 1978). It is important to ensure that traffic cannot bypass the weighing site and that axle loads do not decrease as drivers and vehicle operators become aware of the survey and temporarily reduce the vehicle loads. 6.30 Although traffic induced damage is sensitive to axle loads, once the traffic has been expressed in terms of equivalent standard axles it is found that pavement design thicknesses are much less dependent on traffic load than on subgrade strength. For example, an increase in pavement thickness of ten per cent should enable several hundred per cent more traffic to be carried. Conversely, if the thickness is too low, very rapid failure can be expected.

Variability and uncertainty

6.32 The design must take account of inherent variability in the materials, variability in the quality control, uncertainties associated with climate, in particular rainfall and depth of water table and uncertainties in future maintenance, future vehicle axle loadings and traffic flow levels. 6.33 Subgrade strength. The subgrade strength normally varies both along the road alignment, from season to season and from year to year. Soil properties can change within a few metres, but it is quite impractical to change the structural design over short distances, hence a representative value must be chosen for the subgrade strength for design purposes which reduces the risk of early localised pavement failures to acceptable levels. The more soil testing that is done beforehand, the easier it is to reduce the risk in the design and to produce a cheaper pavement. It is recommended that the value of subgrade strength chosen for design purposes should be the lower ten percentile value for each nominally uniform section of subgrade. 6.34 The variation of subgrade strength with time is more difficult to assess. Underneath the centre of an impermeable road the strength remains reasonably constant and its value can be estimated from knowledge of the depth of water table and easily measured properties of the soil. Problems arise when road maintenance cannot be guaranteed to ensure that the surface is always impermeable. The ingress of water through damaged or aged surfaces and shoulders, and the retention of this water through poor maintenance of the drainage systems has a drastic effect on material strength and road performance.

Materials
6.31 The third factor which influences thickness is the choice of materials for the construction of the pavement layers themselves. This becomes most significant for the design of very heavily trafficked roads and depends on the detailed mechanisms of deterioration for each type of material. The better design methods available take this into account, but the subject is complex and specialist engineering advice should be sought.

35

It is not possible to compensate adequately for such effects by means of more conservative designs. 6.35 Materials. Additional problems of variability arise with the aggregates chosen for road bases and, to a lesser extent, sub-bases. There are numerous ways in which the aggregates can fall outside specification and unless sufficient testing of potential quarry sources is done at the feasibility study stage of a project to ensure that all materials are within specification, problems are inevitable. There are so many factors which affect the performance of a road that it is extremely difficult to evaluate the effects of deviations from the specifications for many of the material properties specified. Lack of sufficient testing is likely to give rise to disputes during the construction phase, often with serious financial consequences. 6.36 However, selection of pavement materials is probably one aspect of structural design where large financial savings can be made in road construction, especially for low and intermediate levels of traffic, by using marginal materials as discussed in para 6.22-24, particularly on more lightly trafficked roads. 6.37 Construction control. The construction process itself is seldom as well controlled as expected or desired. Sources of variability arise in all aspects of the work such as the mix proportions of premixed bituminous materials and the degree of road base compaction achieved transversely across the road. Variation in the thickness of the pavement layers is often a major cause for concern because of the extreme sensitivity of traffic carrying capacity to structural thickness. This sensitivity means that small increases in thickness should ensure that the road carries the traffic satisfactorily despite large

uncertainties in traffic forecasts providing that the natural variations in thickness arising from the construction process are properly accounted for in the design.

Shoulders
6.38 Shoulders are an essential element of the structural design of a road, providing lateral support for the pavement layers. They are especially important when unbound materials are used in the pavement and, for this type of construction, shoulders should be at least two metres wide. Narrower shoulders are acceptable for roads with bound bases (see Section 7). In order to exclude water from the road, at least one metre of the shoulder nearest the road should be impermeable and a surface dressing or other seal should be applied. Unsealed shoulders are not recommended as they often require considerable maintenance if satisfactory performance is to be guaranteed.

Drainage of pavement layers

6.39 Drainage within the pavement layers themselves is an essential element of structural design as the strength of the subgrade used for design purposes depends on the moisture content during the most likely adverse conditions. It is impossible to guarantee that roads will remain waterproof throughout their lives, hence it is important to ensure that if any layer of the pavement, including the subgrade, consists of material which is seriously weakened by the presence of water, the water must be able to drain away quickly. To facilitate this, correct camber should be maintained on all layers that are impermeable and a suitable path for water to escape must

Fig.6.3 Drainage of pavement layers

36

be provided, either by extending a permeable pavement layer right through the shoulder as indicated in Figure 6.3, or by including a permeable layer within the shoulder.

methods. Whilst this is somewhat overstating the case, it is from this area that future improvements in designs will come, and some developing countries are beginning to introduce some of these ideas into their pavement design methods. 6.44 There is an increasing body of evidence that suggests that the mechanism of deterioration of flexible pavements in tropical countries are often quite different to those in temperate climates. In addition, considerably different structural designs are obtained by using different design charts. Total thickness variations exceeding 100 per cent are not uncommon and, for heavily trafficked roads, even larger differences can occur. There are various reasons for this. Each type of structure behaves differently and therefore the same thickness design would not be expected to apply. Furthermore, criteria for determining terminal conditions, which are inherent in each design method, are often quite different. The result of this is that the designs to carry a particular traffic vary quite significantly from method to method. Inbuilt assumptions in the design methods are not normally described in the published manuals and it is often quite difficult to find them described at all. Technical comparisons between structural designs are therefore akvays difficult and economic comparisons often impossible. 6.45 Road Note 31 (Transport and Road Research Laboratory 1977-currently under revision) is a general design guide for bituminous surfaced roads in developing countries and emphasises good engineering practice which applies universally. It is based on research by the Overseas Unit of TRRL, but it cannot encompass all of the conditions likely to be encountered in all countries. In particular, extreme conditions of climate and axle loading are not dealt with. This design guide can be used to prepare or to check the pavement design being put forward as part of a project analysis to ensure that the design being proposed is of the correct order of magnitude.

PREPARATION AND CHECKING OF FLEXIBLE PAVEMENT DESIGNS Collection of information

6.40 In order to estimate pavement costs for a feasibility study, it is necessary to carry out a preliminary pavement design. This task should be carried out by a road engineer. If a paved road is being considered, the cost of the pavement will represent a significant proportion of the construction cost, so comparable effort should be put into the design study. 6.41 For most projects, a pavement design life equivalent to 15 years should normally be used to match that of the project analysis period. This not only simplifies the calculation of the residual value at the end of the analysis period, but reduces the problem of forecasting uncertain traffic trends for long periods into the future. However, shorter design periods do increase the accuracy of the assessment. 6.42 Information from the traffic and axle load surveys should be used to determine the cumulative equivalent standard axle loading that the road is forecast to carry over the design life. Information from the geotechnical surveys should indicate the likely availability of materials and the unit costs for using them in pavement construction. All of this information should be used together to prepare several alternative designs. The alternatives should contain different types of pavement construction and should reflect the uncertainties in traffic forecasts.

Choice of design method

6.43 Most pavement design methods in current use are derived primarily from empirical studies in Europe and North America. These methods have proved reasonably satisfactory, provided the materials, environment and traffic loading conditions do not differ significantly from those which pertained during the original studies on which the design methods were based. However, the extension of these empirical design methods to the different materials, different weights and volumes of traffic and different environmental conditions found in developing countries can pose serious problems. During the last decade, considerable advances have been made in the theoretical understanding of pavement behaviour. It is now claimed by proponents of the theoretical techniques that cheaper and better roads can be designed using these

STRENGTHENING FLEXIBLE PAVEMENTS

6.46 Strengthening overlays are designed using similar empirical or theoretical techniques as for the design of new roads. Usually some method of non-destructive testing, such as dynamic cone pentrometer or benkelman beam deflection testing, is used to assess the 'strength' of the existing road and to determine how much additional strengthening is required. No proven method of overlay design exists for tropical condition so, until one is developed, the method described in the Appendix of TRRL Laboratory Report 1043 (Smith and Jones 1982) is recommended. This method should be used either to prepare overlay designs or to check those submitted as part of project reports.

37

6.47 Problems arise if the road is in poor condition. Under these circumstances, the decision to strengthen the existing road or to rebuild the whole or parts of the road can be difficult. No easy guidelines exist. Conditions along the road will vary so much from place to place that the quantity of pavement layer testing required to assess the structural condition, and the degree of risk attached to overlaying under these circumstances, often mitigates against strengthening in favour of reconstruction. In this situation, engineering judgement plays a major role and risk analysis may be used to help quantify the hkely consequences of error. When assessments are made of roads requiring rehabilitation, it is important that sufficient testing is done to enable statistically meaningful results to be obtained. The results will need to be assessed by an experienced road engineer to determine the best remedies.

7.

GEOMETRIC DESIGN

PURPOSE OF GEOMETRIC DESIGN

7.1 Geometric design is the process whereby the layout of the road in the terrain is designed to meet the needs of the road user. The principal elements of this process are the selection of suitable horizontal and vertical alignments and road widths. The geometric design standards provide the link between the cost of building the road and the costs of the road users. Usually, but by no means always, the higher the geometric standard, the higher the construction cost and the lower the road user costs. Geometric standards are not more than a first approximation to design needs, since it is now accepted that design must be site-specific. The optimal design for a given traffic flow will depend on terrain and other characteristics. Appropriate geometric design standards for use in developing countries are given in Overseas Road Note 6 (TRRL Overseas Unit 1988). 7.2 One of the principal objectives of a feasibility study should be to make recommendations about the geometric design standards for a project such that the optimum balance between road construction cost and road user cost is obtained over the project analysis period. It is vital that decisions are not taken before this is carried out which prejudice the choice of geometric design standard. In the past, insufficient attention has been given to the choice of design standards with the result that roads have been built to standards well in excess of those that are justified by the traffic levels over the life of the project. 7.3 There are few developing countries who have carried out basic research on traffic economics and safety in order to develop their own geometric standards which have therefore been adapted from standards used in industrialised countries. However, the needs of road users in the industrialised countries are usually very different from those in developing countries. In developing countries pedestrians, animal-drawn c&ts, bicycles, autorickshaws etc are often an important component of traffic mix, even on major highways. In Europe and North America, traffic composition is dominated by the motor car, whilst, in developing countries, lorries and buses often represent the largest proportion of the motorised traffic. As a result, it may be necessary to adapt conventional geometric design standards to meet the needs of all road users by' for example, widening the sh3ulders of the road to allow their use by slow-moving traffic.

COSTING
6.48 Costing the design of the new pavement or overlay should be based on final achieved costs in other contracts, as described in Section 4, rather than on current contract rates. Costs are normally specified on a square metre basis for surfacings and on a cubic metre basis for all other layers. However, it is important to ensure that differences in haulage distances and other variables are taken into account, and that realistic prices are allocated for the use of new or modified materials. Local advice should always be sought.

ELEMENTS OF GEOMETRIC DESIGN

7.4 Geometric design covers horizontal and vertical alignments, road width and sight lines. 'Sight distance' is the distance ahead that can be seen by the driver. The sight distance needed for safe stopping from travelling

38

Fig.7.1 Horizontal alignment (plan view of road)

Fig.7.2 Vertical alignment (section through road) speed is the 'stopping sight distance' and the sight distance needed to see ahead for safe overtaking is known as the 'passing sight distance'. In detail, the specification may include the following. 7.5 7.6

Length of maximum gradient Minimum stopping sight distance or passing sight distance on summit curves Length of valley curves.

Horizontal alignment (see Fig 7.1): Minimum radius of curvature Minimum stopping sight distance Minimum passing sight distance. Vertical alignment (see Fig 7.2): Maximum gradient

7.7

Cross-section (see Fig 7.3): Width of carriageway (running surface) Width of shoulders Crossfall, camber and superelevation Width of structures Width of road reserve.

39

Fig.7.3 Cross-section (section through road)

RATIONAL BASIS FOR GEOMETRIC DESIGN

7.8 Traditionally, the first step in choosing geometric standards is to fix the 'design speed' for the road. However, when choosing geometric design standards for a particular situation, it is more important to consider the purpose for which the road is being provided. For geometric design purposes, it is most convenient to consider road projects under one of the following three categories. 7.9 Access roads (lightly trafficked roads carrying up to a few hundred vehicles per day) These provide a basic means of communication between minor centres of population, or between a centre of population and an existing road. The geometric standards of such a road have much less importance than whether a road link exists at all or, if a link exists, whether it is 'passable' at all times. 7.10 Collector roads (traffic volume likely to be in the range of 100 to 1000 vehicles per day) Projects for these roads normally have the purpose of providing additional capacity for low-volume roads in the network. Geometric standards contribute to these projects in the areas of road width and gradient but, for most developing countries, more important factors are whether or not a road is paved or whether it has sufficient structural strength to carry the traffic using it. 7.11 Arterial roads (traffic volume normally greater than about 400 vehicles per day) These will normally carry relatively high levels of traffic and, because of this, the

operational efficiency of the traffic on the road becomes a significant factor with the result that geometric design standards assume their greatest importance for this type of project. 7.12 This Section of the Note gives guidelines on the range of geometric standard which are appropriate for road projects in each of these categories based on concepts of both economics and road safety.

DESIGN RELATED TO TERRAIN

7.13 In flat undeveloped terrain, the cost of a road is almost entirely independent of its alignment, so a high design standard can often be adopted with no cost penalty. Only when the road is in hilly or mountainous terrain will there be any significant costs which are attributable to the alignment chosen. A higher standard of alignment will mean that more cuts and fills are needed resulting in a higher earthworks cost. However, a shorter alignment will result in lower pavement and maintenance costs, and cost savings for road users. The objective should be to produce a design such that any marginal increase in earthworks costs is more than offset by potential savings in user costs over the analysis period for the project. Feasibility studies for roads in hilly terrain should always consider alternative alignments before recommendations are made. 7.14 Consistency of design standards over relatively short lengths of route of say 5 to 15 km will improve road safety by reducing sudden and unexpected changes in road standard. However, when the road passes through

40

TABLE 7.1 GUIDELINES FOR GEOMETRIC DESIGN STANDARDS(1) Road type Normal surface type Approximate range of traffic levels (vehicles per day) Carriageway width (metres) Shoulder width (metres) Crossfall (per cent) Stopping sight distance (metres) Overtaking sight distance (metres)(7) Minimum horizontal curve radius (metres) Minimum crest curve K values (10) Maximum percentage gradient Access road Unpaved/paved Collector road Gravel/paved Arterial road Paved

< 400 2.5 - 5.0 0(3) - 1.5(4)

(5)

100 - 1000 5.0 - 5.5 1.0

(5)

400 - 15000(2) 6.0 - 7.0 1.0 - 2.5

(5)

25(6) - 85 l40(8) - 240 0(9) - 190 0(9) - l6 l5 - 20(11)

50 - 120 140 - 320 60 - 210 5 - 30 10

65 - 230 180 - 590 85 - 450 10- 120 8(12)

Notes (1) Values are based on those reconstnended in Overseas Road Note 6 (TRRL Overseas Unit 1988). (2) For higher traffic volumes, use latest British (Department of Transport 1981) or Australian (NAASRA 1980) standards. (3) On 2.5-3.0 metre carriageways with no shoulders, passing places should be provided. (4) 1.5 metre shoulders are needed on carriageways with widths of 3.0 metres or less. (5) Crossfall = 5.0 per cent on unpaved roads; 3.0 per cent on paved roads. (6) Longer stopping sight distances will be needed on single track roads. (7) It will normally be uneconomic to design roads for full overtaking sight distance. (8) Overtaking sight distances are inappropriate on access roads built to minimum standards. (9) Minimum standard access roads need not have designed horizontal and vertical curves. (10) K = curve length/algebraic difference in percentage gradient. (11) Where design is for animal-drawn carts, gradients may need to be restricted to a maximum of 4 per cent; slippery soils may also restrict the practical maximum value to as low as 5 per cent. (12) Climbing lanes should be considered when the length of road at maximum gradient exceeds about 500 metres. In very hilly country, short lengths of climbing lane of about 200 metres may be helpful.

significant changes in topographic type, it will usually be cost-effective to lower the alignment standards in the more rugged terrain. 7.15 The alignment should be chosen carefully to provide good drainage and to minimise earthworks. When locating roads in erosion-prone regions or areas where instability of slopes is a problem, careful attention must be given to minimising the disturbance of the terrain caused by the road. The location of the road will also be affected by the availability of road-making materials.

given in Table 7.1. There will normally be little difficulty in producing cost-effective designs with horizontal curve radii well above the minimum values in the range. The absolute minimum values should only be required in severe terrain.

HORIZONTAL ALIGNMENT
7.16 Suitable ranges of horizontal alignment standard are

7.17 For access roads, it is assumed that any new roads will carry low traffic volumes and therefore should be built to minimum standards. In steep ground, it is important that all of the vehicle types likely to operate on a road can negotiate curves. The minimum radius should be chosen to be adequate for all likely vehicles to use the road, including 4-wheel drive tipper lorries which tend to have very poor turning circles, but excluding large heavy goods vehicles. Where it is known that the road

41

requirement will only be for say light 4-wheel drive jeeps, this figure can be reduced. However, consideration would need to be given to other special users of the road, such as the Roads Department to allow access during road maintenance. It is most desirable to flatten the gradient at any tight corner. 7.18 For collector and arterial roads, projects will often be for improvement works rather than new construction. In such cases, it is seldom economic to improve individual horizontal curves. Proposals to do this to remove safety hazards may be considered, but should be looked at very critically in terms of benefits and costs. A completely new alignment may be justified where there is a physical constraint to widening the road, where a higher speed road is economic, or where safety and congestion factors arise when the road passes through a settlement. 7.19 It is preferable that lengths of horizontal curves should be close to desirable minimum radius and as short as possible. This type of design provides the maximum length of road where sight distances are not reduced and where overtaking can be carried out. Previous design methods used longer curves to produce 'flowing alignments' and more gentle bends. However, with such designs, sight distances will be restricted on the longer curves and a shorter length of alignment will be available where overtaking is safe.

gradients can be verv difficult to negotiate. On hairpin bends, it is important to keep the gradient on the bend itself as flat as possible. 7.22 Collector road projects will often involve improvement rather than new construction. If recommendations are made to bypass a steep gradient, the proposal should compare the benefits and costs of this course of action and the alternative of providing a climbing lane on the existing alignment. This comparison should be looked at very critically. Steep gradients can bring problems of potentially dangerous downhill speeds for heavy vehicles and can also lead to difficulties climbing hills for old or poorly maintained vehicles such as often found in developing countries. It may be economic to pave steep gradients on a gravel road to reduce maintenance problems. 7.23 For arterial roads, the choice of vertical alignment for the road has a significant effect on both the construction cost and the road user cost. The maximum gradient should be chosen to minimise the sum of these costs. Proposals should support their recommendation for choice of vertical alignment by comparing costs with alternative standard alignments. Road investment models should be used for carrying out such an analysis, and these are described in Section 10. Minimum values for the radii of vertical curves should be based on safety criteria, and the standards in Table 7.1 are based on limits of stopping sight distance which are appropriate. Where traffic flows will lead to congestion on steep gradients, the relative costs of building climbing lanes and flatter gradients should be evaluated. If climbing lanes are adopted, road markings should be used to indicate clearly that two lanes are 'up' and one is 'down'. 7.24 The length of summit curves should be as close to the minimum radius and as short as possible to reduce the length of road where minimum sight distance applies. This is for the same reason as was recommended for horizontal curves.

VERTICAL ALIGNMENT
7.20 The vertical alignment of a road has a strong influence upon the construction cost, the operating cost of vehicles using the road, and the number of accidents. The vertical alignment should provide adequate sight distances over crests and should not present any sudden hidden changes in aiignment to the driver. Gradients need to be considered from the standpoint of both length and steepness, and the speed at which heavy vehicles enter the gradient. They should be chosen such that any marginal increase in construction cost is more than offset by the savings in operating costs of the heavy vehicles ascending them over the project analysis period. 7.21 For access roads, the cost of earthworks is often a substantial part of the cost of total construction, so it is best to consider the maximum gradient that particular types of vehicle can climb safely, rather than to adopt a gradient that can be negotiated by all kinds of vehicles. The basic determinant for maximum gradient is whether vehicles are to be 2-wheeldrive, 4-wheel-dnve or animal drawn trailers. The absolute maximum gradients given in Table 7.1 should then apply. These gradients are extremely steep and, if possible, all gradients should be much less severe than these. On access roads with an earth surface, particular soil types may give rise to slippery conditions in the wet and even moderate

CROSS-SECTION
7.25 Road capacity is a measure of the number of vehicles that are able to use the road at any time and is chiefly a function of road width. As traffic levels approach the capacity of the road, vehicle speeds will fall. 7.26 For access roads where traffic flows are so low that vehicles meet only occasionally, a single track width road with intermittent passing places is the cheapest road to construct. For higher traffic flows, single track roads cause considerable inconvenience to traffic and it is only to be recommended for short roads or in hilly terrain where the cost of construction in side cut and the subsequent haulage cost of materials is high. Most rural

42

access road programmes have provided roads that are sufficiently wide for two vehicles to pass safely. It should be noted that, if the new road is to be constructed by machine, the extra construction cost of building a wider road will often be quite small. If there is a large amount of pedestrian, animal or bicycle traffic, both shoulders should be increased in width. Where visibility is reduced on bends and on summit curves, the carriageway width should also be increased. 7.27 For collector roads, research has shown that, for relatively low traffic volumes, carriageway widths in excess of the minimum values in Table 7.1 cannot be justified in terms of accidents or traffic operation. Further work has shown that there is benefit in widening from this value on bends and summit curves. If there is a large amount of pedestrian, animal or bicycle traffic, shoulder widths should be increased, or it may be appropriate to segregate this entirely from the vehicular traffic. The edges of paved carriageways can be delineated by sealing shoulders with different coloured aggregates, or painted edge marking should be provided. 7.28 One of the objectives of arterial roads is to provide efficient operation of the road network and roads at this level will normally be paved. This has an influence on the appropriate road width, and the width and type of shoulders. For roads expected to carry very high volumes of traffic, multi-lane roads may need to be considered from both a capacity and safety point of view. Proposals for such designs should be supported by good technical justifications and may be based on the latest British (Department of Transport 1981) or Australian (NAASRA 1980) practice. Shoulders should normally be at least partly sealed and should be differentiated from the carriageway by the use of different coloured aggregate or edge markings. Where large amounts of pedestrian, animal and bicycle transport is expected, shoulder widths should be increased. Arterial roads should normally carry centre line markings. 7.29 Crossfall is needed on all roads in order to assist the shedding of water into the side drains. Suitable values are given in Table 7.1. Shoulders should be at the same crossfall as the carriageway.

a loss of land to agriculture or other use; and, secondly, the road authority is sometimes obliged to cut the grass in the road reserve. A third problem may be that, if people graze their cattle in the road reserve, then this can be a traffic hazard. On the whole, the road reserve only needs to be sufficient for the road, its drainage, services, for future widening to dual carriageway, and to control erosion and ribbon development. A wider road reserve will improve sight distance.

JUNCTION DESIGN
7.31 Conflicting vehicle movements at junctions are the largest cause of accidents in many developing countries. A small number of well designed junctions on a route is preferable to a large number of low standard junctions. Simple cross-roads have the worst accident record. Staggered cross-roads or two separated T-junctions will reduce the accident rate. The use of roundabouts, traffic lights and channelisation may be appropriate to improve vehicle flow and safety. Local widening at T-junctions combined with painted channelisation (ghost islands) has proved highly cost effective in industrialised countries, as has the use of low cost traffic engineering devices, such as yellow bar markings on the approach to junctions. Conflicts can be largely eliminated by the expensive solution of grade separation, but it is not normally necessary to design for free-flow conditions at intersections, and their use will not be appropriate in most cases.

SIGNS AND ROAD MARKINGS

7.32 Warning signs should be used to inform the motorist that there is a change in design standard on the road and to reduce approach speeds. On new projects, signs will be needed where the project joins on to the existing road or where standards have been changed on passing into a different terrain type. Road markings will be needed on summit curves on paved roads where overtaking sight distance is not provided. Warning signs should also be erected to inform of other hazards such as road junctions. Mandatory signs indicate where action by the driver is necessary which is enforceable by law, and direction signs help to direct traffic along a route to a destination. If these three types of signs are used properly, they will form an information system which will reduce accidents and minimise confusion and delay. 7.33 Ideally, signs should be reflectorised, but ordinary paint is better than nothing. The use of marker posts and chevron board on bends is also strongly recommended, particularly for roads being designed at the minimum standards being recommended here. As has been mentioned earlier, the use of road markings on the edge of roads, on the centre line and on climbing lanes is also recommended. However, in many countries, maintenance of road markings can be a problem.

ROAD RESERVE
7.30 For new road construction, most countries already have standards of the amount of land that they will buy or acquire for the road. Often this is much wider than in industrialised countries since the pressure on land space is not as great. The total width of the road reserve may be from 10 to in excess of 50 metres. Its purpose is to provide land for future road widening and to provide the fill material for the road construction. There are two problems with having too wide a reserve: first, it implies

43

COSTING EARTHWORKS
7.34 Standard engineering methods should be used to prepare preliminary alignment designs based on the range or geometric standards recommended here. Computer programs are available to assist with this process from a variety of sources including: Highways Directorate Department of Transport 2 Marsham Street London SW1P 3BB United Kingdom 7.35 For feasibility study purposes, alignments designed on the basis of available contour maps are adequate to provide the level of detail required to make cost estimates to an acceptable order of accuracy. For design studies, more detailed terrain information is needed as described in Section 5. 7.36 A significant proportion of the cost of building or realigning a road is the cost of earthworks. This is made up principally of the cost of excavating cuttings, building embankments and hauling material between the two. Additional material may also need to be brought in from pits, and any surplus or unsuitable material will need to be dumped. Standard engineering methods should be used for determining earthworks quantities and costs. Computer programs are available to assist with these calculations including the construction cost submodel of micro-RTIM2 which has been developed specifically for use in feasibility studies in developing countries. Details of how to obtain copies of this program can be obtained from: Overseas Unit Transport and Road Research Laboratory Crowthorne Berkshire RG45 6AU United Kingdom 7.37 Methods of costing based on those in Section 4 should be used.

8. DRAINAGE AND STRUCTURES

8.1 Consideration of drainage is important in a road project analysis because the cost of providing structures can represent a significant proportion of the construction cost of a new road.

THE DRAINAGE SYSTEM

8.2 One of the most important aspects of the design of a road is the provision made for protecting the road from surface water or ground water. If water is allowed to enter the structure of the road, the pavement will be weakened and it will be much more susceptible to damage by traffic. Water can enter the road as a result of rain penetrating the surface or as a result of the infiltration of ground water. The road surface must be constructed with a sufficient camber or crossfall to shed rainwater quickly and the formation of the road must be raised above the level of the local water table to prevent it being affected by ground water. 8.3 Water can also have a harmful effect on shoulders, slopes, ditches and other features. High water velocities can cause erosion which, when severe, can lead to the road being cut. Alternatively, low velocities in drainage facilities can lead to silt being deposited which, in turn, can lead to a blockage. Blockages often result in further erosion. 8.4 A good road drainage system, which is properly maintained, is vital to the successful operation of a road. It has four main functions: to convey rainwater from the surface of the carriageway to outfalls to control the level of the water table in the subgrade beneath the carriageway to intercept ground and surface water flowing towards the road to convey water across the line of the road in a controlled fashion.

The first three functions are performed by side drains and the fourth by culverts, drifts and bridges.

SIDE DRAINS
8.5 The cost of side drains will normally be calculated as part of the cost of earthworks (see Figs 6.3 and 7.3, and Section 7). Side drains should be flat-bottomed if they are to be maintained by hand (Fig 7.3), or 'v'-shaped if they are to be maintained by machine (Fig 6.3). Wide flat drains, known as 'meadow drains', can be used with advantage if there is room. The longitudinal gradient of

44

side drains should always exceed 0.5 per cent to reduce the possibility of silting up. In hilly terrain, providing side drains with the same gradient as the road may result in water velocities that are too high. It may therefore be necessary to reduce the maximum gradient to an acceptable level by the provision of shallow dams or scour checks. These are often constructed of masonry, but can also be constructed in concrete or even timber. Wide drains are preferred to reduce the velocity and so minimise erosion. The provision of turnouts or cut-off drains should also be considered to reduce or control the amount of water in the side drains. Costing may need to take account of these and the need to line drains with masonry or concrete in highly erodible soils.

8.9 In order to determine the maximum likely flow of water to be accommodated by the chosen structure, information may be needed on: water catchment area rainfall characteristics topography vegetation and soils catchment shape available storage in lakes and swamps urban development (if any).

Peak flood volumes can then be estimated using standard hydrological techniques such as those described by Watkins and Fiddes (1984). 8.10 In order to determine the type of structure, consideration may need to be given to the cost of delays to traffic if, for example, a concrete drift is impassable for several days or if a single lane bridge is being considered instead of a two lane structure. Thus, the choice of structure may need to be determined by appraisal. The value of time is discussed in Section 12.

REQUIREMENTS FOR CROSS DRAINAGE

8.6 In order to determine requirements for cross drainage, information must be collected and predictions made about the level of traffic and the likely flow of water passing under the road. This enables decisions to be made about the types of structure that will be necessary and the number required. 8.7 The following types of structure should be considered:

FORDS AND DRIFTS

Ford. This utilises a suitable existing river bed and is appropriate for shallow, slow moving watercourses with little probability of flash floods: traffic volumes up to about 100 vehicles per day. Drift. This is sometimes known as an 'Irish bridge' and consists of a concrete slab constructed in the river bed which would otherwise be unable to carry vehicles and is suitable as a crossing for rivers that are normally fordable but are prone to flash floods; where the river is running most of the year, culverts can be placed in the drift to reduce the frequency and depth of overtopping during flash floods: traffic volumes up to about 100 vehicles per day. Culvert. This usually consists of a concrete or steel pipe, or a reinforced concrete box, placed under the road within an embankment to provide a suitable means of conveying streams, or the contents of side drains, under the road with no restrictions on traffic. Bridge. This may have a superstructure of timber, concrete and/or steel on masonry, concrete, or timber abutments and will be required for crossing streams or rivers where culverts would provide insufficient capacity, or where the road crosses an obstruction such as a railway or canal. There is no restriction to traffic unless the width of the structure is less than the road width.

8.11 The simplest river crossing is a ford. Large stones with flat tops can be placed at the upstream and downstream sides of the ford so that pedestrians can use them as stepping stones rather than having to wade across the river. Gravel or stones can be used to line the bottom of the ford to provide a firm footing for vehicles. Fords should normally only be used for rivers that do not flood as this may cause the ford to be washed away. However, repair or replacement is cheap and this may still provide an acceptable solution. The cost of providing fords is small and can usually be ignored for project analysis purposes, although some additional earthworks costs may be incurred to ensure that the road gradient on either side is acceptable to traffic. 8.12 An improvement on a ford is a concrete drift. This provides a permanent running surface for traffic, although delays may still occur when stream levels are above the level of the carriageway. The gradient of the road on either side of the drift should be not more than about 10 per cent, or 4 per cent where animal drawn traffic is expected. It may be necessary to surface the road where such gradients are unavoidable, even where a gravel surface is otherwise adequate. The width of the drift need be no more than 3.5 to 4 metres, but should be delineated by graduated marker posts to show both the edge of the road and the depth of the water during floods. Culverted drifts may be used to cross perennial streams. Culverts should normally be at least one metre in diameter to reduce the likelihood of blockage and to make them easier to maintain. The cost of drifts can be

8.8 Traffic requirements should be determined using the methods described in Section 3.

45

estimated from the volume of concrete required for construction, but allowances must be made for engineering work required to ensure that the pavement is not eroded or undermined. These costs may be significant. Cost estimates for culvert pipes are made on the basis of the length of pipe required. Unit prices for concrete and for culvert pipes of various diameters which are appropriate to the road being analysed should be readily available.

do not have sufficient capacity to carry the flow, or where drifts are not suitable because of safety considerations, or because resulting traffic delays are unacceptable. The following gives a brief survey of the main issues and choices. Specialist engineering advice should be sought whenever possible.

Bridge design CULVERTS

8.13 The use of culvert pipes to convey surface water under a road alignment is common, and provides a relatively cheap and durable solution. Most countries make concrete pipes of up to one metre diameter and these may be cost effective provided that they can be transported and handled. Corrugated galvanised steel pipes, often known by the trade name 'Armco', are available in larger diameters and are usually more expensive, but lighter and easier to handle. There should be little maintenance required for either material other than an annual inspection and clearing of accumulated silt or debris, although corrosion may occur to metal pipes in some circumstances. Culvert pipes require headwalls to protect the ends of the pipe and to direct water either towards or away from the culvert. The outfall of the culvert must be protected against scour and environmental damage downstream. 8.14 For larger volumes of water, it is possible to use several pipes in parallel under the road. Multiple pipes can also be used where the planned embankment height is insufficient to cover a single pipe of sufficient diameter adequately. However, pipes of less than one metre diameter are not recommended since they are difficult to maintain. 8.15 Reinforced concrete box culverts may also be used either singly or in parallel where relatively large volumes of water must be carried. These are normally cast in place, although smaller sizes may be precast. 8.16 Costs for culverts and for mass and reinforced concrete should be based on final achieved contract costs rather than current contract rates (see Section 4). Culvert pipes are normally costed per unit length, depending on diameter, whereas headwalls and box culverts are normally costed on the basis of volume of concrete used. Local advice on costs should always be sought to ensure that all reasons for cost differences are taken into account. 8.18 Most countries have established bridge design codes which specify the size, type and configuration of loads which the structure must be able to carry safely. Such codes are usually based on, or similar to, codes adopted in the USA (AASHTO 1983) or the United Kingdom (BSI 1972, 1978). It is normally advisable that new structures are designed to the code adopted by the country concerned. This avoids the need to impose weight restrictions, which may be difficult to enforce or, conversely, the provision of a structure which will safely carry loads which cannot be permitted elsewhere on the network. An exception may be made where: A temporary structure is envisaged Provision must be made for known exceptional loads, eg access to a power station or other structures known to require exceptionally heavy plant or equipment.

8.19 The width of a proposed bridge can significantly affect the cost of construction. If two lanes are provided instead of one, material costs will more than double as heavier construction will be required to accommodate the additional traffic loads. Particularly where a long bridge is required, careful consideration should be given to the relative cost of the provision of two lanes and the delays to traffic that would otherwise occur over the life of the structure. The proportion of pedestrians, bicycles, and animal-drawn vehicles should also be considered. In such cases, it may only be necessary to widen the carriageway sufficiently to ensure that motor traffic is unimpeded rather than provide two standard traffic lanes. A cheaper solution may be to provide light footways cantilevered out from the main structure. It is very unlikely that more than two lanes will be required in rural areas in developing countries. 8.20 A significant reason for bridge failure and high maintenance costs in tropical countries is erosion and scour leading to foundation failure. Even at the analysis stage, it is worthwhile to make sure that these aspects have been considered and appropriate protection provided.

BRIDGES
8.17 Bridges will be needed over rivers where high level crossings are essential, where several culverts in parallel

Site conditions
8.21 Temporary structures apart, bridges are normally designed to operate without major structural alteration for very long periods, typically fifty years. It is therefore

46

particularly important that the design takes into account all factors that can reasonably be foreseen and that may affect performance. Such factors must include the characteristics of the river, which will determine the location and design of abutments and piers, and the elevation of the deck or superstructure. Other factors include the extent of river training and scour protection measures that are needed. Of equal importance, will be a knowledge of ground conditions at the selected site. No reasonable estimate of cost can be made without this information and a site investigation should always be undertaken to determine the strength and other characteristics of the soils on which the structure will be founded. 8.22 The cost of providing river training works is often high and, where there is evidence of the river changing its course, it may be preferable to reduce the design life and the cost, and accept the need to rebuild the bridge at a later date. It is also important to consider what the cost of disruption or damage would be if the river were to overtop the bridge. An understanding of the river characteristics, the construction cost and the risk of structural failure is crucially important in deciding the optimum choice of materials and design for the structure.

bridges may be parallel timber logs. Cutting and squaring timber for such crossings is expensive and not normally worthwhile. Ideally, timber should only be used where there is little or no problem with wood-boring insects and a naturally durable species should be selected, or else some form of chemical treatment, such as creosote, applied. To be effective, timber preservation must be done thoroughly and may significantly increase costs. On top of the logs, cross beams should be used to support longitudinal running boards. 8.27 The maximum span that can be used will depend on the species and height of available trees, but spans of up to about 15 metres are feasible. 8.28 Modular timber bridges have been developed for use in developing countries Parry 1981, UNIDO and TRADA undated). These are suitable for spans of 12 to 24 metres and have the following advantages: relatively cheap to build the materials and skills required to build the bridge are available locally in most developing countries the modular design permits prefabrication of the frames in local workshops the frames may be stored for emergency use, and can be assembled to make a bridge on prepared abutments very quickly the bridge components are small enough and light enough to be transported to a remote site if a bridge is required urgently.

Abutments and piers

8.23 Abutments and intermediate piers distribute the vertical and horizontal loads on the bridge to the foundations. Abutments must also resist the horizontal forces of the soil which is constrained. 8.24 Where the ground conditions at a reasonable depth are adequate to support the bridge and traffic loads, it is normal to support abutments on narrow reinforced concrete slabs or footings. Where the soil is too weak to support this type of foundation, piles will be needed to support the abutments and piers. Normally piles are more expensive than concrete footings and require specialist design and construction skills. 8.25 Abutments and piers are usually constructed of reinforced or mass concrete, masonry, brick or timber. The choice between concrete, masonry or brick will be determined by the cost and availability of materials and the skill and experience of the available labour force. Timber should be considered with care because, although accommodating considerable movement without distress, it is prone to rot and insect attack, particularly when used in abutments and retaining soil. Careful selection of species and treatment will help, but maintenance costs may be high and regular monitoring of condition essential.

Although such bridges have some disadvantages, they should be considered for use in appropriate situations.

Concrete decks
8.29 Concrete superstructures are now common practice in most developing countries. Local contractors may be capable and experienced in some of the simpler forms of reinforced concrete. Where cement is locally produced, it may be economic to set up a precasting factory for standard bridge beams. Where these are available, they will often be cheaper and more suitable than steel. Alternatively, the beams may be cast in-situ but, in either case, a concrete slab needs to be cast to provide a running surface. A bituminous wearing course may also be added. An alternative to a beam and slab design would be either a solid concrete slab without beams, or a slab cast with voids to reduce the weight, also without beams. The most cost-effective configuration will depend on the span, width, available reinforcement, concrete strength achievable, and many other factors which the bridge designer should take into account. 8.30 Other, more sophisticated techniques of pre-stressing may be considered.

Timber decks
8.26 In many rural areas which are close to forests, the cheapest construction option for the superstructure of

47

8.31 Post tensioned beams and slab. The deck is constructed in-situ in a similar way to above, but incorporating accurately located steel ducts to accommodate separate wires, strands, or high strength steel bars. When the concrete has hardened, the wires or strands are tensioned by jacks bearing against the concrete faces. The tensile force in the wires imposes a compressive force on the concrete. This condition is maintained by specially designed anchors attached to the ends of the wires. 8.32 Pretensioned beams. This method is applicable mainly to precast elements. Prior to casting the concrete, wires, strands or high strength steel bars are located in the mould and loaded to the required tensile stress. After the concrete has hardened, the load is removed and the tensile stress in the reinforcement applies an equal compressive stress in the concrete through the bond between the materials. 8.33 Both forms of pre-stressing offer advantages over conventional reinforced concrete. A pre-stressed beam or slab is generally free from cracks and is therefore more durable. Much less steel is required, since the weight of high-strength steel in the tendons is only a fraction of the weight of the reinforcement it supersedes. The cross-section is smaller since the concrete is used more efficiently and resistance to shear stress is substantially increased. However, pre-stressing demands high quality concrete, special steels, specialist equipment, and experienced and knowledgeable contractors and designers. Pre-stressing should not be considered if any of these requirements cannot be met. 8.34 Segmental box girder units. This is a technique for use on longer spans. Each separate unit is manufactured using industrial processes either at site or in a factory. High quality concrete is therefore achievable, but specialist construction skills and equipment are required and, in general, this design is unlikely to be appropriate where local contractors are employed.

assembled before delivery to site. In the UK, these are fabricated and designed by Callender-Hamilton and Mabey, who produce both their own 'Universal' design and the 'Bailey' type. Bailey bridges are also available from Messrs Thomas Storey. By varying the number of panels, various spans can be constructed. Although relatively expensive, the panel system is also excellent for the quick erection of bridging at temporary sites. Pontoon type crossings have also been effective on many rivers using standard panel units. 8.38 Box girder bridges are sophisticated structures used for long spans. They require specialist design and construction skills, but are technically very efficient in that they have a high strength to weight ratio. Suspension bridges often incorporate prefabricated box girders.

Arch bridges
8.39 Masonry and brick arch bridges are a traditional form of construction in some countries and may well be competitive where skilled bricklayers or masons are plentiful, and where appropriate materials are available. Despite this, the use of this type of structure is often overlooked in road projects where design is done by engineers from countries where there is little experience of their construction because labour costs make them uncompetitive. Structural analysis of this type of structure is less precise than that possible for steel or concrete, but arch bridge are capable of carrying exceptionally high loads without distress. There is thus more scope for the use of arch bridges in road projects and they should be considered as an alternative to steel and concrete structures in appropriate situations.

Replacement of existing bridges

8.40 If the project is to replace an existing bridge, a technical appraisal should be carried out to ascertain the need. If load restrictions are in force, the benefits of replacing or strengthening will be derived from more efficient freight operations. The age of a bridge should not be the sole criterion for replacement. Replacement should be based on a technical assessment of the bridge's ability to carry the required loads in the future. In many cases, deck reconstruction will be more cost-effective than replacement, particularly where piers and abutments are sound. However, the costs of disruption to traffic should be included in the analysis. In many cases, old bridges, especially stone or brick arches, will carry legally permitted traffic loads even though they were not originally designed for them. A sub-standard bridge located in an important route can have a considerable effect on goods throughout the network and thus reconstruction/replacement may have benefits beyond the immediate vicinity of the bridge which should be considered in the analysis of a project.

Steel decks
8.35 Steel superstructures are of three types.

8.36 Rolled steel beams provide the simplest design consisting of a number of parallel 'I' beams spanning from one fixed abutment to the other, or to intermediate piers. The length is usually limited by handling and transport constraints to about 18 metres but, in many countries, the size of beam available limits the span to about 8 or 10 metres. A timber or reinforced concrete deck is constructed on top of the beams. If a concrete deck is used, this can be more efficient it the steel beam and concrete are designed to act compositely, ie are effectively bonded together. 8.37 Panel bridges are made of steel sections fabricated at a factory to form trusses and may be either part or fully

48

8.41 Another common problem arises where a two lane road includes a series of narrow single lane bridges. Often these have short spans and are of adequate strength, but because of their width, they represent a safety hazard and traffic is delayed by giving way to oncoming vehicles. A widening or replacement programme may be appropriate, but should be tested by the economic analysis. 8.42 Many large rivers in developing countries have an existing railway bridge but no equivalent road crossing. Where both rail and road traffic is light (say up to 10 trains and 250 vehicles per day), the economic feasibility of converting the rail bridge into a rail/road bridge should also be considered.

acute and the use of construction materials near to the site becomes very economic. On low volume roads, the use of whole timber logs may be appropriate as the timber can normally be obtained virtually 'free' whilst cement and other conventional materials are expensive or unobtainable8.48 Costs do not rise gradually, but in a series of steps at particular loadings and spans. Single lane bridges can still be suitable for main roads if the capital cost advantage of doing this is substantial. This will usually be the case on long span bridges where the traffic flow is less than about 250 vehicles per day. 8.49 It is not possible to give definitive rules on which design is 'best' for a particular situation since much will depend on the terrain, river regime, materials, local skills and available plant. 8.50 Bridges can represent a significant proportion of the cost of a construction project and, if large structures are needed, they can outweigh other costs. Thus, even at the project feasibility stage, considerable attention must be paid to bridge costs and this cannot be done unless preliminary designs and a site investigation have been carried out. This design and costing should always be carried out by a structural engineer on the basis of prices obtained locally.

Costing bridges
8.43 There are many factors which affect the design choice and the cost of the structure, and these will vary at each site. In general, the simpler designs will be easier to construct, and hence cheaper, for short span bridges. 8.44 The availability of materials and local expertise will tend to govern the choice between concrete and steel structures. Local contractors may construct reinforced concrete competently; pre-stressed or post-tensioned concrete may be beyond their capability. The cost of concrete bridges, in general, will be relatively insensitive to the load carried whereas, for steel panel bridges, the load at certain span lengths determines the number of units that are needed and hence the cost. For some timber bridges, the load carried may be extremely critical on particular spans as this will determine whether locally available timber is strong enough. 8.45 The choice of optimum span for a large bridge is an important decision since the longer the span, the more expensive and difficult construction becomes, but there is a corresponding saving in cost of foundations and piers. Where the river is permanent, fast flowing, and carrying considerable debris in flood, then the cost of building adequate intermediate supports for the bridge is likely to be high. There will be physical limits on the maximum length of span for a given design that can be constructed, and these will have to be carefully balanced with the river conditions and foundation problems. 8.46 The alignment of a structure is usually determined by the geometry of the approach road. This may result in the bridge being 'skewed' in relation to the river. Both design and construction costs will be higher for a skewed structure than for one at right angles to the river. Local realignment of the approach road should be considered as an alternative to a skewed structure if adequate sight distances can be maintained. 8.47 A major cost may be the transport of materials to the bridge site. In inaccessible areas where existing roads or tracks are non existent, this problem becomes more

MAINTENANCE
8.51 If a structure is to perform adequately over its design life, it must be regularly inspected and maintained. The resources and capabilities of the department responsible for upkeep of the structure should be considered at the design stage if a truly cost-effective solution is to be found. Care taken to ensure that such critical details as bridge bearings and expansion joints are both as maintenance-free as possible, and easily accessible, will ensure that expensive repairs are less likely to be required later. Steel components will require regular painting and the performance of timber will be radically affected by both the quality of the initial treatment and the avoidance of traps for moisture and debris in the design. 8.52 A cost-effective design will be that which most successfully takes account of local skills, materials, location, safety and maintenance capabilities.

FERRIES
8.53 Rivers do not have to be crossed by bridges. Another option is to improve or provide a ferry service. These are particularly applicable where the river channel is constantly changing. Where traffic levels are low and the river wide and slow-moving, they can be a cost-effective alternative and several designs of flat bottomed ferry are available. However, appraisals should take account of the delay to traffic introduced by the utilisation of ferries and both their capital and maintenance cost.

49

9. INTRODUCTION TO THE ASSESSMENT OF BENEFITS

9.1 In order to analyse a project, estimates need to be made, not only of the costs associated with the project, but also of the benefits that are expected to occur. Benefits normally considered are:

9.4 The assessment and valuation of benefits is discussed in more detailed in the following sections which consider vehicle operating cost savings, road maintenance benefits, time savings, reduction in road accidents and economic development benefits. It is expected that transport planners will already have detailed knowledge of the methods used to assess these benefits. The following five sections are therefore written primarily for engineers, economists and administrators who may not have the practical experience of assessing the benefits of road projects.

direct savings on the costs of operating vehicles economies in road maintenance time savings by travellers and freight reduction in road accidents wider effects on the economic development of the region.

9.2 All types of benefits should be considered for all projects although, depending on the type of project, different benefits will predominate. It is possible for some benefits to be negative. Direct savings in vehicle operating costs will normally be the most significant benefit of inter-urban road projects in situations where the value of time is low. These roads carry relatively high traffic flows and their role is mainly to provide for the carriage of goods and people over long distances. Time and accident savings may also be significant for inter-urban projects. At the other extreme, rural access roads represent the grass roots of the road network which feed traffic onto the primary roads linking rural areas to the main network. They are generally short in length, have low traffic volumes and are usually constructed with earth or gravel surfaces. For these roads, economic justification for the investment rests mainly on the expected impact on rural and agricultural development. This benefit often manifests itself as generated traffic (see para 3.22-26) and, if development benefits and generated traffic are both being evaluated, it is important to ensure that 'double counting' of benefits does not take place in the appraisal.

9.3 Many developing countries are experiencing increasing pressure from their agricultural populations for rural road development and rural access roads can assist in the production of real wealth. Although criteria have been established whereby rural access road investment can be justified on grounds other than savings in vehicle operating costs, it should be recognised that the diversion of scarce resources to this sector can be at the expense of the standard and upkeep of the primary network. Interurban routes carry vastly higher traffic flows than the rural access road network and their upkeep is a high priority component of the roads budget. Furthermore, without a properly maintained primary road system, it may not be possible to realise all the potential benefits from rural access road improvements.

50

10. VEHICLE OPERATING COST SAVINGS

Factors affecting vehicle operating costs
10.1 When a road improvement is undertaken, the owners and users of vehicles profit from reduced costs of transport. Higher average speeds can be maintained, and the more even running, with fewer gear changes and braking, may lead to savings in fuel consumption. Tyres last longer on improved road surfaces and there is less wear and tear on the suspension and body. These savings are perceived by road users in the form of lower expenditures. 10.2 Vehicle operating costs depend on the number and types of vehicles using the road, the geometric design standards of the road, particularly the curvature, gradient and road width, the condition of the surface of the road, primarily its uneveness or 'roughness', and driver behaviour. Changes in any of these parameters as a result of a project will result in a change in vehicle operating costs. 10.3 The components of vehicle operating cost with their approximate respective contribution to the total are given in Table 10.1.
TABLE 10.1 RELATIVE CONTRIBUTION OF VEHICLE OPERATING COST COMPONENTS

as a result of road projects. The most well-known of these are 'micro-RTIM2' and HDM-III'. Micro-RTIM2 was developed for ODA by the TRRL Overseas Unit and is designed to be simple to use with a user-friendly input facility. HDM-1Il has been developed by the World Bank and is more comprehensive than the TRRL model. Both models run on a micro-computer. The models simulate the performance of a road over time and under traffic. Costs and benefis are determined by applying unit rates to quantities that are calculated; since these unit rates are supplied by the user, the models are applicable to a wide range of economic and financial environments. The effect on vehicle operating cost of the changing condition of the road surface is taken into account in the determination of costs and benefits. 10.6 The micro-RTIM2 model is normally run first for a 'do nothing' or 'do minimum case. A series of project options may then be run and, as each is completed, the results are compared with those of the 'do nothing' case to determine benefits. Net present values (see para 15.19-23) are then calculated over a range of discount rates (see para 15.9-11) chosen by the user, and the internal rate of return and first year rate of return are determined (para 15.25-27). The relationships in the model allow it to be used to study many aspects of a road investment project such as the optimum maintenance standards for the road, the choice of a gravel or bituminous pavement, and the benefits of adopting any number of different stage construction options. The model will also enable the study of uncertainties in traffic forecasts, in the selected discount rate or in any other variables. 10.7 Copies of the micro-RTIM2 program to run on a range of microcomputers may be obtained on application to: Overseas Unit Transport and Road Research Laboratory Crowthorne, Berkshire RG45 6AU United Kingdom Copies of HDM-III for mainframe or IBM-compatible micros may be obtained from: Transportation Department The World Bank 1818 H Street NW Washington DC 20433 United States of America

Percentage contribution Component Private cars Fuel consumption Lubricating oil consumption Spare parts consumption Vehicle maintenance labour hours Tyre consumption Vehicle depreciation Crew costs Other costs and overheads 10-35 <2 10-40 <6 5-10 15-40 0 10-15 Trucks 10-30 <2 10-30 <8 5-15 10-40 5-50 5-20

10.4 In feasibility studies, all costs and prices should be expressed in 'economic' as opposed to 'market' terms to reflect the use of real resources of the country's economy (see para 15.12-17).

VEHICLE OPERATING COST TABLES

10.8 Tables for estimating vehicle operating costs can be obtained on written application from both TRRL Overseas Unit and the World Bank. These provide

ROAD INVESTMENT MODELS

10.5 Computer models are available for assisting in the calculation of vehicle operating costs under a range of conditions and of estimating vehicle operating cost savings

51

convenient tools for estimating vehicle operating costs on low volume rural roads with free-flowing traffic. Costs have been divided into various components, for different classes of vehicle and for road surface types. The tables are given in terms of physical quantities or non-dimensional ratios and costs are obtained by applying the appropriate unit price to these. They can be applied to any monetary system or used in any appropriate environment and are useful where no computer or microcomputer facility is available.

Relative importance of data items

10.14 In order to determine vehicle operating costs, it is necessary to obtain several items of data. The ease of collection of these items varies as does the relative impact that they have on the final value. The collection of data should be viewed against the background of the accuracy of the relationships used in the investment models and tables which is only of the order of plus or minus 25 per cent. 10.15 Data should be relatively easy to collect for rise and fall, curvature, road width, vehicle weight, power to weight ratio, vehicle and tyre prices, and prices for fuel for existing roads. Good estimates of these should therefore be obtained in all cases. For new roads, estimates must be made based on information collected on related roads and predictions made about any changes in the parameters that are likely to result from the project. Although data on surface moisture content, rut depth and lubricating oil are easy to collect, their effect on total vehicle operating costs is relatively small (see Table 10.1) and, when resources for data collection are severely limited, these values can be ignored or replaced by simple estimates. 10.16 If a change in vehicle operating cost following a road improvement is being determined, then the key variables will be the change in roughness, annual utilisation and vehicle age. These are all difficult items for which to obtain reliable estimates and considerable effort should be made to collect good data in these areas. 10.17 The input parameters are discussed further in the following sub-sections which contain suggestions and, where necessary, guidelines, for their use.

DATA REQUIREMENTS Measured values

10.9 In the TRRL investment model and vehicle operating cost tables, vehicle speed and operating cost have been related to the physical condition of the road and the characteristics of the vehicle, using relationships based on the results of empirical research. The following variables and units are used, and measurements of these should be made for existing roads and estimates made for road projects that are being appraised. 10.10 Road characteristics Rise (m/km) Fall (m/km) Curvature (degrees/km) Roughness (m/km) Road width (metres) Surface moisture content for gravel and earth roads (per cent) Rut depth (mm). 10.11 Vehicle characteristics Free speed in the environment under consideration (km/h) Vehicle weight (tonnes) Power to weight ratio (bhp/tonne) Vehicle age (km, yr) Annual utilisation (km, hr) Vehicle price Tyre price Price of fuel and lubricants per litre) Price of maintenance labour per hr) Vehicle crew cost per hr) Overheads. 10.12 Cilmate Ralnfall (mm/year).

Physical parameters of the road

10.18 Rise and fall. The average rise and average fall per kilometre should be estimated separately. The accuracy of the estimates of rise and fall provided by the user will depend on the amount of information available for the study. In the absence of detailed engineering plans, reasonably accurate estimates can be extracted from large scale contour maps. Also, simple measurements of an existing road can be made using an abney level to measure the amount of rise and fall. 10.19 Horizontal curvature. Horizontal curvature is usually related to gradient, and sections of road which are similar with respect to rise and fall can be expected to have broadly similar horizontal curvature characteristics. The average value of curvature should be expressed in degrees of curvature per kilometre of road. The degree of curvature can be extracted from engineering plans or large scale maps. It can also be measured on an existing road using a prismatic compass.

10.13 It is essential that the user applies the best available estimates of the relevant physical quantities and vehicle descriptors in order to arrive at valid cost estimates whether using the tables or one of the investment models.

52

10.20 Roughness. Many of the operating cost components are influenced by the surface condition of the road, measured primarily in terms of roughness. The measurement of roughness will normally be made with a 'response-type' instrument such as: TRRL integrator unit NAASRA meter Mays ride meter

Therefore, an estimate of road width is required. On existing roads, a reasonable estimate can be obtained by making a few sample measurements with a tape measure. 10.25 Surface moisture content. On unpaved roads, the moisture content of the road surface affects its slipperiness and hence the speeds of vehicles, but its effect on cost is small. The use of values of zero for dry zones (annual rainfall less than 750 mm) ranging up to six per cent for very wet zones (annual rainfall greater than 1750 mm) is recommended. 10.26 Rut depth. The rutting of unpaved road surfaces reduces the speeds of vehicles to a small extent. Rut depth is measured under a two metre straight edge placed across the wheel track. Average values varying from 10 mm for unpaved roads in good condition to 50 mm for roads in very poor condition are common.

The TRRL integrator unit is relatively cheap and easy to install and use but, whatever type of instrument is used for measuring roughness, it will be necessary to calibrate this to ensure that the results obtained are consistent with the standardised values used for vehicle operating cost calculations in the road investment models and vehicle operating cost tables. Equations now exist (Sayers et al 1986) for relating standard roughness values to the absolute longitudinal profile of the road. The response-type instruments used can be calibrated by running them over roads with various levels of roughness. The longitudinal profile can be determined by measuring the road with an engineer's rod and level. However, this method is laborious and the calculations needed to determine the calibrated values are time consuming. An instrument has been developed at TRRL for calibrating response-type roughness measuring devices more easily and this is known as the 'Abay beam'. This contains an on-board microprocessor and, when the machine is used on the road, an automatic readout of calibration roughness is obtained. 10.21 In the past, roughness has been measured in units of 'mm/km' based on a BI trailer towed at 32 km/h. These units can be converted to international standard (IRI) values using the following equation: m/km IRI = 0.0032 (mm/km BI)0.89 10.22 In the absence of any information for assessing the surface condition of the road, the user will require some guidance on estimating the roughness of the road under study. A range of roughness values are given in Table 10.2 below. These values provide very broad guidelines by basically demarcating different pavement types in terms of a range of roughness. 10.23 The value of roughness used should be the average value experienced by vehicles over the period for which operating costs are being calculated. Although estimates need only be made to the nearest 0.5 m/km, the value used will have a significant effect on the magnitude of the operating costs and some effort needs to be made to determine the appropriate value. 10.24 Road width. As noted in Section 7, for free-flow traffic conditions, width has little effect on speed for wide roads but, on narrower roads, the speeds are reduced.

Vehicle descriptions
10.27 Free-speed of vehicles. In order to estimate the average speed of vehicles on the road under investigation, it is necessary to determine the free-speed of each class of vehicle in the environment under investigation. This is defined as the average speeds at which vehicles of different classes will travel on uncongested, flat, straight, smooth and wide sections of road in the particular environment. These speeds have been found to be affected by the general layout of the roads in the area through differences in driver behaviour and vehicle performance. For example, vehicles operating in a hilly or mountainous region are found to have considerably lower free-speeds than those of vehicles operating in flat or rolling open terrain. Similarly, the free speeds of vehicles operating in countries where straying animals are common, or that tend to have slow-moving animal-drawn carts, are found to be lower than in countries where this tends not to happen. 10.28 The user should conduct a vehicle speed survey where the speeds of the various classes of vehicles are recorded on a flat, straight, smooth and wide section of road. The average speed for each class of vehicle will be the free speed for that class for the environment under investigation. 10.29 Vehicle weight. Fuel and tyre consumption are influenced by the weight of light vehicles, trucks and buses. The best estimates of vehicle weights for different classes of commercial vehicle can be found from axle-load surveys conducted on the route under study (see Section 6). In the absence of information from such surveys, a knowledge of the nature and type of goods transported on the route together with the size distribution by carrying capacity of commercial vehicles will help to provide reasonable estimates of the average vehicle weight

53

TABLE 10.2 ROAD ROUGHNESS VALUES FOR DIFFERENT ROAD SERVICEABILITY LEVELS

54

of the different classes of vehicle. The value used should be the total weight of the vehicle over the period for which the operating costs are being determined. This involves making an estimate of the loads carried by the vehicles and the amount of unladen running. 10.30 Power to weight ratio. The power to weight ratio is the net brake horse power per tonne of the gross vehicle weight. If an axle-load survey is carried out this could also be used to provide information on the brake horse power of the various classes of commercial vehicle using the route, and so enable the average brake horse power per tonne of vehicle weight to be calculated. Table 10.3 gives examples of commercial vehicles by type and average brake horse power and the power to weight ratios for different conditions of load. They are representative of the type of vehicles operating in most developing countries. 10.31 Vehicle age. In order to determine vehicle operating costs, vehicle age needs to be determined as the total distance run since new as well as in years. The age in kilometres is a significant factor in determining average vehicle repair costs and the age in years is used to determine depreciation. Normally average values for vehicle classes will be used in the calculation of the vehicle operating cost components. Vehicle age spectra, both in terms of kilometres and years, vary widely from country to country, and it is vital that good information is collected in the field in all cases. This is emphasised because the components of depreciation and repairs normally form a significant proportion of total vehicle operating costs (see Table 10.1). The value of age in kilometres used should be that at the end of the year for

which the operating cost is being evaluated, whereas the age in years should be the mid year value. 10.32 Annual utilisation. Vehicle utilisation in hours per year is used directly to determine annual crew costs and to apportion annually accruing costs such as depreciation and overheads on to a kilometre basis so that they can be related to usage on particular roads. Utilisation in kilometres per year is important because, when multiplied by vehicle age, is used to determine spare parts costs. These costs are normally some of the largest components of vehicle operating cost (see Table 10.1), 50 the need to collect good local data on utilisation is emphasised. 10.33 Vehicle price. Average vehicle price is used to determine depreciation costs and, in the TRRL investment model and operating cost tables, is also used to determine spare parts costs. Both these costs are a significant proportion of the total cost (Table 10.1) and are highly sensitive to the initial vehicle price. Good local data should be collected and this is normally easy to do. Price values should be obtained for the mid point of the year for which costs are being evaluated. 10.34 Tyre price. This obviously affects the cost of tyre wear and local data is easily obtainable. Mid year values should be used. 10.35 Price of fuel and lubricants. Information on prices of these commodities is easy to collect locally. Again, mid year values should be used. 10.36 Price of maintenance labour. This is needed to determine the labour component of fitting spare parts and

TABLE 10.3 SAMPLES OF GOODS VEHICLES SHOWING VEHICLE WEIGHTS AND POWER/WEIGHT RATIOS (Extracted from a range of European and Japanese manufacturers specifications)

55

repairing vehicles, although its contribution to vehicle operating cost is relatively small (Table 10.1). It is important that prices include all supervision and overhead costs otherwise underestimates will be made. Prices will vary between countries and also within a country depending on the size and type of workshop carrying out repairs. A weighted average of unit prices should be used between large and small repair workshops depending on the proportion of vehicles on the project road that are repaired in each. This proportion is easily estimated from information collected in origin and destination surveys (see Section 3). 10.37 Vehicle crew cost. Crew cost is normally included as a vehicle operating cost rather than as a time cost. It is determined by the number of crew that a vehicle carries, their wages and the overheads of employing them. As in the case of maintenance labour prices, the crew cost will depend on the size and type of organisation operating the vehicle, and unit values should be based on weighted averages in a similar way. Crew costs can be a significant proportion of the total cost for commercial vehicles (Table 10.1). 10.38 Overheads. This component of vehicle operating cost includes such items as interest on capital and garaging. Insurance costs may also be included as a surrogate for accident costs. Maintenance labour costs should not be included here to avoid double counting. The costs will vary from country to country and, depending on the size of the transport operator, will vary from company to company. A weighted average of costs should be used to represent typical vehicles on the road under study. Overheads can represent a significant proportion of operating costs and, although good data is sometimes difficult to collect, considerable effort should be made.

information. In this case, the tables of vehicle operating cost may prove useful to determine the change in vehicle operating costs with road geometry, surface condition, etc. 10.40 Road improvements do not necessarily lead to reductions in all components of vehicle operating costs. When a gravel road is paved, lower values of roughness reduce fuel consumption, but increases in speed have the opposite effect, with the result that fuel consumption is often virtually unchanged. Spare parts requirements depend on roughness values and vehicle age in kilometres, hence road improvements that result in increased vehicle utilisation will tend to increase costs whereas roughness reductions will lower costs. Depending on the relative effect of these parameters, the cost of the spare parts replacement component can either increase or decrease. Nevertheless, when all components are considered, road improvements normally reduce vehicle operating costs, but these examples emphasise the need to collect data and determine costs specifically for each individual project. 10.41 For road upgrading projects, vehicle operating costs should be estimated for the existing road and the new project. Savings in vehicle operating cost for normal traffic represent a direct benefit of the project. Where the project results in traffic diversion taking place, all vehicle operating costs on both the road from which the diversion has taken place and on the project road should be considered when determining benefits. Benefits accruing to any traffic generated by the project are discussed in para 14.7-11. For upgrading projects for arterial and collector roads, vehicle operating cost savings for normal traffic will usually be the largest project benefit. For rural access roads, where traffic flows are low, vehicle operating cost savings may be insignificant. Road maintenance projects usually reduce roughness values and, hence, affect vehicle operating costs.

DETERMINING COSTS
10.39 The vehicle operating cost relationships included in the road investment models and tables are appropriate for use in most developing countries with the exception of those for vehicle depreciation which may differ in countries where import policies influence the age spectrum of vehicles. This does not mean that the relationships are 'perfect', but that it is unlikely that economists, planners and engineers carrying out feasibility studies can produce better relationships in the time available during an appraisal mission. However, it is vital that the project report provides evidence of having collected data from vehicle operators to check that the size of the operating costs predicted by the investment models or tables is correct for the country and the projects in question. If the size differs considerably, the appraisals should determine why this has occurred and vehicle operating costs should be computed manually purely from local

56

11. ROAD MAINTENANCE BENEFITS

11.1 Despite the difficulties experienced by many developing countries in carrying out adequate maintenance (see para 2.22-26), savings in road maintenance cost are a potential benefit from many types of project and are particularly welcome because they release scarce resources for maintenance of other roads. 11.2 Maintenance savings can normally be obtained with the following types of project: Paving a gravel road where traffic levels have increased Strenthening or reconstructing a road which has deteriorated badly.

benefits and agricultural benefits may arise as a result of paving gravel roads, but these are difficult to quantify in an economic analysis. 11.7 Where economies in maintenance are made as a result of paving gravel roads, vehicle operating cost savings will also normally be made. These two benefits are linked closely together and road investment models are therefore very appropriate for carrying out the analysis.

STRENGTHENING AND RECONSTRUCTION

11.8 A bitumen road with a rapidly deteriorating surface needs increasing amounts of maintenance if it is to continue serving its intended purpose. A bitumen road may require the patching of pot-holes, repair of eroded edges, and the sealing and repairing of cracked areas. Compared to this, the overlaying or reconstruction of the road can produce immediate savings by eliminating the need for continuous recurrent maintenance, although future periodic maintenance will still be needed. It is, however, important to strengthen pavements before they deteriorate to the extent that their structural integrity is lost. 11.9 Road investment models can be used to assess maintenance benefits in these cases, but their use is limited because the modelling of badly maintained pavements that are in poor condition is not very accurate. Thus, any assessments of maintenance benefits for this situation which have been derived using investment models should be treated with caution. 11.10 The cost of strengthening and reconstructing paved roads is considerably greater than the annual cost of routine, recurrent and periodic maintenance, so it will be unusual for projects of this nature to be justified solely on the grounds of economies in maintenance. Projects will normally be justified principally on vehicle operating cost savings and any maintenance savings will increase the benefits and lead to a higher rate of return.

PAVING GRAVEL ROADS

11.3 In order to keep gravel roads in an acceptable and economic condition, their surface will normally need grading several times a year and regravelling every few years. The frequency at which these activities are needed depends on the level of traffic, the type of gravel material and the climate. As traffic levels increase, the frequency of the maintenance activity needs to be increased and eventually the cost of maintenance is so high, that it becomes cheaper to provide a paved road. 11.4 The actual traffic level at which paving becomes economic should normally be determined using one of the investment models described in Section 10. It is not possible to give recommended traffic levels, because these values will depend on the relative costs of grading, regravelling and paving which, in turn, will depend on local circumstances. The higher the relative cost of grading and regravelling, the lower will be the traffic level at which paving becomes justified. 11.5 A further difficulty is that sources of good road building gravel are becoming scarce in many developing countries with the result that haul distances and costs are increasing. It may therefore be appropriate in appraisal studies to re-estimate the unit cost of regravelling during the life of the project to take account of this. A consequence of this will be that, in some cases, it may be appropriate to pave a road earlier and at a lower traffic level than was previously the case. 11.6 In arid areas, unpaved roads are often affected by dust. Dust is a maintenance problem because it results in the loss of material from the road surface which has to be replaced. It is a contributory factor to road accidents because of the reduction in visibility and it also pollutes the atmosphere close to the road and may reduce the value of crops. Hence, road safety and environmental

CONCRETE ROADS
11.11 Where traffic levels are rising rapidly, and particularly when large increases in goods vehicles can be expected, the provision of a concrete surfacing to an existing gravel road may prove to be economically justified. Similarly, concrete overlays to existing bituminous surfacings are likely to reduce future maintenance costs. Experience of the construction of concrete pavements is limited at present to very few developing countries and experience of concrete overlays is almost entirely limited to Europe and North America. The use of these techniques should therefore be treated

57

with caution, particularly as the investment models currently available cannot assess their viability.

DIVERTED TRAFFIC
11.12 If significant traffic diversion from other roads is expected to take place as a result of a new project (see para 3.19-21), then the changing maintenance needs on the road from which the diversion took place should be considered in the assessment of benefits. Reduced maintenance needs on the existing network will normally result in a small benefit to the project, although this may be offset by an increased cost of maintenance on the project itself.

quoted by roads departments. The quality of field recording of activities and expenditures is usually very poor with the result that the usefulness of the data collected is very doubtful. Many costing systems in use only attempt to provide details of total expenditure for budgetary purposes and it is not possible to identify in detail the activities on which expenditures have taken place. 11.16 Against this background, it is difficult to obtain realistic unit costs which can be used to determine maintenance savings for many countries. However, in most cases, projects will not be justified solely on the grounds of maintenance savings as these will be small in comparison with savings in vehicle operating costs. Nevertheless, maintenance cost estimates are a necessary part of appraisal, including cases where they are a negative benefit, and an attempt to collect good local cost information must be made. Available records in maintenance organisations must be examined to provide the basis of cost estimates, but these should be reviewed in the light of knowledge of how the records are obtained. In all cases, the sensitivity of benefits to large potential errors in the cost estimates should be determined.

TRAFFIC DELAYS DURING MAINTENANCE WORKS

11.13 When large scale maintenance and renewal works take place on heavily trafficked roads, delays to traffic and increased accidents are likely to occur. For project appraisal purposes, where future strengthening or stage construction is being planned, these additional costs should ideally be taken into consideration as part of the appraisal. However, where these works are taking place in the later years of the project's life, the effect of additional costs of delay and accidents on the outcome of the project are likely to be small in present value terms on all but the most heavily trafficked roads because of the effect of discounting (see para 15.2-11). In these cases, lump sum estimates should be made of the additional costs for heavily trafficked roads; additional costs can be ignored on lightly trafficked roads. 11.14 However, where the project is for the upgrading of an existing paved road to provide additional capacity or structural strength, the additional costs will occur early in the projects life and are therefore more likely to influence the choice or timing of the capital investment. The costs of traffic delays will increase if projects are delayed, because traffic levels will be higher. For very heavily trafficked roads, a more rigorous estimate may be appropriate.

DETERMINING COSTS
11.15 Maintenance costing systems that are implemented in organisations are often not accurate enough for determining maintenance cost savings. Typically, costing systems undervalue the costs of owning and operating plant and equipment by a significant amount by failing to include interest charges or even the replacement cost of the equipment. Costing systems seldom include realistic overheads for employing personnel and providing buildings and other facilities. The result is that real costs are commonly more than 100 per cent greater than those

58

12.

TIMESAVINGS

GENERAL CONSIDERATIONS
12.1 Journey time savings can represent a large proportion of a project's benefits. The benefits of shorter journey times will accrue to the vehicle fleet, in that greater vehicle productivity can be achieved, and to the passengers and freight being carried. A general discussion of some of the principles involved in the valuation of time savings is given below, together with a suggested approach to their quantification and incorporation in a feasibility study.

12.4 Looking at the problem from an overall point of view, because of the discrete nature of most activities, the vehicle fleet cannot be productively employed for 100 per cent of the working day. If, after the project is completed, vehicles on average are working for the same proportion of the working day as in the before situation, this is equivalent to saying that time savings are fully used. To assume that, in the long term, time savings should not be costed as if the time were fully used is to imply that there is some special feature of the before situation which gives rise to an efficient use of time which will never be matched in the after situation. 12.5 It may well be that adaptation of current transport activity to take full benefit of the reduction in travel time brought about by the project will not be immediate. However, it would be difficult to judge the true form of the lag between change and benefit on the basis of detailed examination of the activities of individual operators. On the whole, unless other reliable information is available, it is safest to assume that all time benefits are available at once. 12.6 In the case of privately owned cars, then the above discussions are less appropriate. The demand for transport by a car owner is not shared between a number of vehicles but falls just on his own vehicle. If his travel time on a particular journey were to fall, this is unlikely to reduce directly the number of vehicles owned. It may well encourage the car owner to make more journeys, but the treatment of this is separately dealt with in the discussion on traffic generation benefits (para 14.7-11). Taxis should be considered in the same way as other commercial vehicles

VEHICLE FLEET
12.2 Consider first vehicles which are used exclusively for commercial purposes such as buses and lorries. When travel time is reduced, the time saving can in principle be used to make further journeys, and hence productivity per vehicle rises and the size of fleet necessary to support the current demand for transport can be reduced. This reduction in fleet size means a reduction in those elements of the fleet operating costs which are classed as standing costs, notably crew wages, vehicle depreciation and interest on capital. By using appropriate values of vehicle utilisation in the 'with' and 'without' project cases, these cost savings will be determined directly. 12.3 It is often argued that, in practice, time savings cannot be properly utilised and, as a result, will not lead to pro-rata reductions in fleet size. The reasoning for this is that currently most journeys are 'quantised' as round trips, such as a complete circuit of a bus route, or a delivery made by road where the lorry both starts and ends its journey at its base. If travel time on any of these journeys were saved, the chances are that it would be insufficient to permit another round trip during the same working day and, as completion of only part of the trip within a working day is not acceptable, the time saved could not be usefully employed. One of the problems with this kind of argument is that, in some instances, the time saving might just be adequate to allow another round trip and, in these cases, the benefits could be far more than simply prorata. Overall, one has to try and visualise a pattern of use which fairly represents the whole of the current pattern. Unfortunately, it has to allow, for example, for the possibility, in the case of buses, of extending the route, having additional stops, etc and, in the case of lorries, of loading the night before, staying out overnight, etc. Additionally the demand for transport is subject to fluctuations and long-terms trends, and travel times themselves may also be subject to fluctuations and trends for reasons not associated with the project under review. Clearly, providing a long-term realistic and representative picture is overwhelmingly difficult in all but the very simplest of situations.

VEHICLE OCCUPANTS
12.7 Travel time savings for passengers in buses and the occupants of private cars may occur either during working or non-working time. Time savings during working hours can be used for productive purposes to increase the GNP. Nonworking time savings do not increase national production but, since there is evidence that people are prepared to pay for time savings that occur in non-working time, such savings must be perceived as increasing their welfare. 12.8 If working time is spent travelling, the value of that travelling time is clearly equal to the wage rate plus those costs to the employer which are directly associated with the costs of employment. In practice, the situation is not so straightforward. There are imperfections in the labour market, especially where minimum wage legislation exists, where there are high rates of unemployment, or significant levels of under employment. Despite these problems, it is usually assumed that working time savings should be equated to the average wage rate plus

59

overheads associated with employment, such as pensions, insurance, etc, shadowed priced if appropriate (see para 15.12-17). 12.9 The value of non-working time is usually based on perceived cost studies. Most of the research into perceived costs has taken place in the developed world, but similar results have been found in studies undertaken in developing countries. The studies show that the value put by individuals on journey time savings accruing outside working hours is between 25-45 per cent of their earnings and that higher unit values of time saving should be ascribed to higher income groups than to lower income groups. In practice this is rarely done because it is considered inequitable. In the United Kingdom, for example, a flat rate equivalent to 43 per cent of the average hourly earnings is used in the evaluation of nonworking time travel savings for full time adult employees. This value is an average of both commuting and leisure time. Where governments wish to adopt a policy that maximises GDP rather than leisure time preferences, a zero value should be used for leisure time whilst maintaining working time values. To use a percentage of the average wage may lead to an underestimate of time costs in developing countries because only the comparatively wealthy can afford to travel, even by bus, and certainly by car. 12.10 Other problems occur in the valuation of passenger time savings. The distinction between working and nonworking time is not always clear cut, especially when many trips are multi-purpose. Marginal values of time may vary for the same individual, depending on the activities for which the time saved is used. The value of time is normally a function of factors other than a tradeoff between time and cost, such as comfort and convenience, and in the UK, for example, walking and waiting times are valued more highly than travelling times. 12.11 As a general guide, the following approach should be adopted: (1) (ii) Measure time savings separately for working time and leisure time, as a minimum, In the absence of better data, value working time at the average wage rate in the monetised economy, plus overheads, Value non-working time in the range zero to 45 per cent of working time, unless there special reasons for attributing at higher value. It would normally be expected that values would be at the lower end of this range.

represent, costs due to damage or spoilage of perishable goods, and ancillary costs which arise as a consequence of journey time, for example, where a piece of equipment is immobilised while waiting for a spare part. The cost of interest on capital is normally very small compared to the other elements of vehicle operating costs. Costs due to spoilage or damage may be significant, but care must be taken to ensure that a reduction in spoilage or damage of perishable goods is due primarily to reductions in journey time rather than the provision of a smoother road. If it is the latter, and this is more usually the case, then the cost savings should still be credited to the project but, strictly, not be allocated as a time saving. 12.13 Studies of modal choice for goods travelling by road and other modes have suggested that, even for nonperishable goods, consignors are usually willing to pay far more than interest cost on the goods to reduce travel time or to reduce uncertainty in time of delivery. This presumably reflects the size of the ancillary costs mentioned above.

(iii)

FREIGHT
12.12 The cost of delays in moving goods consists chiefly of costs due to interest on the capital which the goods

60

13.

REDUCTION IN ROAD ACCIDENTS

accident reduction resulting from low cost engineering counter measures introduced to improve the safety of specific sites.

FORECASTING ACCIDENT REDUCTIONS Factors leading to reductions

13.1 Although accidents are one of the inevitable costs of road transport, there are nevertheless a number of measures which can lead to their reduction with corresponding reductions in costs. Road accidents should therefore be included in an economic analysis of road improvements. 13.2 The factors contributing to road safety are usually referred to as the three 'E's': engineering, education and enforcement. Both education and enforcement fall outside the scope of this Note and will not be discussed further. However, it must be appreciated that these factors interact, and different combinations of factors are likely to have different impacts. Consequently, it is important that countries adopt a coordinated approach to road safety at national, regional and local levels. 13.3 The engineering factor can itself be broken down into five categories: geometric design road surfaces road markings and delineation road signs, streetlights and other road furniture traffic management.

Effect of highway design

13.7 The majority of highway design and safety studies in developing countries have investigated the correlation between personal injury accident rates and certain geometric design characteristics for rural roads. These have indicated that the number of junctions per kilometre appears to be the most significant factor related to accident rates, followed by improvements in horizontal and vertical curvature. In most of the countries studied, there was little variation in road width, so no conclusions could be drawn on the effect of this. However, in the one country where considerable variation did occur, road width was found to be a significant factor affecting accident rates. The data suggested that, on a range of roads carrying between 200 and 2000 vpd, an increase in width from 5 metres to 7 metres might reduce the accident rate by 40 per cent. 13.8 High accident rates were observed on gravel roads. Among the possible causes of this might be poor geometry, slipperiness of the surface in wet weather and poor visibility caused by dust and high vehicle speeds. Since, in two of the countries studied, accident rates reduced with reduced road roughness, it is likely that by paving gravel roads, accident rates will be reduced. 13.9 Further results obtained, suggest that the construction of a new dual carriageway road carrying up to 20 000 vpd between major towns can reduce all injury accidents by up to 50 per cent and fatal and serious accidents by about 25 per cent. 13.10 The effects of the different improvements in highway design are not additive and there is clearly a limit to the likely benefits that can be obtained from improved design. From the above results, it may be possible to expect up to a 40 per cent reduction in accident rates on an existing road with improvements in geometric design and planning.

13.4 The relationship between the various components of the engineering factor have been widely researched in industrialised countries and, although the findings are not totally clear cut, individual industrialised countries are able to make some predictions for accident reduction. 13.5 However, in developing countries, only a very few studies have been carried out and, at this stage, predictions cannot be made with any certainty. Also, many of the results indicate that the findings from developed countries cannot be directly transferred to the third world because of the different physical, cultural and traffic conditions found in typical developing countries. 13.6 In spite of this scarcity of data, specific comments on the effects of highway engineering improvements on accident rates in developing countries can be made. Examples of where appropriate engineering design can reduce accident rates are given in Fig 13.1. In considering these effects, a distinction can be drawn between: accident prevention resulting from improved standards of highway design and planning, and

Low cost remedial measures

13.11 There is increasing evidence from developed countries that the combination of relatively detailed local accident investigation with low cost engineering remedial measures, can be highly cost-effective. Because driver behaviour and knowledge are much poorer in developing countries than in industrialised countries, it may well be that engineering measures that are 'self enforcing', such as median barriers, guard ralls, pedestrian segregation, etc, prove effective, whilst measures such as improved signs, road markings and speed limits may not. There is

61

Fig.13.1 Examples of effect of engineering design on road safety

62

some evidence to suggest that these measures can be made effective if coupled with improved enforcement techniques.

Effects of traffic
13.12 If accident savings are to be included in highway project analysis, another factor that needs to be taken into account is that of the relationship between accident rates and traffic flow. Work on this subject tends to be inconclusive, even in the developed world. This uncertainty can be largely attributed to the multi-causal nature of accidents. 13.l3 From an appraisal of work carried out in industrialised countries on lightly trafficked rural roads, it would appear that the number of single-vehicle accidents per unit of vehicle distance travelled, tends to decrease with increasing traffic, whilst the number of collisions between two or more vehicles tends to increase. With the scarcity of data available, the most sensible approach when estimating changes in accidents with traffic flow is to assume that they increase at the same rate.

13.17 Practical problems which are encountered when measuring material costs are frequently due to the ambiguity of the definitions used. In other cases, estimates have to be based on numerous assumptions because of limited empirical evidence. Therefore, it is not possible to give a 'correct' figure for accident costs, and estimates of costs, whatever their origin, must be interpreted and used with discretion. Accident cost studies should be used as an aid to decision making and not be regarded as definitive.

Methods available for costing road accidents

13.18 The selection of an accident cost methodology must be related to the objectives being pursued by the agency that will ultimately use the costs obtained. Three broad classes of objectives can be identified which encompass the primary goals of the majority of planners in developing countries: National output objectives such as maximisation of Gross National Product (GNP) or national income Social welfare objectives such as the minimisation of all types of accidents in the interests of the well-being of the community Mixed objectives whereby governments combine the above objectives in various forms; this inevitably leads to the trading-off of one objective against another.

ROAD ACCIDENT COSTS Material and subjective factors

13.14 Whilst it is not easy to attribute monetary values to the losses arising from accidents, estimates of the magnitude of the material costs are an essential aid to decision-making in the road safety aspects of highway engineering projects. Accident costs are also needed for the appraisal and optimisation of road safety countermeasures. Resources for countermeasures are usually limited and, in the absence of accident costs, it would clearly be difficult to make an objective assessment of which proposed projects should be given priority. 13.15 Costs of road accidents arise from three areas: Damage to vehicles and other property Costs of police work, hospital treatment, administration, etc Loss of life and injury.

13.19 All methods that have been used in the past for costing or valuing accidents have included the material or real resource costs of the accident in terms of damage, medical and police reports, etc. The contentious issue of valuing human life has been dealt with in a number of ways, as follows. 13.20 Gross output (or 'human capital') approach. The value of human life is taken as the discounted present value of the victim's future output. 13.21 Net-output approach. This differs from the 'gross output' approach only to the extent that the present value of the victim's future consumption is subtracted from the gross output figure. 13.22 Life insurance approach. The value of life is defined as the amount for which individuals are willing to insure themselves. 13.23 Court award approach. The sums awarded by the courts to the surviving dependants of those killed are treated as indicative of the cost that society associates with a fatality, or the value that it would have placed on its prevention. Real resource costs are then added to this figure to obtain the cost of an accident.

13.16 All losses which result directly or indirectly from the occurrence of road accidents should be incorporated into the cost analysis. Certain losses, such as damage, police and hospital costs, involve material resources and are normally readily defined, even though their values may be uncertain. They can be translated into economic terms without great difficulty. Costs relating to the loss of life and injury are subjective, involving the need to value human life and 'pain, grief and suffering'. It is conceptually awkward to assign monetary values to these.

63

13.24 Implicit public sector valuation approach. This attempts to determine the costs and values that are imp]icitly placed on human life in safety legislation, or in public sector decisions, taken either in favour, or against, investment programmes that affect safety. 13.25 Value of risk-change approach. Each individual has their chance of being involved in a fatal accident reduced by a small margin if a road safety improvement is introduced. Thus, the value of preventing one fatality in one accident is defined as the amount, in aggregate, that all the affected individuals in society are willing to pay for these small risk reductions. 13.26 Whichever method is used, a significant sum may also be added to compensate for pain, grief and suffering. The value of this is also very contentious. 13.27 Not surprisingly these different approaches generate substantially different costs and values for accidents involving one fatality. Depending on which method is used, the cost of an accident involving a fatality can vary by several orders of magnitude.

of data of the appropriate kind and quality. However, in the evaluation of accidents, the choice of objectives and methods should always be viewed as the starting point, with deficiencies of data being identified as a consequence, rather than vice versa. 13.31 The various accident costing methods should be viewed as a system which increases in comprehensiveness and sophistication. This moves from the hard and undeniable material costs of physical damage to vehicles and property, through to the valuation of human life, and onto highly subjective values based on willingness of individuals to pay for a reduction in their perceived risks of death or injury. Each country can operate at different stages of this system of increasing complexity depending on their choice of objectives and availability of data.

Recommended method of costing

13.32 The 'gross output' or 'human capital' approach to accident costing or valuation is recommended where the maximisation of GNP is the most important criterion. In this case, the cost of a traffic accident involving one fatality is treated as the sum of the real resource costs of vehicle damage, medical and police costs, plus the discounted present value of the victim's future output. If accident costs and values are intended for use in conventional cost-benefit analysis, then it can be argued that only the 'value of risk change' approach is unambiguously relevant. In practice, however, this method has proved difficult to use even in the industrialised countries. Output-based methods, provided they include a 'pain, grief and suffering' component may be relevant providing they reflect individuals' attitudes to the prospect of their involvement in accidents. 13.33 Clearly, the valuing of human life is a difficult and often contentious process. No one method can be said to be ideal under all circumstances. However, the adoption of a single gross figure, based on national increase, for use as a broad indicator requires fewer dubious assumptions or unacceptable implications than the alternatives. Many of the objections to valuing life can be answered if it is clearly specified that the sum derived is a minimum that society would find worthwhile spending in order to avoid a fatal accident. 13.34 Failure to associate explicit costs to road accidents will lead to wide differences in the assessment of projects that affect road safety. The resulting inconsistent pattern of treatment could yield implicit costs that might range from zero to a significant proportion of the potential benefits of a project. Consequently, it would be unlikely that overall expenditure on road safety would be optimal in such cases. This would lead to an under investment in road safety at the national level.

The relevance of alternative methods for developing countries

13.28 Typically, government objectives will involve the maximisation of either national output or social welfare. Two accident costing/valuation methods which are directly relevant to these are the 'gross output' and the 'value of risk-change' methods. There are practical problems in recommending that one or other of these objectives be followed without modification. Firstly, reliable data is essential and, secondly, practical and political problems may deter governments and their agencies from implementing certain objectives. 13.29 Since resource costs such as vehicle damage, police and medical costs, together with net output losses, form a common component of both gross output and value of riskchange, these should be regarded as an absolute minimum cost of an accident or value of accident prevention. If the maximisation of GNP is an important criteria for decision takers in developing countries, then a costing/valuation method based upon gross, rather than net, should be used. It should be borne in mind that this method does not take into account society's aversion to death. To resolve the latter, a component for pain, grief and suffering could be included. Most western countries use the gross output approach plus a component for pain, grief and suffering, although there is now some doubt as to whether even this methodology is satisfactory, and a value-of-risk approach is now being considered. 13.30 The feasibility of any of the preferred methods will, in the short term, be conditioned by the availability

64

14. ECONOMIC BENEFITS

DEVELOPMENT

THE IMPACT OF DIFFERENT FORMS OF ROAD INVESTMENT

14.1 The immediate economic consequence of road investment is to lower transport costs. As a result, economic activity will be changed throughout the whole economy as the saved resources are redeployed, as producers adjust to their new cost and price structure, and as consumers adjust their pattern of expenditure. The extent to which the local economy adjacent to the road will benefit from the investment will be dependent on its economic potential, such as unused land and labour, and on the change in transport costs and prices. The effect on the economy is extremely complex and it is virtually impossible to model in detail. 14.2 For most road projects where vehicle access already exists, however rudimentary, the principal benefits from the project should be measured as road user cost savings as described in Sections 10, 12 and 13. In these cases, a 'consumer surplus' approach to assessing benefits should be used as described in para 14.7-11. 14.3 Generated traffic may arise as a result of transport cost savings when roads are upgraded (see para 3.22-26). This traffic is a measurable indicator that economic development benefits have arisen. Where the change in transport costs are relatively small, generated traffic benefits will, in most cases, represent a very small component of total benefits and can often be ignored. By contrast, when the change in transport costs are large, then generated traffic benefits are more likely to be important. 14.4 Generated traffic benefits are usually the most difficult component of benefits to estimate. As a result, the 'producer surplus' approach can sometimes be used as an alternative (see para 14.12-15). This approach predicts agricultural producer benefits resulting from the investment in the vicinity of the road. Other components of generated traffic, such as changes in passenger movements, still have to be estimated separately. The approach is of most use when a large change in transport costs is expected for crops which have low value to weight ratios, like sugar cane, or when new road access is planned to be accompanied by other agricultural development inputs that are transport intensive, such as the introduction of fertiliser. The producer surplus approach is difficult to use and should only be applied when there are good grounds for believing that it will give better results than predicting generated traffic benefits by more conventional means (see para 14.7-11). 14.5 When evaluating generated traffic benefits, it is useful to consider the current traffic composition and the

nature of the proposed investment. Studies have shown that passenger traffic is more sensitive than freight traffic to changes in transport costs. Passenger fares are a direct component of consumers' final demand whereas freight costs represent only a small proportion of the final costs of both the product to the consumer and the revenue to the producer. Upgrading long lengths of inter-urban roads to a high standard may have little effect on freight traffic, but may weli have an important effect on passenger traffic, particularly for private motor car traffic, which is often deterred from using poor quality road surfaces. However, upgrading short lengths of road will change transport costs very little and, as a result, will have little effect on traffic levels or on agricultural production. The only exception to this is when roads are cut for long periods during critical periods of the crop season, or if crops, like bananas for export, are damaged during transit. The majority of rural access road projects involve upgrading roads and tracks of up to about 20 km. For these projects, road user cost savings for forecast normal traffic is the most appropriate method of estimating benefits. 14.6 Providing completely new vehicle access can change transport costs dramatically. For example, the cost of headloading is typically twelve times the cost of motor truck transport. Where it is planned to build access roads to rural communities that previously had to rely on human or animal transport, then transport cost savings (including a valuation of passenger and walking time savings) for normal traffic will often be sufficient to justify the provision of motor vehicle access at MINIMUM standards (see Section 7). Initially, such access will probably require simple bridging and culverts, with the use of gravel surfacing material only in problem areas. Later on, if traffic levels warrant, the road can be upgraded.

CONSUMER SURPLUS
14.7 If reductions in transport cost result from a road project, there will be a direct benefit to road users which equals the product of the number of trips and the cost saving per trip. This cost saving, or consumer surplus, may be vehicle operating costs, time costs or road accident costs, of a combination of the three. Its evaluation has been discussed in Sections 10, 12 and 13. Technically, there is only a consumer surplus if cost savings are passed on to consumers through lower fares and freight charges; otherwise they accrue to vehicle operators as producers' surplus. It is therefore important to assess the prevalling market and make judgements as to how any transport costs reductions are likely to be distributed. 14.8 If the transport cost savings are sufficient, these may result in more trips being made and extra benefits will

65

Fig.14.1 Benefits measured as consumer surplus

accrue as a result of this generated traffic. Thus, generated traffic resulting from a road project is a measure of the extra consumer surplus, and can be used to determine the project's developmental benefits. 14.9 Consumer surplus benefits are best estimated using a demand curve as shown in Fig 14.1. If, before the project is undertaken, t1 trips are made each day at a unit cost of c1, then the transport cost is c1t1 per day. If, as a result of the project, unit transport costs are reduced to c2, then the transport costs of the traffic t1 are reduced to c2t1 per day giving a Benefit to normal traffic = (c1 - c2)t1 per day. This is the benefit discussed in Sections 10-13. 14.10 If additional traffic is generated as a result of the savings in unit transport cost, additional benefits will accrue. The prediction of generated traffic is discussed in para 3.2226. The amount of traffic that is generated will depend on the size of the unit cost reduction and on the ability of the consumer to take advantage of this cost reduction. This ability is known as the elasticity of demand. A demand curve is shown in Fig 14.1. In this case, a cost reduction from c1 to c2 will result in an increased number of trips from t1 to t2: the greater the cost reduction, the more trips that will be generated. The demand curve can normally be approximated by a straight line whose gradient is related to the elasticity of demand. The area under the demand curve less the transport cost of the generated traffic, c2(t2 - t1), gives the Benefit to generated traffic =0.5 (c1 - c2)(t2 - t1) per day.

14.11 In areas where there already is considerable economic activity and traffic levels are relatively high, the consumer surplus approach should normally be used to provide an estimate of the total development benefits associated with a road project. For more complex situations, consumer surplus benefits can be estimated using the methods developed by the World Bank and described by Van der Tak and Ray (1971).

PRODUCER SURPLUS
14.12 In situations where no conventional road exists and a substantial improvement in vehicle accessibility is planned to help develop an area, the producer surplus approach may be the most appropriate way of estimating agricultural benefits arising from road investment. For this method to be used requires a great deal of knowledge of the agricultural production function such as might be the case in a rural development project. 14.13 The predicted benefits arising from the reduced transport cost of agricultural produce will normally be the same as that predicted by a consumer surplus approach. However, when the producer surplus method is used, passenger benefits and other non agricultural cost savings still need to be estimated separately. 14.14 The forecast increase in agricultural production and the size of producer benefits are predicted from. (i) (ii) the rise in farmgate prices brought about by the decline in costs of transporting produce to market the decline in transport costs of agricultural inputs.

66

14.15 Unfortunately, the practical application of the agricultural production approach in the field, has been poor. The empirical justification for estimating changes in agricultural production has been weak and a failure to consider all the relevant costs of production has often led to the benefits being grossly over valued. It is not recommended to use the approach unless there is a great deal of knowledge about agriculture and its likely supply response to changes in input and output prices. More details on using the producer surplus approach are given in World Bank Staff Working Paper No.241 (Garnemark et al 1976).

15. COST-BENEFIT ANALYSIS

PRINCIPLES
15.1 This section describes the standard techniques used by economists for cost-benefit analysis. It is included for completeness for the benefit of engineers, transport planners and administrators who may not be familiar with these techniques. More details on cost-benefit analysis will be found in the ODA guide to the economic appraisal of projects in developing countries (Overseas Development Administration 1988). 15.2 The purpose of carrying out cost-benefit analysis is primarily to ensure that an adequate return in terms of benefits results from making a capital investment. An additional purpose is to ensure that the investment option adopted gives the highest return of those considered in terms of such things as the choice of route, the design and structural standards, and the timing of the project. 15.3 For economic appraisal, the assessment is made in terms of the net contribution that the investment will make to the country as a whole. Thus, the analysis differs from that which would be undertaken by private companies in appraising commercial ventures in that it attempts to evaluate economic costs and benefits rather than financial ones. The essential characteristics is to use 'opportunity costs' as a measure of resource rather than market prices. 15.4 Each project is unique and has features that prevent analysis following an identical pattern, although the same overall approach can usually be followed. It is normal to determine the costs and benefits which will be incurred over the analysis period if no investment is made, and compare these with the costs and benefits arising as a result of making an investment. Costs should be determined as described in Sections 4 to 8 and benefits should be determined as described in Sections 9 to 14. These costs and benefits can then be compared as described in para 15.18-31 to determine whether the investment is worthwhile and to identify which is the best of the alternatives being considered. The alternative in which no investment takes place is sometimes known as the 'baseline' or 'do nothing' case. However, it is unusual for future investment in such cases to be absolutely zero, as there is normally an existing road or track in existence which in the future will at least require some expenditure or maintenance. If traffic on the existing road is expected to grow rapidly in the future, perhaps because of some complementary investment, then relatively large capital investments may be needed just to prevent the road from becoming impassable. In cases such as this, the 'do minimum' alternative should be considered as the most realistic baseline case against
15.5

67

which alternative improvement projects should be evaluated. The choice of an appropriate 'do minimum' case is an extremely difficult decision and has a very large influence on the size of economic return obtained from a project. Considerable attention should therefore be given to its selection. 15.6 The project analysis and appraisal must always be carried out in terms of achieving the project's objectives as discussed in para 1.53-54.

15.10 The value of the discount rate used will clearly have a considerable influence on the balance between the effect of capital costs, which are typically spent early in the project life, and that of benefits obtained in the future. Discounted benefits may exceed costs at one discount rate, but not at another. The choice of discount rate is therefore crucial to the outcome of an appraisal in many cases. 15.11 The discount rate normally used is the government accounting rate of interest (ARI) which is the rate at which the value of uncommitted goverment income in constant price terms falls over time. The ARI is the opportunity cost of capital in the public sector, ie the rate of return on marginal public sector investments. The discount rate to be used in an appraisal will normally be provided by the planning authority responsible for the project. The method of determining its choice is described in the ODA guide to project appraisal, but is beyond the scope of this Note. In the absence of other information, figures of around 10 per cent are often used.

PRICES
15.7 In order to carry out an economic analysis, it is necessary to make adjustments to costs and prices to ensure that they are all measured in the same units and that they represent real resource costs to the country as a whole.

Inflation Shadow prices

15.8 A first step in this is usually to remove the effect of inflation to enable values to be compared on the same basis over time. Costs and prices are normally expressed in constant monetary terms, usually for the first, or base year, of analysis. In most cases, it can be assumed that future inflation will affect both costs and benefits equally, and hence its effect can be ignored. However, there may be exceptions to this and, in these cases, different costs and prices will need to be assumed for different elements at different times in the project analysis period. 15.12 If investment in the project is to improve the rate of economic growth through the reallocation of scarce resources, the taxation component of all prices should be deducted to give the economic price which should be used in the project analysis. This is because these charges do not reflect a demand on real resources, but represent a transfer of spending power from those benefiting from the project to the government. Other transfer charges include such items as vehicle licence fees which should also be excluded from the analysis. 15.13 Other distortion in the price system may arise through quotas, subsidies and through imperfect competition. Where market prices are fixed by institutional forces which cause them to be higher than would be expected in a completely deregulated market, resource costs would be exaggerated in the appraisal. The converse is also true. To overcome this problem, shadow pricing is used. Thus: Economic price = market price - transfer charge + effect of other distortions 15.14 Many developing countries control the value of foreign exchange to keep it lower in relation to domestic currency than is justified by the goods and services priced in domestic currency. Because the official exchange rate overvalues domestic currency, imported items appear too cheap and domestic items too expensive with the result that it tends to encourage overinvestment in imported capital items. This can be overcome by valuing all resources at their border prices. Imports are valued at the international price inclusive of cost, insurance and freight,

Discounting
15.9 It is also necessary to factor costs and benefits to take account of the different economic values of investments made at different times during the project's life. When money is invested commercially, compound interest is normally paid on the capital sum. The interest rate comprises inflation, risk and the real cost of postponing consumption. Thus, money used to invest in projects in the roads subsector could be invested elsewhere and earn a dividend. By using capital to invest in a project, the dividend is foregone and this should be taken into account in the analysis. To do this, all future costs and benefits are discounted to convert them to present values of cost using the formula: PVC =ci / (1 +(r / l00))
i

where ci = costs or benefits incurred in year i r = discount rate expressed as a percentage i = year of analysis where, for the base year, i=0. Since the inflation element is dealt with separately (para 15.8) and risk also needs separate treatment, the discount rate used will differ from market interest rates.

68

but excluding import duty (c.i.f.), and the exports free of any export duty (f.o.b.). This approach will tend to reallocate resources towards a mix of output where a country will only import goods that cannot be produced more cheaply at home, paying for them by exports which can be produced comparatively cheaply.

Where n i bi ci r

= the project analysis period in years = current year, with i=0 in the base year = the sum of all benefits in year i = the sum of all costs in year i = the planning discount rate expressed as a percentage.

15.15 Many developing countries have minimum wage laws or other regulations and inflexibilities which result in the wages being paid not correctly measuring the opportunity cost of labour. Where significant unemployment or underemployment exists, this results in the real cost of labour being much less than actual wage rates. A good estimate of the annual shadow wage for agricultural labour is the number of days in the year when most rural labour can expect to find employment, multiplied by the daily wage rate at such time. This may need to be modified further to account for overvalued domestic currency.

15.20 A positive NPV indicates that the project is economically justified at the given discount rate and, the higher the NPV, the greater will be the benefits from the project. If there are budgetary constraints, then the choice between projects should be based on NPV. 15.21 The NPV can only be calculated from a predetermined discount rate which needs to be the same for each project being compared. The NPV should only be quoted in conjunction with the discount rate that has been used. The rate used should normally be the government's own estimate of the minimum acceptable rate of return on public investment (as in para 15.11). 15.22 One problem with the use of NPV is that, other things being equal, a large project will have a larger NPV than a smaller one, and on this criterion would always be chosen. This can cause difficulties when only two or three projects are being compared. However, if all projects that could be undertaken with available public investment were appraised and ranked according to the size of NPV, the best choice would be that collection that maximised overall NPV. In this event, several smaller projects which in aggregate had a higher NPV would be chosen over a single larger project. 15.23 A range of NPV's should always be quoted to reflect the range of scenarios being investigated by the feasibility study. It is also important to consider the results of the financial, social and environmental appraisals when deciding which is the best project.

15.16 On the other hand, it would also appear that the real costs of skilled labour may be greater than the wages paid. The shadow price for this is difficult to estimate and advice should normally be sought from the relevant local ministry or commission. For both skilled and unskilled labour, shadow pricing should also be used when assessing benefits. If laboursaving equipment is introduced as part of a project, the real benefit is substantially less if the replaced labour remains unemployed for a significant period during the economic life of the equipment.

15.17 For further explanation of these issues see the ODA project appraisal guide.

COMPARISON OF ALTERNATIVES
15.18 In order to determine whether an adequate return in terms of benefits results from making a capital investment, cost-benefit analysis must be carried out. This can be done using either the 'net present value' or 'internal rate of return' decision rules. These rules may also be used for helping to determine which investment option gives the highest return of those considered. In addition, the 'first year rate of return' rule can be used to assess whether the project is timely.

Internal rate of return

15.24 This is the discount rate at which the present value of costs and benefits are equal; in other words, the NPV = 0. Calculation of IRR is not as straightforward as for NPV and is found by solving the following equation for r.

n1

Net present value

15.19 This is simply the difference between the discounted benefits and costs over the project analysis period.

b i - ci =0 (1 + (r / 100))i

i=0

n-1

NPV =

i=0

b i - ci (1 + (r / 100))i

Solutions are normally found graphically or by iteration. The IRR gives no indication of the size of the costs or benefits of a project, but acts as a guide to the profitability of the investment. The higher the IRR, the better the project. If it is larger than

69

the planning discount rate, then the project is economically justified

the other methods should still be presented to provide a broader picture. 15.31 The timing of a project should always be tested by evaluating the FYRR and, if this suggests that the project is premature, a range of investment timings should be investigated to determine which produces the highest NPV.

Project timing
15.25 Cost-benefit analysis should also be used to assist in determining the best time that a project should start. Even if the analysis shows that the project is worthwhile, there may still be a case for delaying the start whilst traffic continues to grow to increase the rate of return to a more appropriate level. The best way of determining the timing of the start is to analyse the project with a range of investment timings to see which produces the highest NPV. However, for most road projects, where traffic continues to grow in the future, the first year rate of return criterion can be used. 15.26 The FYRR is simply the sum of the benefits in the first year of trafficking after project completion, divided by the present value of the capital cost, grossed up by the discount rate to the same year and expressed as a percentage. Thus the FYRR is given by:
j-1

FYRR = 100.bj /

ci (1 + (r / 100))j - i

i=0

where j = first year of benefits, with j = 0 in the base year and other notation as before. 15.27 If the FYRR is greater than the planning discount rate, then the project is timely and should go ahead. If it is less than the discount rate, but the NPV is positive, the start of the project should be deferred and further rates of return should be calculated to define the optimum starting date.

Recommended approach
15.28 In most cases, the NPV and IRR will give consistent results and will produce the sarne ranking of alternatives according to their attractiveness. However, in a few cases, the use of IRR will give a different ranking to that recommended by using NPV. 15.29 In general, where the government is using a target, or minimum cut-off rate of return on capital, maximising NPV should be the criterion. As already mentioned, the IRR method is particularly useful where discount rates are highly uncertain. Normally, both methods should be evaluated for a project and, in cases of conflict, other factors will usually indicate which of the methods is most appropriate in the particular circumstances. 15.30 Some sponsoring agencies dictate which method they require to be used and, clearly, recommendations should be based on that result. Nevertheless, results from

70

16. RURAL ACCESS ROADS

THE NEED FOR SPECIAL CONSIDERATIONS
16.1 The methods of determining benefits for rural access road projects were discussed in Section 14. Since the cost per kilometre of constructing rural access roads is relatively low, the expenditure of significant amounts of time and money to determine the detailed costs and benefits of those projects cannot normally be justified. Nevertheless, appraisals should still be carried out for all rural access road projects, but a different approach may be needed to that used for arterial and collector roads. 16.2 Investment in rural access roads has the following objectives: To maintain existing levels of access To provide new vehicle access to rural settlements To help promote rural/agricultural development To supply transport at minimum cost to new development projects To upgrade the quality of existing roads and tracks

construction and maintenance. The preparation of a rural access road project frequently requires inputs from several agencies and from the communities to be served. In order to address these institutional issues, there is a need to consider the project in three separate ways: It is necessary to consider the place of rural access roads within the broader policy environment and the issues which may need to be addressed at this level. It is necessary to consider the issues that relate to the choice among organisational options for project execution. The role of local participation in the project must be considered.

An approach for taking these three considerations into account when appraising rural access roads is given in World Bank Staff Working Paper No 748 (Cook et al 1985). 16.8 Where roads are planned to form a minor part of the investment programme of an integrated development project, then it is sensible to plan the road investment to meet the forecast transport demand at minimum cost. In addition, it may also be necessary to check that the total project investment of roads plus other investments is viable. 16.9 Building completely new roads through previously inaccessible areas can produce dramatic effects on the local economy and the environment. Not all projects of this sort are successful: the land needs to be fertile, a mobile workforce must want to work there and the environmental effects may be substantial. In order to predict traffic, agricultural activity and the effect on the environment for this type of road investment, it is necessary to seek advice of experts in many fields and to look at similar recent experience elsewhere.

16.3 Most rural access road programmes will be planned to meet a number of these objectives. When the project is being formulated at the outset, it is useful to specify its main objectives so that all concerned can design the project accordingly. 16.4 Upgrading existing vehicle access will have little impact on rural development and, in general, it is recommended that, where rural development is the prime goal, that other constraints to development should also be investigated. If necessary, a total package of road investment and other inputs should be prepared. 16.5 Evidence from the evaluation of past projects indicates that rural access roads have, in general, not been able to stimulate the agricultural benefits that had originally been anticipated. This is not very surprising if the relationships between transport costs for a given agricultural product and its selling price are investigated. Seldom are transport costs more than 10 per cent of the price at the market, so any savings made will give little incentive to the farmer to increase production. In general, rural development will be difficult without vehicle access, but vehicle access on its own will not necessarily promote rural development. 16.6 In most instances, the very low traffic levels encountered on rural access roads will only justify the minimum standards of construction. Expensive gravel or bitumen surfaces should not be provided without very good reasons. 16.7 Rural access roads are different to arterial and collector roads in ways that affect the choice of an appropriate institutional setting for their planning,

NON ECONOMIC CONSIDERATIONS

16.10 All investment decisions have political, social and environmental consequences, besides the economic effects that are the principal focus of attention in this guide. The relative importance attached to these different considerations will depend upon the nature of the road investment. In planning the investment of rural access roads, the consequences of providing new vehicle access will be very different from upgrading existing roads. 16.11 In planning main road investment, economic/ engineering implications are usually paramount in the decisions to upgrade existing road surfaces. In principal, there is little difference with the decision to upgrade existing rural access roads and tracks. By contrast, when new access is provided, not only will the impact on the local economy be more important, but other considerations may also need to be considered. For

71

example, a new road will affect local drainage and may well interfere with crop irrigation. Likewise, there may be a substantial social/medical benefit if people can travel to hospital or clinic by vehicle rather than undertake an arduous walk.

Upgrading and maintaining existing access

16.17 If the final decisions on upgrading existing roads and tracks are to be on economic grounds, then any proposals to simplify the data collection exercise should be directed towards identifying and measuring the key parameters which figure in a cost-benefit analysis. These are: Current traffic levels Future traffic growth Current condition of existing roads The costs of upgrading the road surface.

16.12 For social, political and economic reasons, a number of governments have defined target minimum levels of accessibility for their rural populations. These targets are usually defined in terms of the minimum distance to roads of different standards for villages of different sizes. Another approach is to use more comprehensive ranking and screening criteria to take account of the non economic effects of rural access roads (Carnemark et al 1976, Beenhakker and Lago 1983). An alternative approach (Bovill 1978) is to value non economic benefits directly and then to incorporate them directly into the cost benefit framework.

16.13 If it is anticipated that the road investment will have major social or environmental consequences, then it is important to seek appropriate advice. Whatever method of appraisal is used, a statement should be prepared relating to those effects which have been omitted from the analysis so that they can be considered as part of the final decision.

16.18 It can be time-consuming and expensive to collect traffic data for each road section under consideration. However, provided some reliable traffic figures can be collected for a sample of roads, then estimates of traffic on the remainder may be made by relating traffic levels to factors such as adjacent rural population or agricultural output, if these are available. 16.19 If the areas to be investigated are fairly uniform, then there will be little need to provide separate estimates of traffic growth. However, if the areas are likely to have differences in their rates of growth of population or economic activity (eg because of new rural investment projects), then it may be useful to make different traffic projections for different areas. 16.20 The current condition of each road is usually extremely difficult to assess without making a field visit. If the roads are motorable, then in most cases a brief inspection of road condition can be carried out reasonably quickly. However, because of the serious consequences of a road or track becoming impassable, it is very important to look for those factors such as weak bridges, or broken or clogged culverts, which could cut vehicle access. Some likely clues to the condition of the road may be found from any existing road maintenance records and knowledge of weather, soils and terrain. 16.21 From an inventory of each road and knowledge of the terrain, soils and likely location of suitable roadmaking materials, estimates of the likely costs of upgrading the road surface can' be made. Again, there is considerable scope for making simplifying assumptions and grouping together likely common factors to help estimate these costs.

SCREENING AND THE SIMPLIFICATION OF DATA COLLECTION

16.14 Typically, investments in rural access roads are planned by preparing an investment programme on an area wide basis. Because of the costs of collecting data direct from all road projects under consideration, it may be useful to simplify the data collection exercise by collecting together data common to the proposed road investment in each local area.

16.15 By using fairly simple relationships between known parameters (eg population and traffic), estimates of the key parameters used in an economic appraisal may be made without carrying out a full collection of data for each road. From these estimates, an appraisal of each road investment may be made and a ranking of each road project achieved. Net present values can also be determined, and this task is simplified when many road investments are being considered if access is available to a computer.

New road access

16.16 Where the key components of an economic analysis cannot be estimated easily and cheaply, then a first screening of likely road projects can be made from data which is easily available to enable projects that are not promising investments to be eliminated prior to any further field investigations. 16.22 When new road access is planned, the total costs of existing means of transport need to be matched against the costs of transport by inotor vehicle. As above, estimates must be made of traffic growth and of the costs of road construction. In addition, existing volumes of non

72

motorised traffic and the current costs of exisfing transport need to be found. 16.23 The following data estimates are required: The distance of the centres of population from the road network The volumes of goods moved by current means including headloading, animal transport and by agricultural tractor The costs of moving goods by the alternative means The volumes of movement by people, the opportunity costs of their time and the other costs involved (eg the costs of animal transport or of tractors used to transport people) The extent of diversion of existing traffic to motor vehicle once the new road access is provided.

17. ANALYSIS OF UNCERTAINTY

SCENARIO AND RISK ANALYSES
17.1 As noted in para 2.2-7, the data and parameters used in the analysis of a road project can be prone to substantial errors and it is important to recognise that these exist and to take steps to minimise them. Because of this, the results of a feasibility study are subject to uncertainty and there will be a risk associated with pursuing any course of action suggested by the appraisal. 17.2 It was recommended in para 2.5 that scenario analysis should be used for projects that are not well defined or at the early stages of the project cycle. In such cases, a range of scenarios should be examined covering future possibilities that might reasonably be expected to occur. For such scenarios, which will often be covering political, economic and social uncertainties, projects should be examined for their robustness in being able to deliver a satisfactory NPV over the range of scenarios considered. 17.3 Where projects are well defined, risk analysis is more appropriate and, in these cases, the effect on the NPV of combinations of uncertainties in the project's most sensitive parameters should be examined. Ideally, an approach based on probabilities should be used and the remainder of this section describes how this should be carried out.

16.24 There is again considerable scope for estimating the likely transport demand from a few limited surveys of typical areas and extrapolating the results to the rest of the area. Use can be made of detailed maps, census returns and agricultural production data. 16.25 Road access should first be planned to connect rural settlements to the road network. in most circumstances, this will provide a useful degree of direct road access to farming areas. In the early stages of development of commercial agriculture, there is little need to provide additional access to farming areas because agricultural demand is usually spread extremely thinly. In addition, farmers tend to store and process crops for sale in the vicinity of their home, and the typical small load sizes and short distances between farm and home (usually less than about 8 km) means that it is normally uneconomic to use conventional motor transport for this first movement of crops, even if a motorable road is available. Exceptions to these general rules can be made for crops with very low value to weight ratios like sugar cane, coconuts or melons, where farmers will be keen to transport the crops in large quantity straight from field to market.

EXPECTED VALUES
17.4 The basic calculation of net present value should incorporate the best estimates of the variables and parameters that determine the cost and benefit streams. The estimates should be the 'expected values' obtained, in principle by weighting each possible value by the probability of its occurrence. Using expected values ensures that the estimates are unbiased providing that the formation of the probability function of values is unbiased. In the absence of probability data, mid points of the range of expected values should be used. Biased estimates, such as conservative estimates of costs (on the high side) and of benefits (on the low side), should be avoided, since they distort the comparison of alternative projects. 17.5 In view of the uncertainties present, it will usually be preferable to show the results of a project analysis as a range of values reflecting major uncertainties, rather than as a single figure, as this aids the choice of the most robust project. It is therefore necessary to identify: those sensitive areas which have a critical importance on the success or failure of the project

73

ways of improving the project by making it less risky, more cost-effective, or both.

17.6 Sensitivity analysis is appropriate for initial identification of sensitive inputs or parameters, but risk analysis is also relevant where correlated sets of inputs need to be identified, and to demonstrate clearly the range of possible outcomes for a project, even in the absence of correlated inputs. Sensitivity analysis is described in para 17.12-22 and risk analysis is described in para 17.23-26.

variables fixed. By looking at higher and lower figures than those expected, it is possible to determine how sensitive the net present value is to such changes. The variables that are chosen for testing are a matter of judgement but, for most road schemes, the following should be considered as possible candidates, approximately in order of presentation here.

Traffic
17.13 The difficulty of obtaining reliable estimates for traffic and of forecasting future growth rates has been discussed in Section 3. In most projects, traffic levels will have a major affect on the level of benefits obtained. Thus sensitivity analysis should be carried out, both of baseline flows and of forecast growth. For baseline flows, ranges of values of up to plus or minus 50 per cent of the expected value should be examined for low traffic flows and up to plus or minus 25 per cent for high flows. Similarly, the effect of 'optimistic' and 'pessimistic' traffic growths of up to about 25 per cent for low growth rates and 50 per cent for high growth rates should be examined.

CONTINGENCY
17.7 It is usual to include in the estimates of capital costs a separate allowance to cover contingencies. These are of two types. 17.8 Expected costs. Allowances should be included to cover costs which have not been separately identified, but which experience indicates must inevitably occur during the construction period. A lump sum contingency allowance to take account of all the constituent parts should he used in such cases to cover a variety of items. 17.9 Tolerances. This form of contingency allowance is an estimate, usually based on past experience, of the probability of unforeseen costs arising and of their probable magnitude. Tolerances reflect the fact that costs may overrun due to physical contingencies, such as unexpected poor ground conditions or lack of finance which prolongs construction time. The best estimate of the allowance should be regarded for appraisal purposes as part of the cost of the project, even though it may not have to be spent. 17.10 Expected contingency allowances of up to about 25 per cent of the construction cost are normal for road projects in developing countries. 17.11 It is not necessary to make allowances in an economic appraisal to cover price increases due to inflation during the construction period providing that all prices are expressed in terms of constant base year values as described in para 15.8. However, any such price increase will affect the project's cash flow and will need to be estimated for budgetarv purposes in the financial analysis. When preparing the project budget, it may also be necessary to consider separately the prices of imported items from those affecting the cost of local labour and materials.

Project costs
17.14 Project costs are always difficult to estimate accurately in developing countries. Sensitivity to uncertainties in the project cost of plus or minus 25-100 per cent should normally be investigated. Note that the risk of price escalation should normally be taken into account in the financial analysis rather than through sensitivity testing in the economic analysis.

Delay
17.15 A major risk to be tested in the sensitivity analysis is delay in implementation. A test should therefore be carried out on a one year delay in implementation or with construction costs spread over one extra year. For very large projects, longer delays may be possible.

Generated traffic
17.16 As noted in Section 14, the forecasting of the level of economic development associated with a road project is extremely difficult. When dealing with arterial and collector roads, with relatively high traffic levels, the project outcome should be considered with and without benefits due to generated traffic. If the project is heavily dependent on generated traffic to provide a positive NPV, its acceptance should be viewed with some caution. For rural access roads with relatively low traffic flows, the project's sensitivity to variations in developmental benefits of up to about plus or minus 50 per cent should be considered.

SENSITIVITY ANALYSIS
17.12 Sensitivity analysis is carried out by varying the magnitude of the more important variables, normally one at a time, whilst keeping the values of the remaining

74

Time and accident savings

17.17 For arterial and collector roads in rural areas, projects that are heavily dependent on time and accident savings to ensure a positive NPV should be viewed with caution in the same way as for generated traffic benefits. In such cases, the sensitivity of the project to variations in time values, accident rates and costs of up to about 25 per cent should be considered.

happen and risk analysis should normally be undertaken (see para 17.23-26).

Investment models
17.22 Providing that investment models, such as microRTIM2 or HDM-III, are being used to assist with the project appraisal, sensitivity analysis is easy and quick to do. Indeed, one of the principal advantages of such models is that they enable this to be done at low cost in terms of time. When investment models are not available, a limited amount of sensitivity testing should still be done. The effect of uncertainty in traffic should always be investigated even though the manual calculations needed may be time consuming. The effect of uncertainty in project cost is easy to evaluate by hand and should therefore normally be determined. In all cases, for arterial and collector roads, the NPV should be evaluated separately for: benefits to normal traffic benefits to generated traffic time and accident savings.

Shadow prices
17.18 In projects where the shadow prices used differ markedly from the market price minus transfer charges (see para 15.12-17), the sensitivity to uncertainties in the shadow price of up to about 25 per cent should be considered.

Maintenance
17.19 The importance of good management of a project, particularly during the operational phase has been emphasised in para 2.22-26 where the need to use the expected value of maintenance performance in project assessment was also stressed. It is difficult to examine the effect of uncertainty in this directly, but its consequences can be inferred by examining the sensitivity of the project to uncertainty in the rate of road deterioration and its effect on vehicle operating costs. For road improvement projects, vehicle operating cost savings should also be evaluated for a higher range of roughness levels, such as those in Table 10.2, for any particular road type. The vehicle operating cost figures obtained may then be used to determine the effect on project benefits.

This will greatly assist the interpretation of results.

RISK ANALYSIS
17.23 Sensitivity analysis should indicate which of the parameters examined are likely to have the most significant effect on the feasibility of the project because of inherent uncertainty, but does not show the combined net effect of changes in all variables or the likelihood of changes occurring together. Where several parameters are identified whose estimated accuracy is critical to the successful outcome of the project, risk analysis may be appropriate. 17.24 Risk analysis, in its simplest form, requires specifying the probability of an individual input variable attaining a range of values. Using this, the probability distributions of the NPV and other output parameters can be determined. An approach for dealing with this is given in the ODA project appraisal guide (Overseas Development Administration 1988). Examples of the application of risk analysis to road projects are given by Pouliquen (1970). 17.25 Risk analysis provides a better basis for judging the relative merits of alternative projects, but it does nothing to diminish the risks. it is time consuming to carry out with projects that are as complex as roads, and such analysis must therefore be reserved for a very few variables in highly critical cases. Some risks identified by the sensitivity analysis can be reduced by carrying out further field investigations and redesign which may or may not be worthwhile depending on the cost of the

Special factors
17.20 It may be that there is uncertainty about future events which could have an important bearing on the project. A dam, for instance, might be built which would flood the valley in which the road was built. If the dam project were to go ahead, then the road would have to be relocated in the future, although the cost of this would be included as part of the cost of the dam. A railway may be under consideration which, if constructed, would significantly affect the design requirements of the proposed road. Roads built in unstable hilly terrain are always at risk from landslide activity, even when planned using the principles described in Section 5, and these may need to be partially realigned and rebuilt in the future. Structures may be damaged in areas subject to flooding. 17.24 In such cases, or where there is doubt about the implementation of other major development projects which will affect the benefits of the road project, it is normally appropriate to carry out analyses based on the alternative assumptions that the event will or will not

75

investigation and the expected reduction in the risk. Risk may also sometimes be reduced by a more flexible approach to design and construction such as is possible under costreimbursement or target price contracts as described in Section 2.5.1. 17.26 However, for a small amount of effort, even roughand-ready forms of risk analysis are likely to improve the quality of decision making considerably.

18. THE FEASIBILITY STUDY REPORT

PREPARATION
18.1 Decisions must be made at various stages throughout the project cycle. The early decisions on a project, however apparently innocuous have a disproportionate effect on the final shape of the scheme. At each stage, careful preparation and presentation are necessary to reveal and justify decisions taken or recommendations made. The feasibility study report marks the end of the appraisal process and should recommend whether the project should go ahead, and to what standards it should be built. The report may wish to recommend alternative designs or approaches to the project that would increase the rate of return in those areas where the original project is not viable. 18.2 In addition to these decisions about the nature of the project, the way in which the project is presented can be important for future projects of a similar kind, and for the future monitoring and evaluation of the project. It can, for instance, through scenario or sensitivity analysis, show the crucial factors which will make or break the project. These can give important signals to those concerned with checking the progress and reviewing the results of the project in the future. 18.3 Once the need for a project, and its objectives, have been identified, the extent of further investigation will depend on a number of considerations. The political, managerial, economic, technical and financial aspects need to be covered adequately in every case, but depending on who the report is being written for, some aspects have to be covered in greater depth than others. 18.4 Where reports are prepared for aid donors, each will have its own different requirements. An analysis carried out for a development bank will have to cover financial aspects very thoroughly. Projects prepared for aid agencies normally dwell heavily on socio-economic factors. The World Bank, for instance, has a highly formal, elaborate and thorough process of approving projects through its executive board, necessitating extremely careful and comprehensive preparation. The British Overseas Development Administration, likewise, imposes on itself a well-delined and rigorous procedure for approving large aid projects. Other development banks may have simpler procedures requiring briefer preparation, relying more on their judgement of the calibre of associated institutions, partners, or sponsors. 18.5 The tearn assigned to prepare the project should normally contain a range of prolessionals such as engineers, transport planners and economists. Sector specialists can be added where the size and complexity of

76

the project require, such as agronomists, engineering geologists, environmental specialists, etc. Where, as is often necessary, members of the project team are from an international consultant, the local government should participate as fully as possible in the investigations, and this normally requires the allocation of local professional staff to the project team. The finance and planning ministries should be made fully aware of progress and recommendations, although the promoting ministry should take responsibility for the detailed professional work.

1. 2.

Summary and conclusions Brief description of project Objectives Project type Main features Preliminary considerations History and background to the project Political factors Method of project execution and technology to be used Managerial, administrative and maintenance capability for implementation Assessment of demand Consideration of alternative routes, standards, modes Current traffic levels and forecast growth Diverted and generated traffic Determining costs Geotechnical considerations Design and costs of: pavement alignment (earthworks) drainage and structures Assessment of benefits Vehicle operating cost savings Road maintenance benefits Time savings Reduction in road accidents Economic development Economic analysis Cost-benefit analysis Analysis of uncertainty Financial aspects Costs of construction Inflation, contingencies and arrangements for cost overruns Operation and revenues Foreign exchange implications and exchange rate assumptions Sources of funds: capital and recurrent Other aspects Environmental impact statement Social consequences, etc

3.

PRESENTATION
18.6 The particular approval procedure to be used affects the way in which the project is presented. Some agencies insist on standardised presentations with bulky supporting documentation, while others prefer shorter and more sharply focused reports. 18.7 Whatever the nature of the approving body, there must be an assumption that the majority of the people who have to take the decision are non-specialists and busy. This argues for a clear and simple document with the accent on objectivity and brevity, and containing the more detailed discussion of technical and specialist aspects as annexes to the main document. It should contain a summary and conclusions. A map of the project location is usually essential, together with other visual aids like diagrams and bar charts. Where vahies are expressed in foreign currency, a conversion rate into local currency should be included. 18.8 In principle, the paper should be in a form that can be made available to other parties involved such as a foreign government providing the loan or aid, the local authority that will have to implement the work, etc. To this end, the document could be divided into two sections, one that can be distributed and the other containing information and views meant for the approval body only. 18.9 It is helpful if the submission clearly draws out the effects of the project on different parties who may be affected and on the wider economy of the country. Benefits and costs should be shown individually and the appraisal methodology used should be indicated. Likewise, the economic discussion should include scenario analysis, or sensitivity and risk analysis, in order to accentuate the most important factors governing the success or failure of the project. This analysis should be consistent with government policies of pricing, tariffs, procurement, incomes policies, etc, where they are likely to have influence on the outcome of the project. 18.10 One possible approach for presenting the feasibility study report is to follow the general order of topics as in this Note: 4.

5.

6.

7.

8.

9.

10. Implementation Responsibility for implementation Arrangements for construction Maintenance 11. Plans for monitoring and evaluation

77

12. Annexes (these must be 'keyed in' to the main text, otherwise they may be ignored). 18.11 The conclusions in the project report should ensure that the following aspects of the project have been considered and are reflected in the final recommendations: the opttons investtgated have been selected from the full range available the results for each option are presented as a range of values in terms of NPV, etc the main assumptions and sensitivity of the result to them are clearly identified the result may need to be interpreted, not in terms of profit, but as cost savings or benefits which are available for alternative use.

19.

CHECKLIST OF KEY POINTS

19.1 This checklist is designed to assist those submitting or appraising project reports to check quickly whether all of the key issues have been included. References are given to the paragraphs where items can be followed up in the main text.

OBJECTIVES
19.2 What are the project's objectives (1.53-54)?

19.3 What is the nature of the project: new construction, upgrading (1.35-37), reconstruction rehabilitation (1.33-43), stage construction (1.44-45), road maintenance (1.46-51, 2.22-34), bridge construction (8.17-51)? 19.4 What stage of the project cycle has been reached (1.4-18)?

BACKGROUND
19.5 What alternatives to the project have been considered in terms of mode, route, standard, timing (1.26-34, 1.53-54, Sections 5-8, 15.25-27)? 19.6 Has the project been set against the background of a transport sector or road plan (1.53)? 19.7 What are the relevant features of terrain, relief, climate, vegetation, drainage, soils, rock, etc (5.4-19)? 19.8 What are the major economic activities (3.1-3, 3.1426, Sections 9-14)? 19.9 How does the project complement the existing network (1.52)? 19.10 Have socio-economic considerations been taken into account (2.35-40)? 19.11 Have environmental considerations been taken into account (2.41-62)?

INSTITUTIONAL AND MANAGERIAL ASPECTS

19.12 Is the institutional framework conducive to the success of the project (2.16-17)? 19.13 Is there a project component for improving institutional development (2.18-21)? 19.14 What is the roads organisation's capability for carrying out maintenance (2.22-34)? 19.15 What form of contract will be used and will there be a supervising consultant (4.67-72)?

78

TRAFFIC
19.16 What is the volume of existing traffic, how is it classified and how were the estimates obtained (3.1-13)? 19.17 What is the projected traffic growth and how has this been derived 3.14-18)? 19.18 What is the estimate of diverted and generated traffic (3.19-26)?

REFERENCES
AASHTO, 1974. AASHTO Interim guide for design of pavement structures 1972. Washington DC: American Association of State Highway and Transportation Officials. AASHTO, 1983. Standard specifications for highway bridges. Washington DC: American Association of State Highway and Transportation Officials, Thirteenth edition. ADLER, H A, 1987. Economic appraisal of transport projects. A manual with case studies. Baltimore: Johns Hopkins University Press for the World Bank, Revised and expanded edition. ALLPORT, R J et al, 1986. The use of scenario techniques to formulate transport strategy for an urban area. In: PTRC. Transportation Planning Methods. Proc of Seminar M, PTRC Summer Annual Meeting, University of Sussex, 14-17 July 1986. London: PTRC Education and Research Services, 229240. BEENHAKKER, H L and A M LAGO, 1983. Economic appraisal for rural roads. Simplified operational procedures for screening and appraisal. World Rank Staff Working Papers No.610. Washington DC: International Bank for Reconstruction and Development. BOVILL, D I N, 1978. Rural road appraisal methods for developing countries. Supplementary Report 395. Crowthorne: Transport and Road Research Laboratory. BRIDGER, G A and J T WINPENNY, 1983. Planning development projects. London: HMSO for the Overseas Development Administration. BSI, 1972. Specification for steel girder bridges. BS 153: Parts 1-3: 1972. London: British Standards Institution. BSI, 1978. Steel, concrete and composite bridges. BS 5400: Parts 1-10: 1978-1982. London: British Standards Institution. CARNEMARK, C et at, 1976. The economic analysis of rural road projects. World Bank Staff Working Paper No.241. Washington DC: International Bank for Reconstruction and Development. COOK, C C et at, 1985. Institutional considerations in rural road projects. World Bank Staff Working Papers No.748. Washington DC: International Bank for Reconstruction and Development. CPCA, 1984. Thickness design for concrete highway and street pavements. Ottawa: Canadian Portland Cement Association.

DESIGN
19.19 What are the design features: Pavement (5.6-12, 6.1-47) Geometrics (5.13-16, 7.1-33) Structures (5.17-19, 8.1-42)? 19.20 What is the design life (1.56-57, 6.41-42)? 19.21 Have axle loads been measured and forecast (6.28-30)? 19.22 Is the method of construction appropriate (2.8-15)?

COSTS
19.23 What general methods of costing have been adopted (4.1-66)? 19.24 How have costs been determined for: Pavement (6.48) Earthworks (7.34-36) Structures (8.11-12, 8.16, 8.43-50)?

19.25 What are the estimates for recurrent costs (11.15-16)?

BENEFITS
19.26 What are expected direct benefits due to: Vehicle operating cost savings (Section 10) Economies in road maintenance (Section 11) Time savings (Section 2) Reduction in road accidents (Section 13)? What are the indirect and induced benefits (Section

19.27 14)?

19.28 What are the results of the cost-benefit analysis (Section 15, 16)? 19.29 How dependent is the rate of return on generated traffic and time savings (17.16-17)? 19.30 How has uncertainty and risk been dealt with 2.2-7, 17.1-26)?

79

INTERNATIONAL BANK FOR RECONSTRUCTION AND DEVELOPMENT and SCOTT WILSON KIRKPATRICK & PARTNERS, 1978. Guide to competitive bidding on construction projects in labor-abundant economies. Washington DC: World Bank. KENIBER-SMITH, J, 1984. Temporary inductive road loops for automatic traffic recording. Application Guide I. Crowthorne: Transport and Land Research Laboratory.

TRRL, 1978. Terrain evaluation for highway engineering and transport planning. Supplementary Report 448. Crowthorne: Transport and Road Research Laboratory.

TRRL OVERSEAS UNIT, 1985. Maintenance techniques for district engineers. Overseas Road Note 2. Crowthorne: Transport and Road Research Laboratory, Second edition. TRRL OVERSEAS UNIT, 1987. Maintenance management for district engineers. Overseas Road Note 1. Crowthorne: Transport and Road Research Laboratory, Second edition. TRRL OVERSEAS UNIT, 1988. A guide to geometric design. Overseas Road Note 6. Crowthorne: Transport and Road Research Laboratory. UNIDO and TRADA, undated. Pre-fabricated modular wooden bridges, Parts 1-5. Place of publication not given: United Nations Industrial Development Organisation and Timber Research and Development Association. VAN DER TAK, H G and A RAY, 1971. The economic benefits of road transport projects. World Bank Staff Occasional Papers Number Thirteen. Baltimore: Johns Hopkins Press for International Bank for Reconstruction and Development. WATKINS, L H and D FIDDES, 1984. Highway and urban hydrology in the tropics. London: Pentech Press.

LAWRANCE, C J and P J BEAVEN, 1985. Remote sensing for highway engineering projects in developing countries. In: KENNIE, T J M and M C MATTHEWS (Eds). Remote sensing in civil engineering. Glasgow: Surrey University Press, 240-268. MAYHEW, H C and H M HARDING, 1987. Thickness design of concrete roads. Research Report 87. Crowthorne: Transport and Road Research Laboratory.

NAASRA, 1980. Interim guide to the geometric design of rural roads. Sydney: National Association of Australia State Road Authorities. OVERSEAS DEVELOPMENT ADMINISTRATION, 1988. Appraisal of projects in developing countries: a guide for economists. London: HMSO. PARRY, J D, 1981. The Kenyan low cost modular timber bridge. TRRL Laboratory Report 970. Crowthorne: Transport and Road Research Laboratory. POULIQUEN, L Y, 1970. Risk analysis in project appraisal. World Rank Staff Occasional Papers Number Eleven. Baltimore: Johns Hopkins University Press for the World Bank. SAYERS, M W et al, 1986. Guidelines for conducting and calibrating road roughness measurements. World Bank Technical Paper Number 46. Washington DC: International Bank for Reconstruction and Development. SMITH, H R and C R JONES, 1982. Early performance of some experimental bituminous overlays in Kenya. TRRL Laboratory Report 1043. Crowthorne: Transport and Road Research Laboratory TRANSPORT AND ROAD RESEARCH LABORATORY, 1977. A guide to the structural design of bitumensurfaced roads in tropical and sub-tropical countries. Road Note 31. London: HMSO, Third edition, from TRRL Overseas Unit (currently being revised). TRANSPORT AND ROAD RESEARCH LABORATORY, 1978. A guide to the measurement of axle loads in developing countries using a portable weighbridge. Road Note 40. London: HMSO, available from TRRL Overseas Unit.

80

APPENDIX A ROAD UPGRADING PROJECT

COMPONENTS OF THE STUDY
A.l A feasibility study has been carried out for ODA by consultants to look at the upgrading of a 103 km road in Africa from a gravel to bitumen surface. The existing road was not all weather, although it was possible to drive the whole length in the dry season. An earlier design for an upgrading had been carried out by different consultants, but the high cost of construction to this design had proved not to be justified in terms of benefits generated. The project analysis therefore had four main components: to show savings in construction costs coupled with harmonisation of highway design throughout the project to design a road which conformed to the high standard of bitumen road construction which had been established in the country to ensure that the road when built would truly serve the population of this rich agricultural region of the country, bearing in mind its importance to the country as a whole to adopt design standards relevant to the speed and numbers of vehicles through and between the many centres of population along the route.

levels along the existing road. Information on vehicle loading was obtained from axle load measurements carried out both along the existing gravel road and 6n a nearby bituminous road that was thought to be more representative of the type of loadings that the road would carry after upgrading. A.3 Moving observer counts were carried out along the existing road to determine what were the general flow patterns along its length and, as a result, four survey stations were set up for counting traffic. At the first station, a seven day classified count was carried out, for 24 hours on two of the days and for 16 hours on the remaining days. Roadside interviews were carried out on two of the days. At each of the other three sites, classified counts and roadside interviews were undertaken for two days. The results of these counts were considered in conjunction with the results of more extensive surveys carried out by other consultants for ODA some six months previously. Further counts were also carried out at other locations in order to give a more comprehensive picture of traffic movements. A.4 The annual average daily traffic (ADT) was estimated from the results of the surveys carried out by the two consultants and is shown for different classes of vehicle in Table Al. A.5 The objective of the roadside interviews was to sample vehicles to determine their type, start and end points of their journey, trip purpose, the number of passengers and any commodity carried. It proved to be very difficult to obtain a suitably representative sample

TRAFFIC SURVEYS
A.2 Classified counts and roadside interviews were carried out to determine the magnitude and variation in flow

TABLE A1 ESTIMATED ADT ON EXISTING ROAD

81

and the results did not give a clear picture of the variation in traffic flows or loads carried, either between stations or between weekdays and weekends. Since a very large amount of data was collected, it was theoretically possible to derive trip matrices for each survey station by type of vehicle and trip purpose for each hour of the day. However, the disaggregation of the data was such as to make such an exercise virtually worthless, so results were not analysed fully. However, the interview data was used to help derive equivalence factors for vehicles and to establish both occupancy rates for each vehicle type and trip purpose for estimating average values of time for use in the economic analysis. A.6 Measurements of vehicle wheel load were carried out using a portable weighbridge in connection with the roadside interviews. Due to delays in clearing the weighbridge through customs, the time available for obtaining information was limited. The traffic counts indicated that a large number of lorries were travelling very early in the morning and at night, so special surveys were undertaken to weigh these vehicles. The mean equivalence factors derived from the axle load surveys and the roadside interviews are shown in Table A2. From the limited number of measurements made, there is clear evidence of overloading. TABLE A2 ESTIMATES OF MEAN EQUIVALENCE FACTORS 2 axle Road section Outward Section 1 (StartStation 1) Section 2 (Station 1Station 3) Section 3 (Station 3-end) Inward Outward Inward 3 axle

Data in these areas proved to be neither comprehensive nor reliable. A.9 These socio-economic parameters suggested traffic growth rates for individual vehicle types for normal traffic varying from 2.0 to 8.0 per cent per annum and 2.5 to 10.7 per cent per annum for the growth of both normal and generated traffic (see Table A3). These could be set against a background of economic growth for the whole country in terms of GDP quoted variously between 7.5 and 8.5 per cent per annum. A.10 Historical evidence in the country suggested that, when a new road opened, there was likely to be an immediate generation of traffic of the order of 10 per cent and this figure was assumed for this study. The actual generated traffic figures were obtained by determining the total traffic from Tables Al and A3, and subtracting the normal traffic figures from the totals for each year. The growth rates for generated traffic were reduced to zero towards the end of the project analysis period because there was some uncertainty as to whether the earlier increases in generated traffic flows could be maintained. The resulting forecast traffic for the final year of analysis (2003) is shown in Table A4. A.11 A further check was carried out on the plausibility of growth rates for generated traffic by considering vehicle operating cost savings and demand elasticities. The TABLE A3 ESTIMATES OF TRAFFIC GROWTH RATES With generation from upgrading 2.0 8.5 0.5 8.0

3.0

2.1

4.0

4.9

Socio-economic parameters

Base case 1.6 5.0 6.0

2.7 4.0

3.7 7.1

3.0 6.7

5.3 6.1

Population (A) Vehicle (B) Income (C) Agricultural growth (D) Traffic 1. Car/taxi (B) 2. Pickups/LGV's (AXCXD) 3. Minibuses (AxC) 4. 2 and 3-axle trucks based on growth in: - consumer goods - consumption (A x C) - (D) - construction materials (mainly AX C but also affected by economic growth) Weighted average

A.7 There was considered to be limited opportunity for traffic diversion and, since the numbers involved were expected to be small, they were ignored for the analysis.

5.0 8.0 2.0 4.0

8.5 10.7 2.5 5.0

TRAFFIC GROWTH
A.8 Traffic growth in terms of both trip numbers and trip lengths were determined based on several factors, the most significant of which were: Population size Vehicle fleet and levels of motorisation Disposable income Economic activity and growth Costs and patterns of current vehicle operations.

5.0

8.0

82

TABLE A4 FORECAST 2003 ADT VALUES Section Base traffic Generated traffic Total traffic 2 Base traffic Generated traffic Total traffic 3 Base traffic Generated traffic Total traffic Private cars 1109 418 1527 416 157 573 446 168 614 Taxis 2027 923 2950 925 422 1347 646 294 940 LGV's 370 0 370 224 0 224 158 0 158 Minibuses 320 108 428 234 74 308 139 44 183 Lorries 2-axle 182 58 240 234 74 308 95 30 125 Lorries 3-axle 1604 605 2209 277 105 382 356 134 490 Buses 0 42 42 0 25 25 0 18 18 Total 5612 2154 7766 2310 857 3167 1840 688 2528

TABLE A5 TRAFFIC GROWTH BASED ON ELASTICITIES Elasticity Passenger traffic = 1.0 Goods traffic = 0.1 to 0.2 1.0 0.575* 1.0 0.15 0.15 1.0

Vehicle type Private cars Pickups Minibuses 2 axle trucks 3 axle trucks Taxis

Cost saving (%) 43.95 56.94 55.47 45.50 44.22 57.94

Growth (%) 43.95 32.74 55.47 6.825 6.633 57.94

*Average value for passenger and freight These growth figures compare with a flat figure of 10 per cent used in the study in the first year of operation of the new road.

vehicle operating costs per kilometre for each vehicle type were derived for the existing and the new road. The cost savings from using the new road were expressed as a percentage of the costs on the existing road and multiplied by the elasticity to give the likely growth following upgrading as shown in Table A5. Elasticities were assumed to be constant across the range of cost savings.

TABLE A6 CUMULATIVE STANDARD AXLES (Standard axles/day) 1989 565 633 554 Million standard axles over 15 years 5.3 msa 5.9 msa 5.2 msa

Section 1 2 3

Direction Outward Inward Inward

ENGINEERING DESIGN
A.12 The traffic used as the basis of the pavement design for the 15 year analysis period is shown in Table A6. On the basis of these figures, the same payment structure was designed for the entire road length consisting of a double surface dressing on 200 mm of crushed rock road base with a 225 mm laterite sub-base. It was planned that,

after one year, the pavement would be overlaid with 50 mm of asphaltic concrete. A.13 When carrying out the geometric design, the following factors were taken into account:

83

the existing alignment should be followed wherever possible as few houses as possible should be destroyed the alignment should stay as close as possible to the existing ground level.

the stability of cutting and side slopes the California bearing ratios used for the pavement design the nature of soft ground crossings.

A.14 The geometric standard used in the redesign can be summarised as follows: Generally, standards for a 60 km/h design speed were recommended for crest and sag curves, but earthworks savings were made by reducing the radius of selected crest curves nearer to the minimum standard In some of the more rugged sections, the minimum radius for a 60 km/h design speed could not be achieved so, in isolated instances, design speed was reduced to 40 km/h Earthworks savings were obtained by the use of steeper gradients, with a maximum of 10 per cent in the most rugged sections of the route; where a long climb in open country could be avoided by a shorter climb of up to 10 per cent gradient, these were included Climbing lanes were used except where long straights enabled safe overtaking without a climbing lane Special urban cross-sections were used where the road passed through villages. The minimum standards used are shown in Table A7.

A.18 Site work consisted of inspecting existing earthworks, digging of trial pits and the use of a mackintosh probe to investigate subsoil conditions. Laboratory tests were carried out to analyse samples that were collected. An investigation was carried out of materials available from cuttings and borrow areas needed to allow construction of the proposed road. A.19 Rainfall data were collected from records and were used with field observations to determine hydrological and drainage requirements. Runoff was estimated using a variety of methods to arrive at appropriate values. Culverts were designed for a 10 year return period and larger culverts were checked for the possible consequences of a 25 year return flood period. Where it was considered that this larger flood might seriously damage the road or structure, then culvert capacity was increased. Bridges were designed for a 50 year return period of flood.

A.15

CONSTRUCTION COSTS
A.20 The estimate of construction cost that was based on the engineering design and used for the economic analysis is shown in Table A8. A separate financial cost analysis was also carried out. TABLE A8 ESTIMATE OF CONSTRUCTION COST FOR THE ECONOMIC ANALYSIS (1986 prices) Section Section Section 1 2 3 6.2km 71.3km 17.5km 106 1691 476 24 281 70 169 1994 492 88 112 165 100 764 76 210 1050 1013 1304 1800 1212 9295 930 2557 12782 256 332 705 350 2681 268 737 3683 Total 95km 2273 375 2655 1357 1748 2670 1662 12740 1274 3504 17518

TABLE A7 MINIMUM GEOMETRIC STANDARDS ADOPTED Design speed (km/h) Absolute minimum horizontal radius (m) (maximum superelevation 7%) Normal minimum horizontal radius (m) (maximum superelevation 5%) Maximum grade Minimum crest vertical curve (m) Minimum sag vertical curve (m) 80 60 40

240

120

40 Length

450 250 10% 10% 4500 1600 2200 1500

110 10% 500 700

A.16 In areas where detailed mapping did not exist, topographic surveys were carried out to provide data for input to a computer model used to derive cross-section information and to estimate earthworks quantities. A.17 Considerable geotechnical investigation had been carried out by the previous consultant and further work was undertaken to augment this and, to a limited extent, verify the earlier work. Further geotechnical studies were also carried out wherever realignments were proposed. In particular, the aim of these studies was to investigate:

Earthworks Gravel sub-base Crushed rock base Double seal surface dressing Asphaltic concrete surfacing Drainage Miscellaneous (15%) Prelim & general (10%) Contingencies & dayworks (25%) Total

84

A.21 It was anticipated that the construction would be divided into two lots and would start in 1987, lasting for four years. Expenditure through this period used for the economic analysis is shown in Table A9. TABLE A9 DISTRIBUTION OF CONSTRUCTION COSTS FOR THE ECONOMIC ANALYSIS Year 1987 (30%) 1988 (45%) 1989 (15%) 1990 (10%) Totals Lot 1 2752 4128 1376 917 9173 Lot 2 2503 3755 1252 835 8345 Total 5255 7883 2628 1752 17518

A.25 Costs of maintenance were obtained from the road maintenance organisation in the country.

ROAD USER COSTS

A.26 Data on vehicle operating costs and operating patterns were gathering for typical vehicle classes: Private car: Toyota Cressida Break Pick-up: Toyota Hilux LN6S Double Cab Standard (Four wheel drive) Minibus: Toyota Hiace 15 Seater Two axle truck: Toyota DA 110 Benne Three axle truck: Volvo N1033 6x4 Taxi: Nissan Sunny

A.22 In order to determine the residual value of the project at the end of the 15 year economic analysis period, the following assumptions were made: All earthworks were assumed to be satisfactory and their cost treated as part of the residual value Large drainage structures would have many years of useful life ahead of them, but the small steel pipe culverts might be starting to fall into disrepair; it was assumed that 50 per cent of the capital cost of drainage works would form part of the residual value None of the pavement construction would have any value after 15 years although, if maintenance were to be carried out satisfactorily, much of it might still be useful.

A.27 Vehicle and tyre costs were obtained from distributors. Overheads costs and operating patterns were established from interviews with local businessmen, operators and officials, and from published sources, such as insurance and licencing price lists. A.28 Values of passenger time were obtained by factoring vehicle occupancy (based on observations and counts) by standard wage rates for typical passengers, obtained from national wage rate charts, and making allowances for the proportion of passengers travelling for business and personal reasons found from traffic surveys. A.29 Some adjustments were made to the pattern of vehicle operation and costs in the case of the road being upgraded. It was not thought likely that there would be a decline in the number of four wheel drive vehicles, as access would still be required to outlying parts but their proportion of the total car/pickup fleet might drop. A.30 The sizes of buses and trucks would almost certainly increase. Interviews with vehicle operators indicated that the paving and realignment of the road would lead to: a rapid replacement of 15-18 seat minibuses with 20-25 seat midibuses: this was taken into account in the traffic forecasts by assuming a progressive transfer over the analysis period from one type of bus to the other. a progressive change from 2-axle to 3-axle trucks for general utility purposes, as old vehicles were retired: this did not appear to make a great difference in terms of road deterioration and was ignored in the traffic forecasts. fitting of trailers to 3-axle trucks for heavy haulage: again, this did not appear to affect the road deterioration and was ignored in the traffic growth calculations.

As a result of the above, a figure of 25 per cent of the capital cost of construction was taken as the residual value of the road.

MAINTENANCE COSTS
A.23 Experience has shown that, for a road as heavily trafficked as this, costs of maintenance of both a baseline and a study case would be extremely low when compared with the vehicle operating costs. The results of an economic analysis are thus insensitive to large variations in maintenance costs. This was borne out by the results of this study. A.24 At the time of the fieldwork for the project, the study team were impressed by the success of maintenance being carried out, both on the existing gravel road and on other roads. The road was regraded at least twice a year and minor pavement reconstruction and recompaction was normally carried out on one of these occasions. For the baseline case of the economic analysis, it was assumed that this policy would continue.

A.31 In terms of maintenance practice, it was thought likely that urban taxis would start to use remould tyres;

85

originally they operated part of their journeys over the unpaved road and therefore invested in new tyres. The cost differences were reflected in the unit tyre prices assumed. A.32 Another outcome of road improvements would be a growth in the number of two-wheeled vehicles. These would mainly be in addition to, rather than instead of, the number of motor vehicles discussed above, and inconsequential in terms of road deterioration. Costs and benefits in terms of operating costs and user time savings were not taken into account for these vehicles.

Operating cost benefit to base traffic (existing traffic that would be on the road plus the projected growth in traffic even if the road were not improved) Operating cost benefits to base traffic plus the additional traffic generated by improvements to the road Operating cost benefits to these two sets of traffic plus the value of time savings by passengers in the vehicles on the road.

Results are shown in terms of NPV and IRR in Table A.l0. A.35 It can be seen that each section of the project and the project overall (at an IRR of 20.6 per cent) all appear to be economically viable even on the basis of only operating cost savings to base traffic. Once the generated traffic and time savings are included, the IRR rises to 23.9 per cent for the whole road. It should be noted that, in every case, the addition of the benefits to generated traffic and from passenger time savings made relatively little difference to the rates of return obtained from considering only the base traffic; the savings in vehicle operating costs to base traffic alone yield a firm basis for assessing the viability of the project.

COST-BENEFIT ANALYSIS
A.33 The project was analysed using the TRRL RTIM2 computer program. The basic analysis used the most likely forecasts of the various parameters to compare the situation over the years 1989 to 2003 for the case where the existing situation was to be maintained with its present characteristics, and with the situation which would follow the upgrading of the road. A.34 Analysis was carried out separately for the three sections of road and for three separate benefit streams:

TABLE A10 ECONOMIC ANALYSIS RESULTS OF THE MOST LIKELY CASE

Base traffic only Road section 1 (6.2km) 2 (71.3km) 3 (17.5km) Total project NPV (at 10% discount rate) 3000 8143 882 12025 IRR (%) 46.8 20.0 14.0 20.6

Base plus generated traffic NPV (at 10% discount rate) 3554 10658 1406 15618 IRR (%) 50.1 22.3 17.1 23.0

Traffic benefits plus time savings NPV (at 10% discount rate) 3776 11398 1687 16861 IRR (%) 52.4 23.1 17.1 23.9

TABLE A11 DEPENDENCE OF NPV ON DISCOUNT RATE

Base traffic only Section 7.5% 1 2 3 Total project 3898 12231 1723 17852 12.5% 2327 5148 275 7750

Base plus generated traffic 7.5% 4612 15466 2397 22475 12.5% 2763 7126 687 10576

Traffic benefits plus time savings 7.5% 4887 16380 2744 24011 12.5% 2945 7733 917 11595

86

A.36 Clearly, the most heavily trafficked section of the road (Section 1) shows the highest internal rate of return, but the least trafficked (Section 3) shows an IRR of 14.0 per cent for operating cost savings to basic traffic, and 17.1 per cent when benefits to all traffic and the value of time savings are also included. The highest NPV is, not surprisingly, generated by the longest section of road (Section 2) in all cases. A.37 Overall, the project is clearly viable with the assumptions considered to be the most likely for each of the input parameters, and even the least trafficked section is viable on the basis of benefits to existing base traffic only. The robustness of the project's viability is further demonstrated by the sensitivity tests which were subsequently carried out.

Traffic growth rates Construction cost Road maintenance

Discount rate
The project was analysed for discount rates 25 per cent above and below the test discount rate of 10 per cent. The effect on NPV is shown in Table All. It will be seen that, at all discount rates, all individual sections of road, as well as the total project, always produced a positive NPV.

Variations in ADT SENSITIVITY TESTS

A.38 Because of the uncertainties surrounding many of the base year data and, therefore, the scope for uncertainty in many of the forecasts, extensive sensitivity testing was carried out. The parameters considered were as follows: Discount rate ADT A.40 Sensitivity tests were carried out for ADT levels 25 per cent above and below the estimated values used in the most likely cases, which were derived from traffic counts. The sensitivity of NPV is shown in Table A12 and that of IRR is shown in Table A13. The use of a discount rate of 10 per cent led to the NPV for Section 3, including benefits to base traffic only, becoming negative; the IRR was below 10 per cent. All other cases had a positive NPV. The economic return for the project as a whole was satisfactory when considering benefits to normal traffic only and its viability seemed robust.

TABLE A12 DEPENDENCE OF NPV (AT 10% DISCOUNT RATE) ON ADT Road section 1 2 3 Total project Base traffic only Low 2701 3816 -21 5866 High 3912 12451 1769 18132 Base plus generated traffic Low 2490 5726 374 8590 High 4598 15566 2418 22582 Traffic benefits plus time savings Low 2656 6283 585 9524 High 4875 16491 2769 24135

TABLE A13 DEPENDENCE OF IRR ON ADT

Road section 1 2 3 Total project

Base traffic only Low 36.7 14.9 9.9 15.4 High 56.4 24.8 17.7 25.6

Base plus generated traffic Low 39.7 17.0 11.7 17.5 High 60.0 27.7 19.8 28.0

Traffic benefits plus time savings Low 41.4 17.6 12.6 18.2 High 62.8 28.2 21.1 29.2

87

Traffic growth rates

A.41 Three variations in traffic growth rates were considered in the sensitivity tests: Zero traffic scenario: no growth at all in either base traffic or generated traffic Low traffic scenario: 3 per cent growth in base traffic and 5 per cent growth in generated traffic High traffic scenario: 7 per cent growth in base traffic and 11 per cent growth in generated traffic.

Construction cost
A.43 The sensitivity of the project's viability to variations in the construction cost was tested by assuming costs 25 per cent higher and 25 per cent lower than the most likely case. Even in the high cost case, all road sections remain viable, as shown in Tables A16 and A17. A.44 The effect of assuming changes in construction cost had more effect on the economic return of the project than had been exhibited in the sensitivity tests on traffic levels and growth. However, where the low cost assumptions produced a marked increase in the rates of return, the high cost assumptions did not generally depress the IRR's to the same levels as the sensitivity tests on traffic.

In each case, the percentages given for traffic growth were weighted averages across the different vehicle types which comprised the total traffic flow. The results of these analyses are shown in Tables A14 and Al5. A.42 It was demonstrated that the viability of the project was robust under varying traffic assumptions, which was plausible given the high rates of return demonstrated for the most likely case and the satisfactory results for lower than estimated ADT. Even in the case of zero traffic growth, the project remained extremely viable overall, at 16.1 per cent IRR on benefits to base traffic only (18.0 per cent if all the benefits are included). The small variation in economic return between the low and high traffic growth scenarios again illustrated that a significant element of the net benefit accrued from the cost savings to existing base traffic.

Maintenance capability
A.45 The economic analysis was carried out on the assumption that maintenance would continue at existing levels observed in the country. Sensitivity testing was carried out to try and predict the consequences of future levels of maintenance falling from these levels. It was not possible to investigate this directly with the investment model so, instead, it was assumed that poor maintenance would result in a pavement that was effectively weaker than it otherwise would be. The best estimate of

TABLE A14 DEPENDENCE OF NPV ON TRAFFIC GROWTH RATES

Road section 1 2 3 Total project

Base traffic only Zero 2053 3658 -26 5685 Low 2571 6081 471 9123 High 3512 10563 1374 15449

Base plus generated traffic Zero 2191 4301 107 6599 Low 2903 7569 785 11257 High 4445 14835 2255 21535

Traffic benefits plus time savings Zero 2364 4861 330 7585 Low 3103 8241 1040 12384 High 4692 15652 2566 22910

TABLE A15 DEPENDENCE OF IRR ON TRAFFIC GROWTH RATES

Road section 1 2 3 Total project

Base traffic only Zero 42.0 15.4 9.9 16.1 Low 44.9 18.1 12.3 18.8 High 48.7 22.0 15.8 22.6

Base plus generated traffic Zero 44.0 16.3 10.6 17.1 Low 47.6 19.7 13.7 20.4 High 52.8 25.1 18.6 25.7

Traffic benefits plus time savings Zero 46.4 17.1 11.7 18.0 Low 49.9 20.5 14.8 21.4 High 55.0 25.9 19.6 26.6

88

TABLE A16 DEPENDENCE OF NPV ON CONSTRUCTION COST Road section 1 2 3 Total project Base traffic only Low 3204 10624 1600 15428 High 2796 5661 164 8621 Base plus generated traffic Low 3758 13140 2124 19022 High 3350 8177 688 12215 Traffic benefits plus time savings Low 3980 13880 2405 20265 High 3572 8917 969 13458

TABLE A17 DEPENDENCE OF IRR ON CONSTRUCTION COST Road section 1 2 3 Total project Base traffic only Low 59.6 26.5 19.0 27.3 High 38.6 15.8 10.6 16.3 Base plus generated traffic Low 63.4 29.0 21.3 29.8 High 41.7 17.9 12.5 18.5 Traffic benefits plus time savings Low 66.2 30.0 22.6 31.0 High 43.5 18.6 13.4 19.3

TABLE A18 EFFECT ON NPV AND IRR OF A REDUCED MAINTENANCE CAPABILITY Road section 1 2 3 Total project Base traffic only NPV 2782 6190 631 9603 IRR 46.1 18.3 13.0 19.1 Base plus generated traffic NPV 3301 8458 1126 12885 IRR 49.5 20.6 15.0 21.5 Traffic benefits plus time savings NPV 3522 9187 1406 14115 IRR 51.8 21.5 16.2 22.5

pavement strength was that it would have a 'modified structural number' of 4.0, so the effect of reducing this to 3.2 was examined in order to study the implications of a reduction in maintenance capability. The result are shown in Table A18. A.46 In each case, the economic return was very slightly lower than in the most likely case because of the increased vehicle operating costs, but were still more than acceptable in all cases.

because RTIM2 does not allow for their inclusion, or because the base data were not available to permit their ready quantification. However, although the project appeared robust without their inclusion, they are described below.

Time savings for freight

A.48 The benefits from saving time in the transport of freight were seen as the prevention of deterioration of produce and other perishable commodities. It was thought that greater damage to produce was caused by it having to be hauled over a rough road, rather than it perishing due to long journey time. However there would be an accumulation of small benefits from saving time in

OTHER BENEFITS
A.47 In addition to the quantified factors included in the analysis, a number of benefits were omitted, either

89

freight haulage, in terms of getting produce to markets more quickly, which were not included in the economic analysis.

Diverted traffic
A.49 Turning movement counts and interviews from freight hauliers suggested that there would be some limited diversion from other roads. As it was difficult to quantify the extent of the current use of these roads and to predict the degree of diversion after the paving of the road, diverted traffic was omitted from the analysis. A.50 It is clear, however, that there would be two sets of benefits from such diversion: Traffic diverting on to the new road would experience vehicle operating cost saving since that which did not, would not divert There would be maintenance cost savings on the existing unpaved roads if the level of use of them by trucks decreased.

agricultural produce, much of it high value/low weight. There were good markets in the country's capital for this but, more importantly, good possibilities for air freight export as black hauls, as is currently practised from other African countries. Produce of this type could not be transported without considerable and uneconomic damage and loss.

Road safety
A.54 It proved impossible to determine the number of accidents along the road, or to establish the total number of road accidents even at the provincial level, due to incomplete recording of damage-only accidents and the absence of a reliable data-base on injury accidents. Neither were data readily available to enable road accidents to be costed in a scientific manner. It did appear, however, that the prevailing level of road accidents is, rightly, of major concern to the government. It was clear that accidents throughout this country were contributed to by poor road alignment and inadequate sightlines but, in the absence of any thorough investigation of the cause of accidents, it was difficult to express a view on the matter, and any costs or benefits in this respect were omitted from the analysis.

There would be disbenefits in terms of road deterioration and speed/flow effects if the level of diverted traffic were very high but, as a proportion of the total traffic on the new road, it was likely to be small.

Socio-economic integration
A.5 1 The study suggested that a major, but largely unquantifiable benefit, from the proposed scheme would be the enhanced socio-economic integration of the province in which the road was sited into the rest of the country. Many of these effects would be manifested as generated traffic, but it was likely that there would be other beneficial effects from a governmental and public perception of the province becoming more accessible. For example, it should prove easier to attract doctors and teachers to work in the remoter areas, and banks which were currently inhibited from establishing new branches along the road would feel more secure in transporting money along its length.

Improved consumer choice

A.52 The possibility of using larger trucks on an improved road would lead to more economic distribution of consumer items, and thus to a greater choice and availability of local and imported manufactured goods in the markets along the road.

Increased export earnings

A.53 There appeared to be considerable potential for diversification into high altitude, temperate climate

90

APPENDIX B ROAD SHORTENING PROJECT BACKGROUND TO THE STUDY

B.1 The TRRL road investment model was used as part of an appraisal of a road project by ODA. The existing gravel road traversed an area of dense population and high agricultural potential. The topography in the area was rugged and included 10 mountain rivers and 30 streams. The new alignment replaced the old road and shortened the direct route length by 30 km, from 84.40 to 54.38 km. The principal benefits considered were the economic value of road user savings resulting from shortening the route and improving the road surface to bitumen standard. These were principally vehicle operating cost savings for the expected 'normal' traffic over the life of the project, but the value of time savings for this traffic was also considered. A further source of benefits was a reduction in road maintenance costs as a result of the project. The effect of generated traffic was also considered. B.2 The existing road followed the contours of the ground, having a tortuous horizontal alignment, but relatively flat gradients. The new road was to have gentle horizontal curves and steeper gradients and involved the construction of substantial earthworks. On completion of the project, the existing road was to have been abandoned. Predicted traffic levels suggested that the ADT for normal traffic in 1985 would be just under 200 vpd rising to more than 600 vpd, with generated traffic rising from 300 to over 1200 vpd after 20 years.

ROAD USER COST SAVINGS

B.3 The vehicle operating costs of traffic on the road would be affected by the severity of the road geometry and the condition of the road surface (roughness). On the existing road, it was predicted that roughness would vary from 5.0 rn/km on average in the first year up to about 11.0 m/km with the same level of maintenance in the final year. On the new project, roughness would rise steadily from 2.5 rn/km in the first year of trafficking to 3.5 rn/km in the last. Vehicle operating costs were also affected by the age of vehicle. It was estimated that vehicles would achieve higher utilisation on the new project which implied that, on average, vehicles would be older in terms of the total number of kilometres travelled since new. The differences in road user costs for the 'with' and 'without' project situations are illustrated in Table B 1 which shows costs for the first year of traffic and the final year of analysis. B.4 There would be a reduction in vehicle operating costs per kilometre for all vehicle types following the construction of the project in both the first year of trafficking and the final year of analysis. The reduction ranged from 10 to 43 per cent for different vehicle types and in different project years. These changes in cost were caused by the reduction in road roughness and the increase in speed and vehicle utilisation. All these savings would be increased further because of the shortening of the road length.

B.5 The assessment of time savings per kilometre was particularly interesting. The new road had lower roughness levels than the original, but had steeper gradients, and vehicle speed, and consequently time savings, would depend on the relative magnitude of

TABLE B1 ROAD USER COST SAVINGS PER VEHICLE KM

Vehicle operating costs Existing road New project Saving (%) First year of trafficking Car Lt. Goods Truck Bus 1.03 2.29 4.87 3.32 0.68 1.44 4.38 2.47 34 37 10 25 Final year of analysis Car Lt. Goods Truck Bus 1.28 2.72 5.59 3.84 0.80 1.54 4.57 2.67 38 43 18 30 1.59 0.22 0.31 1.68 0.96 0.13 0.12 0.73 Existing road

Time New project Saving (%)

0.47 0.07 0.17 1.02

51 46 -42 -40

0.48 0.08 0.18 1.06

70 64 42 37

91

TABLE B2 Annual costs for new project

92

these components. In the case of cars and light goods vehicles, the effect of reduction in roughness outweighed the increased gradients in the first year of traffic and considerable time savings were predicted. This saving was increased further in the last year of traffic. However, in the case of trucks and buses, the balance of these effects caused speed to fall rapidly in the first year of trafficking which resulted in an increased time cost. This increase was so large that, even allowing for the shorter road length, it would be quicker for trucks and buses to use the old route. By the time of the final year of analysis, it was predicted that roughness levels on the old road would have outweighed the effect of steeper gradients, and positive time savings per kilometre would occur on the new project. B.6 Tables B2 and B3 show the undiscounted cost streams for all cost components for both the new project and the existing road over the analysis period. The vehicle operating cost and time cost columns show clearly the gradual increase in these costs over time for both normal and generated traffic. This increase is due both to the growth in traffic over time and to the effect on vehicle operating costs of the deterioration of the road surface condition with time.

B.l0 The costs in Tables B2 and B3 are not discounted and it is interesting to note that the maintenance cost of the new project was less than 3 per cent of the construction cost and less than 2 per cent of the total cost (construction plus maintenance plus vehicle operating cost). For the existing road, maintenance still represented less than 4 per cent of the total life cycle cost.

BENEFITS
B.11 The undiscounted annual net benefits for the project are shown in Table B4. The benefits due to construction, maintenance, and vehicle operation and time for normal traffic were found by the investment model by subtracting the costs in Table B2 from those in Table B3. The vehicle operating cost and time benefits due to generated traffic were evaluated using the consumer surplus method as described in para 14.7-11. The investment model has determined these automatically. Road accident savings were not included in the analysis in this case. B. 12 The discount rate used for this study was 12 per cent and, at the bottom of Table B4, the various cost stream totals have been discounted at this rate and at rates three per cent on either side of this. It was predicted that the project would produce direct savings in vehicle operating costs of 52 million currency units, savings in maintenance costs of 2.6 million, but at a discounted capital cost of 150 million. These values were used to determine the NPV's shown in Table B5. Both NPV, IRR and FYRR values have been presented separately for: (i) (ii) (iii) construction, maintenance and vehicle operating cost benefits to normal traffic only as (i), but with the addition of vehicle operating cost benefits to generated traffic (consumer surplus) as (ii) but, with the addition of time saving benefits.

ENGINEERING COSTS
B.7 Table B2 shows how the construction cost was spread over the five years 1981-1985, with different percentages being spent in each of these years. For the existing road, in Table B3, a regravelling operation was programmed for 1984 in the do-nothing or do minimum case. B.8 The inputs of periodic maintenance can be clearly seen in the 'Maintenance' columns of Tables B2 and B3. Table B2 shows the cost peaks where surface dressing would be carried out in 1992 and 2000. The cost of 79856.70 represents the cost of drainage and shoulder maintenance plus the cost of maintenance overheads. The slight escalation of maintenance cost in 1998 and 1999 represented the cost of patching cracks and pot-holes. Notice that the need for this was eliminated following a surface dressing. Table B3 shows that regravelling was planned to be carried out in 1984, 1988, 1992, 1996, 2000 and 2004. This was based on the estimate of regravelling needs because of the rate at which gravel would be worn away by traffic and rainfall. The remaining costs represented drainage and overhead costs plus the cost of grading the road which was expected to be carried out at the same frequency throughout the life of the project. B.9 A benefit of 200 million currency units was included in the final year of the project to cover the residual value of the earthworks and drainage structures at the end of the analysis period.

B.13 It is clear from these figures that, at a 12 per cent discount rate, the project was not justifiable. At a 9 per cent rate, it was worth going ahead providing generated traffic and time benefits were included, but the timing of the start of the project should be reassessed.

SENSITIVITY TESTING
B. 14 The road investment model was used to carry out sensitivity analysis of the key variables. The sensitivity to discount rate, developmental benefits (generated traffic) and time savings were carried out during the initial analysis using expected values of data. The sensitivity of the result to other variables was determined by carrying out additional analysis with the model.

93

94
TABLE B3 Annual costs for existing road

TABLE B4 Annual benefits for project

95

Discount rate, developmental benefits and time

B.15 The sensitivity of the result to these variables is shown in Table B5. The discount rate was varied to examine the NPV at discount rates of 9 and 15 per cent. Table B5 enabled the NPV to be compared with and without generated traffic, and with and without time savings. B.16 Considering the benefits to normal traffic alone, the NPV was negative at all discount rates tested. This was because the relatively large construction cost involved was not offset by the vehicle operating cost savings obtainable. The internal rate of return was less than half the test discount rate. A large amount of generated traffic was expected on the project and it was only when the benefits from this were considered in conjunction with a discount rate of 9 per cent that the project produced a positive NPV. In this case, the inclusion of time savings doubled the NPV. However, time savings were insufficient to produce a positive NPV at any other discount rate. Even including both benefits to generated traffic and time savings, an internal rate of return of less than 11 per cent was expected. B.17 Consideration of the FYRR, with values ranging from 3.5 to less than 8 per cent, compared with a discount rate of 12 per cent, confirmed that the project was premature.

Traffic
B.18 The sensitivity to both baseline traffic levels and traffic growth was examined for values of plus or minus 50 per cent of that expected. The results are shown in Table B6.

B.19 Clearly, with a negative NPV, reducing either baseline flows or growth rates would reduce the value of the NPV further. However, even increasing the ADT or growth rate by 50 per cent could not produce a positive NPV unless generated traffic was taken into account, in which case benefits of 30-40 million currency units were anticipated. The inclusion of time savings more than doubled this value.

Construction cost
B.20 As expected, increasing the construction cost by 25 per cent reduced the NPV of the project even further. Reducing costs by this amount produces a marginally positive NPV when generated traffic benefits were

96

TABLE B6 SENSITIVITY ANALYSIS Normal traffic only NPV (millions) Expected values Discount rate -25% Discount rate +25% ADT 50% ADT +50% Traffic growth 50% Traffic growth +50% Construction 25% Construction +25% -86 -63 -94 -115 -48 -103 -54 -48 -124 Per cent change +27% -9% -34% +44% -16% +37% +44% -44% Plus generated traffic NPV (millions) -35 +15 -60 -93 +41 -70 +29 +3 -74 Per cent change +59% +117% +30% -12% +148% +19% +134% +103% +13% Plus time savings NPV (millions) -23 +34 -52 -88 +87 -62 +79 +15 -61 Per cent change +73% +140% +40% -3% +201% +23% +192% +117% +29%

included. Considering time savings as well gave an NPV of 15 million currency units with this level of reduction in construction cost.

Relative sensitivities
B.21 Table B6 shows the changes in NPV for various percentage changes in discount rate, traffic, construction cost, and for the inclusion of developmental benefits (generated traffic) and time savings. The results from Table B6 show that the effect on NPV of reducing the discount rate was much more significant than increasing the discount rate. Changing either the ADT or traffic growth rate by plus or minus 50 per cent caused a smaller change in the NPV, and a slightly larger change was caused by increases, rather than decreases, in traffic. The NPV was increased dramatically when the effect of generated traffic was included. A 25 per cent change in construction cost caused a much larger change in NPV and this change was the same for both increases and decreases in values. B.22 The parameter whose sensitivity had most effect on the NPV for normal traffic benefits was that of construction cost, but when generated traffic benefits were included, the baseline flow and traffic growth rate parameters were those whose sensitivity was most likely to give rise to a positive NPV. Reductions in NPV were most likely to be caused by an increase in construction cost.

97

NOTES

98

NOTES

99

NOTES

100

A guide to road project appraisal

ORN 5

Transport and Road Research Laboratory Overseas Unit Department of Transport Overseas Development Administration

Overseas Road Note 6

A guide to geometric design

Overseas Unit Transport and Road Research Laboratory Crowthorne Berkshire United Kingdom 1988

ACKNOWLEDGEMENTS This note has been produced for the Overseas Unit of the Transport Research Laboratory by Roughton and Partners, Consulting Engineers. The Project Manager for TRL was Dr R Robinson of the Overseas Unit

First Published 1988; Reprinted 1998

TRL is committed to optimising energy efficiency, reducing waste and promoting re-cycling and re-use. In support of these environmental goals, this note has been printed on recycled paper, comprising 1OO% postconsumer waste, manufactured using a TCF (totally chlorine free) process.

OVERSEAS ROAD NOTES

Overseas Road Notes are prepared principally for road and road transport authorities in countries receiving technical assistance from the British Government. A 1imited number of copies is available to other organisations and to individuals with an interest in roads overseas, and may be obtained from Transport Research Laboratory Crowthorne, Berkshire, RG45 6AU United Kingdom

Crown Copyright 1988 Limited extracts from the text may be produced provided the source is acknowledged. For more extensive reproduction, please write to Electronic Publishing Manager, Transport Research Laboratory ISSN 0951-8797

CONTENTS

1.

INTRODUCTION Purpose of this Note Approach to design Selection of design standard Cross sectional considerations Design speed Safety Economic design Road function Traffic flow Design flow Composition Capacity Terrain Curvature standards The design process Basic parameters Select Design Class Determine trial alignment Design Class standards Approach speed estimation Economic consequences Relaxation of standards Economic return CROSS-SECTION Basic considerations for determining widths Carriageways and shoulders Passing places Curve widening Lateral and vertical clearances Crossfall Carriageway markings Provision for non-motorised travellers Rights-of-way HORIZONTAL ALIGNMENT Circular curves Adverse crossfall Superelevation Transition curves Other considerations VERTICAL ALIGNMENT Components of the vertical alignment Crest curves Sag curves Gradient Climbing Lanes ECONOMICS AND SAFETY Economic Assessment Safety Non-motorised traffic Driver safety

page 1 1 1 1 1 1 1 1 2 3 3 4 4 4 4 5 5 5 5 5 6 6 6 6 7 7 7 8 8 9 9 10 10 10 11 11 11 12 12 13 14 14 14 17 19 20 21 21 21 21 21

2.

3.

4.

5.

REFERENCES APPENDIX A : GLOSSARY OF TERMS APPENDIX B ESTIMATION OF VEHICLE SPEED APPENDIX C : PHASING OF THE VERTICAL AND HORIZONTAL ALIGNMENT Defects in the alignment due to misphasing Types of misphasing and corresponding corrective action Insufficient separation between curves The vertical curve overlaps one end of the horizontal curve Both ends of the vertical curve lie on the horizontal curve The vertical curve overlaps both ends of the horizontal curve The economic penalty due to phasing

page 23 23 25 28 28 28 28 28 28 28 29

INTRODUCTION
PURPOSE OF THIS NOTE
1.1 This Note gives guidance on geometric design and the setting of geometric design standards for single carriageway rural (inter-urban) roads in developing countries. It is aimed at government officials who are responsible for formulating policy on geometric design and at engineers who are responsible for preparing road designs. It will also be of interest to personnel in aid agencies and consultancies who are responsible for the preparation and design of road projects. Many countries will have existing standards different from those described in this guide. This should not preclude the use of the standards in this guide, although where good local cost and benefit information is available, some aspects may need to be reviewed. 1.2 Geometric design is the process whereby the layout of the road in the terrain is designed to meet the needs of the road users. The principal geometric features are the road cross-section and horizontal and vertical alignment. The use of geometric design standards fulfills three inter-related objectives. Firstly, standards are intended to provide minimum levels of safety and comfort for drivers by the provision of adequate sight distances, coefficients of friction and road space for vehicle manoeuvres; secondly, they provide the framework for economic design; and, thirdly, they ensure a consistency of alignment. The design standards adopted must take into account the environmental road conditions, traffic characteristics, and driver behaviour. 1.3 The derivation of the standards recommended in this Note, and summarised in Tables 1.1 and 1.2, is described in TRRL Contractor Report 94 (Boyce et al 1988). 1.4 A glossary of terms in this guide is given as Appendix A. types of flow, as well as geometric elements. Additional design consideration or special signing will only be necessary where the information available to the driver may lead to incorrect interpretation and consequent danger.

Cross-sectional considerations
1.7 Cross-section parameters are related to traffic flows of all types, and will vary with the requirements of vehicular traffic and with the needs of pedestrians and non-motorised vehicles. In many developing country situations, it will be necessary to consider cost effective ways of segregating nonmotorised traffic at the earliest stage in the design process.

Design speed
1.8 Design speed is used as an index which links road function, traffic flow and terrain to the design parameters of sight distance and curvature to ensure that a driver is presented with a reasonably consistent speed environment. In practice, most roads will only be constrained to minimum parameter values over short sections or on specific geometric elements.

Safety
1.9 There is very little information from developing countries on the effects of changes in standards on accident rates. Indeed. equivalent information from developed countries is also limited. Highway engineering safety is usually assumed to be optimised by linking geometric elements to a design or operating speed, so that the resulting geometry has a consistency which reduces the likelihood of a driver being presented with an unexpected situation. This concept of driver expectation forms the basis of this set of design standards.

Economic design
1.10 Designs should be justified economically, and the optimum choice will vary with both construction and road user costs. Construction costs will be related to terrain type and choice of pavement construction, whereas road user costs will be related to level and composition of traffic, journey time, vehicle operation and road accident costs. Methods of determining these costs are given in Overseas Road Note 5 (TRRL Overseas Unit 1988). 1.11 The most economic designs will often not involve the use of minimum standards, as levels of traffic may be such that the additional vehicle operating cost, accident, and travel time saving benefits from wider, straighter and shorter roads may more than offset the extra construction costs needed.

APPROACH TO DESIGN Selection of design standards

1.5 The section of design standards is related to road function, volume of traffic and terrain, with additional procedures for the recognition and appropriate treatment of potential hazards (Tables 1.1 and 1.2). Opportunities for the relaxation of standards have also been identified. 1.6 A basic assumption in the approach is that drivers receive clues about the standard of the road from local surrounding features such as the terrain, levels and

1.12 As flows increase, vehicle-to-vehicle interactions become more important and congestion may result in increases in journey times and accident risk if additional lanes are not added. The scope of this Note has been limited to single carriageway roads, and consideration of the possible introduction of dual carriageways should be made when flows approach 15,000 vehicles per day.

ROAD FUNCTION
1.13 Each inter-urban road may be classified as being arterial, collector or access in nature as shown in Figure 1.1.

1.14 Arterial roads are the main routes connecting national and international centres. Traffic on them is derived from that generated at the urban centres and from the inter-urban areas through the Collector and Access road systems. Trip lengths are likely to be relatively long and levels of traffic flow and speeds relatively high. Geometric standards need to be adequate to enable efficient traffic operation under these conditions, in which vehicle-to-vehicle interactions may be high. 1.15 Collector roads have the function of linking traffic to and front rural areas, either direct to adjacent urban centres, or to the Arterial road network. Traffic flows and trip lengths will be of an intermediate level and the need for high geometric standards is therefore less important. 1.16 Access roads are the lowest level in the network hierarchy. Vehicular flows will be very light and will be aggregated in the Collector road network. Geometric standards may be low and need only be sufficient to provide appropriate access to the rural agricultural, commercial and population centres served. Substantial proportions of the total movements are likely to be by non-motorised traffic. 1.17 Whilst this hierarchy is shown simplistically in Figure 1.1, in practice there will be many overlaps of function and clear distinctions will not always be apparent on functional terms alone. This hierarchy should not be confused with the division of administrative responsibilities which may be based on historic conditions.

Fig.1.1 Road hierarchy and function TABLE 1.1: ROAD STANDARDS

* The two way traffic flow is recommended to be not more than one Design Class step in excess of first year ADT. + For unpaved roads where the carriageway is gravelled, the shoulders would not normally be gravelled; however, for Design Class D roads, consideration should be given to gravelling the shoulders if shoulder damage occurs.

TABLE 1.2 : SPEED RELATED DESIGN PARAMETER

1.18 For the lowest Design Class of road, it is inappropriate to design on the basis of geometric standards, and the sole criterion of acceptability will be the achievement of an appropriate level of access. Design, in these situations, should be based on minimum values of radii, width and gradient for the passage of a suitable design vehicle.

1.20 Design Classes A to F have associated bands of traffic flow as shown in Table 1.1. The range of flows extends from less than 20 to 15,000 motorised vehicles per day, excluding motorcycles, and covers the design conditions for all single carriageway roads. 1.21 Although the levels of flow at which design standards change are based on the best evidence available, the somewhat subjective boundaries should be treated as approximate in the light of the uncertainties inherent in traffic estimation and economic variability. Therefore, design flows should normally be constrained to be no more than one Design Class step higher than the annual average daily traffic (ADT) in the first year of trafficking. Thus, a road with a first year traffic flow of 390 vehicles per day rising to 1,100 vehicles per day should be constructed to Design Class C rather than Design Class B geometry (see Table 1.1). The design flow band in this case is therefore 400-1000 vehicles per day. Design to the higher Design Class would result in an overdesigned facility during

TRAFFIC FLOW Design flow

1.19 The functional hierarchy is such that traffic is aggregated as it moves from Access to Collector to Arterial road, and levels of flow will normally be correlated to road type. However, flow levels will vary between countries and regions and it is important that the designation of a road by functional type should not give rise to overdesign for the levels of traffic actually encountered. Uneconomic designs reduce the likelihood of roads being built and result in wastage of often scarce national resources.

almost the whole of the life of the road and may provide a solution that was less than the economic optimum. If the initial flow were 410 vehicles per day, design would still be to Design Class C. It is particularly important that roads are not overdesigned on the basis of high traffic growth rates which normally incorporate considerable uncertainty.

Level (0-10 five metre ground contours per 1.27 kilometre). Level or gently rolling terrain with largely unrestricted horizontal and vertical alignment. Minimum values of alignment will rarely be necessary. Roads will, for the most part, follow the ground contours and amounts of cut and fill will be very small. 1.28 Rolling (11-25 five metre ground contours per kilometre). Rolling terrain with low hills introducing moderate levels of rise and fall with some restrictions on vertical alignment. Whilst low standard roads will be able to follow the ground contours with small amounts of cut and fill, the higher standards will require more substantial amounts.

Composition
1.22 Although, in some situations, heavy vehicles have a greater effect on congestion than light vehicles, no attempt has been made to use passenger car unit (pcu) equivalent values. The relative effects of heavier vehicles vary with level of flow, geometry, and vehicle performance and consistent values that are well researched are not available for the range of flows covered in this design guide. All flows are therefore presented as ADT values. However, where there are very high percentages of heavy vehicles in a traffic stream, consideration may be given to the enhancement of standards, and particularly of carriageway width.

1.29

Mountainous (Greater than 25 five metre ground contours per kilometre). Rugged, hilly and mountainous with substantial restrictions in both horizontal and vertical alignment. Higher standard roads will generally require large amounts of cut and fill.
1.30 In general, construction costs will be greater as the terrain becomes more difficult and higher standards will become less justifiable or achievable in such situations than for roads in either flat or rolling terrain. Drivers should also expect lower standards in such conditions and therefore adjust their driving accordingly, so minimising accident risk. Design speed will therefore vary with terrain.

Capacity
1.23 Congestion increases with increased traffic flow when there is a lack of overtaking opportunity. The result is high journey times and vehicle operating costs, often accompanied by more accidents as frustrated drivers take risks. 1.24 Practical capacity is usually estimated to have been reached when the level of congestion becomes "unacceptable". Capacity reduces with increased proportions of heavy vehicles, greater unevenness in directional flows, reduced overtaking opportunities, animal drawn vehicles and pedestrian activity. Normally acceptable practical capacity will be about 1500 to 2000 vehicles per hour, but may be increased substantially by the provision of short sections of climbing and overtaking lanes. 1.25 Capacity is only likely to be approached for road Design Class A, or at the higher flow levels, road Design Class B, particularly in the more rugged terrain if adequate overtaking opportunities are unavailable.

CURVATURE STANDARDS
1.31
Minimum horizontal and vertical curvatures are governed by maximum acceptable levels of lateral and vertical acceleration and minimum sight distances required for safe stopping and passing manoeuvres. These design parameters are, in turn, related to the vehicle speeds assumed in the design. Curvature standards are thus either explicitly or implicitly dependent on an assumed design speed. 1.32 Within this guide, the adopted design speeds are explicitly stated and, as shown in Tables 1.1 and 1.2, have been taken to vary with both terrain and level of traffic flow. However, it must be emphasised that these speeds are intended to provide an appropriate consistency between geometric elements rather than as indicators of actual vehicle speeds at any particular location on the road section.

TERRAIN
1.26 A simple classification of "level", "rolling" and "mountainous" has been adopted and is defined by both subjective description and by the average ground slope. The average ground slope is measured as the number of 5 metre contour lines crossed per kilometre on a straight line linking the two ends of the road section. (The slope may be interpolated using other contour intervals on a proportional basis).

1.33

The use of lower design speeds in the more difficult terrain is intended to incorporate an element of reduced driver expectation and performance as well as the need to keep construction costs to acceptable levels. As flows increase, the level of benefits from reduced road length also increase and generally support higher standards with more direct and shorter routes.

THE DESIGN PROCESS

1.34 The design process is shown in Figure 1.2 with the main features detailed below. The emphasis throughout is on the need to obtain best value for money. scale of the map or photograph. Consideration will be given to gradient by reference to the contours of a map, or by relief when using stereo photographs. Several alternative alignments should be tried. The design process should be carried out in conjunction with on-site inspections and surveys. One or two of the alignments should be chosen for additional studies in more detail and be subject to further design and assessment prior to possible construction. 1.39 On two lane roads, the horizontal alignments should be designed to maximise overtaking opportunities by avoiding long, continuous curves. Instead, relatively short curves at, or approaching, the minimum radius for the design speed should be used in conjunction with straights or gentle, very large radius curves. Conversely, an alignment of flowing curves may reduce real overtaking opportunities, thus encouraging injudicious driver behaviour. On two lane single carriageway roads in developing countries, the provision of adequate overtaking opportunities may be particularly important because of the large proportions of slow moving vehicles. 1.40 Often a new road will be built to replace an existing facility. The structural features of the existing road, including bridges, embankments and cuttings may have substantial residual value and influence alignment choice. 1.41 The geometric standard of individual elements of the road will vary with the terrain. It is necessary that elements of lower geometric standard are identified to ensure that they will not result in unacceptable hazards to approaching vehicles. These elements will be readily identifiable from the preliminary horizontal and vertical curvature profiles. The tests for the necessary consistency are simple, as described below, and should be carried out if there is any doubt as to the acceptability of an element.

Basic parameters
1.35 Initially, the basic parameters of road function. traffic flow and terrain type are defined.

Select Design Class

1.36 On the basis of the above estimates, a Design Class is selected from Table 1.1. Values of Design Class boundaries are for guidance only, and the lower Design Class should be chosen in borderline cases.

Determine trial alignment

1.37 A road consists of a series of discrete geometric elements of horizontal and vertical curvature. Contiguous groups of these elements combine to form sections. In this guide, the minimum length of a road section is considered to be about one kilometre.

FIG. 1.2 : The design process

Design Class standards

1.42 It is recommended that, where the standard of a geometric element falls substantially below that on the approach section, its adequacy should be checked by estimating approach speed from the relationships given in Appendix B. Geometric elements should not normally be designed to a Design Class more than a one Design Class step lower than the approach speed to that element. However, two Design Class steps may be achieved by successive reductions from a design speed of, for example, 85 km/h in rolling terrain to 70 km/h and then 60 km/h (see paras 1.48-51). If this is not possible, consideration must be given to redesign of the element or alterations to the geometry of the approach section to obtain this speed reduction.

1.38 The initial stage in selecting an alignment for a new road is to sketch a route on a contoured map or aerial photograph. A similar process can be carried out when investigating the upgrading of an existing road. By reference to the standards, the designer will have some knowledge of appropriate minimum radii for the

Approach speed estimation Relaxation of standards

1.43 The speeds of freely moving vehicles on an interurban road usually conform to a normal distribution within which percentile values of speed are approximately related as follows: 1.2 x 15th percentile speed = 50th percentile speed 1.2 x 50th percentile speed = 85th percentile speed 1.2 x 85th percentile speed = 99th percentile speed. 1.48 The standards summarised in Tables 1.1 and 1.2 are intended to provide guidance for designers rather than to be considered as rigid minima. The justification for construction of a particular road will almost always be based on a detailed economic appraisal, and relaxations of standards may be essential in order to achieve an acceptable level of return on investment. In other circumstances, an already acceptable rate of return may be increased substantially by the inclusion of a short section of substandard road where achievement of the design standard would be expensive, although the safety implications of this would need serious consideration. 1.49 Relaxation of one Design Class step implies design to the 50th percentile rather than the 85th percentile speed. Relaxation of two Design Class steps reduces the design to the 15th percentile speed. Experience in the UK has shown that reduction of design parameters by one step, equivalent to a 17 per cent reduction, is likely to have little effect on safety. Normally, a relaxation of two steps, equivalent to a 30 per cent speed reduction, should not significantly increase risk where appropriate signing or other warning measures, such as bend marker posts, are provided. On low flow roads where most of the drivers will be regular users, the increased risk will be less significant and the resultant number of accidents should be negligible. Greater care and consideration should be given to relaxations on high flow/high speed alignments. 1.50 In special circumstances, where standards have been reduced on successive design elements, further relaxations may be made based on those reduced approach speeds. Sight distances, and the potential accident risk as a result of driver error, would need to be considered on a site-specific basis. 1.51 Reductions in standards should only apply to stopping distances and curvature, and suitable values have been included in Table 1.2. Widths should not be reduced as they are particularly flow related, and additional widening may be required on curves with the tighter radii.

1.44 The 85th percentile value of speed has been used as the basis of design in this guide. Thus, 15 per cent of the vehicles could be considered to be exceeding the design speed on any section of road. It also follows from the above that, for example, with a design speed of 100 km/h: I per cent would be exceeding 120 km/h; 50 per cent would be exceeding 85 km/h; and 85 per cent would be exceeding 70 km/h. Each such speed change has been taken to represent a consistent design step in Table 1.1. in which rounded values have been used. 1.45 A driver's ability to negotiate a geometric element safely will depend on his approach speed relative to a safe speed on the element. As it is not possible to predict speed profiles accurately, it is recommended that estimates of approach speed are made using the relationships described in Appendix B. These relationships produce estimates of 85th percentile speed. Speeds are modified by geometric characteristics and estimates of approach speed will be based on the geometry of about one kilometre on both approaches to the geometric element under consideration. These approach sections should include complete design elements ie complete horizontal or vertical curves and gradient lengths. There are considerable uncertainties in the accuracy of speed estimation relationships and the results should therefore be treated as approximate.

Economic consequences
1.46 If a geometric element fails to achieve the standard chosen for design, the economic consequences of upgrading to the standard must be considered. The economic consequences should generally be measured as additional cost of construction either in absolute terms or as a proportion of the overall cost. If this cost is small, the road alignment should normally be redesigned. If the cost is large, consideration should be given to further relaxation of standard as described in paras 1.48-51. 1.47 In general, the higher the class of road, and hence volume of traffic, the more likely will benefits from vehicle operating and time cost savings lead to the justification of a shorter, straighter route.

Economic return
1.52 All road design projects should be subject to an economic appraisal as recommended in Overseas Road Note 5 (TRRL Overseas Unit 1988). It is essential that those responsible for design should investigate whether amendments to an alignment will produce significant increases in economic rates of return.

2. CROSS-SECTION
BASIC CONSIDERATIONS FOR DETERMINING WIDTHS
2.1 Road width should be minimised so as to reduce the costs of construction and maintenance whilst being sufficient to carry the traffic loading efficiently and safely. Recommended values are given in Table 1.1. 2.2 For Access roads with low volumes of traffic (<100 ADT), single lane operation is adequate as there will be only a small probability of vehicles meeting, and the few passing manoeuvres can be undertaken at very reduced speeds using either passing places or shoulders. Provided sight distances are adequate for safe stopping, these manoeuvres can be performed without hazard, and the overall loss in efficiency brought about by the reduced speeds will be small as only a few such manoeuvres will be involved. It is not cost-effective to widen the running surface in such circumstances and a basic width of 3.0 metres will normally suffice. In some situations, 2.5 metres will allow effective passage.

2.3 On roads with medium volumes of traffic (1001000 ADT), the numbers of passing manoeuvres will increase and pavement widening will become worthwhile operationally and economically. However, in view of the generally high cost of capital for construction in developing countries and the relatively low cost of travel time, reductions in speed when approaching vehicles pass will remain acceptable for such flow levels and running surface widths of 5.0 and 5.5 metres are recommended. For Arterial roads with higher flows (> 1000 ADT), a running surface 6.5 metres wide will allow vehicles in opposing directions of travel to pass safely without the need to move laterally in their lanes or to slow down. 2.4 Typical cross-sections are shown in Figure 2.1.

CARRIAGEWAYS AND SHOULDERS

2.5 Shoulders are recommended for all but the lowest Design Class and will normally be paved when the carriageway is paved (Figure 2.1.). They are intended to perform three main traffic functions: To provide additional manoeuvring space on roads of lower classification and traffic flows To provide parking space at least partly off the carriageway for vehicles which are broken down To enable non-motorised traffic to travel with minimum encroachment on the carriageway. 2.6 Additionally, it may be desirable to provide sufficient width for two way movement during roadworks. 2.7 Clearly, these functions are not wholly compatible and detailed design recommendations have been based on the following logic. 2.8 Design Class F This class of road provides basic access only and the motorised traffic flows are so low that shoulders are not required. All road users will share the 3.0 (2.5) metre carriageway and passing places will be provided as appropriate. The width should be just sufficient to allow the occasional vehicles to traverse the road and design to specific geometric standards will be inappropriate. 2.9 Design Class E 1.5 metre shoulders have been recommended for this class of Access road as they will allow a total road width of 6.0 metres, sufficient for two trucks to pass with 1.0 metre clearance. Shoulders may also be used by non-motorised traffic and pedestrians, and potential conflicts will be acceptably low with the few motorised vehicles on the road. In difficult terrain, and elsewhere where construction costs are high, 1.0 metre shoulders may be acceptable, particularly when the carriageway and shoulders are paved or where the flow of non-motorised traffic is small.

Fig.2.1 TypicaI cross-section (Dimension in mm)

7

2.10 Design Class D On paved roads, 1.0 metre paved shoulders are recommended to provide a total paved width of 7.0 metres. this will allow approaching vehicles some lateral movement where necessary, albeit at a reduced speed. If the shoulders are unsurfaced, a high level of maintenance will be necessary to avoid damage and the resulting break up of the edge of the pavement. A minimum of 1.0 metre of surfaced shoulder will also encourage pedestrians and non-motorised users to use the shoulders, rather than the carriageway. Shoulder delineation is particularly important. It is most unlikely that non-motorised traffic will justify the construction of additional width with the levels of motorised vehicles on this class of road. However, the justification for surfaced shoulders or special provision will become greater as flows of traffic of all kinds rise (paras 2.37-42). Conversely, full shoulders may not be necessary in mountainous areas when construction costs are high and non-motorised vehicle flows are low. Where this is the case, the minimum paved width should be 5.5 metres, and side drains may need special consideration for safety reasons (para 5.11). 2.11 Design Class C Roads in this category will normally be paved. Recommendations for 1.0 metre surfaced shoulders are similar to those for Design Class D, but the extra 0.5 metre carriageway width to give a total paved width of 7.5 metres will allow easier passing. Full shoulders may be omitted in mountainous or difficult terrain where the costs of achieving desired cross-sections are very high. 2.12 Design Class B The carriageway of 6.5 metres will allow vehicles to pass with sufficient clearance for there to be little speed reduction or lateral movement. The minimum 1.0 metre shoulder will allow easier overtaking of stopped vehicles as well as the movement of some non-motorised traffic. Shoulders should be paved to provide a total paved width of 8.5 metres. At high levels of flow, where there are substantial traffic movements of wide non-motorised vehicles, such as bullock carts, it may be advisable to increase shoulder width in some circumstances up to a maximum of 2.5 metres, or provide special segregated facilities. 2.13 Design Class A Levels of traffic flow will be such that stopped vehicles blocking any part of the carriageway will be likely to cause a significant hazard. Hence, normal practice will be to provide a 2.5 metre shoulder, at least one metre of which should be paved. However, shoulder width may be reduced to 1.0 metre in difficult terrain where construction costs are high. The shoulder would also be available for non-motorised traffic and should be paved. However, in view of the potentially high levels of service and

associated speeds on this class of road, it is recommended that non-motorised traffic be discouraged and alternative segregated facilities provided where possible. 2.14 Dual carriageway construction should be considered where design flows approach about 15,000 vehicles per day. Design of dual carriageways is outside the scope of this guide and reference should be made to the Australian (NAASRA 1980) and British (Department of Transport 1981) standards. The flow value of 15,000 is arbitrary and, in industrialised countries, wider single carriageway roads have been found to carry up to 20,000 to 30,000 vehicles per day, albeit with some reduction in speeds.

PASSING PLACES
2.15 The lowest Design Class with a width of 3.0 (2.5) metres will not allow passing and overtaking to occur and passing places must he provided. The increased width at passing places should be such as to allow two trucks to pass, ie a minimum of 5.0 metres total width, and vehicles would be expected to stop or slow to a very low speed. 2.16 Normally, passing places should be located every 300 to 500 metres depending on the terrain and geometric conditions. Account should be taken of sight distances, the likelihood of vehicles meeting between passing places and the potential difficulty of reversing. In general, passing places should be constructed at the most economic locations as determined by terrain and ground condition, such as at transitions from cut to fill, rather than at precise intervals. 2.17 The length of individual passing places will vary with local conditions and the sizes of vehicles in common use but, generally, a length of 20 metres including tapers will cater for most commercial vehicles on roads of this type. 2.18 A clear distinction should be drawn between, passing places and lay-bys. Lay-bys may be provided for specific purposes, such as parking or bus stops, and allow vehicles to stop safety without impeding through traffic.

CURVE WIDENING
2.19 Widening of the carriageway on low radius curves will be essential to allow for the swept paths of larger vehicles, and the necessary tolerances in lateral location as vehicles follow a curved path. 2.20 Widths should be increased on horizontal curves to allow for the swept paths of trucks and to allow drivers

to manoeuvre when approaching other vehicles. The required amount of widening is dependent on the characteristics of the vehicles using the road, the radius and length of the curve, and lateral clearances. Carriageway widening is also necessary to present a consistent level of driving task to the road users, to enable them to remain centred in lane and reduce the likelihood of either colliding with an oncoming vehicle or driving onto the shoulder. 2.21 The following levels of widening are recommended. 2.22 Single lane roads (3.0m basic width) Curve radius (m) Increase in width (m)
20 1.50 30 1.00 40 0.75 60 0.50

2.27 On lower Design Class roads, E and F, which have substantial curvature requiring local widening, it may be practical to increase width over a complete section to offer a more consistent aspect to the driver. This enhancement of the standards should be undertaken where other advantages such as easier construction or maintenance can be identified and where the additional costs are acceptably small. This argument may also be appropriate for sections of lower curvature on roads of Design Classes C and D.

LATERAL AND VERTICAL CLEARANCES

2.28 Typical maximum truck heights are 4.2 metres and, to allow adequate vertical clearance and the transport of abnormal loads, a 5.0 metres vertical clearance should generally be allowed for in the design. 2.29 Lateral clearances between roadside objects and the edge of the shoulder should normally be 1.5 metres. This may be reduced to 1.0 metre where the cost of providing the full 1.5 metres is high. 2.30 Much smaller clearances will sometimes be necessary at specific locations such as on bridges, although a minimum of 1.0 metre will remain desirable. Minimum overall widths in such circumstances should be sufficient to allow the passage of traffic without an unacceptable reduction in speed, which will depend on the length of the reduced width section and levels of motorised and non-motorised traffic flow. Separate facilities should be provided for pedestrians where possible.

These values for widening on tight low speed bends have been based on a typical two-axle truck with an overall width of 2.5 metres, wheel base of 6.5 metres and overall length of 11.0 metres. This type of truck is typical of the two-axle vehicles to be found in most developing countries. Articulated vehicles have not been considered explicitly as they are not common on Access roads. 2.23 Two lane roads
<50 50-149 1.00 150-299 0.75 300400 0.50

Curve radius (m)

Increase in width (m) 1.50

CROSSFALL
2.31 Crossfall should be sufficient to provide adequate surface drainage whilst not being so great as to be hazardous by making steering difficult. The ability of a surface to shed water varies with its smoothness and integrity. On unpaved roads, the minimum acceptable value of crossfall should be related to the need to carry surface water away from the pavement structure effectively, with a maximum value above which erosion of material starts to become a problem. 2.32 The normal crossfall should be 3 per cent on paved roads and 4 to 6 per cent on unpaved roads. Shoulders having the same surface as the carriageway should have the same cross slope. Unpaved shoulders on a paved road should be 2 per cent steeper than the crossfall of the carriageway. The precise choice of crossfall on unpaved roads will vary with construction type and material rather than any geometric design requirement. In most circumstances, crossfalls of 5 to 6 per cent should be used, although the value will change throughout the maintenance cycle.

2.24 The above values are guidelines only and there will be many situations in which widening is neither necessary nor cost-effective. 2.25 Widening should be applied on the inside of a curve and be gradually introduced over the length of the transition. 2.26 On the narrower two lane roads of Design Class C and D, particularly if there are high flows of trucks, it may be desirable to widen the roads on crest vertical curves. Widening of 0.5 metre should be considered where K values are within one Design Class step of the minimum for safe stopping sight distance.

CARRIAGEWAY MARKINGS
2.33 Carriageway marking should be provided on all two-way paved roads. 2.34 The edge of the carriageway should be delineated by continuous lines and may be supported by surfacing road studs. or other features. The lines should be situated on the shoulder immediately adjacent to the running surface and should be at least lOOmm in width. Alternatively or additionally, delineation can be provided more permanently by sealing the shoulder with a different coloured aggregate to the running surface. (If, contrary to these recommendations, an unsealed shoulder is adopted, the first 150mm should be sealed for marking purposes). 2.35 Centre line markings are also recommended on roads of at least 5 metres width designed for two lane operation in order that a driver may correctly locate his lateral position. These markings should be 100mm wide and normally be discontinuous. except where overtaking is restricted. and may be supported by the use of road studs. 2.36 Within the requirements for centre line and edge markings, local standards and manuals should be used or developed to provide uniformity of marking throughout a national road network. All road markings should conform to international standards.

2.41 There may be substantial movements of pedestrians and non-motorised vehicles which will generally be attracted by the surface quality and all weather properties of roads. Special provisions should be made in situations where such flows are significant with respect to the level of motorised vehicle movements. Some localised shoulder improvements may be appropriate as non-motorised traffic generally increases near towns and villages. Two features which are recommended where large numbers of non-motorised users travel on the shoulders are: The shoulders should be sealed They should be clearly segregated by the use of edge of carriageway surface markings or other measures.

Special crossing facilities should be provided where possible and necessary. 2.42 On high speed roads with substantial flows of motorised vehicles, non-motorised traffic should be given a separate segregated by a physical barrier such as a kerb. Crossing movements should also be concentrated at specific locations and special crossing facilities provided. Traffic approaching these facilities should be given adequate warning and stopping sight distances which are greater than minimum values should be provided where possible.

RIGHTS-OF-WAY
2.43 It is recommended that the rights-of-way should extend to a minimum of three metres from the edge of the road works. This right-of-way should normally be marked by a fence for road Design Classes A and B, and as appropriate for the lower Design Classes.
2.44 The right-of-way must include the acquisition of land necessary for the provision of special facilities for pedestrians and other non-motorised road users. Consideration should also be given to the acquisition of land for short cuts and paths for pedestrians where they exist away from the road. 2.45 Rights-of-way may be reserved for future upgrading of the alignment, although this would not be normal practice.

PROVISION FOR NON-MOTORISED TRAVELLERS

2.37 Consideration needs to be given to the movement of pedestrians. cyclists and animal drawn vehicles either along or across the road. Measurements or estimates of such movements should be made, where possible, to give a firmer basis for making decisions on the design. 2.38 At very low flows of motorised traffic, the problem of interaction is likely to be small. However, care must be taken to ensure that adequate sight distances and/or warnings are given to a driver as he approaches any area of high activity such as a village.

2.39

As flows become greater, the conflicts between slow and fast moving traffic will increase and additional widths of both shoulder and running surface may be necessary. The increase in width will vary with the relative amounts of traffic, their characteristics and the terrain, and should be related to the needs of individual countries and regions as well as individual sections of road. In view of the relatively high costs normally involved in widening, care should be taken to ensure that only those sections of shoulder are widened which are justified by local demand. 2.40 Recommendations for shoulder widths in these situations are given in paras 2.5-2.13.

1 0

10

3. HORIZONTAL ALIGNMENT
CIRCULAR CURVES
3.1 When vehicles negotiate a curve, a sideways frictional force is developed between the tyres and road surface. This friction must be less than the maximum available friction if the bend is to be traversed safely. For any given curve and speed, superelevation may be introduced to enable a component of the vehicle's weight to reduce the frictional need. The general relationship for this effect is: 2 V R= 127 (e+f) where: R = Radius of curve (metres) V = Speed of vehicles (km/h) e = Crossfall of road (metres per metre) f = Coefficient of side friction force developed between the vehicles tyres and road pavement. The value of e may represent the simple removal of adverse crossfall or include superelevation.

3.5 Where only small numbers of specialist vehicles are involved and the costs of improving the alignment are high, not all vehicles can expect to traverse a curve on a single lane road in a single manoeuvre and reversing may be necessary.

ADVERSE CROSSFALL
3.6 The normal crossfall on a road will result in a vehicles on the outside lane of a horizontal curve needing to develop high levels of frictional force to resist sliding; the amount of increase being dependent on speed, curve radius and crossfall. In order to achieve the necessary cornering stability, it is recommended that adverse crossfall is removed. The identification of speed and radius combinations at which this should occur is rather subjective as there is no evidence linking adverse crossfall to accident risk. A side friction factor of 0.07 has been taken as giving suitable minimum radii below which adverse crossfall should be removed. With a normal crossfall of 3 per cent, this value results in a minimum radii shown in Table 3.1. Values for unpaved roads are based on a 4 per cent crossfall which is the minimum crossfall that should be allowed before maintenance is carried out if effective cross-drainage is still to be provided.

3.2

The side friction factor may be considered to be the lateral force developed by the driver on a level road. The technical evidence indicates that lateral accelerations, and hence side friction factors, increase with reduced radii of curvature and increased speed. The range is considerable and values of "f' found from public road measurements have varied from just over 0.1 for high speed roads to over 0.5 on lower speed roads. The results of empirical studies have indicated 0.22 as a value of "f' above which passengers experience some discomfort. The much higher values found on low radius curves indicate that drivers and passengers have a much higher tolerance in these situations. The values of "f' chosen to calculate minimum radii requirements in this guide range from 0.15 to 0.33. A substantial reserve exists between these comfort and control related values, and those at which the vehicle would start to slide sideways. 3.3 In this guide, it is recommended that curves are designed such that it is necessary for vehicles travelling at the design speed to steer into a bend. 3.4 The minimum radii values shown in Table 1.2 were derived on the basis of sideways friction factors and superelevation. In some situations with minimal lateral clearances, sight distance will be the factor controlling minimum radii. Sight distances may be improved by increasing curve radius or sight distance across the inside of the curve.

TABLE 3.1: MINIMUM RADII OF CURVES BELOW WHICH ADVERSE CROSSFALL SHOULD BE REMOVED

* Values in the brackets are the design speeds in km/h with zero lateral accelerations for 3 per cent crossfall ie the speeds at which curve can be negotiated with hands off (approximately one speed design step lower). 3.7 The values shown in the table are approximate and cut-off levels should be varied to offer consistency to the driver. For example, two adjacent horizontal curves on a road link, one of which is marginally above the cutoff whilst the other is marginally below the minimum radii given, should be treated in a similar manner in the design.

11

F ig.3. 1 Superevelation design curves 3.11 Where transition curves are used (paras 3.14-19),
3.8 Removal and restoration of adverse crossfall should take place over similar distances to superelevation as described in the following Section. superelevation should be applied over the length of the transition curves. Otherwise it should be introduced such that two thirds are applied prior to the start of the circular curve. 3.12 For curves with radii above the minimum values, but below the values at which adverse crossfall should be eliminated, it is advisable to improve passenger comfort by introducing superelevation and reducing the sideways force. Intermediate values of superelevation are given in Figure 3.1. 3.13 On paved roads with unsealed shoulders, the shoulders should drain away from the paved area to avoid loose material being washed across the road.

SUPERELEVATION
3.9 For small radius curves and at higher speeds, the removal of adverse crossfall alone will be insufficient to reduce frictional needs to an acceptable level, and crossfall should be increased by the application of superelevation. A minimum radius is reached when the maximum acceptable frictional and superelevation derived forces have been developed. These minimum radii values are identified in Figure 3.1 for maximum levels of superelevation of 10 per cent. These relate to paved roads only. Although this percentage is rather arbitrary, it is widely considered to be a value above which drivers may find it difficult to remain centred in lane as they negotiate a bend. 3.10 On unpaved roads, the crossfall is designed to remove rainwater quickly and effectively, and will be dependent on local conditions and materials. Values of superelevation lower than the value of the crossfall will fail to drain the surface, whist higher values will be likely to result in erosion. On unpaved roads, the maximum superelevation will therefore be the elimination of adverse crossfall (see Table 3.1).

TRANSITION CURVES
3.14 The characteristic of a transition curve is that it has a constantly changing radius. Transition curves may be inserted between tangents and circular curves to reduce the abrupt introduction of the lateral acceleration. They may also be used to link straights or two circular curves.
3.15 In practice, drivers employ their own transition on entry to a circular curve and transition curves contribute to the comfort of the driver in only a limited number of

12

situations. However, they also provide convenient sections over which superelevation or pavement widening may be applied, and can improve the appearance of the road by avoiding sharp discontinuities in alignment at the beginning and end of circular curves. For large radius curves, the rate of change of lateral acceleration is small and transition curves are not normally required. 3.16 Several methods exist for the calculation of transition curves and any may be used in most situations. The rate of pavement rotation method has been adopted here. The rate of pavement rotation is defined as the change in crossfall divided by the time taken to travel along the length of transition at the design speed. The length of transition curve is derived from the formula: e.V Ls = 3.6n

OTHER CONSIDERATIONS
3.20 For small changes of direction, it is often desirable to use large radius curves. This improves the appearance of the road by removing rapid changes in edge profile. It also reduces the tendency for drivers to cut the comers of small radius curves. Providing the curve radii are sufficiently large, visibility should not be restricted enough to prevent safe overtaking. 3.21 The use of long curves of tight radii should be avoided where possible, as drivers at speeds other than the design speed will find it difficult to remain in lane. Curve widening reduces such problems. In such situations, it will usually be more important to provide adequate overtaking opportunities with longer straights and tighter curves, and to overcome terrain constraints, than to allow for detailed operational problems. 3.22 Abrupt changes in direction from successive curves should be avoided where possible by the inclusion of a tangent section in between. This will allow appropriate changes to be made in crossfall and superelevation. 3.23 Successive curves in the same direction should also be separated by an appropriate tangent, as drivers are unlikely to anticipate what may be an abrupt change in radial acceleration.

where Ls = Length of transition curve (metres) e = Superelevation of the curve (metres per metre) V= Design speed (km/h) n = Rate of pavement rotation (metres per metre per second)

3.17 The same values of rate of change of pavement rotation should be used to calculate the minimum length (Lc) over which adverse camber should be removed on a tangent section prior to the transition: e .V n Lc = 3.6n

Where Lc =Length of section over which adverse camber is removed en =Normal crossfall of the pavement (metres per metre).

3.18 The length of transition curve (Ls) is used to apply the superelevation, with the adverse camber removed on the preceding section of tangent (Lc). The change from normal cross-section to full superelevation at the start of the circular curve is achieved over the superelevation run-off distance which is the sum of Ls and Lc. 3.19 Several relationships are available to calculate the coordinates of a transition curve. The shift, ie the offset of the start of the circular curve from the line of the tangent, should be at least 0.25 metres for appearance purposes. The transition should be omitted if the shift is less than this value.

13

4. VERTICAL ALIGNMENT
COMPONENTS OF THE VERTICAL ALIGNMENT 4.1 The two major aspects of vertical alignment are vertical curvature, which is governed by sight distance and comfort criteria, and gradient which is related to vehicle performance and level of service 4.2 Vertical curves are required to provide smooth transitions between consecutive gradients and the simple parabola is recommended for these. The parabola provides a constant rate of change of curvature, and hence visibility, along its length and has the form: G.L y= 200 L x 2

where sight distance is less than the length of the vertical curve, and the second is where sight distance extends beyond the vertical curve. Consideration of the properties of the parabola results in the following relationships for minimum curve length to achieve the required sight distances: G.S For S<L: Lm=
2

200 (h1 + h2)

200 (h1 + h2) For S>L: Lm = 2S G where

where y = vertical distance from the tangent to the curve (metres) x = horizontal distance from the start of the vertical curve (metres) G = algebraic difference in gradients (%) L = length of vertical curve (metres)

Lm = minimum length of vertical crest curve (metres) S = required sight distance (metres) G = algebraic difference in gradients (%) h1 = driver eye height (metres) h2 = object height (metres)

CREST CURVES
4.3 The minimum lengths of crest curves have been designed to provide sufficient sight distance during daylight conditions. Longer lengths would be needed to meet the same visibility requirements at night on unlit roads. Even on a level road, low meeting beam headlight illumination may not even show up small objects at the design stopping sight distances. However, it is considered that these longer lengths of curve are not justified as high objects and vehicle tail lights will be illuminated at the required stopping sight distances on crest curves. Vehicles will be identified by the approaching illumination and drivers should be more alert at night and/or be travelling at reduced speed. 4.4 The greater sight distances required to provide safe overtaking opportunities are not easily provided on crest curves. If full overtaking sight distance cannot be obtained, the design should aim to reduce the length of crest curves to provide the minimum stopping sight distance, thus increasing overtaking opportunities on the gradients on either side of the curve. 4.5 Two conditions exist when considering minimum sight distance criteria on vertical curves. The first is

4.6 For S<L, the most common situation in practice, L = K.C where K is a constant for a given design speed (minimum safe stopping speed), eye and object heights. 4.7 Eye height (h1) has been taken as 1.05 metres, and object heights have been adopted of 0.2 metres above the road surface and to the road surface itself. The need to see the road surface is only applicable in particular circumstances such as a vertical curve on the approach to a ford or drift where a driver may have to stop because of the presence of surface water. 4.8 Two approaching vehicles on a single lane road require twice the distance in order to stop safely and avoid collision, and in this instance an object height of 1.05 metres has been used. The K-values relating change in gradient to minimum vertical curve length are given in Table 1.2 for the various object heights. 4.9 Charts of required lengths of vertical curves for safe stopping for an object on the road, safe overtaking and for meeting vehicles on a single lane, are shown in Figures 4.1, 4.2 and 4.3 respectively. Minimum values have been derived from considerations of appearance. 4.10 Sight distances have been based on the characteristics of car drivers as, although braking distances are greater with trucks, they will usually be travelling more slowly and the eye height of truck drivers is about 1.0 metre higher. Requirements are related to rates of deceleration available with an emergency stop. Skid resistance values are dependent

14

Algebraic difference in gradient (A)(%)

Fig.4.1 Length of crest vertical curves for safe stopping sight distance

15

Algebraic difference in gradient A (%)

Fig.4.2 Length of crest vertical curve for overtaking sight distance

16

Algebraic difference in gradient A (%)

Fig.4.3 Length of crest vertical curve for safe stopping sight distance for meeting vehicles
on tyre, road surface conditions and speed, and vary substantially. The values for available longitudinal friction in this guide are given in Table 1.2. A reaction time of 2.0 seconds has been assumed. Drivers will react more quickly when alert and in a situation where action is expected and, in practice, reaction times normally vary from about 0.5 to 1.7 seconds.
2

For S < L:

Lm =

G.S

200 (h1 + S.tan)

200 (h1 + S.tan) For S > L: Lm = 2S G where h1 = headlight height (metres) = angle of upward divergence of headlight beam (degrees) Appropriate values for h1 and are 0.6 metres and 1.0 degrees respectively.

SAG CURVES
4.11 It has been assumed that adequate sight distance will be available on sag curves in daylight. However, at night, visibility is limited by the distance illuminated by the headlamp beams, and minimum sag curve length for this condition is given as:

17

Algebraic difference in gradient A (%)

Fig.4.4 Length of sag vertical curve for adequate riding comfort

4.12 The use of these equations can lead to requirements for unrealistically long vertical curves as, especially at higher speeds, sight distances may be in excess of the effective range of the headlamp beam, particularly when low meeting beams are used. Thus, the only likely situation when the above equations should be considered for use is on the approaches to fords and drifts and other similar locations where flowing or standing water may be present on the road surface. Most of these structures occur on low speed road where headlamp illumination is more likely to reach the full sight distances. 4.13 It is recommended that, for most situations, sag curves are designed using the driver comfort criterion of vertical acceleration. The values used are given in Table 4.2 and the resulting curve length values are shown in Figure 4.4. with minimum length values for satisfactory appearance.

TABLE 4.2 : MINIMUM LEVELS OF ACCEPTABLE VERTICAL ACCELERATION

18

GRADIENT
4.14 Vehicle operations on gradients are complex and depend on a number of factors: severity and length of gradient; level and composition of traffic; the number of overtaking opportunities on the gradient and in its vicinity.

4.15 For very low levels of traffic flow with only a few four-wheel drive vehicles, the maximum traversable gradient is in excess of 20 per cent. Small commercial vehicles can usually negotiate an 18 per cent gradient. whilst two-wheel drive trucks can successfully tackle gradients of 15-16 per cent except when heavily laden. These performance considerations have formed the basic limiting criteria for gradient as shown in Table 1.1.

Gradient (%)

Fig.4.5 Estimated speed increases with climbing lanes

Note: 1) The above results are estimations based on simulation. Vehicle performance and driver characteristics will vary from country to country and the assumptions incorporated here should be considered as coarse approximations. 2) Climbing lanes on gradients of up to 100 metres in length were shown to have little effect. 3) Varying the percentage of heavy vehicles (HGV) from 20 percent to 40 percent has little effect on the mean speed reduction. 4) The above curves are based on directional flows of 200 vehicles per hour. For lower flows. the benefits of a climbing lane were small, although for higher flows of 400 to 600 vehicles per hour. the benefits were found to increase by about 25 percent on a 300 metre gradient, and by about 60 percent on a 600 metre long gradient. 5) The speed increases shown in the above Figure are values averaged over a 1.0 kilometre section of road which contains the gradient section.

19

4.19
4.16 Gradients of 10 per cent or over will usually need to be paved to enable sufficient traction to be achieved, as well as for pavement maintenance reasons. There will often be considerable non-motorised movement on the lower Design Class roads and, whilst pedestrian and animal movements are possible on very steep inclines, some laden animal drawn carts may find steep grades difficult to traverse because of a lack of grip. 4.17 As traffic flows increase, the economic disbenefits of more severe gradients, measured as increased vehicle operating and travel time costs, are more likely to result in economic justification for reducing the severity and/or length of a gradient. On the higher Design Classes of road, the lower maximum recommended gradients (Table 1.1) reflect the economics, as well as the need to avoid the build up of local congestion. However, separate economic assessment of alternatives to long or severe gradients should be undertaken where possible or necessary. An estimation of vehicle operating cost savings may be made from relationships such as those incorporated in the TRRL road investment model (micro-RTIM2), or the World Bank's highway design and maintenance standards model (HDM-III).

Benefits from the provision of a climbing lane accrue because faster vehicles are able to overtake more easily, resulting in shorter average journey times and reduced vehicle operating costs. Benefits will increase with increases in gradient, length of gradient, traffic flow, the proportion of trucks, and reductions in overtaking opportunities. The effect of a climbing lane in breaking up queues of vehicles held up by a slow moving truck will continue for some distance along the road. 4.20 The effects of a climbing lane on the mean operating speed of a traffic stream have been estimated with a simulation model and are given as Figure 4.5 for guidance. These mean speeds should be used with local values of travel time savings. appropriate vehicle operating costs savings, and the additional costs of construction, to estimate overall economic returns of the alternatives to enable the most cost-effective solution to be determined. With the generally low values of travel time found in developing countries and excluding accident considerations, climbing lanes are unlikely to be justified other than on a small proportion of Arterial roads with very high flows. In view of the uncertainties associated with simulation, local data should be used where available. 4.21 As climbing lanes will be used largely by trucks and buses, they must be a minimum of 3.0 metres in width. They must be clearly marked and, where possible, should end on level or downhill sections where speed differences between different classes of vehicle are lowest to allow safe and efficient merging manoeuvres.

CLIMBING LANES
4.18 A climbing lane may be introduced as a more cost effective alternative to reducing a gradient.

20

5. ECONOMICS AND SAFETY

Non-motorised traffic
ECONOMIC ASSESSMENT
5.1 All road schemes must be worthwhile economically. However, a road scheme with higher than necessary geometric standards may achieve a target rate of return without giving best value for money. In developing countries, finance is usually scarce and it is particularly important that minimum effective designs are used to enable the funds saved to be applied elsewhere. The standards recommended in this guide are intended to encourage the identification of such minimum effective designs. 5.2 In most developing countries, the economic benefits from road schemes are mainly derived from vehicle operating cost savings. Other benefits are savings in travel time and reduced accident rates. All three types of benefit are increased by reducing route length and, for higher flow roads, such savings may well outweigh the additional costs of straightening a tortuous alignment. Whilst this process is reflected in the design steps shown in Figure 1.2, care must be taken to ensure that a proper range of alternative alignments is considered to achieve the best economic return. 5.3 The choice of an appropriate unit value of travel time is often contentious and, in most situations, the extent and value of potential accident savings is difficult to define. A recommended approach to costing in both of these areas is given in Overseas Road Note 5 (TRRL Overseas Unit 1988). 5.7 This traffic should normally be segregated onto sealed shoulders of appropriate width. Clear delineation is essential and may be achieved by road markings, use of different coloured surfacing, surface texture, or kerb features. 5.8 Kerb features may include edge strips or intermittent placement of slightly raised blocks, sufficient to deter drivers from travelling over them at speed, but not so raised as to be likely to cause loss of control or damage to the vehicles. Such features should not be more than about 20mm high. They must be clearly marked and should only be introduced on roads of Design Class A and B with running widths of 6.5 metres. 5.9 Traffic on the approach to crossing facilities, or through villages where many crossing pedestrian movements are concentrated, may be slowed down by the use of road humps or other pavement features. Road humps should be designed as shown in Figure 5.1. Short sharp humps are not recommended as they can damage tyres and suspensions, and lead to loss of control. They must not be placed on high speed roads. Warning signs must be provided on the approaches to road humps and where crossing movements are concentrated. Proper maintenance of the profiles of road humps is essential. Alternative effective features include "rumble strips", which are short sections of road with a coarse surface texture.

SAFETY
5.4 The operating conditions on roads in developing countries are normally very different from those in developed countries. Principal areas of difference are the substantial variations in vehicle performance and condition, the often large amounts of non-motorised traffic, and low levels of training and control of road users. 5.5 Road accident rates in developing countries are high and result in substantial economic loss as well as pain, grief and suffering. However, in view of the uncertainties of accident prediction, it has not been possible to evaluate the specific effects of the geometric design parameters recommended in this guide. Therefore, accident rates must be monitored accurately to identify the need for specific remedial treatment, and to form a basis for future local amendments to the design procedure. 5.6 In general, designers should be aware of the need to consider safety, and should make use of opportunities which may arise at design or construction stages, and which may result in substantial benefits at little additional cost. The following factors should be considered when designing for safety.

Fig.5.1. Road hump cross-section and dimensions (Dimensions in mm)

5.10 Shoulder width must reflect the characteristics of the non-motorised traffic using it and additional width may be required where there are substantial flows of pedestrians, cycle rickshaws, bullock carts, etc. Segregation is recommended where such flows coincide with heavy flows of motorised vehicles, or where it may be achieved through low cost measures.

Driver safety
5.11
Occasionally a driver will lose control and swerve off the road. Design features should be such that the effects of such a manoeuvre will be minimised:

21

 Steep open side drains should be avoided as these will increase the likelihood of vehicles overturning; trees should not be planted immediately adjacent to the road Because of their high costs of installation and maintenance, guard rails should only be introduced at sites of known accident risk.

5.12 The detailed design of junctions and accesses is beyond the scope of this guide. However, they should he situated at locations where full safe stopping sight distances are available. 5.13 A check list of engineering design features that affect road safety is given in Figure 5.2.

Fig.5.2. Example of effect of engineering design on road safety

22

REFERENCES
BOYCE, AM et al, 1988. A review of geometric design and standards for rural roads in developing countries. Contractor Report 94. Crowthome: Transport and Road Research Laboratory. DEPARTMENT OF TRANSPORT, 1981. Highway link design. Departmental Standard TD9/81. London: Department of Transport. NAASRA, 1980. Interim guide to the geometric design of rural roads. Sydney: National Association of Australian State Road Authorities. TRRL OVERSEAS UNIT, 1988. A guide to road project appraisal. Overseas Road Note 5. Crowthorne: Transport and Road Research Laboratory.

APPENDIX A
GLOSSARY OF TERMS Access Road. The lowest level of road in the network
hierarchy with the function of linking traffic to and from rural areas, either direct to adjacent urban centres, or to the Collector road network; a feeder road or tertiary route (see Fig 1.1). Adverse crossfall. Crossfall on a horizontal curve that tilts away from the centre of the curve. Annual average daily traffic (ADT). The total annual traffic in both directions on a road link divided by 365. Arterial road. A main route connecting national or international centres: a primary route (see Fig 1.1). Capacity. The maximum practicable traffic flow in given circumstances. Carriageway. That part of the road constructed for use by moving traffic, including auxiliary lanes, climbing lanes and passing places (see Figure 2.1). Climbing lane. An auxiliary lane provided on an up gradient for use by slow moving vehicles and to facilitate overtaking. Coefficient of friction. The ratio of the frictional force on the vehicle, and the component of the weight of the vehicle perpendicular to the frictional force. Collector road. A road that has the function of linking traffic to and from rural areas, either direct to adjacent urban centres, or to the Arterial road network; a secondary route (see Fig 1.1). Crest. A peak formed by the junction of two gradients. Cross-section. A vertical sect ion of the road at right angles to the centre line. Crossfall. The difference in level measured traversely across the surface of the carriageway. Design Class. The classification of roads for geometric design purposes according to traffic and road function (see Table 1.1). Design Speed. The 85th percentile speed of vehicles on any particular section of road. Dual carriageway. A road having two separate carriageways for travel in opposite directions.

23

24 Edge strip. A flush border of stone, concrete or other material laid or formed at the edge of the carriageway to delineate the shoulder. Eye height. An assumed height of drivers' eyes above the surface of the carriageway used for the purpose of determining sight distances. Free speed. The speed at which a vehicle travels on uncongested, flat, straight, smooth and wide sections of road in the particular environment under consideration. (Road) function. The objective of providing a particular road link in terms of being "Arterial", "Collector" or "Access". Geometric element. An individual horizontal or vertical curve, transition curve, gradient, or straight section of road. Geometric (design) standards. Guidelines for limiting values of road alignment and cross-section design. Gradient. The rate of rise or fall on any length of road, with respect to the horizontal. Guard rail. A continuous barrier erected alongside a carriageway to minimise the consequences of vehicles running off the road. "Hands off" speed. The speed at which, for a particular combination of horizontal curvature and superelevation, a vehicle will follow the curve without any necessity to steer to the left or the right. Horizontal alignment. The direction and course of the centre line in plan. Horizontal curve/curvature. A curve/succession of curves, normally circular, in plan. K-value. The ratio of the minimum length of vertical crest curve in metres to the algebraic difference in percentage gradients adjoining the curve. Kerb. A border, flush or up-standing, of stone, concrete or other material laid or formed at the edge of a carriageway, shoulder or footway. Lane. A strip of carriageway intended to accommodate a single lane of moving vehicles, frequently defined by carriageway markings. Level (terrain). Flat or gently rolling terrain with largely unrestricted horizontal and vertical alignment; the road line crosses 0-10 five metre ground contours per kilometre (see para. 1.27). Mountainous (terrain). Terrain that is rugged and very hilly with substantial restrictions in both horizontal and vertical alignment; the road line crosses more than 25 five metre ground contours per kilometre (see para. 1.29). Network (hierarchy). The classification of roads according to Arterial, Collector and Access (see Fig. 1.1). Normal distribution. A symmetrical hell-shaped curve relating the probability of occurrence of an event to the range of events possible; for the mathmatical formulation of the normal distribution, any standard text on statistics should be consulted. Object height. An assumed height of a notional object on the surface of the carriageway used for the purpose of determining sight distance. Passenger car unit (pcu). A unit for converting the equivalence in terms of effect on capacity of different vehicle types in terms of one normal passenger car. Passing place. A local widening of a narrow carriageway to enable vehicles to pass or overtake each other. Pavement. The part of a road designed to withstand the weight or loading by traffic. Percentile. The percentage of the total below which the given number of values fall. Rate of pavement rotation. The ratio of the change in crossfall to the time taken to travel along the length of a transition curve when travelling at the design speed. Residual value. The value of a road which remains at the end of the economic evaluation period; normally taken as the difference in cost between rebuilding the road at the end of its life using the structure remaining from the initial project, and the building cost if the first project were not to take place. Right-of-way. The physical extent of the right of access that is granted in association with a road. Road hump. A physical obstruction, normally of semicircular profile, placed transversely on the surface of the carriageway for the purpose of reducing traffic speed. Rolling (terrain). Terrain with low hills introducing moderate levels of rise and fall with some restrictions on vertical alignment; the road line crosses 11-25 five metre ground contours per kilometre (see para. 1.28).

Rumble strip. A strip of coarse textured surfacing material placed on the surface of the carriageway for the purpose of altering drivers. Sag (curve). A depression formed by the junction of two gradients. Shoulder. That part of the road outside the carriageway, but at substantially the same level (see Figure 2.1). Sight distance. The distance at which' an object becomes visible to an observer, the height above the carriageway of observer and object being specified. Speed environment The speed below which 85 percent of vehicles are driving on the longer straights and large radius curves of a section of road where speed is not constrained by traffic or geometric elements. Superelevation. The inward tilt or transverse inclination given to the cross-section of a carriageway throughout the length of a horizontal curve to reduce the frictional requirements between the vehicles' tyres and the road surface. Superelevation run-off. The length of road over which superelevation is reduced from its maximum value to zero. Taper. The transition length between a passing place, auxiliary lane or climbing lane and the standard carriageway. Transition curve. A curve in which the radius changes continuously along its length, used for the purpose of connecting a straight with a circular curve, or two circular curves of different radii. Vertical alignment. The direction and course of the centre line in profile. Vertical curve/curvature. A curve/succession of curves, normally parabolic, in profile.

APPENDIX B
ESTIMATION OF VEHICLE SPEED
B.1 In order to estimate the 85th percentile approach speed of a particular class of vehicle on the road under investigation, it is necessary to determine the "free speed" of that class of vehicle. This is defined as the speed at which vehicles of different classes are observed to travel on uncongested, flat, straight, smooth and wide sections of road in the environment under investigation. These speeds have been found to be affected by the general layout of the roads in the area and overall characteristics of driver behaviour which are not, at this stage, amenable to modelling. For example, vehicles operated in predominantly hilly regions are found to have considerable lower free speeds than those in rolling terrain. Similarly, the free speeds of vehicles operated on roads where straying animals are common, or that tend to have slow-moving animal-drawn carts, are lower than those of vehicles operated on roads which are free of such obstacles B.2 The following tables enable estimates of speed to be made over any section of road. Where possible, free speeds should he determined by field measurement on a flat, straight, smooth and wide section of road. Where it is not possible to determine a local value, the mid value for the range given in the tables should be used. Starting with the estimate of 85th percentile free speed, reductions from this are successively made to take account of road rise, fall, horizontal curvature, width, etc, for the road section approaching the geometric element being designed. In most cases, the design vehicle will be a car, and Table B 1 should be used. B.3 These tables give only crude estimates of the 85th percentile speed and SHOULD BE USED

WITH CAUTION

25

TABLE B1:CAR SPEEDS

26

TABLE B2 : TRUCK AND BUS SPEEDS

INCREASE IN SPEED DUE TO POWER TO WEIGHT RATIO

27

APPENDIX C
PHASING OF THE VERTICAL AND HORIZONTAL ALIGNEMNT DEFECTS IN THE ALIGNMENT DUE TO MISPHASING
C.1 Phasing of the vertical and horizontal curves of a road implies their coordination so that the line of the road appears to a driver to flow smoothly, avoiding the creation of hazards and visual defects. It is particularly important in the design of high-speed roads on which a driver must be able to anticipate changes in both horizontal and vertical alignment well within his safe stopping distance. It becomes more important with small radius curves than with large. C.2 Defects may arise if an alignment is misphased. Defects may be purely visual and do no more than present the driver with an aesthetically displeasing impression of the road. Such defects often occur on sag curves. When these defects are severe, they may create a psychological obstacle and cause some drivers to reduce speed unnecessarily. In other cases, the defects may endanger the safety of the user by concealing hazards on the road ahead. A sharp bend hidden by a crest curve is an example of this kind of defect.

28 separation between the curves

The vertical curve overlaps one end of the horizontal curve

C.7 If a vertical crest curve overlaps either the beginning or the end of a horizontal curve, a driver's perception of the change of direction at the start of the horizontal curve may be delayed because his sight distance is reduced by the vertical curve. This defect is hazardous. The position of the crest is important because vehicles tend to increase speed on the down gradient following the highest point of the crest curve, and the danger due to an unexpected change of direction is consequently greater. If a vertical sag curve overlaps a horizontal curve, an apparent kink may be produced. This visual defect is illustrated in Fig C1(c). C8 The defect may be corrected in both cases by completely separating the curves. If this is uneconomic, the curves must be adjusted so that they are coincident at both ends, if the horizontal curve is of short radius, or they need be coincident at only one end, if the horizontal curve is of longer radius.

Both ends of the vertical curve lie on the horizontal curve

C.9 If both ends of a crest curve lie on a sharp horizontal curve, the radius of the horizontal curve may appear to the driver to decrease abruptly over the length of the crest curve. If the vertical curve is a sag curve, the radius of the horizontal curve may appear to increase. An example of such a visual defect is illustrated in Fig C1(d). The corrective action is to make both ends of the curves coincident, or to separate them.

TYPES OF MISPHASING AND CORRESPONDING CORRECTIVE ACTION

C.3 When the horizontal and vertical curves are adequately separated or when they are coincident, no phasing problem occurs and no corrective action is required. Where defects occur, phasing may be achieved either by separating the curves or by adjusting their lengths so that vertical and horizontal curves begin at a common chainage and end at a common chainage. In some cases, depending on the curvature, it is sufficient if only one end of each of the curves is at a common chainage. C.4 Cases of misphasing fall into four types. These are described below together with the necessary corrective action for each type.

The vertical curve overlaps both ends of the horizontal curve

C.10 If a vertical crest curve overlaps both ends of a sharp horizontal curve, a hazard may be created because a vehicle has to undergo a sudden change of direction during passage of the vertical curve while sight distance is reduced. C.11 The corrective action is to make both ends of the curves coincident. If the horizontal curve is less sharp, then a hazard may still be created if the crest occurs off the horizontal curve because the change of direction at the beginning of the horizontal curve will then occur on a downgrade (for traffic in one direction) where vehicles may be increasing speed. C.12 The corrective action is to make the curves coincident at one end so as to bring the crest on to the horizontal curve.

Insufficient separation between the curves.

C.5 If there is insufficient separation between the ends of the horizontal and vertical curves, a false reverse curve may appear on the outside edge-line at the beginning of the horizontal curve, or on the inside edgeline at the end of the horizontal curve. This is a visual defect. It is illustrated in Figs Cl(a) and Cl(b). C.6 Corrective action consists of increasing the

C.13 No action is necessary if a vertical curve that has no crest is combined with a gentle horizontal curve. C.14 If the vertical curve is a sag curve, an illusory crest or dip, depending on the hand of the horizontal curve, will appear in the road alignment. C.15 The corrective action is to make both ends of the curves coincident or to separate them.

THE ECONOMIC PENALTY DUE TO PHASING

C.16 The phasing of vertical curves restricts their movement and fitting to the ground so that the designer is prevented from obtaining the lowest cost design. Therefore, phasing is usually bought at the cost of extra earthworks and the designer must decide at what point it becomes uneconomic. He will normally accept curves that have to be phased for reasons of safety. In cases when the advantage due to phasing is aesthetic, the designer will have to balance the costs of trial alignments against their elegance.

Fig C.1. Examples of some combinations of horizontal and vertical curves

ISSN 0951-8797

OS-E

29

A guide to geometric design

ORN 6

Transport and Road Research Laboratory Overseas Unit

Department of Transport Overseas Development Administration

Overseas Road Note 7- Volume 1

A guide to bridge inspection and data systems for district engineers

Overseas Unit Transport and Road Research Laboratory Crowthorne Berkshire United Kingdom 1988

ACKNOWLEDGEMENTS

This Note was drafted by Mr J.D. Parry and Mr D.M.Brooks of the Overseas Unit, T.R.R.L. It is based on an original text produced by Rendel, Palmer and Tritton, Consulting Engineers and is published by permission of the Director, Transport and Road Research Laboratory

First Published 1988.

OVERSEAS ROAD NOTES

Overseas Road Notes are prepared principally for road and road transport authorities in countries receiving technical assistance from the British Government. A limited number of copies are available to other organisations and to individuals with an interest in roads overseas, and may be obtained from: Transport and Road Research Laboratory Crowthorne, Berkshire, RG} 1 6AU, United Kingdom Crown Copyright 1988 Limited extracts from the text may be produced provided the source is acknowledged. For more extensive reproduction, please write to: Head of Overseas Unit, Transport and Road Research Laboratory ISSN 0951-8787

CONTENTS 1. 1.3 1.8 2. 2.2 2.4 2.8 2.9 2.10 2.14 2.20 2.21 2.22 3. 3.2 3.4 3.7 3.9 3.12 3.18 3.20 3 23 3.24 3.25 3.26 3.28 3.30 3.40 4. INTRODUCTION The purpose and uses of bridge records Data gathering and recording BRIDGE RECORDS Types of record The Bridge Inventory Drawings and calculations Inspection reports Maintenance records Computers: The Inventory (bridge record cards) Inspection reports and maintenance records Computer field recording BRIDGE INSPECTION The bridge inspector The bridge inspector's handbook The role of the engineer The role of the inspector Selecting inspectors Training inspectors Inspection frequency Types of inspection: Informal inspections General inspections Major inspections Special inspections Organising the inspection Major or unusual bridges REFERENCES Page 1 1 1 2 2 2 2 3 3 3 3 3 4 4 4 4 4 5 5 5 6 6 6 6 6 6 7 7 8

APPENDICES Appendix A - PREPARING AN INVENTORY Appendix B - SAFETY AND EQUIPMENT Appendix C - BLANK INSPECTION REPORT FORM

9 15 16

1 INTRODUCTION
1.1 The purpose of this note is to provide a district engineer with guidance on the establishment and operation of an effective bridge and culvert record system. 1.2 The two main elements of this programme are: - the bridge data and recording methods. - bridge inspection.

effective, even when it is not yet complete. 1.9 Chapter two of this note describes permanent and changing bridge records and recording systems. Chapter three describes the inspection of bridges, both where engineering staff levels are adequate and the alternative found in many countries, where there is a shortage of qualified bridge engineers. Under the latter circumstance it is recommended that less qualified employees should be trained to carry out routine inspections, using the pro-forma provided in Appendix C. 1.10 The companion volume to this note, the Bridge Inspector's Handbook, (TRRL Overseas Unit 1988), has been compiled specifically for the instruction of these inspectors, and to be used for reference by them during the inspection.

THE PURPOSE AND USES OF BRIDGE RECORDS

1.3 Effective management of bridges requires the regular collection and recording of accurate bridge condition data. These data, together with design and construction records, are used to: prepare maintenance and replacement programmes. identify design weaknesses. prepare itineraries for unusual vehicles. maintain a current road inventory.

1.4 A district engineer may be required to prepare a bridge replacement or upgrading programme, taking into account industrial and agricultural development projects planned for the near or midterm future. To do this, accurate data are required on the current condition of each bridge and the rate of deterioration. 1.5 Design weaknesses can be very costly. Information from district engineers to the authority's bridge designers, detailing faults or material failures, etc is the only way of preventing the continuation of design faults in new bridges 1.6 Current bridge data are also required for the preparation of suitable routes for unusually high, wide, or heavy loads. There is pressure in some countries to increase the size or weight of vehicles generally permitted on the highway, or on certain routes. The effect of such changes in legislation can only be predicted using accurate, current bridge data 1.7 As part of the road inventory, bridge data are important to local and national government planners, police, armed forces, emergency services, transport operators and private developers of natural resources such as timber and minerals.

DATA GATHERING AND RECORDING

1.8 A complete and regularly updated record of all bridge data on computer, with appropriate programmes for different types of data retrieval may be a long-term aim, but this is not always suitable or desirable. A simple card based system can be very 1

2. BRIDGE RECORDS
2.1 A record system must suit the specific requirements of the authority it is intended to serve. For this reason the system outlined here should be considered as a source of ideas, rather than as a set of rules. It is recognised that the existing records may not be complete and it may be impossible for the district engineer, because of the limited resources available to him, to assemble an ideal operational system within a short period of time. Appendix A is a guide intended to assist in establishing a new inventory, or modifying an existing one.

referred to as unchanging, the bridge record cards will require updating if: major maintenance is carried out which modifies the structure, when drawings must also be modified to show changes. an analysis for the passage of an abnormal vehicle is carried out. following an inspection, it is decided that a bridge has deteriorated to a point where its capacity must be reduced. a bridge is widened or strengthened, when drawings must also be modified and the bridge record card may need renewing. a new or replacement bridge is built, when new drawings and a new bridge record card must be produced.

TYPES OF RECORD
2.2 There are four primary elements in the bridge system: record

Bridge inventory

Drawings and calculations Inspection reports

Maintenance records

One record card per bridge containing all basic information about the bridge. Construction and as-built drawings together with design calculations. All inspection reports are retained to record the changes in each bridge over time. Estimates, drawings of repairs, cost records and details of routine maintenance, such as painting etc.

2.7 It must be remembered that some of the data on the bridge record card are copied onto the road inventory If the bridge record card is updated, the matching data on the road inventory should be updated at the district office and at headquarters.

DRAWINGS AND CALCULATIONS

2.8 The authority's bridge department usually stores bridge drawings and calculations at headquarters. When they are needed, copies are sent to the district offices. Unfortunately, the original drawings and calculations of many older bridges may no longer exist. It should also be borne in mind that the reliability of drawings and calculations is variable. The different types of drawings, and their potential shortcomings are as follows: Standard drawings. Where a bridge is built to a standard' design it is very likely that the structure as-built will vary considerably from the drawings. Unrecorded alterations on site are likely in order to accommodate site conditions. Drawings known to be original design drawings but not marked 'as-built'. Some changes during construction are possible. 'As-built' drawings. These will show the bridge as originally built, but may not show modifications since completion. 'As-built' drawings with all changes since completion shown and dated. These are the most reliable.

2.3 In addition to these primary elements, it is appropriate to include some of the bridge data on the road inventory as described in Overseas Road Note I (TRRL Overseas Unit 1987). The type of details to be included on the road inventory are. Chainage along the road (kilometres) Bridge name. Bridge number. Over/under (name of obstacles, eg stream, road crossed). Load restriction. Width restriction. Height restriction. Total bridge length. Number of spans. Type of superstructure. Alternative crossing (name, number, location, type, capacity), where available

THE BRIDGE INVENTORY

2.4 This is the basis of the record system and frequently takes the form of a series of cards, each of which records the data for a particular bridge which does not change over time. The originals of the cards are held at the headquarters bridge department, and each district office has copies of the bridge cards for the area which it covers. 2.5 Most district engineering offices will have an established format for the bridge record card. They will need only to ensure that they cover adequately all the aspects mentioned here. Those setting up a new system will find useful notes on the preparation of a bridge inventory and the sample record card, in Appendix A. 2.6 Although the data on the bridge inventory have been

Drawings based purely on a site inspection. (Where no original drawings exist). These should be dimensionally accurate, but are likely to be incomplete Reinforcement details, for example, will almost certainly be missing.

INSPECTION REPORTS
2.9 A recommended form of inspection report is provided in Appendix C. When a report has been completed in the field, it should be used to provide a record of defects over a period of time.

2.16 In planning such a bridge record system, it is necessary to decide which bridge data are to be included. Possible alternatives are: The inventory only (bridge record cards) The inventory and inspection reports. The inventory, inspection reports and maintenance records. 2.17 The ultimate file sizes and software required should be estimated initially, to enable decisions concerning appropriate development and or equipment to be made. 2.18 Two principal systems exist for computer equipment: A central computer with links to district offices. A microcomputer at headquarters and in each district office. 2.19 A central computer is the best option in terms of general access to one central file. It resolves the problem of ensuring that all users are dealing with current and consistent data. Microcomputers, however, may be cheaper, and are less vulnerable to unreliable electricity supplies, telecommunications problems, and hardware failure.

MAINTENANCE RECORDS
2.10 In most maintenance organisations, bridge maintenance records are kept using a simple chronological filing system with one file per bridge. This is perfectly adequate, as working through the file will tell the history of all work carried out on the bridge. 2.11 The maintenance record should include routine correspondence and, for each repair, should include: Estimates of work proposed. Details of work carried out. Date of completion of repair. Supervisors' quality control reports. Actual cost of repairs. 2.12 Where significant repairs are carried out, the drawings and calculations used for the repair work should be copied to the bridges department records and both sets of bridge record cards (headquarters and district) amended to show the repair. 2.13 Some authorities keep bridge maintenance records on files which cover more than one bridge, and perhaps all the bridges on a long section of road. This is Rot recommended, since it makes it difficult to follow the history of any one bridge. Although the one-file-per-bridge system involves more files, mistakes are less likely to occur.

THE INVENTORY (BRIDGE RECORD CARDS)

2.20 This is the simplest and possibly the most useful element of the bridge records to record on computer. An inventory data base has a wide variety of potential users with widely different requirements. Some examples of queries received from users might be: 'List all bridges with load limits less than lOT.' 'List all bridges on the N23 road which are too narrow for a 4.5 metre wide load.' 'List all bridges in the National Park.' 'List all bridges under the control of the '11am' maintenance authority.' List all bridges built by 'Barton Construction' between 1970 and 1976.' 'List all bridges in an area bounded by the following map references.

COMPUTERS
2.14 A bridge inventory may contain a large amount of data, and a variety of organisations may need information derived from it. Computers are a powerful tool for assisting in the handling of data. Proprietary database systems exist for most modern computers. Such a system could form the basis of a bridge record system but would still require additional software development. It must be stressed that all software development is both expensive and difficult. 2.15 The costs and benefits of implementing a computer based system should be evaluated before its introduction. Proprietary systems, if available, are likely to provide the cheapest solution, but the assistance and advice of professional consultants should always be sought at a very early stage.

INSPECTION REPORTS AND MAINTENANCE RECORDS

2.21 If inspection and maintenance data are computerised, it will still be necessary to keep copies of inspection and maintenance reports. Some of the data on these reports cannot readily be computerised (sketch maps, references etc.). These systems have a narrower application than the basic bridge record system, as the data are concerned mostly with maintenance. The system could provide the basis for a series of regularly updated reports, which would be a valuable management tool for the district engineer. Reports could include: 3 A

Regular summaries of urgent work highlighted by inspections. Summaries, defect by defect, which may highlight incorrect maintenance techniques, inappropriate materials, etc. A maintenance diary of work needed and work carried out, listing for example, bridges inspected this month, bridges to be inspected next month, etc. Work schedules for different gangs, provided either on an area by area basis, or a work type by work type basis.

BRIDGE INSPECTION

3.1 In some districts there will be few bridges and die district engineer may be able to carry out the inspection work himself, or he may have other qualified staff to do the inspection on his behalf. In this case, the organisational aspects covered in this chapter will be of interest, but not those parts referring to unqualified inspectors. However, the inspection report form presented in Appendix C is a comprehensive list of items to check, and can be used by all inspectors, whatever their experience.

COMPUTER FIELD RECORDING

2.22 Microcomputer-based loggers are not recommended for bridge inspection. When an inventory is very well established and all inspectors are familiar with its operation, it is possible that such a system could be considered, provided suitable training is available and equipment can be maintained. Even then, such a system could not fully replace the manual one recommended here.

THE BRIDGE INSPECTOR

3.2 Where there are many bridges and few bridge engineers, the district engineer has the choice of either: making infrequent inspections himself, or entrusting routine general inspections to others, and concentrating his own time on larger difficult bridges, and where die routine reports suggest that there are problems.

3.3 Staff suitable for routine bridge inspection are likely to be already in the employ of the district engineer. They may be junior engineers, technicians, or road foremen. Both older, experienced road men and younger, better qualified technicians have performed well in field trials of the procedures advocated here.

THE BRIDGE INSPECTOR'S HANDBOOK

3.4 Volume two of this note, the Bridge Inspector's Handbook, has been compiled specifically for inspectors who are not engineers, and who may not have English as their first language. It has been written in simple terms to teach the basic information needed to understand the task, and to guide the inspector through his work. 3.5 Part [1] of the handbook gives instruction on bridges. It can be used for formal, classroom instruction or private reading. 3.6 Part [2] guides the inspector through the report form presented in Appendix C of this note, item by item, explaining how relevant measurements or observations are made. it can be carried and used for reference, if required, during an inspection

THE ROLE OF THE ENGINEER

3.7 The purpose of the general inspection is to provide the engineer with detailed information about the condition of each bridge to enable him to: - assess maintenance requirements. - determine if a more detailed inspection is required.

3.8 -

He should therefore: ensure that each inspector fully understands his duties and responsibilities. ensure that each inspector has access to all necessary equipment, including safety equipment. ensure that prior to an inspection, the inspector is given the necessary documents. (see paragraph 3.30). occasionally carry out inspections with each inspector; this will help ensure consistent marking by inspectors. read each inspection report carefully so that the significance and possible inter-dependence of isolated defects can be assessed. For example, the inspector may have noted cracking in the road approaches, and elsewhere he may have noted damage to the abutment ballast wall. These maybe unconnected faults, but together they may indicate forward movement of the abutment, the damage to the ballast wall having been caused by thrust against the deck. be prepared to seek expert advice, particularly if the report indicates that the bridge may have potentially serious defects. District engineers will rarely have sufficient knowledge or experience to assess me importance of all defects reported ensure that the records are updated. determine what maintenance is required, categorise priorities, and prepare work schedules.

SELECTING INSPECTORS
3.12 In selecting suitable personnel for training as bridge inspectors, me district engineer should look for the following qualities:

RELIABILITY
3.13 The engineer must be able to depend on his inspectors to carry out a task according to instructions given, and to inform the engineer promptly ofany unusual circumstances.

TECHNICAL SKILLS
3.14 Although inspectors may not necessarily have any formal technical training, they should possess practical experience of construction. They must be able to observe and measure accurately and to draw simple sketches of any deficiencies found.

LANGUAGE SKILLS
3.15 Inspectors must be able to read the simple technical English, or other language, used on the inspection report; they must also be able to write clearly in the same language.

HEALTH
3.16 Inspectors should be physically fit, so they are able to inspect difficult places without taking unnecessary risks; it is advisable to check that the inspector has good eyesight, as he will be required to detect small defects in conditions of varying light and shade.

THE ROLE OF THE INSPECTOR

3.9 The inspector must record as clearly as possible any defects found at a bridge so that the district engineer may decide on the appropriate action 3.10 The inspector should not, in general, have to take decisions beyond giving his view of the severity and extent of problems he finds, if he is not a qualified engineer. Leaving any significant decisions to the inspector could lead to serious mistakes and places an unfair responsibility on him. 3.11 The inspector should be given specific responsibilities according to the discretion of the engineer. If the inspector finds a defect which he believes is so serious that the bridge may collapse, he should be empowered to close the bridge immediately. The inspector may of course be wrong, but closing a bridge unnecessarily is better than allowing it to collapse and possibly cause injury or loss of life. The inspector may be empowered to order or undertake specific maintenance tasks. Typically these tasks could include clearance of blocked drains or movement joints; minor repairs to parapets; repairs or replacement of signs; removal of debris in the river; and repairs to damaged gabions, rip-rap or stone pitching, etc. The defects should still be recorded on the report form, and noted as rectified.

ATTITUDE
3.17 Good inspectors are methodical in their work and must be willing to inspect items thoroughly and critically; they must also be continually aware of safety so that they do not take risks and do not endanger the safety of others.

TRAINING INSPECTORS
3.18 Training of inspectors is crucial to the successful operation of a bridge inspection system and should be tailored to the ability and experience of the trainees. Training should, ideally, be carried out initially by a skilled instructor from the authority's training school. The Bridge Inspector's Handbook can be used as a textbook for formal classroom instruction of inspectors. 3.19 The next step in training an inspector is for the district engineer or an instructor to go to several different bridges with the trainee inspector, or a group of trainee inspectors, and to go through the inspection step by step. When completing the inspection report form, an inspector must make decisions about How bad?" and How much?", and on-site training periods are needed to teach the inspector how the engineer would answer these questions and to develop his note-taking and sketching skills. As the inspector will probably be dealing with several different types of bridge, he should inspect at least one of each type with an engineer before working on his own.

INSPECTION FREQUENCY
3.20 The district engineer is responsible for establishing the frequency of inspection of bridges under his management, although his highway authority may issue guidance. No fixed rules can be given, but if the bridge inspection is carried out by a road maintenance foreman, the engineer should consider the benefits of coordinating the inspection of roads and bridges. 3.21 How often a particular bridge should be inspected depends on a number of factors. These include: age, type of construction, traffic volume and type, structural condition, climate, and availability of resources for carrying out inspections. For example, an old timber bridge now carrying heavy traffic and known to be in a poor condition should be inspected much more frequently than a new short span concrete slab bridge. In an area known to suffer extremely heavy seasonal rainfall, it may be sensible to carry out an inspection both before and towards the end of each wet season, even if only those parts of the bridge likely to be affected by flood are inspected. 3.22 Limited human resources are likely to be a serious constraint on how often bridges can be inspected. However, an appropriate frequency can be based on the use of different types of inspection, as described in paragraphs 3.23 to 3.29.

MAJOR INSPECTIONS
3.26 Major inspections should also be carried out on a regular basis, although less frequently than the general inspections, to provide more detailed reports on the structures than the general inspections. For example, at a major inspection all bolts on a steel structure might be checked for tightness, whereas at a general inspection, the engineer may ask the inspector to check bolts only at selected joints. If possible, the district engineer should arrange for a qualified engineer to visit the bridge during a major inspection, or go himself, to ensure that the inspector's assessment of problems is the same as his own. Parts of Appendix C may also be used for major inspections. 3.27 The time interval between major inspections for a particular bridge should be established initially by an engineer, based on the age, condition and construction materials employed. From the rate of deterioration which he subsequently observes, an optimum time interval can then be determined. If there are a number of bridges of one type and age, a common time interval for the group can be used.

SPECIAL INSPECTIONS
3.28 Special inspections may be needed either when a bridge is at risk, or when there is a problem which requires further investigation. Additional risk may derive from the bridge being in poor condition, or from a series of unusual loading events. The frequency of special inspections can vary from daily to six-monthly, and the inspection report form can be used in all or in part. Some examples of when this type of special inspection may be required are: - A bridge in poor condition, but considered safe for a relatively short time, is due for reconstruction in 2 or 3years time. - A low capacity bridge on a minor road which is normally adequate, but has to carry diverted heavy traffic when a bridge on the main road is closed - An old steel bridge which is believed to be at risk from fatigue due to increases in traffic volume and vehicle weight, in this case, the inspector concentrates on the fatigue risk areas. - A bridge which has been damaged by a major flood, but immediate repair is prevented by a continuing high water level; inspection might need to be daily in this case. Low cost, temporary structures.

TYPES OF INSPECTION
3.23 Inspections can be divided into four basic types: Informal General Major Special

INFORMAL INSPECTIONS
3.24 All personnel involved with bridge maintenance (particularly the inspectors) should be encouraged to keep their eyes open whilst travelling around, and to report anything unusual about any bridge. Telling inspectors and engineers to stop and have a general look when crossing a bridge' is the best way to obtain progress reports on potentially serious defects, such as a river changing its path and undermining an abutment in the interval between routine inspections.

GENERAL INSPECTIONS
3.25 This is the type of inspection that can be accomplished by a less experienced inspector using the pro forma in Appendix C. General inspections should be carried out on a routine bases at least once a year, preferable at a time of low water to permit inspection of the substructures, the river bed and any bed protection. In areas where there is a very pronounced wet and dry season it is recommended that, as well as the inspection during the dry season, an inspection for flood damage is carried out at the end of the wet season.

3.29 Sometimes, during a general or major inspection, a problem is discovered whose cause is not immediately obvious. It may then be necessary to carry out a further special inspection in order to investigate the cause of the problem. Often this will involve inspection by experts, and specialised materials sampling and testing. In this case, a special report.

will be prepared, not using the standard inspection report form. Examples of situations when this type of special inspection is required are: Paint on a repainted steel structure is showing signs of premature deterioration. 'Pop-outs' and 'map-cracking' on a concrete structure. Cracking of prestressed concrete beams.

3.34 The identification and construction details enable initial identification of the correct bridge. By checking each of the construction details at every inspection, the inspector reduces the risk of confusion between similar bridges. This risk cannot be overstated. It is very easy, even for experienced bridge engineers, to be confused between two similar bridges. 3.35 Re-reading the construction details will help the inspector understand the structural principles of the bridge: this is particularly true of 'foundation type' and 'movement'. Such an understanding will greatly assist him in properly completing the inspection report. 3.36 The sketch on page one of the form is for two purposes:

ORGANISING THE INSPECTION

3.30 The district engineer should supply the inspector with the following documents prior to undertaking an inspection: Appropriate parts of a blank: inspection report form (Appendix C). A copy of the previous inspection if any. A copy of the bridge record card. A list of maintenance work that has been completed since the last inspection. 3.31 The district engineer uses the inspection report form to: instruct the inspector to inspect a particular bridge. transmit any special instructions to the inspector. receive and record inspection data. receive any urgent notes from the inspector. 3.32 When the inspection of a bridge is due, the district engineer, or a member of his staff, compiles sufficient copies of the appropriate pages of the form. The form page numbers and the required number of each page are recorded on the bridge record card. Any sections of a page which are not required should be crossed out. 3.33 The following sections of the first two pages should be filled in before the form is given to the Inspector. Page 1 The identification and location details, abstracted from the bridge record card. Any special instructions to the inspector. Page 2 The construction details, abstracted from the bridge record card. Details of services abstracted from the bridge record card. Details of signs abstracted from the highway signs inventory, or the bridge record card. It enables the inspector to mark points of particular interest, eg an area of scour. It helps to establish a reference system if the inspector records adjacent town names. 3.37 It is vital that a clear reference system is used in reporting, if the district engineer is to understand which part of the bridge is referred to in each part of the report. This point must be strongly emphasised to each inspector. 3.38 The inspector should be encouraged to make additional notes and sketches or take photographs. These should be referenced in the final column. Very brief notes may be written in this column. 3.39 The district engineer should review the inspection report as soon as possible after submission by the inspector. He should ensure that he understands any notes or sketches and that the form has been properly completed. Whenever a potentially serious problem has been reported, the district engineer should instigate the appropriate action without undue delay and inform the inspector that he has done so. Inspectors' enthusiasm and commitment will be greatly improved if a quick response results from his efforts.

MAJOR OR UNUSUAL BRIDGES

3.40 Major bridges need special consideration for the following reasons: They represent a large capital investment. The collapse of a major bridge would cause an unacceptably long disruption to traffic since the design and reconstruction of a major bridge is a long process. The inspection of a major bridge or a bridge of unusual construction, eg moving bridges, is beyond the technical

capability of an unqualified inspector. Such structures must be inspected by an engineer with sufficient relevant experience. 3.41 Major or unusual bridges are outside the scope of this Note. Inventory systems must, of course, cover bridges of all types, but a major or unusual bridge will need a special version of the bridge record card and the inspection report. The guidelines in this Note on bridge record cards and inspection reports may assist the engineer in preparing such special versions.

4.

REFERENCES

Bridge Inspector's Handbook. TRRL OVERSEAS UNIT, 1988. Transport and Road Research Laboratory, Crowthorne, Berkshire, RG45 6AU, United Kingdom. Maintenance management for district engineers. TRRL OVERSEAS UNIT, 1987. Overseas Road Note 1. Transport and Road Research Laboratory, Crowthorne, Berkshire, RG45 6AU, United Kingdom. Maintenance techniques for district engineers. TRRL OVERSEAS UNIT, 1985. Overseas Road Note 2. Crowthorne, Berkshire, RG45 6AU, United Kingdom

APPENDIX A PREPARING AN INVENTORY

A.1 It is strongly recommended that, prior to introducing an inventory, discussions take place with all potentially interested parties. As well as the maintenance staff, these should include the highway planning department and may include such bodies as government departments responsible for industrial and agricultural developments, the police, the armed forces, road haulage groups, etc. By soliciting the views of all potentially concerned parties, the inventory can be set up so that the maximum use may be made of the data obtained. Unless engineers recognise its usefulness, the necessary updating may not take place, and within a few years the inventory will be inaccurate.

sub-divisions will be determined by the road density and the total number of structures. Minor road networks often include numerous intersections. It is, therefore, not so critical that bridge reference numbers are sequential along any given road and the addition of a new bridge will not disrupt the system. A.6 Identification in the field will be greatly eased if the bridge number is painted conspicuously on each bridge.

BRIDGE NAME
A.7 Often the names used for rivers and bridges on maps differ from the name used by the villagers in the area. A decision needs to be made as to whether 'map names' (easy to work with in the office) or 'local names' (easy to work with in the field) are to be used, or both.

THE BRIDGE RECORD CARD

A.2 A suggested format for the bridge record card is show in Figures Al and A2. The format shown here may require modifications to meet the specific needs of the authority. If changes are made, care should be taken to ensure that both the bridge record card and inspection report are changed so that they still match, and that suitable notes for guidance are prepared and issued to all inspectors. A.3 In deciding a suitable size for the bridge record cards, available storage facilities must be considered. The card shown in Figures Al and A2 has been designed as A3 size, because it is large enough to carry the required data, sketches and photographs, and a copy will fold readily into the more common A4 size files. Notes on the individual entries on the bridge record card are given in paragraphs A.4 to A.41.

MAINTENANCE AUTHORITY
A.8 This should be the name of the local department responsible for any remedial or maintenance work required.

ROAD
A.9 For minor roads. this should be specified by the names of the towns, villages, or road junctions nearest to the bridge. Major roads should be subdivided into convenient sections (between major towns). The entry on the card should be the road number, followed by the names of the towns or junctions at the ends of the section. e.g. N3 - (MALAN TO GOROKAN section).

KILOMETRE (OR CHAINAGE)

A.10 The distance to the bridge from the start of the section. Note that this will be the same as the bridge number for the first section of each road, but not for subsequent sections. A.11 This entry should identify the name or number of the map on which the bridge occurs. In some countries maps of several different scales may be available in complete series. In others, the choice may be limited. 1/100,000 or 1/50,000 are suitable scales for this purpose.

BRIDGE NUMBER
A.4 Some authorities use elaborate bridge numbering systems, which in themselves indicate information about the bridge, services carried, responsible authorities, etc. This is unnecessary except where there is a large number of bridges and an advanced computerised reporting system. A.5 The following system is recommended. For major roads or roads identified as specific routes: The road number followed by the distance from the road origin (to the nearest tenth of a kilometre) e.g. N3/31.5. A simple sequential system could be used, but if a new bridge is added, such a system becomes difficult to modify. For minor roads which cannot easily be identified by route: Subdivide the area of the authority into small units, which may be local council or village areas or simply an area defined by map references. Use as the bridge number the designated area number followed by a simple sequential number e.g. A15/83.The most appropriate size of the selected

REFERENCE
A.12 This should be the map reference of the bridge location using the map referred to in A. 11.

ROAD CLASSIFICATION
A.13 Most countries have an established nomenclature for their road network and this should be strictly adhered to.

OVER/UNDER
A.14 This identifies both the obstacle crossed and whether the bridge carries the road, or crosses the road. e.g. OVER River Teem or UNDER Tara-Kennan railway. A.15 Road over road bridges are a special case. For the bridge record card, the bridge should be assigned to the road carried.

WIDTH
A.18 This is the minimum width of the carriageway between the kerbs or parapets. For bridges over roads, the width of the road crossed should also be recorded.

HEIGHT RESTRICTION
A.19 This item is only applicable to bridges over roads and braced truss or braced arch bridges. In both cases, care must be taken that the height recorded is for the full carriageway width and, for road over road bridges, to which road the restriction applies. When it is considered useful to record a height over a reduced width of carriageway, this should be clearly marked on the elevation or cross section drawing and noted under 'height restriction'.

LENGTH
A.16 The total length of the bridge between abutment movement joints.

NAVIGATION RESTRICTIONS
A.17 For navigable rivers, a bridge may form an obstruction to shipping, both in terms of width of waterway and clearance to the underside of the bridge. The authority responsible for navigation should be consulted for advice regarding the water level from which heights should be measured. If the restriction involves lengthy descriptions it may be necessary to continue on the NOTES section on the back of the card.

LOAD RESTRICTION
A.20 Where a bridge has a capacity below the legal limit this should be stated here.

ABNORMAL VEHICLE
A.21 Many bridges have a reserve of capacity beyond the normal maximum legal vehicle weight. Such capacity is

Fig. A3 Height restrictions

12

usually only determined when a specific heavy vehicle has need to cross the bridge. After such calculations, the special vehicle and the payload found to be acceptable should be noted on the card or reference made to a specific set of calculations.

MOVEMENT
A.29 This entry should state where the superstructure is 'fixed' to the substructure, and where it is 'free' to move. The fixed and free bearings should be noted on the sketches on the back of the card.

CONSTRUCTION DETAILS
A.22 This section of the card is intended only to record general details of the type of construction, not full section sizes, etc. The details recorded help to identify the bridge and to explain how the bridge works.

SERVICES CARRIED
A.30 Note here all services (electricity, water, sewer, etc.) known to be carried by the bridge.

SPAN(S)
A.23 Note here the individual span length (or lengths) measured between bearing centres. The total of the span lengths may be different from the bridge length.

ROAD SIGNS
A.31 List the number and type of road signs associated with the bridge.

RUNNING SURFACE
A.24 Concrete, mastic asphalt, galvanised steel plate, etc, are typical entries. If there is a waterproof membrane on the bridge, it should be noted here.

LOCATION PLAN
A.32 This is not intended to give detail, but simply to assist in identifying the bridge. In some cases, it may be useful to show associated works such as river training, which cannot, at a reasonable scale, be included in the plans on the back of the card. All road signs relating to the bridge should be shown.

SUPERSTRUCTURE DESIGNED BY
A.25 This item is intended to identify the type of span construction. Typical entries would be: Prestressed concrete beams, concrete deck. Steel beams with concrete deck (note if composite action was assumed in the design). Bailey bridge with timber deck, etc. A.33 The name of the organisation which designed the bridge (e.g. Bridge dept., Wigan & Partners consultants), if known.

CONSTRUCTED BY
A.34 The name of the department or contractor who built the bridge, if known.

PIER(S)
A.26 The form of the pier, for example: R.C. columns with cross head. R.C. wall. Steel trestle etc.

YEAR OF COMPLETION
A.35 Both the date of original construction and the date of any major modification should be recorded.

PHOTOGRAPHS
A.36 Two photographs of the bridge should be fixed to the card, one view taken from the river and the other from the road. A.37 If there is no sign containing the bridge name adjacent to the bridge, it will help identification if a board bearing the bridge number is included in the photograph taken from the road.

ABUTMENT
A.27 The form of the abutment, for example: Masonry. R.C. wall. R.C. bank seat. Gabions.

FOUNDATION TYPE
A.28 Typical entries would be: Piled. Spread footing. Caisson.

DRAWING NUMBERS AND CALCULATION REFERENCES

A.38 Record all drawing numbers, noting if they are 'as-built' drawings or not, and references to original calculations, plus any later calculations such as for heavy load movement

NOTES
A.39 Any special points of interest, modifications, problems, etc. should be recorded here. 13

INSPECTION REPORT PAGE NUMBERS

A.40 The inspection report form covers a wide variety of bridges. Only a limited number of its pages would be used on any bridge. To assist in preparing blank inspection forms before an inspection, the numbers of the pages to be used should be noted

but will certainly be cheaper. It may also eventually produce a better inventory. As ideas on systems can develop as work proceeds Method b) is likely to be the most practical for developing countries unless special resources and finance are made available. A.46 The following sequence of operations is recommended in creating an inventory: 1) Determine from all available sources the number and location of all bridges believed to exist. Such sources include maintenance records. Information from maintenance personnel or others familiar with each road, drawings, maps and old inventories. Set up a road numbering system, if one does not already exist, and a bridge numbering system (see paragraphs A.4 to A.6 Produce a preliminary card for each bridge containing all information which can be derived from existing sources Carry out the initial general inspection and complete the preliminary bridge record cards, as follows: Verify the data on the preliminary bridge record card. Fill in the missing data from observations on and around the bridge. Photograph the bridge and make sketches of river training works. services, etc. Complete the inspection report form, paying particular attention to any deterioration which may influence the load carrying capacity of the structure.

PLAN, ELEVATION AND CROSS SECTION

A.41 Space is provided on the bridge record card for plan, elevation and cross-section sketches. Where there are thought to be useful (See Fig A2) Such drawings should not attempt to provide full details of the bridge, useful general details and principal dimensions are sufficient. Such drawings can usefully indicate The bridge skew, if any. Bridge bearings. whether fixed' or 'free'. The location of services and their supports

2)

3)

4)

CREATING NEW BRIDGE RECORDS

A.42 The preparation of the bridge record cards involves handling a lot of data from a number of different sources. It is very easy, even for experienced bridge engineers, to confuse one bridge with another, care must be taken to provide cross checks to ensure that the correct data are being recorded for each bridge This is particularly true when many similar bridges exist along a section of road. It is suggested that a system be utilised which involves always cross checking a few easily identifiable items of data such as: Chainage along the road. Note this can be prone to error due to an inaccurate zero setting at the survey start and because of vehicle odometer errors. Span length(s) Skew angle. Number of beams or some other very obvious construction feature. A.43 There are two ways of creating a complete set of bridge record cards: (a) Form a survey team (or teams) to complete the work in one continuous programme (b) Prepare the cards over a longer period. intermittently, with field inspections being fitted in with other necessary journeys. A.44 Method a) has much to commend it, as the completed set of cards will be ready for use more quickly. Greater control of data is also possible as the work wilt proceed continuously. The disadvantage is that special teams need to be established and the extra cost may be high A.45 Method b) will take longer and may be harder to control, 14 a) b) c) d)

5) In the office: a) Resolve anomalies arising from differences in old and observed data b) Carry out a load capacity evaluation for all bridges, except those where the capacity is clearly stated on the drawings and the condition survey revealed no serious deterioration. c) Make out the final bridge record cards.

THE INITIAL GENERAL INSPECTION

A.47 Whereas data on the bridge record cards do not normally change with time, the data on the inspection report are changeable. However the two sets of data interrelate, so it is sensible to carry out the first bridge inspections at the same time as the bridge record cards are being finalised This is the initial general inspection. A.48 If at all possible, this particular bridge inspection report should be completed by a bridge engineer working road by road, at the same time as the record cards are being finalised (step 4(d), para A 46). The reasons for this are: a) Only an experienced bridge engineer can determine if the

nominal load capacity must be reduced because the bridge is in poor condition; thus a bridge engineer has to inspect each bridge before the load capacity can be finalised. Completing the initial general inspection report will not take much additional time b) The inspector who carries out subsequent inspections may not be experienced in bridge inspections. If the initial general inspection report has been completed by an engineer, it will help the inspector considerably, and uniformity of approach will be more likely. A.49 To provide useful training, potential inspectors should accompany the bridge engineer on these initial general inspections whenever possible.

APPENDIX B
SAFETY AND EQUIPMENT
B.I The safety of bridge inspectors and road users during the inspection is the responsibility of the district engineer. It is important that he trains the inspectors in safety procedures during the formal instruction period, and by example when lie accompanies them on the occasional joint inspection B.2 Possibilities of injury arise mainly from: Traffic on the deck or road approaches A fall from the superstructure to the river or road below. Insects, snakes, crocodiles, etc below the deck B.3 All these dangers can be at least partly alleviated by the presence of an assistant, such as a driver. He can also help when measurements are required. B.4 In addition, the inspector should: carry and use standard road signs and traffic cones Instruction in their use is detailed in Overseas Road Note 2. (TRRL Overseas Unit 1985). use a rope and harness whenever a fall could cause injury. beware of dangerous insects and reptiles. not enter an enclosed space without ensuring that fresh air can enter. and without keeping in constant contact with his companion

B.5 The inspector should be encouraged to exercise caution at all times When he finds that parts of a bridge are not safely accessible without special equipment, such as a boat, he must understand that he should not take risks to complete the report form and he should not guess at the answers required The responsible action is to record which parts of the report form were not possible to complete and return on another occasion with the required equipment to complete the inspection. B.6 The inspector will assess the importance of the work he does by the interest paid by the engineer and the quality of the equipment issued to him The equipment should be inspected regularly, and kept up to complement. B.7 The inspector should keep his equipment in good order and take care that potentially dangerous items, particularly chemical preservatives and paints. are kept away from children and are not discarded or spilled where they can harm animals or fish.

INSPECTION EQUIPMENT
B.8 The following equipment should be provided for all inspections: First-aid box

15

Clipboard for inspection report forms protected from rain. Waterproof marker, suitable for concrete, steel or timber. Flashlight (torch). Straight edge (at least 2m long). String line. Measuring Tapes (3m or 5m and 30m long). Bush knife or similar tool to clear vegetation. Small shovel or trowel. Wire brush. Small paintbrush and paint, for painting areas damaged during the inspection. Hammer (350 - 450g). Ranging rod or other similar long pole. Set of spanners. Crack measuring gauge. B.9 The following additional equipment will be required when inspecting timber bridges Square-ended spike. Hand drill and 5mm diameter drill bits. 5mm wooden plugs. Saw. Wood preservative. B.10 The following safety equipment should always be used. Traffic warning signs and road cones. High visibility, fluorescent waistcoat. Safety helmet. B.11 In certain circumstances it may also be necessary to provide: About 20m of light rope. A safety harness. Ladders. A small boat to examine river piers and abutments.

APPENDIX C BLANK INSPECTION REPORT FORM

C.1 This appendix contains a complete blank inspection report form, listing all the items to be checked during a bridge inspection. For each bridge, only those pages corresponding to the construction type and materials used will be required. Some pages will be required more than once, e.g. for abutments and multi-span bridges.

16

NOTES

Guidance on the use of this handbook is given in the complementary publication: Overseas Road Note 7, (Volume 1) A Guide to Bridge Inspection and Data Systems for District Engineers.

Published by The Transport and Road Research Laboratory.

OVERSEAS ROAD NOTES

Overseas Road Notes are prepared principally for road and transport authorities in countries receiving technical assistance from the British Government. A limited number of copies are available to other organisations and to individuals with an interest in roads overseas. Enquiries should be made to: Transport and Road Research Laboratory Crowthorne, Berkshire RG11 6AU England.

ACKNOWLEDGEMENTS

This handbook was drafted by Mr J. D Parry of the Overseas Unit, TRRL. It is based on an original text by Rendel, Palmer and Tritton, Consulting Engineers and is published by permission of the Director, Transport and Road Research Laboratory.

CROWN COPYRIGHT 1988

Limited extracts from the text may be reproduced, provided the source is acknowledged. For more extensive reproduction please write to: The Director, Transport and Road Research Laboratory.

FOREWORD In many developing countries there is a shortage of trained bridge engineers. Where this is so, other personnel must be used for routine bridge inspections, or the bridges are neglected and deteriorate. Using this handbook, a person with experience, but little formal technical training, such as a road maintenance supervisor, should be able to carry out routine bridge inspections on the majority of bridges. This handbook utilises an inspector's report form, which also provides a check list of items to be inspected. For each item, detailed but simple, advice is given on what to look for. Large or unusual bridges, such as suspension bridges, are beyond the scope of this handbook, and should therefore be inspected by an engineer. Both technical content and language have been simplified to make this handbook usable by people whose knowledge of engineering, and/or English, is not highly developed. It must be stressed that this handbook aims only to support and guide a suitable person in how to inspect a bridge on behalf of the engineer. The final responsibility must lie with the engineer.

CONTENTS PART 1 - ABOUT BRIDGES To the bridge inspect Introduction TYPES AND PARTS OF BRIDGES General Superstructure Abutments, wing walls and retaining walls Piers MOVEMENT General Bearings and joints Page No [1] - 1 [1] 5 [1] - 6 [1] - 6 [1] - 7 [1] - 13 [1] - 18 [1] - 20 [1] - 20 [1] - 21

DAMAGE AND PROTECTION AGAINST DAMAGE [1] - 26 lmpact damage [1] - 26 River damage [1] - 28 Protection against scour [1] - 33 Slope protection [1] - 33 Bed protection [1] - 36 River training works [1] - 37 Other natural causes of damage [1] 38 Water [1] 38 Debris, dirt and vegetation [1] 40 Earthquakes [1] - 42 Landslides [1] 42 BRIDGE MATERIALS Concrete Steel Masonry Timber Gabions CULVERTS [1] - 43 [1] - 43 [1] - 56 [1] - 67 [1] - 72 [1] - 80 [1] - 83

PART 2 THE INSPECTION Page No [2] 1 [2] 3 [2] 6 [2] 8 [2] 15 [2] 18 [2] 27 [2] 32 [2] 34 [2] 41 [2] 44 [2] 59 [2] 64 [2] 70 [2] 75 [2] 82 [2] 96 [2] 105 [2] 110 [2] 119

Introduction The report form : Page 1 The report form : Page 2 The report form : Page 3 onwards Road approaches and deck Bridge surfaces and footpaths Parapets, railings and guard rails Expansion joints The river Superstructure Main beam, girders, trusses and bracing Underside of deck Bearings Masonry archers Bailey bridges Abutments, wing walls, and retaining walls Embankments and fill in front of abutments Bed protection Piers Culverts

APPENDICES

A. WORD LIST A1 B. INSPECTION EQUIPMENT LIST B1 C. SAFETY DURING INSPECTIONS C1 D. EXAMPLE OF A COMPLETED REPORT FORM

TO THE BRIDGE INSPECTOR Often, a bridge inspector is a qualified engineer. But sometimes, there are not enough engineers to do this work, and other people become bridge Inspectors. Inspecting a bridge is a very responsible job. It is not an easy job, and there are many things to learn before you can inspect a bridge properly. This handbook explains the things you need to know to inspect a bridge
BEFORE YOU INSPECTA BRIDGE, YOU MUST READ PART [1] OF THIS HANDBOOK AND UNDERSTAND EVERYTHING IN IT.

Part [1] of this handbook tells you the names of different parts of bridges. It explains the different kinds of bridges and tells you how bridges can be damaged. It also gives you basic information about concrete, steel, masonry and timber, as they are used in bridges. Part [2] of this handbook helps you to inspect a bridge and fill in an inspection report form. You must read Part [1], 50 that you will understand the engineering terms used in Part [2]. Before you inspect a bridge, you should read Part [2]. When you go to inspect a bridge, take this book with you, as it will tell you what things must be checked and how to fill in the inspection report form. If you have some experience of bridges, the information in the book will not be difficult for you.

[1] - 1

If you speak English in your job, you will already know many of the engineering words which are used. If not, you will need to learn some new words. To help you there is a word list in Appendix A. Appendix A is at the back of this handbook on pages A 1 to A - 9. It is a list of the engineering words and their meanings. The first time each of these words is used in the handbook, it is printed in CAPITALS, so you will know you can find it in the word list. A small handbook like this one cannot teach you everything about bridges. Sometimes you must ask an engineer to explain things to you. The engineer you are working for should be willing to help you with any problems Appendix B is a list of equipment you will need when you are inspecting bridges. You may think of other useful things to add to this list. You should keep all these things in good condition and check them after each inspection, so that you will have a full set of tools and equipment when you arrive at the next bridge. Appendix C is about safety and is very important. Please read this carefully. Appendix D is an example of a completed inspection report form. When you inspect a bridge, your report form will have ticks in different places. The number in the first column, on the left side of the page, tells you which page to look at in Part [2] of this book, if you need help with that part of the inspection

[1] - 2

This handbook does not tell you about suspension bridges, cable-stayed bridges or movable bridges, and it does not tell you about very large bridges. These bridges must be inspected by a bridge engineer. A bridge inspector must know many things and he must also have many good qualities There are 3 qualities which are very important 1. A bridge inspector must be a careful worker You must always look at a bridge carefully and fill in the inspection report form carefully It is better to be slow and careful. 2. A bridge inspector must be a safe worker. You must be in good health and fit, so you can climb over parts of a bridge safely. Safety is very important and you must never take unnecessary risks. 3. A bridge inspector must be a responsible person A good inspector can help prevent road accidents and keep the roads open to traffic.

[1] - 3

[1] - 4

PART (1) - ABOUT BRIDGES

INTRODUCTION Bridges can be the weak links in a road network. They must be very well maintained in order to keep the roads open to traffic. Most of the bridges you will inspect carry a road over a river. Culverts are included too, because they are like small bridges and if a culvert collapses the road may have to be closed. As an inspector you will be looking for damage which has already happened, or will soon happen. You will be looking at the bridge, the material it is made from, and also the river The river is important because it can damage the bridge. To understand about damage to bridges, you must understand something about the materials used in bridges The materials most commonly used for bridge construction are concrete, steel, masonry and timber You will find notes on these materials at the end of Part [1].

[1] - 5

TYPES AND PARTS OF BRIDGES GENERAL The simplest bridge is a single SPAN. Look at this picture of a single span bridge over a river. Study it carefully and learn any new words

SINGLE SPAN BRIDGE

We call the part of the bridge which carries the road over the river the SUPERSTRUCTURE. The superstructure sits on the ABUTMENTS. The BEARINGS are between the superstructure and the abutments. Along both edges of the superstructure there is a PARAPET The parapet stops people and light vehicles from falling off the bridge. As well as supporting the superstructure, the abutments also retain (hold back) the APPROACH EMBANKMENTS. The RETAINING WALLS next to the abutments also retain the approach embankments. The APPROACH ROAD runs along the top of the approach

[1] - 6

SUPERSTRUCTURE If we could cut through the superstructure it may look like this:

[1] - 7

[1] - 8

If a bridge is long, the superstructure needs to have more than a support at each end. These extra supports are called PIERS. In this book abutments and piers are called SUBSTRUCTURES This picture shows a 3-span bridge over a river:

3 SPAN BRIDGE CONTINUOUS CONSTRUCTION

In the drawing above, the superstructure is one continuous piece. Only one set of bearings is needed at each pier. The superstructure of a multi-span bridge (that is a bridge with 2, 3, or more spans) may be constructed of separate sections as shown below:

MULTI- SPAN SIMPLY SUPPORTED BRIDGE

In this case, the superstructure needs 2 sets of bearings at each pier, to allow each section to expand and contract.

[1] - 9

The bridge shown above is called a TRUSS BRIDGE The trusses are the 2 main GIRDERS which are made from a number of steel beams joined together The joints are usually fixed with rivets or bolts. The trusses are connected by CROSS-GIRDERS and BRACING. The cross-girders and deck bracing connect the bottom BOOMS of the trusses, and bracing connects the top booms above the roadway The road is supported on STRINGERS which span between the cross-girders. This truss bridge is called a 'through truss' because the traffic goes through the bridge at the same level as the trusses.

[1] - 10

The bridge in the picture below has no top bracing. It is a half-through truss bridge.

On some bridges the trusses are underneath the roadway. These bridges are called deck type truss bridges. One is shown on page [1] - 78.

[1] - 11

MASONRY ARCH bridges are made from brick or stone masonry. The picture below shows a masonry arch and names the important parts of the bridge:

The arch BARREL is supported on FOUNDATIONS. The FILL which carries the roadway over the arch is contained by SPANDREL walls. The spandrel walls usually continue above the roadway as parapets.

[1] - 12

ABUTMENTS, WING WALLS AND RETAINING WALLS The weight of the superstructure and the abutments is carried by the abutment foundations Abutments can have 3 types of foundations. They can have a SPREAD FOUNDATION directly on the soil or rock; they can have PILES; or they can have CAISSONS. These pictures show the three types of foundations:

[1] - 13

Sometimes abutments sit high up on the river bank. These abutments are called BANK SEAT abutments. They may Sit on piles as in the picture below, or on spread foundations as shown in the picture opposite.

BANK SEAT ABUTMENT ON PILES

[1] - 14

BANK SEAT ABUTMENT ON SPREAD FOUNDATION

Because their foundations are above river level, bank seat abutments can easily be damaged by the river SCOURING the bank beneath them, or by rain water running down the bank and ERODING the bank away. You will read more about this on pages [1] - 28 to [1] 37.

[1] - 15

The superstructure rests on bearings on the abutment BEARING SHELF.

This picture shows a steel girder superstructure resting on an abutment:

[1] - 16

The part of the abutment which holds back the approach embankment above the bearing shelf is called the BALLAST WALL. WING WALLS are attached to the abutment. They retain (hold in) the approach embankment. This picture shows an abutment with wing walls.

Sometimes retaining walls are used to hold back the embankment. They are separate from the abutment, and look like this:

[1] - 17

PIERS Piers, like abutments, can rest on spread foundations, piles, or caissons:

CONCRETE PIER ON SPREAD FOUNDATION

CONCRETE PIER ON PILES

[1] - 18

CONCRETE PIER ON CAISSON

On some bridges the piers are made by carrying piles or caissons from below ground up to the underside of the superstructure. The photograph on page [1] - 51 shows this. On other bridges, such as the one in the photograph on page [2] - 110, Concrete piers are supported on piles

[1] - 19

MOVEMENT GENERAL When the air temperature changes, the superstructure changes its length a little. It is usual that a bridge 50 metres long will change its length by about 25 mm (about 1 inch).

In areas with big temperature differences, such as deserts and mountain areas, the change in length (or expansion) will be bigger.

If the superstructure is fixed to both abutments, it will damage them when it changes length. On all big bridges the superstructure rests on bearings. The bearings carry the weight of the superstructure and allow it to move a little:

[1] - 20

BEARINGS AND JOINTS If the superstructure is free to move at both ends, the whole superstructure could fall off its bearings To avoid this, usually one end of a span is fixed to the abutment or pier The other end is free to move:

Fixed bearings usually have a pin or bolt fixing the beam to the support through the bearing. On concrete bridges it is not possible to see the pins used in the fixed bearings, so they cannot be checked. On steel girder bridges the heads of the bolts or pins can usually be inspected. This is one type of fixed bearing, called a rocker bearing:

[1] 21

It is important that you know which bearings are fixed, and which are free. To find this out, look on the bridge record card before you go to the bridge There are many different types of free bearings, but movement is usually allowed in one of three ways: change of shape of rubber bearings sliding between special surfaces rolling on steel rollers

Many modern bridges have rubber bearings. Some rubber bearings have steel PLATES inside. If you can see the steel inside the rubber, then the fault is bad:

[1] - 22

Metal bearings can be made of two plates, usually one steel plate and one bronze plate. Modern bridges may have a slippery material called PTFE between the sliding plates. This PTFE looks like wax and can be easily damaged because it is not very hard. Some big bridges have roller bearings. One type of roller bearing is shown here:

[1] - 23

At the road surface. between the DECK and the ballast wall, there will be an expansion joint There are many different types of expansion joints. The simplest joint is made by using steel angles in the end of the deck, and in the top of the abutment ballast wall

Sometimes a rubber WATER BAR is used to try to stop water and dirt from going through the expansion gap.

[1] - 24

So far the bridge spans used as examples have been either simply supported (supported at two points only), or continuous (supported at more than two points). In each case the bearings and expansion joints are at the piers or abutments. Another common form of construction is this:

In this type of construction, there are bearings at the abutments and piers, and also there are bearings where the SUSPENDED SPAN sits on the CANTILEVERS. There are joints in the road at both abutments and at the ends of the suspended span Expansion joints in the bridge decks are often a problem, so you will have to look at them carefully during your inspections. On some bridges (usually short span bridges) the bituminous surfacing is carried right over the joint so that you cannot see the joint itself. These are called buried joints When you are inspecting bridges with buried joints, because you cannot see the joint you must look for signs that the joint is not working. You are told how to do this in part [2] of this book.

[1] - 25

DAMAGE AND PROTECTION AGAINST DAMAGE Bridges are damaged by vehicles, by the river, or by other natural causes like CORROSION and rotting.

IMPACT DAMAGE Vehicle impact causes a lot of damage to bridges. When a vehicle hits a parapet, the parapet will be damaged If a heavy vehicle hits a parapet, then the main structure of the bridge may also be damaged. Through-truss and through-girder bridges can be seriously damaged by vehicle impact.

[1] - 26

If a vehicle Is too tall, it may damage the top bracing of a through-truss Sometimes a bridge carries a road over another road. If a vehicle leaves the road below, it may hit the bridge pier, or if the vehicle is too tall, it may hit the superstructure.

When the bridge is over a river, large logs or boats may hit the pier or the superstructure If an impact damages any part of the bridge, except the parapet, then an engineer must inspect the bridge. He will know how important the damage is.

[1] - 27

RIVER DAMAGE When a river flows very fast it picks up material from the river BED or banks and washes it away. This is called SCOUR. Sometimes scour causes large holes in river beds or washes large sections of the bank away. Many bridges have been destroyed by scour. These 3 pictures show different types of damage caused by scour:

Scour of the river bank at the side of an abutment.

[1] - 28

Scour of a river bank in front of a bankseat abutment.

[1] - 29

Scour causing a retaining wall to move forward.

[1] - 30

Rivers can easily damage or destroy bridges. Usually bridges are damaged when the river is too big to go through the waterway under the bridge, or when the river changes its path. There are 3 reasons why a river may not be able to go through the waterway of a bridge. 1. A river can grow and become too big for the waterway. 2. The waterway under the bridge can be blocked by parts of old bridges, trees, fences and other debris. 3. The waterway under the bridge was not made big enough.

If there is a flood which is too big for the waterway under the bridge, the river may do 3 things: 1. Wash away the bridge. 2. Wash away the road embankment and the road, and go round the bridge 3. Wash away the fill in front of the abutments, and scour big holes in the river bed. If the waterway is too small, another bridge or some culverts may be needed to carry the extra floodwater.

[1] - 31

Rivers can change their path slowly or very quickly. Change of path can, after a time, cause damage to a bridge. For example, this picture shows how a new island can form around a large log and other debris:

A new island UPSTREAM of the bridge can make a river change its path and scour around an abutment, or under a pier. To stop the river changing its path, training works are used. These are described on page [1] - 37

[1] - 32

PROTECTION AGAINST SCOUR If the river is causing scour, then the road embankment, the abutments and the piers can be protected with slope protection and bed protection. Slope Protection

There are 4 common ways to protect a slope: Piled walls Stone pitching Gabions Rip-rap

PILED WALLS are made from timber or steel. These timber or steel piles are hammered into the ground at the bottom of the slope. Sometimes the river scours in front of the piled walls, and the walls move forward. This is not serious if the walls are still protecting the slope. [1] - 33

STONE PITCHING is stones set in mortar on the slope. Stone pitching is a good way to protect a slope from water running down it. Stone pitching can be damaged by scour at the base of the embankment or by scour or erosion at the edge

This photo shows the problem

[1] - 34

GABIONS (or RENO MATTRESSES) are wire baskets filled with stones. They are often used as slope protection. Because they can change shape and settle a lot without any damage, gabions are good for protecting slopes

RIP-RAP is large rocks or blocks of concrete placed against the slope. For rip-rap to work, the rocks must be too heavy for the river to wash them away. Stones found in the river close to the bridge cannot be used as rip-rap. They will not be heavy enough. If the rip-rap is being washed away, it must be replaced with larger rocks.

This new rip-rap slope protection has been placed in a scour hole in front of a bank Seat abutment.

[1] - 35

Bed Protection Sometimes, to protect the bridge from scour, part or all of the bed of the river at the bridge is covered with stone pitching, concrete or gabions (or reno mattresses) When all of the river bed under a bridge is covered by bed protection, the bed protection is called an INVERT. For fast flowing rivers, it is sometimes necessary to carry this bed protection a long way DOWNSTREAM of the bridge or culvert. Bed protection carried downstream of a bridge is called an APRON.

This stone pitched invert has been damaged by river scour

[1] - 36

RIVER TRAINING WORKS River training works are used to keep a river on its path. There are 4 common ways of river training: SHEET PILED WALLS of steel or timber (picture on page [1] -33) embankments protected by rip-rap or gabions groynes made from steel or timber piles or gabions trees protected by gabions

GROYNES are lines of piles or gabions which are placed part way across the river from the river bank. If trees can be grown, their roots help to keep the bank in place. Gabions are sometimes used to protect the young trees. River training works can be made of many different types of material and with different methods of construction. If you are not sure about the methods used in your district ask the engineer.

[1] - 37

OTHER NATURAL CAUSES OF DAMAGE Water As well as the damage caused by water to the river bed, water damages bridges in many ways, for example: corrosion of steel in steel bridges (see page [1] - 58). corrosion of REINFORCEMENT or PRESTRESSING in concrete bridges (see page [1] - 51). DECAY of timber (see page [1] - 77). damage to masonry or stone pitching by water running down it (see page [2] - 111). abutments and retaining walls can be pushed forward if drains are blocked and water is held behind the wall. water running down embankments can wash the fill away. This sort of erosion can be a very serious problem on some types of abutment. Good DRAINAGE on the approaches to the bridge, and drainage and waterproofing on the bridge, help to avoid these problems.

[1] - 38

Some of the drainage is easy to see, such as drains on the deck The next picture shows how badly placed deck drains can cause corrosion on a steel girder;

You cannot see drainage behind an abutment, but it is important. If water builds up behind an abutment, the pressure of the water may push the abutment forward. Usually a bridge abutment has a drainage layer behind the abutment wall, and WEEPHOLES through the front face of the abutment wall. Weep-holes are shown on page [1] - 17. Sometimes there is also a drain along the bottom of the drainage layer which collects the water and carries it out to the side of the abutment.

[1] - 39

Debris, dirt and vegetation When dirt or DEBRIS collect on a structure they hold water, and the dampness causes deterioration. If large plants grow in these pockets of dirt, their roots can damage the structure:

[1] - 40

If debris carried by the river collects against a pier or abutment, it can block the waterway The river may then wash out the road embankment. If large amounts of debris collect against a pier or the bridge superstructure, the force of the water on the debris can badly damage the bridge. For many bridges, only part of the waterway is covered by the river for most of the time. Vegetation grows in the areas the river does not use. Grasses and light vegetation (such as rice) are good, as they hold the soil in place. Trees, large bushes and large plants (such as bananas) are bad, as they will block the waterway In desert areas, wind-blown sand can block culverts and bridges over dry river beds. When the river flows, the sand may be washed away but the bridge or culvert might be badly damaged.

[1] - 41

Earthquakes

Bridges are sometimes damaged by earthquakes There are 2 common types of damage caused by earthquakes. foundation failure causing movements of the abutment or piers. the superstructure moving off its supports. Some bridges in earthquake zones have the superstructure held down to stop it falling off.

Landslides

Another danger to bridges is LANDSLI DES. If there is a landslide which blocks the river upstream from a bridge, the water will build up behind it. After some time, the river may break through and wash the bridge away This does not often happen, but it is always helpful to talk to local people when you inspect a bridge. They can tell you about changes in the river that you might not see from the bridge site.

[1] - 42

BRIDGE MATERIALS

The main materials used in bridges are concrete, steel, masonry and timber. This section describes the problems to look for in each material. In the steel notes, you will also see notes about paint and GALVANISING. At the end, you will find notes on gabions.

CONCRETE There are 5 main problems with concrete on bridges: cracking of the concrete SPALLING of concrete corrosion of reinforcement or prestressing steel poor quality concrete chemical attack

These notes on concrete start with some general information about concrete in bridges. Then they tell you about each of these problems

[1] - 43

General Concrete for bridges is usually made from stones, sand, cement and water. As new concrete dries, it shrinks. This shrinkage is very small.

Concrete is strong in COMPRESSION but weak in TENSION.

So when concrete has to carry tension, it needs to be made stronger. To give it more strength, steel reinforcement or steel prestressing is fixed inside the concrete.

[1] - 44

There are 3 ways of using concrete in bridges;

TYPE

CONTAINS STEEL?

OFTEN USED FOR:

1. MASS CONCRETE

No

Abutments Piers Retaining walls Footways KERBS

2. REINFORCED CONCRETE

Yes: Steel bars

Abutments Piers Decking SLABS BEAMS BOX-GIRDERS

3. PRESTRESSED CONCRETE

Yes: Tensioned steel bars or wires

Slabs Beams Box-girders

Most problems with concrete come from water and air getting into the concrete. Water and air together can cause reinforcement or prestressing steel to CORRODE, but good concrete can protect the steel. Sometimes water and air can carry chemicals which damage the concrete, or corrode the steel more quickly.

[1] - 45

If water can lay on a concrete bridge deck, it will get into the concrete. For example a blocked drain can keep water on the deck, which can cause a lot of damage to the concrete. If deck drains are not properly made and water can get down the side of the drain, or through the concrete around the drain, then the beams underneath may be damaged This is shown in the next photograph:

[1] - 46

Cracking of Concrete You must try hard to look for cracks all over the bridge, but the engineer will know that you cannot look at some parts. Make a note of the parts you cannot see and the engineer will arrange for a special inspection. Most concrete has cracks in it. Large cracks are always important, but fine cracks may not be a problem. The next photograph shows fine cracks due to shrinkage. These cracks are not important

During your training, the engineer or an instructor will show you the kinds of cracks that you should report or sketch. The engineer may tell you to report all the cracks that you see in some places on a bridge because they are important places, or because the concrete is a special kind, such as prestressed concrete.

[1] - 47

To report on cracks, look carefully at them and follow this guide:

1. Measuring Cracks: You need to measure the length and maximum width of the important cracks, and draw them on a sketch of the part you are inspecting. A crack is important if it is 1 mm wide at the widest part, or if it is found in an important place, like the crack on page [2] - 89. You should carry a simple gauge to measure the width of the cracks. A gauge can be made from steel, like the one in the drawing below.

[1] - 48

2. If you find a lot of cracks over a large area, you should draw them on a sketch too, even if they are less than 1 mm wide.

3. If you see any rust staining or any white deposits along the line of cracks, make a note

[1] - 49

4. On the bridge use a waterproof pencil to mark the ends of the serious cracks and write the date of your inspection.

5. If the ends of the cracks were marked at the last inspection, look carefully to see if they are longer now and put a new mark at the end of the crack, and the date.

[1] - 50

Spalling of Concrete Spalling means that some of the concrete has fallen away from the structure. Spalling is commonly caused by corrosion of the reinforcement. When steel corrodes, the rust is much thicker than the original steel So when a bar corrodes inside concrete, it breaks pieces of concrete away:

[1] - 51

Corrosion of Reinforcement or Prestressing This is the most important problem with concrete bridges. This can cause the bridge to fail. Corrosion can be caused by: not enough concrete around the reinforcement. a break in the concrete due to serious cracking, spalling or HONEYCOMBING (see the 'Poor quality concrete section on page [1] - 53). poor quality concrete.

Corrosion will happen more quickly when the concrete is in, or near, salt water. Signs that the reinforcement may be corroding are: you can see the reinforcement at the surface of the concrete you can see cracks or rust stains along a line where you think there is reinforcement. you can see areas where concrete has spalled. Where the concrete has spalled, you should measure two things: 1. Measure the original COVER to the bars. To do this, put a straight-edge across the spalled area, so you can see where the concrete surface was before, and then measure from the straight edge to the nearest part of the reinforcement.

[1] - 52

2. Make an estimate of how much of the bar is corroded away. If you find signs of corrosion, draw a sketch which shows where the corrosion is. When possible, show the reinforcing bars on your sketch. Poor quality concrete There are 3 problems to look for with poor quality concrete water and air can go through the concrete too easily; you can see large holes in the surface of the concrete. These holes are called honeycombing; chemicals, which you cannot see, in the stream or river may damage the concrete It is not easy to know if the concrete is poor without special tests. But if water can get into the top surface of a deck, you may see dampness on the bottom of the bridge deck. This can mean that the concrete is poor or that the drainage is bad Whatever the cause, the water should be stopped. If the water is not stopped, the deck reinforcement will corrode very soon. Honeycombing is caused during construction when the wet concrete does not flow properly and air gets trapped in it. The concrete then has holes where the air was trapped. If there is honeycombing, then the concrete cover to the reinforcement will be much thinner than it should be, and the reinforcement may corrode quickly.

[1] - 53

This picture shows honeycombing under a bridge deck:

[1] - 54

Chemical Attack If chemicals are damaging the concrete the surface of the damaged concrete might feel soft or there might be lots of small hollows in the surface of the concrete. If you think the concrete is being damaged by chemicals write a note for the engineer. This picture shows bad attack to a concrete pier by chemicals in the river water:

[1] - 55

STEEL Steel bridges need to be inspected for 5 main problems: deterioration of paint or galvanising corrosion damage (bends) to steel parts loose fixings cracking

Deterioration of Paint and Galvanising Steel will corrode if it is not protected from air and water Steel can be protected from corrosion by paint or galvanism. Sometimes, when the risk of corrosion is high, steel is first galvanised and then painted Galvanising is a thin layer of zinc on the surface of the steel. It is put on the steel by a special process. In air, galvanising stops the steel from rusting for a longer time than paint. But in salt water, galvanising soon comes off and the steel starts to rust. Paint or galvanising does not last for many years When paint or galvanising deteriorates, the steelwork needs new protection Painted steelwork can be painted again, and galvanised steelwork can be painted with a special zinc-rich paint, or with some other paint made for galvanised steel. Before the steelwork can be painted again, the old paint or galvanising must be very well cleaned and all rust removed, or the new paint will not last long.

[1] - 56

Paint deteriorates when the steel starts to rust. Often, the first signs of failure are small spots of rust in the paint surface. These spots of rust allow water to get under the rest of the paint. This causes more rust and the paint starts to come off. Paint deteriorates more quickly where the paint is thin, e.g. at corners or sharp edges in steelwork. Chemicals in the air (from factories) can also cause paint deterioration to happen quickly. Galvanising deteriorates by corrosion of the zinc. If you see white spots on the surface of the zinc then it is corroding. If the paint or galvanising on a steel bridge is not properly MAINTAINED the steel will rust. When you see signs of paint or galvanising failure, make careful notes and sketches. If repairs are carried out quickly, the corrosion can be stopped.

[1] - 57

Corrosion (rust) Corrosion, or rust, is a chemical change which happens to steel when it is in contact with air and MOISTURE. This photograph shows corrosion at a joint:

If corrosion becomes very bad, the edge of the steel plate can look as if it has split into thin layers. This is called LAMINATION. When this happens, the steel has almost no strength left. It is very serious and you must make a special note to the engineer (see introduction to Part [2]). Rust is much thicker than the steel that it comes from. Where two pieces of steel are bolted or rivetted together, rust can push the steel sections apart, bend the steel and may even break the bolts or rivets.

[1] - 58

On the bridge shown in the photograph below, rust has pushed the concrete deck away from the steel beam:

Usually the worst corrosion happens under the deck. Here there is often water from CONDENSATION and sometimes from poor deck drainage. Dirt from the deck and other debris, such as birds' nests, hold the water against the steel. Bird droppings cause corrosion.

[1] - 59

Damage to Steel Parts When a STRUCTURAL part of a steel bridge is bent, for example by vehicle impact, it may seriously weaken the structure. You must measure the bend. Use a straight edge or a string line as shown in these two pictures. Make a sketch for the engineer, showing which part of the bridge has been damaged. Only the engineer will know if the bend is serious or not.

[1] - 60

There are four types of steel deck used on bridges stiffened steel decks, open panel decks, trough decks and steel jack arches. Stiffened steel decks have STRINGERS welded to the underside of the steel plate. The plates usually span 2 or 3 metres between cross girders. Stiffened deck plates may be bent by OVERLOADED vehicles. Bends in these plates are serious. Open panel decks are made of mesh panels or MARSDEN MATTING Because the panels usually span very short gaps and are often used only as a running surface, small bends in these panels are not so important.

[1] - 61

Trough decks and steel jack archers span between the main beams or girders and usually have concrete or gravel filling on top of them. If water leaks through the concrete or the bituminous surface, the water gathers at the bottom of the trough or at the edge of the buckle plate where two arches join. This is a common problem and can cause corrosion. Loose or Broken Fixings Steel parts are joined by fixings such as rivets and bolts, or by welding the parts together. Some years ago, most joints were fixed with rivets. Many old rivetted bridges still exist. In some countries rivets are still used, but bolting or welding is more common now

[1] - 62

All rivets and bolts must be tight and not broken. It there is corrosion between two pieces of steel which are fixed together, the fixings can break. This is because rust is much thicker than the steel it comes from. This picture shows rivets that could be broken by rust.

[1] - 63

You can easily check that rivets are tight and not broken. Put your finger on one side of the rivet head so that your finger touches both the plate and the rivet head Then, hit the other side of the rivet head firmly with a light hammer:

If the rivet is loose or broken, your finger will feel the rivet move.

[1] - 64

There are two basic types of bolts; bearing bolts and friction grip bolts. You can tell which are friction grip bolts from markings on the head.

Friction grip bolts are unlikely to work loose, so you do not usually have to check them. Bolts holding joints near the bottom of a truss or near the road deck are very important. You can check that bolts are tight by using a spanner

[1] - 65

You will not have to check every bolt or rivet on a bridge at every Inspection. The engineer will tell you which joints you must check at each inspection. An example of a joint which should always be checked is where the cross-girders meet the trusses on a half-through truss. Cracking of Steel Sometimes, but not often, steel members crack This can be caused by many heavy loads crossing the bridge, or by problems with welds, or by faults in the steel Look carefully near all welds, holes etc. This is where cracks can start. If you think you can see a crack, make a sketch for the engineer, showing the crack and where it is on the bridge. It may only be a crack in the paint, but it should be checked. A crack often has a thin line of rust along it.

[1] - 66

MASONRY There are 4 main problems you may find with masonry: - cracking - bulging - poor POINTING - deterioration of the bricks or stones General Masonry is bricks or stones with sand and cement mortar in the joints between them It is strong in compression and weak in tension, like mass concrete. Masonry is used for constructing abutments, piers and retaining walls. Because it has no steel reinforcement, it is not easily damaged by dampness. Well-maintained masonry structures carrying light vehicles, have lasted hundreds of years. But heavy vehicles can damage masonry by VIBRATION of masonry arches, or by impact to masonry parapets. Cracking Cracking is an important sign that something might be wrong with masonry. Cracking can be caused by overloading, vibration or impact from traffic, by failure of the foundation, or by temperature changes or wetting and drying.

[1] - 67

Cracking weakens masonry, allows water to run through and allows soil to enter the structure After a short time, plants and even small trees can start to grow in the cracks. As the plants get bigger, they may make the cracks bigger. When you see cracks, it is not always easy to know how they happened. All masonry, like concrete, expands and contracts with changes in temperature. Masonry built with some types of brick will also expand as the bricks get wet and contract as the bricks dry out. Cracks caused by temperature or moisture changes often run through the mortar only. Cracks which go through the bricks or stones of the masonry are very serious. These cracks probably mean that the foundations have failed or that the masonry has been overloaded. In your inspection, note and make a sketch of all large cracks (wider than 5 mm), showing where they are on the bridge; all cracks near the bearings, and all cracks where there is a step caused by cracking in the face of the masonry

[1] - 68

Cracking is 'very serious' if: there is cracking near the bearings, as in the picture above there is a step in the face of the masonry the cracks are as big as 10 mm

[1] - 69

Bulging Bulging is a change in shape or bending of the face of a masonry wall, usually due to soil behind pushing part of the face outwards. It can happen to abutments, retaining walls, or the barrel or spandrel wails of masonry arch bridges. Bulging of masonry parapets can be caused by vehicle impact. The force from the soil behind a wall can increase due to extra soil being put on top, or the water level in the soil rising (perhaps due to blocked weep-holes), or compaction and vibration due to heavy vehicles, or shaking by an earthquake. Also, as the mortar pointing becomes old, it may become weak, allowing the masonry to bulge. Make a sketch of any bulges, and measure them with the help of a straight edge or string line Poor Pointing Pointing is the mortar between the bricks or stones. The mortar can be worn away by the river or by rainwater running down the face of the masonry. Pointing is usually weaker than the stones or bricks and it will deteriorate with age. If the mortar pointing is worn away or is in poor condition, the bricks or stones may move or even fall out. This weakens the structure. During inspections, look carefully for poor pointing.

[1] - 70

Deterioration of the Bricks or Stones Not all masonry will last a long time Many types of bricks, and some stones, can be worn away by rain or the river, and by the effects of heating or cooling. To check if the bricks or stones have deteriorated, tap the surface (not too hard) with a hammer. If pieces break off, it may be necessary to protect the surface with a layer of hard mortar, called RENDERING.

[1] - 71

TIMBER Timber has 2 main problems: decay and insect attack. Notes on the special problems with joints in timber bridges come at the end of this section on timber. General A tree trunk has 3 layers. On the outside there is bark, which has almost no strength. Next, there is a layer called the sapwood, and in the middle of the tree is the heartwood. The sapwood is usually softer and lighter in colour than the heartwood.

Section through a tree

Decay and insect attack happens more often in the sapwood, because it is softer. On some log bridges the bark has not been removed. This is bad, because the bark stays damp and this causes decay. The bark is also a home for many insects. When a log bridge still has bark, you should remove as much of the bark as possible before you inspect it

[1] - 72

Decay Decay is caused by a FUNGUS which attacks damp wood Decay makes the timber go soft and lose its strength. You should look carefully at those places on the bridge which are in contact with both water and air. For example. parts in contact with the ground (piles, ends of beams, logs, etc.), places where dirt, debris and water collect and vegetation grows (bridge deck and joints in a truss); around fixings. Water can sometimes get to the middle of the timber through holes for fixings. It will be difficult to see this type of decay,

[1] - 73

around splits in the timber. Splits are common and will only lead to decay if water can stay in them, as in splits in the top of horizontal timbers.

Look at the picture below. Split 1 will cause decay, water will not go into split 2.

Sometimes, timber is treated with chemicals to prevent decay and insect attack. The chemical will not get into the middle of the timber so, even if the outside of the timber is good, decay may still happen in the middle.

[1] - 74

Decay below the surface You must look carefully and test to see if there is decay in the middle of the timber.

When you are inspecting for decay, you must look for: stains on the timber caused by water; soft areas on the surface; soft areas which are split into small blocks separated by cracks. This is a sign of very bad decay;

[1] - 75

a flat fungus growing on the surface of the timber. It is a sign of bad decay deep inside the timber Do not remove the fungus.

This type of debris keeps the timber wet and causes decay and fungus attack. Insect Attack Holes and tunnels in timber made by insects or worms can seriously weaken a bridge. Insect holes usually have dust in them or near them. A few small holes (less than 5 mm diameter) are not usually serious. If there are many larger holes, the problem is serious.

[1] - 76

A number of insects attack timber. The most damaging are forest longhorn beetles, which make large holes, and termites, which make large tunnels through the timber If you see termite nests near a timber bridge you will know there is a danger of attack to the bridge. In salt water, a worm called the teredo can attack any area below the high tide level. Teredo worms make large holes and can cause very serious damage. You must check all piers and piles in salt water. To test timber in salt water, use the Hammer Test (see below). Hit the timber just above the water. Tests for Decay and Insect Attack There are 3 simple tests for decay and insect attack: Spike Test To test the surface layer of the timber, push a spike into the wood to find soft spots. Use a square spike so the next inspector does not think your spike holes were caused by insects. Hammer Test To test for decay or insect attack inside the timber, hit the timber with a hammer. Decaying timber sounds different from solid timber. Drill Test If you think there is decay inside the timber, drill a small hole into the wood. The different feel of drilling into soft wood will tell you if there is decay. Be careful not to make too many holes near joints. The pin or bolt holes at joints are already the weakest part of the timber. A 5 mm drill is big enough for this test.

[1] - 77

After using the spike or drill test, put some creosote. or other preserving chemical, into the hole. Fill drill holes with wood plugs. Joints in Timber

There are 3 points to remember about joints in timber: Shrinkage in timber or vibration from traffic can make bolts in joints loose All bolts should be checked for tightness. Some bridges, like the timber deck truss bridge shown above, have decks which are 'nail laminated'. The timbers are all nailed together with no gaps. These decks are often used on timber truss bridges.

[1] - 78

For some bridges with nail laminated decks, the connection between the deck and the truss is very important for the strength of the bridge This connection must be checked from below, to see that the deck is tightly nailed to the top member of the truss. The joints in timber are often made with steel plates, bolts and pins. These steel parts must also be checked. You should look for loose pins and damaged or corroded bolts, pins or plates.

[1] - 79

GABIONS Gabions are wire baskets filled with stones. Most gabion baskets are 2 metres x 1 metre x 1 metre in size. For scour protection a thinner, larger gabion is sometimes used. This is called a reno mattress. Reno mattresses are usually not more than 0.5 metres thick. The baskets can be tied together with wire to form retaining walls. Gabion retaining walls can bulge without damage to the gabions. Baskets can be tied together to form mattresses to cover large areas of slopes or river beds Gabion mattresses can change shape a lot without damage. This makes them specially useful for protecting abutments, piers and road embankments against scour. As the river scours, the gabions settle into a new position, but still protect the abutment pier or embankment. For simple bridges, gabion baskets can be tied together to make an abutment. If the gabions change shape, this may not damage the gabion abutment, but any SETTLEMENT of the abutment can damage the deck. The next picture shows a simple log bridge being built on a gabion abutment:

[1] - 80

There are 2 main problems with gabions: the wrong sort of stones in the baskets; corrosion or damage to the basket wires or the tie wires.

The wrong sort of stones in the baskets If the stones inside the baskets are too small, the wrong shape, or the wrong material, the baskets will not hold them and the gabions will fail.

[1] - 81

Corrosion or damage to basket wires or tie wires Basket wires and tie wires are galvanised or coated with plastic, but they will corrode after some time. If they corrode badly, they may break. Basket wire and tie wires can also break if the gabions change shape too much. Wires and ties can also be broken by debris in the river. If the wires and ties break, the stones may be washed away. However, plants may grow in soil between the stones and help to keep stones in place, even when the wires have corroded.

[1] - 82

CULVERTS

A culvert is used to take a stream or small river under a road. The picture below shows a simple culvert with one barrel.

Sometimes, culverts have 2, 3 or more barrels, as in the photograph shown on page [2] - 11 9. Culverts can be made from timber, concrete, masonry or steel, and are inspected just like small bridges, but using the CULVERT page of the report form.

[1] - 83

Because culverts are small they can easily be blocked by debris. Unlike bridges, culverts can run full of water during a flood. The force of the water may then do a lot of damage to the HEADWALLS and aprons. Culverts may be made of small pipes, like the one in the picture below. They may have no headwalls and no aprons. Even small pipe culverts are important because they are part of the drainage system that keeps the road base strong.

[1] - 84

PART (2) THE INSPECTION INTRODUCTION Now that you have read and you understand Part [1] of this book, you are ready to inspect bridges. For each inspection you will use a report form. Your first inspections should be with an engineer, so that you can ask questions and learn how to use the report form. BEFORE YOU GO Before you go, look at a copy of the bridge record card for the bridge you will inspect and the last inspection report form, if there is one. The bridge record card has details of the bridge you are to inspect: where it is, how long it is, and what type of bridge it is. It will also have a drawing or photograph to help you know it when you get there. Some details on the bridge record card must be written on the report form before you leave the office. These are listed on page 2 of the form. You will check that they are right when you get to the bridge. The last report form will tell you what problems were found during the last inspection. You should look carefully to see if these problems have got worse. The notes and sketches will help you to measure the problems. You should check in the office to find out if any maintenance has been done on the bridge since the last inspection. You should look on the bridge or in the river to see if the maintenance has been done well, and to see if it has solved the problem.

[2] - 1

There may be other notes in the office that be useful to you. For example, there may be a report of an earthquake or a flood in the area of the bridge, or a complaint from drivers that some damage has been done. You should make a note and remember these things when you make the inspection. Before you go, make sure that you have all the equipment that you will need Appendix B has a list that you can use. Change the list to suit the way you work and use it each time before you go.

[2] - 2

THE INSPECTION The following notes follow the order of the inspection report form in Appendix D at the end of this book. Appendix D is a completed report form to show you how a finished form will be. Look at it to see how an inspector should work, but remember that you must fill in your report form to show what you have seen. The ticks you make may be in different boxes to the ticks shown in Appendix D. Report form page 1: Arriving at the bridge It is easy to make a mistake and inspect the wrong bridge. CHECK that you are at the right bridge: Check the bridge number Check the bridge name Check the distance from the beginning of the road, or use the kilometre posts

These three things to check are at the top of page 1 of the report form.

* * * Read carefully the INSTRUCTIONS FROM THE ENGINEER TO THE INSPECTOR. * * *

[2] - 3

On the small sketch of the bridge write the names of the towns or villages on each side of the bridge. Mark which way the river flows, and write numbers on the spans and piers (if there are more than one). Call abutments by the names of the town or village nearest to that end of the bridge. The engineer must know which part of the bridge you are reporting on. The sketch will help. As an example the sketch might look like this:

[2] - 4

At the end of the inspection When you have finished inspecting the bridge, look carefully at your report Have you filled in all the sections? Have you made all the notes and sketches that you need to? Have you numbered all your notes and sketches? Have you written these numbers in the correct place on the form? Will the engineer be able to understand? Are there any problems not on the form, which you should tell the engineer about by writing a note?

When you are sure you have finished, sign the report form at the bottom of page 1. Also write the date of the inspection and the number of pages in your report. Make sure that the pages, including any notes and sketches, are fixed together Then take it back to the office for the attention of the engineer.

[2] - 5

Report from page 2: Construction details, Services and Signs CONSTRUCTION DETAILS CHECK each construction detail as well as you can and tick it as CORRECT: YES or NO * SERVICES The report form lists any SERVICES fixed to the bridge. These are electricity or telephone cables (as in the picture below), or gas, water, oil or sewage pipes. * *

CHECK that the service is still there. Tick under YES if it is, or tick under NO if the service has been taken away. * * *

[2] - 6

Check for new services. If you find any new services that are not written on the form, write a note to say what new services are there and say if they cause a problem. CHECK for damage to services. If there is a problem make a note on the form, especially if it can damage the bridge. For example a leak from a waterpipe can cause deterioration. Write a note if you can see damage to the bridge by the Service Authority. SIGNS Signs are very important. They may give limits on height, width, weight and speed. The bridge can be badly damaged and may fall if a vehicle hits part of it or if a very heavy vehicle overloads it. Drivers must be able to read the signs. Page 2 of the report form has a list of all the signs which should be on or near the bridge. CHECK that each of the signs is still there. Tick under YES if the sign is still there or tick under NO if you cannot find it. CHECK for damage to the signs. Make sure each sign is still fixed firmly and that it can still be read by drivers. Make a note of any damage on the report form. Make a note if you find any signs that are not written on the form.

[2] - 7

Report form - Page 3 onwards The main part of the form, from page 3 to the end, is a list of problems you must look for. These problems come under different sections. For example, in 'The River' section you will report on Scour, Blockages in the Waterway and Change of River Path. You must fill in the report form at the same time as you inspect the bridge. To fill in the report form, you do not have to write very much. Each line will tell you what to look for. Read it carefully and then examine the bridge and decide which box to tick for that line. You may also need to write a note or draw a sketch, so that the engineer will understand what you have seen. For every line on the report form there is a note in Part [2] of this book. To the left of each problem, in the first column on the page, you will see a page number. This tells you where to look for help in your inspection of that problem. If you need more help, you can look back at Part [1] again. The CONTENTS page at the beginning will help you find which chapter you want in Part [1].

[2] - 8

To fill in the form, the first thing you must do is tick either 'Yes' or 'No' to the right of the problem you are looking at, like this:

Remember, all the things listed on the form are problems If there is no problem tick the 'NO' column. If you have ticked 'Yes' (yes, there is a problem), the next thing you must do is decide how bad this problem is There are 3 possible answers to this question. The problem can be either very serious', 'bad', or 'not very bad'. You must tick one of them, like this:

In the last example the problem was not very bad, so the tick goes under 'not very bad'.

[2] - 9

The next thing you must do is decide how much of this problem there is. There can be 'a lot', 'some', or 'not very much'. You must tick one of them like this:

In this example there was a lot, so the tick goes under 'a lot'.

The answers under 'How bad' and 'How much' depend on your experience and your judgement. It is Important that the engineer understands what you mean. Your first inspection should be with the engineer Make sure you learn from him how to answer the 'How bad?' and 'How much?' questions.

[2] - 10

For example, it you find a lot of cracks over a large area on the face of an abutment, fill in the form like this:

If you find that a small part is very badly damaged, fill in the form like this'

[2] - 11

On parts where not very bad very ' damage is not important at all, the box has been covered up. In this case you need only tick when you see 'bad' or 'very serious' damage

Sometimes you will need to write a note as well. For example: writing the location of a problem (where it is). If you can, write this on the same line as the problem, like this:

[2] - 12

If there is not enough room for your note in this space, write the note on the back of the form page, and give it a number. Write this number in the space on the form, on the same line as the problem. If you need to draw a sketch, draw it on the back of the page, or a new piece of paper Give the sketch a number and write this on the form, on the same line as the problem. The form might look like this.

[2] - 13

Take your time and fill in each line of the report form carefully as you come to it. Remember that you must fill in the form at the time you are looking at the problem. Do not wait until later. Sometimes it is very difficult to know the answers, even experienced inspectors may not be able to tell. If you do not know how bad or how much, or if you are not sure if there is a problem, make a note. Each section of the form has a heading such as UNDERSIDE OF BRIDGE. When you have answered all of the questions under the heading, tick the 'ALL CHECKED? - YES' box. If you could not properly check all of this part, answer as many questions as you can, but also tick the 'NO' box after 'ALL CHECKED?' to show that this part is not completely checked.

All checked

[2] - 14

ROAD APPROACHES AND DECK

ROAD SURFACE NEAR BRIDGE

If the road surface near the bridge is very bad, vehicles bounce up and down as they cross the bridge. This bouncing can damage the bridge. CHECK for bumpy road surface. On the report form make a note if the road surface is bumpy for up to 50 metres on either side of the bridge, Look carefully at the area behind the ballast wall of the bridge. There is often a pothole here.

[2] - 15

DRAINAGE

Drainage is important because if water runs on to the bridge deck it may damage it. Water can also carry small pieces of stone which may fall into the expansion joints and damage them. Water from the road can erode the approach embankment. CHECK for badly built road drainage near the bridge. Look on both sides of the bridge. Note where the drainage is bad. * * *

[2] - 16

CHECK for blocked or damaged road drains. * * *

CHECK to see if there is water lying on the deck. * * * If water remains on the deck, it can be a safety hazard and may cause corrosion and decay. CHECK for blocked or damaged deck drains. If you can see deck drains under the deck but not on top, they could be covered by bitumen surfacing Make a note if the deck drains are blocked by road surfacing. * * * * *

[2] - 17

BRIDGE SURFACE & FOOTPATHS The road surface on the bridge can be sealed with bitumen, or made of concrete, steel or timber. There is a different section of the report form for each surface. If the footpath is the same surface as the road, report on the road and footpath together. If the footpath is a different surface from the road surface, then report on the road surface and footpath surface separately. Note on the form which is the footpath. BITUMEN SURFACE At deck joints where very small movements are expected, the surfacing may cover over the joint. Often the sealed surface cracks or breaks-up over these joints. Surfacing is often damaged at joints * * * CHECK if the surface is breaking-up or lifting off the concrete underneath. Look very carefully near the expansion joints and near the drain holes. * * *

[2] - 18

CHECK for cracking of the surface above buried joints.

Some modern bridges have a waterproof layer between the concrete and the sealed surface. The bridge record card will tell you if a bridge has a waterproof layer. If the bridge deck surface is in good condition, you cannot see this waterproof layer. When the sealed surface is broken, the waterproof layer may be damaged. If you see any signs of damage to the waterproof layer, write a note for the engineer. * * *

[2] - 19

CONCRETE SURFACE CHECK for cracking of the concrete. Look again at the notes on page [1] - 47 if you are not sure how to do this. On long span truss bridges look carefully for cracking of the concrete decks Any cracks which begin in spalled areas of the deck are serious * * * CHECK for spalling of the concrete.

CHECK if reinforcement is exposed. If it is measure and make a note of the original concrete cover to the reinforcement bars. Also, measure and make a note of how much of the reinforcing bar diameter has been lost (More notes on this are on page [1] - 52 in the first part of this book.) * * *

[2] - 20

Check for poor quality concrete. Look for honeycombing and other signs of poor quality concrete. * * * If the road to the bridge is not surfaced with bitumen, small stones in vehicle tyres can damage the surface of the concrete. CHECK for wear of the surface due to small stones in tyres. * * *

[2] - 21

STEEL SURFACES There are two types of steel surface: Stiffened steel panels. Grid or open panels such as Marsden matting. This is shown in the photograph.

CHECK to see if the fixings are loose or damaged. * * [2] - 22 *

CHECK for bends in the steel panels. For Marsden matting, only make a note of bad bends. On other decks note all bends. * * * CHECK for corrosion of the steel surface. Look carefully at areas near kerbs and deck drains. * * *

[2] - 23

TIMBER SURFACE Dirt, debris and plants in corners and between the boards hold water in the timber. This will damage the deck. CHECK for dirt or plants growing between the boards. Look carefully close to kerbs or running strips. * * * CHECK for decay of the deck timber Look carefully near the deck ends and against kerbs and running strips. * * * Check for signs of insect attack. Look at all parts of the timber deck. * * * CHECK for splitting of timbers. For deck timbers, small splits are not important, but larger splits should be reported. * * * Loose deck timbers are very dangerous on any bridge and could cause an accident. On some bridges, the deck timber helps to strengthen the whole bridge. On these bridges it is very important that the deck fixings are good. The engineer will tell you if you are inspecting this kind of bridge. CHECK for loose or damaged fixings. * * *

[2] - 24

TIMBER RUNNING STRIPS Timber running strips are often damaged, and must be replaced with new running strips Damaged or loose running strips can cause bad accidents. CHECK for damage (decay, insect attack, splitting) to the running strips.

This timber deck shows bad decay in some cross members and running strips in very bad condition. * * * CHECK for loose or damaged fixings * * *

[2] - 25

RAILWAY OR TRAM RAILS Rail or tram tracks sometimes cross road bridges. The rails can work loose, especially near expansion joints. If this happens, the bridge deck can be badly damaged as the train or tram bounces over the loose rail. CHECK for loose rail fixings. * KERBS Kerbs separate the road from the footpath. They are often hit by vehicles. On some bridges high kerbs are used to stop vehicles driving onto the footpath. These high kerbs are sometimes used instead of guard rails. CHECK for damaged or loose kerbs. * FOOTPATHS Damaged paving slabs or boards can be a danger for people walking across the bridge. CHECK for damaged footpaths. * * * * * * * * *

[2] - 26

PARAPETS, RAILINGS AND GUARD RAILS

If the bridge you are inspecting does not have a parapet or railing, tick the 'ALL CHECKED - NO' box and write a note to say if it was not built, or if it was taken away.

Parapets and crash barriers can be damaged by vehicles. Sometimes, the bridge is also damaged near the parapet CHECK for impact damage to parapets. Make a note if the bridge next to the parapet is also damaged. Look at holding down bolts at post bases, joints in rails, and joints between rails and posts. * * *

[2] - 27

CHECK for loose or damaged fixings. * * * CHECK for loose post bases where steel, aluminium or timber parapets are set in holes in concrete. * * * The next sections are about parapets made of particular materials. Sometimes, parapets are made of two materials. For example, if there are concrete posts with steel rails, report on the posts using the concrete section and the rails in the steel section.

STEEL or ALUMINIUM PARAPETS CHECK for damaged galvanising or paint. * CHECK for corrosion. * * *

[2] - 28

CONCRETE PARAPETS CHECK for cracking. * * * CHECK for spalling of the concrete. Look carefully at the corners of the posts and rails. * * * CHECK for corrosion of reinforcement. Look where concrete has spalled. Look for rust stains from cracks, especially near the corners of the posts and rails. * * CHECK for poor concrete. * * *

[2] - 29

TIMBER PARAPETS CHECK for decay, especially at the base of the posts and at joints between posts and rails. * * CHECK for insect attack. * * CHECK for splitting of timber. * * *

[2] - 30

MASONRY PARAPETS Masonry, whether brick or stone, damages easily if vehicles hit it. Often it cracks and may be pushed outwards. Small cracks in masonry are not usually important. CHECK for serious cracking. * * * CHECK for outward movement or bending of the parapet. * * CHECK for poor pointing. *

* * * CHECK for deterioration of the bricks or stonework of the parapet. * * * * *

[2] - 31

EXPANSION JOINTS

There are many different sorts of expansion joints that can be used on bridges. It is a good idea to check the bridge record card in the office, before going out to the bridge, to find out what sort of joint is used and where the joints are on the bridge. If you look at the bridge record card, you must also look carefully for other joints on the bridge. The bridge record card could be wrong. The next five checks can be done from above, as you stand on the deck, but remember the joints when you go under the deck. Look for signs of problems and listen when vehicles pass over the bridge. You might hear the sound of loose plates or see signs of damage.

[2] - 32

CHECK for damage to the concrete of the deck end or ballast wall near to the joint. * * * Joints on long bridges sometimes have several moving parts. Stones or other debris can stop some of the parts from moving. CHECK for debris or vegetation in the joints. * * * CHECK for loose or damaged fixings. With steel parts, this is a common problem. * * * CHECK for damage or corrosion to metal parts of the joint. * * * Some joints have a rubber water-bar to stop water and debris from the deck falling through the joint. Stones falling into the joint can cut the water bar. CHECK for damage to the rubber water-bars. * * * * *

[2] - 33

THE RIVER You must look very carefully for changes in the river. You will be able to see most problems from the deck of the bridge or standing under it by the abutments. Sometimes you may have to walk along the banks for a short distance to check what is happening to the river

BLOCKAGES IN THE WATERWAY The waterway under a bridge should be clear. Any blockage is bad because: 1. it can make the water scour a hole in the river bed and 2 debris can get stuck on it and make a bigger blockage. CHECK that debris carried by a flood has not piled up against piers or abutments blocking the waterway. * * *

[2] - 34

Sometimes, when a new bridge is built, the old bridge is not all taken away. Parts of the old bridge can catch debris or cause the river to scour holes around the new abutments.

CHECK that the waterway is not affected by the remains of old bridges under the bridge or just upstream of the bridge. * * *

[2] - 35

For many bridges, some of the waterway is usually dry, except in a flood. Sometimes, people put fences and buildings on the dry areas of the waterway. These must be taken away because they will block the waterway. The dry areas of the waterway are often covered with grass and small plants, such as rice. These are not a problem. Large plants and trees, such as bananas, can block the waterway in a flood. They must be taken away.

CHECK that there are no fences or buildings of any sort under the bridge, or just upstream in the waterway. * * * CHECK that there are no trees or bushes growing under the bridge or just upstream in the waterway, or where a flood can wash them under the bridge. * * *

[2] - 36

CHANGE OF RIVER PATH If the river changes its path it can destroy the bridge.

CHECK if the river is changing its path upstream from the bridge. To find out if a river is changing its path: Ask the local people. Look at the bends in the river upstream from the bridge. Look at the river banks on the bends.

[2] - 37

Usually you can see far enough upstream from the bridge. Sometimes you may need to go upstream a short distance. If the bank is steep and there are trees at the edge of the bank, but nothing is growing on the bank, then the river is moving towards that bank. When there is a lot of water flowing and the bank is not steep and has small plants growing on it and some mud or small stones on it, then the river is moving away from that bank * * * You must also look for other changes upstream of the bridge. If new islands form, then the river may change its path and may damage the bridge. CHECK to see if new islands are forming. Look to see if there is debris in the river. Debris can cause a new island to form. * * *

[2] - 38

RIVER TRAINING WORKS Two things can go wrong with river training works. The river can move beyond the upstream end of the river training works and attack the end of the works. The materials of the works may be damaged.

CHECK for river attack beyond the upstream end of the river training works. * * * Piles may move forward with scour or be damaged by floating logs. Timber piles may decay, steel piles may corrode. CHECK for damage to sheet piled walls. * * * Rip-rap can sink into a soft river bed, or be washed away. CHECK for loss of rip-rap. * * * Groynes made of gabions, or fencing can be washed away or damaged. CHECK for damage to gabions, timber fencing, etc. * * *

[2] - 39

One type of river training uses trees to train the river The trees are protected with gabions while they grow. Dead or damaged trees must be replaced. CHECK for damage to trees used in river training works. * * * * *

[2] - 40

SUPERSTRUCTURE The report form asks you to look at some general problems of bridge superstructures first You then look at the main supports (beams, slab, girders or trusses), whether they are above or below the deck. Sometimes, to inspect a bridge properly, you will have to look at the same part of the bridge (such as a truss) from above the road and below the road. Finally, you look at the problems under the bridge deck. You must repeat this inspection and fill in a new part of the report form for each span.

[2] - 41

GENERAL Bridges are often damaged by vehicles hitting the girders or trusses. Even small bends in girders and trusses can be serious. Sometimes, during a flood, a boat, log or other floating debris hits a bridge and damages it. CHECK for impact damage to beams, girders, trusses or bracings by vehicles, boats or logs Make a note of how you think the damage was caused. * * * CHECK for debris or vegetation on beams, girders, trusses, bracings or in joints. These can cause serious damage to the bridge. * * * If the underside of the bridge deck is wet, this can damage the bridge.

CHECK that there is no water coming through the bridge deck, except through proper drains

If the underside is not wet, there may still be a problem. You need to look for signs that the underside has been wet before: On steelwork, look for rusty marks. On timber, look for dark water run marks. On concrete, look for dark areas and a build up of white material on the surface. * * *

[2] - 42

CHECK that water from the deck drainage does not flow on to the girders, trusses, beams or bracings. * * * For a bridge which carries a road over another road, it is important that there is enough headroom. When a road is resurfaced, too much bituminous surfacing may be put on, so that the headroom to an overbridge is too small. If the road has been resurfaced since the last inspection: CHECK for enough headroom for an overbridge Measure the headroom where it is smallest, from the surface of the road below to the underside of the bridge superstructure. Write the headroom in the box on the form * * * * *

[2] - 43

MAIN BEAMS. GIRDERS,TRUSSES AND BRACINGS

CONCRETE BEAMS

CHECK for cracking of the concrete. If you find cracks, measure the length of the crack and its width at the widest point. Make a sketch of the crack and include it with the report. Mark and date the end of the crack. Look to see if any cracks have got bigger since the last inspection. Look carefully for 1. Cracks at the beam ends which spread up from the bearings as in the picture below. These are a serious problem.

2. Cracks on the face of beams which are parallel to the bottom of the beams. These cracks can happen when the corner of the beam is starting to spall. Look carefully at these cracks and tap the concrete firmly to see if it is spalling. Mark the ends of the crack and make a sketch of it. * * *

[2] - 44

CHECK for spalling of the concrete. * * *

This picture shows spalling of the concrete due to corrosion of the reinforcing steel. CHECK for signs of corrosion of reinforcement. If any reinforcement is visible, measure and make a note of the concrete cover to the reinforcement. Also, measure and make a note of how much of the reinforcing bar has been lost. * * *

[2] - 45

CHECK for signs of poor quality concrete. Look especially at the bottom corners of deep concrete beams:

[2] - 46

STEEL GIRDERS AND BRACINGS This drawing names the parts of a steel girder.

CHECK for deterioration of paint or galvanising, on all main girders and cross girders * * *

[2] - 47

CHECK for corrosion of the steel. Look everywhere, but especially at any wet areas On many bridges there is debris on the bearing shelf and this keeps the area wet. Look very carefully at the girder ends. Many steel girder bridges corrode badly at the ends near the bearings. The next photograph shows bad corrosion at the end of a girder:

Two very serious problems caused by corrosion are: Laminations in steel (see page [1] - 58). Steel sections joined by rivets or bolts being forced apart. If this has happened check all bolts or rivets in the joint. Write a note if you see lamination or other damage caused by corrosion. * * *

[2] - 48

A big overload can cause a bend along most of the girder You can see these bends by looking along the edge of the girder flange. Bends in the beams or girders are serious. Also look for bends in the STIFFENERS above the bearings. Make a sketch of the damage and show which part is damaged. CHECK for bends in webs and flanges of girders, and in stiffeners and bracings. If possible, measure the size of all bends. Place a straight edge or string line next to the damaged member and measure from this to the girder. * * * CHECK for loose bolts or rivets. On the structure mark all loose bolts or rivets with a waterproof pencil or paint. Make a sketch of the girders and mark the joints which have loose bolts or rivets. * * * CHECK for cracking, especially at welds and holes. If you find any cracking write a note to say where the cracking is on the girder and draw a sketch. Cracking can be very dangerous. * * *

[2] - 49

STEEL TRUSSES This section covers through trusses and half-through trusses, but not Bailey bridges.

CHECK for deterioration of paint or galvanising. Look carefully at all places like the ones circled on the picture above. * * *

[2] - 50

CHECK for corrosion, especially at the joints in the truss. The bottom boom joints often corrode first. If there are signs of corrosion, especially near joints, check that plates are not laminated (see page [1] - 58). Look for signs that corrosion is forcing the plates apart * * * CHECK for bends in truss members. Make a sketch of the truss and mark any bent members on it. Using a straight edge or string line, measure the size of the bend and mark it on the sketch. * * * CHECK for bent or damaged joints. Make a sketch of the truss and mark any damaged joints on it. * * * CHECK for bent or damaged bracings. Measure the bend if possible and make a careful sketch of the pattern of the bracing system Mark the damaged member on your sketch. * * *

[2] - 51

CHECK for loose bolts or rivets. Check all bolts or rivets in damaged or corroded joints. On the structure mark all loose bolts or rivets with paint. On a sketch of the truss, mark the joints with loose fixings. If you find loose, damaged or missing bolts or rivets in a joint, look carefully along the line of the truss to see if the top has bent. If you see that it has bent, measure the bend with a line as shown in the pictures on page [2] - 81 and write a note. * * * Cracking is very serious as it can lead to rapid collapse. The engineer will tell you which areas must be checked for each inspection. CHECK for cracking of the steel, particularly at welds and holes. If you find cracking, write a note or sketch to show where the crack is. * * *

[2] - 52

TIMBER BEAMS

This section covers jog bridges, bridges made of sawn timber beams, like the one below, and bridges made of laminated timber beams.

Log bridges are often poorly maintained and have a short life. But log bridges can last a long time if they are properly constructed and well maintained.

[2] - 53

CHECK all timber for decay. Look very carefully at damp areas and around nails or spikes If the bridge beams (or logs) Sit on the soil, or if the beams ends come into contact with the soil, then the ends often decay. Check this area carefully Look very carefully at logs with bark still in place, and when possible remove the bark. * * * CHECK all timbers for insect attack. If insect attack or decay is found, find out how much of the timber is affected. The notes on page [1] - 77 tell you about three tests using a spike, a hammer or a drill. Using one or more of these tests, keep checking until you find good timber. Make a sketch of the beam or log showing the decay or insect attack. Note on the sketch the size of the bad area and the depth. * * * CHECK for splitting of timber. There are 2 kinds of split which are important. Make a note if you find either of these: 1. Very large splits which may be getting bigger. 2. A split which will allow water to get into the beam or log and not drain out. Splits on the top, or on the upper part of the sides, of a beam or log, are a problem. Splits on the bottom of a beam, or on a vertical member will drain and are not a problem. * * *

[2] - 54

Sometimes beams are made by sticking together smaller pieces of timber. These are called glue laminated beams. If the glue fails, the laminations (the small pieces which make up the beam) may separate This could be serious. CHECK for separation of the laminations of glue laminated beams.

On beam bridges, the main beams are often connected by timber cross beams which are nailed or spiked to the main beams. The nails or spikes should be tight. On long span log bridges, sometimes many logs are needed These are often held together by wire ropes. Check that the ropes are not corroded and are tightly fastened. CHECK for loose or corroded nails, spikes or fixing wires.

[2] - 55

TIMBER TRUSSES Deck type truss bridges like this one have many small corners where dirt and debris collect:

CHECK for decay all over each truss. Look very carefully at joints and other areas where dirt or vegetation can collect. The pin or bolt holes in the timber at joints already make those parts the weakest places on the timber So when you use the spike and drill tests for decay do not make too many holes at joints because this will weaken the timber. * * [2] - 56 *

CHECK for insect attack. * * CHECK for splitting of timber. *

Some splits can be serious. There are 2 kinds of split you must report: large splits at joints which are close to bolt or pin holes. If you find these, make a sketch of the split, show the length of the split and the distance to any holes. This will help the next inspector to see if there is a problem. If the last inspector noted a split near a joint, check to see if it has got worse. splits where the wood is wet, causing decay inside the wood.

* * * CHECK to see if the deck to truss connection is loose. This is very important on bridges with nail laminated decks. * * * CHECK for loose or corroded bolts or pins at the truss joints. As timber dries, it shrinks. This can cause bolts in joints to be loose. Bolts can also loosen because of vibration. * * *

[2] - 57

CHECK for bends in truss members. Make a sketch of the truss and mark any bent members on it. Measure the size of the bend and mark it on the sketch. On bridges with the trusses above the road, look along the line of the truss to see if the top has bent. Measure the bend and make a note. Small bends can be measured in the way shown on pages [[1] - 60 and [1] 61. Long bends in a truss are measured in the way shown on page [2] - 81. * * * CHECK for damaged or corroded steel parts (joint plates, bearing plates, etc.), and mark their location on a sketch. * * * * *

[2] - 58

UNDERSIDE OF DECK CONCRETE CHECK for cracking of concrete. Make sketches of any cracks wider than 1 mm, or cracks over a wide area. Mark and date the ends of important cracks, note any white deposits, and check if cracks have got any bigger since the last inspection. * * * CHECK for spalling of concrete. * * * CHECK for corrosion of reinforcement. Note any rust staining. Where concrete has spalled, note the cover and how much of the original bar has been lost by corrosion. Make a sketch to show where you have found signs of corrosion of reinforcement. Mark on it the size of areas of spalling, the pattern of cracking or the pattern of rust staining. * * * CHECK for poor concrete. Make a sketch showing the location and size of honeycombed areas and areas which are damp. * * * CHECK for not enough cover to reinforcement If the cover is very small, you can see the pattern of reinforcement on the surface of the concrete. * * *

[2] - 59

STEEL This section is for the underside of stiffened steel decks, for steel arch decks (buckle plates) and for trough decks. CHECK for deterioration of paint or galvanising. * * * CHECK for corrosion of steel. Look everywhere, but check any wet areas especially. Look carefully at the bottom of troughs and the edge of steel arch decks. If you find laminations or other bad corrosion you must write a note on the form. * * * CHECK for bends in stringers or plates. On steel decked bridges if you notice a bend in the surface, look very carefully at the underside of the deck, below the bend. The stringers which carry the deck may have bent. This is serious. On steel arch decks bends in the plate of the arches are very serious. * * * CHECK for loose bolts or rivets. * * * CHECK for cracking of the steel, especially at welds and holes. * * *

[2] - 60

TIMBER This section is for the underside of timber decking on any type of main beams.

On log bridges, when the logs are close together it is not possible to inspect the underside of the deck timber. You can only check on the top surface of the bridge and the ends of the deck timbers.

[2] - 61

CHECK for decay, look especially at any areas where water from the deck flows onto the timber. Remember to seal and plug any drill holes you make. If the timber deck is on a steel bridge look at the places where the timber sits on the steel beams. Also look for decay where the bolts or spikes enter the timber. * * CHECK for insect attack. *

* * * CHECK for split timbers. When you inspected the top surface you will have seen any splits which go through the deck timbers. Splits across the grain of the wood, which can be seen on the underside but not on the top surface, can happen when the timber is starting to break, possibly because of a heavy vehicle. If this has happened, make a note in your report that the timber must be replaced. * * * CHECK for loose or badly corroded bolts or pins. * * *

[2] - 62

MASONRY JACK ARCH DECKS CHECK for change of shape of the arch. This can happen because of an overload or a failure in the material of the arch. If the change of shape is serious, the arch must be rebuilt. * * * CHECK for cracking or spalling of the bricks or masonry. * * * CHECK for poor pointing of the arches. If the mortar between the bricks or stones is in poor condition the arches might need repointing. * * * * *

[2] - 63

BEARINGS This section covers all types of bridge except Bailey bridges. Sometimes you will not be able to see all the parts of a bearing. If you cannot check all of a bearing, tick the 'ALL CHECKED - NO' box and do not tick the 'YES' or 'NO' box next to the questions you cannot answer. First find out which bearings are fixed, and which are free. The bridge record card will give you the location of the fixed and free bearings. If you have difficulty, look back at pages [1] -21 to [1] -23. Bearing shelves are very often wet places in which debris and vegetation collect. CHECK for debris or vegetation around the bearings and on the bearing shelf. Remember, even small amounts of debris on the bearing shelf may cause damage to the bearing, especially if the drainage to the bearing shelf is not working properly. * * * CHECK for poor drainage to the bearing shelf. Look for signs that water stays on the shelf in wet weather. * * *

[2] - 64

CHECK to see if there is not enough room for the bridge span to move when the temperature changes. This is not easy. Tick 'YES' or 'NO' only if you are sure. * * * CHECK to see if the bearings are not seated properly on the abutment. * * * CHECK to see if the bridge span is not seated properly on the bearings. * * * CHECK for damaged bedding mortar. Look for cracks around the edges, and look for pieces of mortar which are broken off or crushed You will need to clean all the soil and plants away from the bearing to see this. * * *

[2] - 65

RUBBER BEARINGS Some rubber bearings have steel plates inside. If you can see the steel inside the rubber, the fault is bad. CHECK for splitting, tearing or cracking of the rubber. * * * Fixed bearings usually have a pin or bolt fixing the beam to the support through the bearing. On concrete bridges, it is not possible to see the pins used in the fixed bearings. They cannot, therefore, be checked. On steel girder bridges the heads of the bolts or pins can usually be inspected. CHECK for damaged or loose bolts or pins at fixed bearings on steel beam bridges. * * *

[2] - 66

METAL BEARINGS Look carefully at the bearing and decide which type of metal bearing you are inspecting. This is a typical simple metal sliding bearing Note the corrosion caused by debris on the bearing shelf:

CHECK if any of the bearing parts are not properly seated. * * * If pieces of gravel or other debris are caught in the moving parts of a bearing, they can damage it or stop it moving. CHECK if any moving parts are not free to move. Look to see if anything is caught in the bearing. Look at the surface where the bearing should move. You can usually see a narrow clean area where the bearing moves. There will be a thin line of dirt at the edge of the area the bearing moves over.

[2] - 67

Some rocker or roller type bearings, and bearings where steel slides on bronze, have LUBRICATION SYSTEMS. CHECK for problems with the lubrication system. If there is a lubrication system and you find a fault with it, write a note on the form. * * * For a sliding bearing to work, the sliding surfaces must be in good condition. CHECK to see if the sliding surfaces are damaged. Look very carefully at the sliding surfaces. A small amount of corrosion on a sliding surface will cause large forces. The bridge could then be damaged. Look carefully at the PTFE but be careful It is easily damaged. * * * CHECK for cracks or bends in metal parts. * * * CHECK for corrosion of metal parts. * * *

[2] - 68

EARTHQUAKE RESTRAINTS On modern bridges in earthquake areas, there may be an earthquake restraint. An earthquake restraint stops the span from falling off the abutment or pier during an earthquake. There are many types of earthquake restraint. Here are tour common types: 1. A large bolt or bolts which hold the span down. 2. Extra strong wing walls which stop the span falling off. 3. A block of concrete between the bridge girders. 4. Special frames of steel or concrete, at the end of the bridge, which stop it falling off in an earthquake.

CHECK for damaged or loose earthquake restraints. * * * * *

[2] - 69

MASONRY ARCHES CHECK for change of shape of the arch. You must: 1. Stand a long way from the arch. Compare the shape of the arch left of centre, with the shape of the arch right of centre. Any difference means the arch has changed shape. 2. Look at the centre stone on the face of the arch. If this has moved out of line, the arch will be badly weakened. 3. Go underneath the arch and check the barrel of the arch. Look to see if parts of the barrel have changed shape differently from the face of the arch. (This often happens.) 4. Look at the other face to see if the arch has changed its shape or if the centre stone has moved. Mark the location of any change of shape on a sketch or on the overall plan on page 1 of the report form. Estimate the amount of change of shape, if possible, and write it on your report.

[2] - 70

In this picture there is a crack in the spandrel wall and the arch has changed shape:

CHECK for cracking of the arch barrel. On the plan on page 1 of the report form, sketch any long cracks which run through the arch. * * *

[2] - 71

If you find vehicle damage to the parapets, check that the spandrel walls under the parapet have not been pushed out or cracked by the impact. Where a road crosses an arch bridge there is sometimes a hump. When heavy vehicles hit this, they can push the spandrel walls outwards. Two things can happen: 1. The spandrel wall can crack and bulge. 2. The spandrel wall separates from the arch and is pushed outwards. If this happens, you can see a step in the masonry between the arch barrel and the spandrel wall. CHECK for cracking or bulging of the spandrel walls. If you see a crack or bulge make a sketch of each face of the arch and spandrel walls. Mark any cracks or bulges on your sketch. * * *

[2] - 72

CHECK to see if the spandrel wall has separated from the arch, as in the picture above. * * * CHECK for spalling of the stones or bricks of the arch and spandrel walls. * * CHECK for poor pointing. * * *

[2] - 73

The materials of an arch will gradually deteriorate with the effects of water and weather. Most arches leak some water through them. A small amount of water is not a problem but if the leak is bad then the bridge will be damaged. If a lot of water is leaking through the arch, you will see white stains near the joints and sometimes clear, wet areas, where the water has been running. CHECK for water leaking through the arch. * * * All masonry arch bridges have spread foundations If an arch foundation settles, the bridge will be badly damaged. Scour can cause settlement. CHECK for scour under the arch foundation. Use a long pole to feel if there is scour If you can push the pole into the ground under the arch foundation there is a bad problem. * * * * *

[2] - 74

BAILEY BRIDGES

Many different kinds of bridge have been made using the basic Bailey panels Design details have changed over the years, and different strengths of steel have been used. It is not possible to describe all the variations of a Bailey bridge in this chapter, 50 you will need help from the engineer, until you get to know the bridges in your area. There are no extra parts on a Bailey bridge, All the parts are needed. If some parts are removed, the bridge may still look safe, but it could fall when a heavy vehicle crosses it. You need to check very carefully to make sure no parts are missing.

[2] - 75

The drawing below shows most of the parts of a typical Bailey bridge and gives their names. Not all Bailey bridges are exactly like this.

On single span Bailey bridges, panel pins usually stay in place. On continuous bridges (two or more spans), when the bridge moves under traffic, panel pins can be moved The panel pins can even work loose and fall out if the SAFETY PIN is missing This is most likely to happen about one quarter of the span away from a pier. CHECK for missing safety pins. * * *

[2] - 76

CHECK for missing panel pins. * * * CHECK for missing or loose bolts. The bolts that hold on the CHORD reinforcement and hold together DOUBLE STOREY BRIDGES are very important. * * * CHECK for missing rakers and tie plates. Some bridges use vertical bracing frames in place of rakers. * * * CHECK for missing or loose sway braces Check that the pins are in place and that the lock nuts are screwed up. * * * Some Bailey bridges have horizontal bracing frames on the top chords of the panels. CHECK for missing, loose or damaged horizontal bracing frames. * * * CHECK for missing or loose transom clamps. Some modern Bailey bridges do not use transom clamps. For these, note on the inspection form - 'no transom clamps'. * * * Bailey bridges move a lot when vehicles cross them. This causes wear, where the stringers sit on the transoms. CHECK for wear at the stringer to transom SEATING.
* * *

[2] - 77

The original Bailey bridge panels sometimes cracked and some bridges fell down because of this. Modern Bailey bridges are less likely to crack, but unless you are told not to, you should: CHECK ALL Bailey bridges for cracking. This picture shows where you should look for cracks:

All these places must be checked on each panel at each inspection. * * *

[2] - 78

Parts of the bridge can get bent. CHECK for bends in bridge members. Note where the damage is and sketch the damage. * * * CHECK for deterioration of paint or galvanising. * * * CHECK for corrosion. There are some areas on a Bailey bridge where dirt and moisture collect. Look especially at: 1. The bearings. 2. The base of the rakers 3. The bottom chord near the SPACER PLATES and where the diagonal members (diamonds) join. 4. The panels and transoms near where the transom sits on the bottom chord. Bailey bridges are made of lots of panels, so it is not easy to make a sketch showing where you have found a problem. If you find a panel that is damaged, or has any problem, mark it with paint and give it a number. When you fill in the report form you can write the number in the column for notes. * * * CHECK for settlement of the bearings. * * *

[2] - 79

CHECK for damage to the bearings and baseplates. * * * All Bailey bridges sag a little at the middle of the span, and some are bent sideways. Look at the photograph and the two drawings. They show you how to measure the vertical sag and the horizontal bend. Look along the top of the panels on one side of the bridge. You will need an assistant to help measure the sag and the bend with a rule or measuring tape, taking instructions from you. Write on the report form the biggest measurement of vertical sag and the biggest measurement of horizontal bend on the upstream side of the bridge. Then measure the sag and bend on the downstream side of the bridge and write these measurements on the report form.

[2] - 80

[2] - 81

ABUTMENTS, WING WALLS AND RETAINING WALLS.

GENERAL The most common major problem with abutments is scour caused by the river. If the river scours the foundations or the bank in front of the abutment, the abutment will move and the bridge may even fall. This photograph shows how a river can quickly scour under an abutment, causing it to move and displace the deck. Scour should be reported as soon as it starts, so that the damage can be repaired before it gets this bad.

[2] - 82

Bank seat abutments can easily be damaged if the bank under them is scoured by the river or eroded by rain water, so you must look very carefully for scour or erosion near bank seat abutments. The picture below shows scour under a piled abutment:

CHECK for erosion and scour near the base of the abutment, or scour of the bank in front of the abutment. If the water is not low enough or clear enough to see, then use a long pole to feel if the river has caused scour. * * *

[2] - 83

CHECK for damage to caissons, or damage or corrosion to piles, if you can see them. * * * CHECK for movement of the abutment. This can be a serious problem. Look for disturbance of the ground around the abutment. If the abutment has moved, there are often cracks in the soil. There may even be cracks in the road behind the abutment. * * * In a flood, debris against the abutment can cause damage to the abutment or the bridge superstructure. Debris can also cause the river to scour the soil around the abutment. CHECK that there is no debris on, or against, the abutment. * * *

[2] - 84

CHECK for vegetation growing on or in the abutment. Look in cracks or drains in the abutments, or in cracks between the abutment and wing walls or retaining walls * * * CHECK for scour near to retaining walls. * * *

[2] - 85

CHECK for signs of movement (sliding forward or settlement) of retaining walls. Look at the picture on page [1] - 30. Look for:

a step where the retaining wall meets the wing walls or abutment. damage to the road or the embankment supported by walls

* * * CHECK for signs of water leaking down through the expansion joint. * * *

[2] - 86

DRAINAGE SYSTEM Look at the abutment and the retaining walls. Look for signs that the drainage system is not working properly. CHECK to see if there are not enough weep-holes. As a general guide, weep-holes should be less than 2 metres apart, measured both horizontally and vertically. If the water coming out of the weep-holes washes away the soil in front of the abutment then there are not enough weep-holes. * * * CHECK to see if the weep-holes are not working. In wet weather, check that the bottom weep-holes have carried water recently. In dry weather, look carefully to find out if the weep-holes carried water when it was raining. Look for signs such as: if there are insect nests in the weep-holes, then the weep-holes may not be working. if there are water stains below the weep-holes, then the weep-holes are probably working. * * *

[2] - 87

Water may find other ways through the abutment if the drainage system is blocked. CHECK for water leaking through the abutment. Look for dampness or white staining at cracks or construction Joints. * * *

[2] - 88

CONCRETE ABUTMENTS, WING WALLS AND RETAINING WALLS CHECK for cracking of the concrete. Make a sketch showing where important cracks are Show only those cracks which are wider than 1 mm or those where water has come through. Look carefully for cracks: in pilecaps. where wing walls connect to the abutment. around parapet posts. near bearings even thin cracks can be important, so draw all cracks near bearings.

[2] - 89

CHECK for spalling of the concrete. Mark spalled areas on a sketch. * * * CHECK for corrosion of reinforcement. Look for rust stains from cracks and look where concrete has spalled. Where concrete has spalled, measure the concrete cover to reinforcement, and measure how much of the bar has been lost. * * * CHECK for poor quality concrete. Look for honeycombing and chemical attack. * * *

[2] - 90

MASONRY ABUTMENTS AND RETAINING WALLS,

CHECK for cracking of the masonry. Sketch all cracks wider than 3 mm at the widest place and where there is a step across the crack. At abutments, look carefully for any cracks which run underneath the bearing. If you find a crack which goes all round the bearing seating, make a note of it. This is very serious as the bridge may fall down. * * *

[2] - 91

Bulges in abutments are more serious than bulges in retaining walls. Any bulge, especially if there are horizontal cracks near it, can be serious. CHECK for bulging of the abutment or retaining walls. Make a sketch showing the area which is bulging. If you can, put a straight edge against the bulging part and measure how much it has moved out of line. * * * CHECK for poor pointing of the masonry. * * * CHECK for deterioration of the bricks or stones. * * *

[2] - 92

GABION ABUTMENTS AND RETAINING WALLS Gabions are often used as retaining walls and as abutments for simple bridges For example, they are sometimes used as abutments for timber bridges, Bailey bridges and steel beam bridges with timber decks.

When gabion abutments settle or bulge, the bridge deck may be damaged, or the road may be damaged. CHECK for settlement or bulging of the gabions. For abutments, look for any change which allows the bridge to settle. Make a note of the amount of settlement. For retaining walls look for any change of shape which allows cracks or hollows to form in the road.

* * * The gabion wires and the tie wires can be torn by debris carried by the river, or by large changes in shape of the baskets. They may corrode. CHECK for damage to the gabion wires or ties.

[2] - 93

TIMBER ABUTMENTS AND RETAINING WALLS

Abutments made of logs tied or spiked together are often used for log bridges. Retaining walls are sometimes made in the same way. Abutments for timber bridges are also sometimes made using timber piles and a timber cross beam. Because the timber is in contact with the ground, it will decay Insect attack is also very common. CHECK for decay. Look near the beams of the span. In timber piles, look near the ground or at water level. * * *

[2] - 94

CHECK for insect attack. Look where the beams sit on the abutment. * * CHECK for splitting of timber. *

* * * CHECK for loose or corroded binding cables. * * * CHECK for loose or corroded fixing spikes. * * * * *

[2] - 95

EMBANKMENTS AND FILL IN FRONT OF ABUTMENTS GENERAL If the base of the embankment is being washed away by the river, the slope will become unstable. CHECK for scour at the base of slopes. * * *

Where scour happens, the embankment fill may slip down, as in the photograph above. CHECK for slip of embankment fill. * * *

[2] - 96

If rainwater on the approaches to the bridge is not properly drained, water may run down the face of the embankment. After some time it might wash the fill away. This can damage parts of the abutment or wing walls. CHECK for erosion of the fill near the abutment. * * * Cracking of the road, or embankment edge, behind a retaining wall, might be a sign that the wall has moved. CHECK for cracking of the road or embankment edge. * * *

[2] - 97

If there is a hole deep in the embankment behind the abutment then, as soil drops into the hole, a small circular hole, called a 'pipe' will appear at the road surface, or in the embankment edge, as in the photograph below. This is called a 'piping failure' and shows that material is being lost from around the abutment. CHECK for piping failures of the fill behind the abutment.

[2] - 98

PILED WALLS Piled walls are used at the bottom of embankments to prevent the river scouring away the fill around the abutment. Piled walls can be made of steel (usually sheet piles), or timber (usually logs), and sometimes concrete. CHECK for forward movement of the piles caused by river scour. If large sections of the wall have moved forward, the problem is serious. Small movements are not serious. * * * CHECK for deterioration of the piles: Steel piles: Timber piles: Steel and timber piles: Concrete. look for corrosion. look for decay.

look for damage caused by debris. look for cracking and exposed reinforcement. * * *

[2] - 99

STONE PITCHING SLOPE POTECTION Stone pitching is one way of protecting the surface of a slope from being washed away by rain, or by water running off the road. If the embankment settles, the stone pitching may crack. Small cracks are not serious. CHECK for cracking of the stone pitching. Make a sketch of any cracks wider than 5 mm. The next inspector can check if the crack is getting wider or longer. * * * If the mortar pointing is poor, then the stone pitching will break up. CHECK for poor pointing. * * *

[2] - 100

The bank behind the stone pitching may be scoured by the river, or eroded by water running down the slope. CHECK for scour or erosion at the edges of the stone pitching.

* * * If the stone pitching has been badly undermined, pieces may break off. CHECK for pieces broken off the stone pitching. * * *

[2] - 101

GABION SLOPE PROTECTION Gabions or reno mattresses are often used to protect slopes from being washed away by rain water running off the road, or by the river.

Gabions can move a lot without being damaged, but the slope protection must still be complete. The picture above shows too much movement of the gabions. CHECK for too much movement of gabions.

[2] - 102

Make a note it: the slope protection has settled so far that the top of the slope is not protected (the road edge may have cracked). there is bad scour at the bottom of the slope.

* * * The wires which make up the gabion boxes and the wires which tie the boxes together must be in good condition. CHECK for damage to the gabion wires and ties. * * *

[2] - 103

RIP-RAP SLOPE PROTECTION For rip-rap to work, the rocks must be too heavy for the river to move them. CHECK if the rip-rap is being washed away. * * *

As the river scours the embankment the rip-rap will sink into the river bed. CHECK for bad settlement of the rip-rap. * * * * *

[2] - 104

BED PROTECTION It is important that the river bed under a bridge is not damaged by the river. CHECK for large holes in the river bed under or near the bridge. * * * If the river flows fast and the river bed is soft, the bed may be protected with: stone pitching concrete gabions rip-rap

[2] 105

STONE PITCHING OR CONCRETE BED PROTECTION AND APRONS The river often scours the river bed away at the edge of the bed protection. Small scour holes are not important. If a hole goes under the bed protection, you should report it. CHECK for scour at the edge of the bed protection. Make a note of where the scour is and draw a sketch. * * * Cracks in the surface of bed protection are usually because of settlement. CHECK for cracking. For mass concrete and stone pitching, only report large cracks. For reinforced concrete, report all cracks. To find cracks, first use a spade to scrape off any sand or debris. Then look carefully for any cracks. If you find cracks, draw a sketch of them. * * * CHECK for any signs of spalling of concrete or stones missing from stone pitching. * * *

[2] - 106

Stone carried by the river often cause erosion of the surface of the concrete. CHECK for erosion of the surface of the concrete or stone pitching. * * * For reinforced concrete bed protection, if there is spalling or erosion, the reinforcement may be corroding. CHECK for corrosion of reinforcement. * * *

[2] 107

GABION BED PROTECTION AND APRONS

Gabions or reno mattresses used as river bed protection have a very big advantage. They are flexible, so that when scour takes place, the edge of the gabion drops down to fill the hole. Because of this, scour is not usually a problem when the gabions are in good condition. CHECK for signs that the gabions have broken away from the pier or abutment. If this happens the river could destroy the bed protection very quickly. * * * CHECK for damage to gabion wires or ties. Look carefully for broken wires or corrosion of the wires. * * *

[2] 108

RIP-RAP BED PROTECTION AND APRONS Rip-rap is often used to prevent scour damage to piers It is sometimes necessary to put down more rip-rap to replace rip-rap lost during floods, or which has settled into the river bed. CHECK for loss of rip-rap. If you think that some rip-rap has been lost, make a sketch of the rip-rap you can see. Is this less than at the last inspection? If so, more rip-rap might be needed. * * * * *

[2] - 109

PIERS GENERAL Different kinds of piers are shown in this handbook on pages [1] - 18 and [1] - 19. For each bridge inspection, you will see what kind of pier you are to inspect by looking at the Construction Details on page 2 of the report form. To inspect the piers on some bridges you will have to use a boat.

CHECK for scour near the base of the pier. If the water is not low enough or clear enough to see, then use a long pole to feel if the river has caused scour. If you can feel under the base of the pier then the scour is serious. The photograph above shows bad scouring under piled piers * * [2] 110 *

If the caissons or piles are exposed they can be damaged, or the piles can corrode. CHECK for damage to caissons or piles, or corrosion of piles. * * * A pier will move if its foundation fails due to scour, or due to overloading of the bridge. It may also move due to impact from a boat or due to an earthquake. Movement of the pier can cause: one expansion joint to close and the other expansion joint to open. damage to a continuous superstructure.

CHECK for movement of the pier. To do this: look along the line of the piers, if there are other piers, to see if one is out of line. look along the top of the bridge to see if the bridge is a smooth line. If there is a hollow at a pier then the pier has settled. look carefully at the expansion joints and bearings.

* * * Boats or floating logs can damage river piers. Vehicles can damage the piers of bridges over roads. CHECK for impact damage due to boats, floating logs, or vehicles. * * *

[2] - 111

Debris can damage river piles

CHECK for debris against the pier. Light branches are not a problem unless there are a lot of them. * * * CHECK for vegetation growing on the pier. * * *

There may be an expansion joint in the bridge deck above a pier. If water leaks down past the expansion joint in the bridge deck and gathers near the bearing, it can do a lot of damage to the bridge. You will only be able to see if water is leaking through the expansion joint when you are under the bridge. CHECK for water leaking past the expansion joints. * * *

[2] - 112

CONCRETE PIERS CHECK for cracking of the concrete. Make a sketch showing where important cracks are. Look carefully for cracks. around bearings. in pilecaps. in beams over columns.

* * * CHECK for spalling of the concrete. Mark spalled areas on a sketch. * * * CHECK for corrosion of reinforcement. Look for rust stains from cracks and look where concrete has spa lied. Where concrete has spalled measure the concrete cover to reinforcement, and measure how much of the bar has been lost. * * * CHECK for poor concrete. Look for honeycombing and chemical attack. * * *

[2] - 113

MASONRY PIERS Cracks in masonry piers are sometimes caused by failure of the foundation. These cracks are usually long and go deep into the pier. Cracks around bearings may be caused by traffic loads or by the 'free' bearing not being able to move. A dangerous crack under a bearing is shown on page [1] - 69

CHECK for cracking of the masonry. Note especially any long and wide cracks, and any cracks near bearings. If you find a crack which goes all round a bearing, write a note on the report form. This is very serious as the bridge may fall down. * * *

Brickwork and masonry in a pier can lose some of its mortar pointing because of the flow of the river. Later, bricks or stones may be washed out. This can be serious, as the pier is then not so strong. CHECK for poor pointing. * * [2] - 114 *

The flow of water past a pier and the constant wetting and drying can, over a long period of time, damage the material of the pier. CHECK for deterioration of bricks or masonry. Test the surface by hitting it (not too hard) with a hammer. If pieces of the surface fall off, the bricks or stones are in poor condition. * * *

[2] - 115

STEEL PIERS Floating vegetation, such as branches and grasses, can catch on the steelwork and collect in the joints. This keeps the joints wet and makes them corrode. CHECK for debris in the joints of the steelwork. * * * CHECK for deterioration of the paint or galvanising. * * * CHECK for corrosion, especially near the water level If the pier is in salt water, look especially at the part of the pier between high tide and low tide. * * * Steel piers may be damaged by boats, or logs carried by the river in flood. If debris collects, then the river can push the pier very hard. CHECK for bends in steel members or joints. * * * CHECK for loose bolts or rivets. * * * CHECK for cracking. Look carefully near the bearings for loose bolts or cracks in the steel. * * *

[2] 116

TIMBER PIERS Floating vegetation and debris, can catch on the timber and collect at the joints. This keeps the joints wet and makes the timber decay. CHECK for debris in the joints. * * * CHECK for decay. Look carefully at the joints. Look carefully near the water when the pier is in fresh water. * * * CHECK for insect attack. Look carefully below the water level when the pier is in salt water. Attack by teredo worms is very common here. * * * CHECK for splitting of the timber. Note large splits, splits near bolt holes, and any splits where water can sit and cause decay. * * * CHECK for loose bolts or pins at joints. * * * The timbers of the pier may be bent if the pier is hit by a log or boat. They may also bend if the bridge is overloaded. CHECK for bends in timbers. * * *

[2] - 117

Most timber piers have some parts made of steel, such as plates at the joints. Often these steel joint plates and fixings corrode at river level. CHECK for damaged or corroded steel parts. Look for: Deterioration of paint or galvanising. Corrosion. Bends. Cracking, especially near welds. * * * * *

[2] - 118

CULVERTS GENERAL Because culverts are small, they are often blocked by debris or vegetation. In desert areas, culverts can be blocked by sand:

CHECK for debris, vegetation, etc, in or near the culvert. * * * Sometimes, when a culvert is built, the ground is not properly compacted If the ground is not properly compacted, parts of the culvert may settle. When this happens the water may not flow properly through the culvert, or there may be a dip in the road above. CHECK for settlement of parts of the culvert. * * *

[2] - 119

Flood water can scour the ends of the culvert. or the edge of the apron, or the road embankment. CHECK for scour at the ends of the culvert, or at the edge of the apron. Mark where there is scour damage on a sketch of the culvert.

On very small culverts, like the one shown on page [1] 84, the last three checks are all that you can do The following checks are for larger culverts and culverts with wing walls and aprons.

[2] - 120

CONCRETE CULVERT BARRELS If you need them, there are notes on inspecting reinforced concrete on pages [1] - 43 to [1] - 55 CHECK for cracking. * CHECK for spalling. * * * CHECK for corrosion of reinforcement. * * * CHECK for poor concrete. Make a sketch to show any cracking, spalling, corrosion or poor concrete. * * * * *

[2] - 121

CORRUGATED STEEL CULVERTS Corrugated steel culverts can be different shapes. Mostly they are circular, oval or arched. A culvert can change shape because it was badly constructed, or because the traffic above it is too heavy. If a corrugated steel culvert changes shape it can collapse. Even a small change in shape of a culvert can make it collapse. CHECK for change of shape of the culvert barrel. * * * Corrosion can be a problem with corrugated steel culverts. Most culverts are galvanised, but some are protected by paint. Stones and other debris can damage the galvanising and paint. If this happens, there will be corrosion. CHECK for damage or deterioration to paint or galvanising. * * * CHECK for corrosion of steel Corrosion can also begin on the outside of the culvert, which you cannot see. A common place to find corrosion is next to the bolts between sections of the culvert. * * * The bolted joints are very important to the strength of the culvert. All bolts must be tight and in good order. CHECK for loose or corroded bolts. * * *

[2] - 122

CULVERT APRONS Like bridges, culverts may be protected from scour by aprons. These aprons can be made of gabions or reno mattresses, concrete, or stone pitching. Sometimes, scour causes the apron to settle at the edge Concrete or stone pitching may crack and break up. CHECK for cracking and damage to concrete or stone pitching. * * * Gabion or reno mattress wires or ties may corrode or be damaged by debris. The stones in the gabions may then be washed away and the gabion will not do its job. The gabion baskets can settle or move due to scour. CHECK for damage to gabions. * * *

[2] - 123

HEADWALLS Headwalls retain the embankment fill at the ends of the culvert. They can be made of concrete or masonry. If a headwall settles or is pushed forward, water can wash away the embankment fill at the joint between the headwall and the culvert barrel. This can damage the embankment. CHECK for movement of the headwalls. Look carefully at the joint between the headwalls and the culvert barrel. * * For Concrete HeadwalIs. *

CHECK for cracking, spalling, corrosion of the reinforcement, or poor concrete. Make a note of the kind of problem you find, and make a sketch to show where the problem is, and how big the problem is. * For Masonry Headwalls. * *

CHECK for cracking, poor pointing or deterioration of the bricks or stones. Make a note of the kind of problem you find, and make a sketch to show where the problem is, and how big the problem is. * * * * *

[2] 124

APPENDICES

A. WORD LIST. B. EQUIPMENTLIST. C. SAFETY DURING INSPECTIONS. D. EXAMPLE OF A COMPLETED REPORT FORM.

APPENDIX A WORD LIST This is a list of engineering words that are used in this book, and their meanings. The first time these words appear in the book, they are printed in CAPITAL LETTERS, so that you will know that they are explained here. In some of these notes you are asked to look at a page where there is a picture or some more information that will be helpful. The picture on page [2] - 76 shows most of the parts of a Bailey bridge. The Bailey bridge parts not shown in the picture are described in this word list. ABUTMENT: end support of a bridge, see pages [1] - 6 and [1]-17 APPROACH EMBANKMENTS: the earthwork that carries the road up to the bridge. APPROACH ROAD: the road near the bridge. APRON: bed protection at the mouth of a culvert, see page [2]-120 ARCH: curved bridge structure, see page [1] - 12. BAILEY BRIDGE: a type of steel truss bridge designed for quick assembly, see page [2] - 75. BALLAST WALL: abutment wall above bearing shelf, see page [1]-17.

A1

BANK SEAT ABUTUMENT: an abutment set well up the river bank; above the usual river level, see pages [1] - 14 and [1] -15. BARREL (of arch): the main part of an arch which supports fill and road, see page [1] - 12. BARREL (of culvert): the pipe or box part of a culvert where water flows. BEAM: narrow member which spans between supports; for example a log in a log bridge. BEARING: this part is between the deck and the abutment or pier. It carries the weight of the deck, and slides or hinges to allow the deck to move. See pages [1] - 16, [1] - 20 and [1] - 21. BEARING BOLT: type of fixing for joints in steel or timber, see page [1] - 65. BEARING SHELF: part of the abutment, where the superstructure rests, see page [1] - 16. BED (RIVER BED): the bottom of the river. BED (AS IN 'BEDDED IN MORTAR'): fix in place with (mortar). BOOM : top or bottom part of a truss, see page [1] - 1 0. BOX GIRDER: a hollow beam with a box shape, see page [1]-7 BRACING: parts of a bridge which help to keep it stiff and not change shape, see page [1] - 10.

A-2

BULGING: where a flat place changes shape and is bent. CAISSONS: type of masonry or concrete foundations, built like a tube, see page [1] - 13. CANTILEVER: a beam, fixed at one end and free to move at the other. CHORD: On a Bailey bridge, a chord is a long steel plate bolted to the truss to make it stronger COMPRESSION (STRONG IN COMPRESSION): being pushed together, see page [1] - 44. CONDENSATION: water from the air that settles on a cold surface. CORRODE, CORRODED (STEEL): to be attacked by rust. CORROSION: damage done to steel (or other metal) by air, water, salts, etc. CORRUGATED STEEL: thin sheets of steel which have been shaped to make them strong. COVER (COVER TO THE BAR): the thickness of the concrete over the reinforcement at that place. CROSS GIRDERS: steel beams that connect two trusses. CULVERT: a tube or box to carry water under a road, see pages [1]-83 and [1]-84. DEBRIS: rubbish and other unwanted things.

A-3

DECAY: rotting of wood, making it soft and weak; caused by dampness and fungus. DECK (DECKING): top of bridge superstructure, see page [1]-7. DETERIORATE, DETERIORATED: to become worse; to get into a bad condition. DOUBLE STOREY BRIDGE: bridge made of two trusses, one on top of the other, each side of the bridge. DOWNSTREAM: where the river flows away from a bridge DRAINAGE: system for taking waste water (usually rain water) away. EMBANKMENT: soil bank which supports the road, see pages [1] - 6and [1] -17. EROSION, ERODING: removal of earth by wind, rain or flowing water. FILL (FILL IN FRONT OF THE ABUTMENT): soil placed in front of the abutment. FIXED (FIXED BEARING): not able to move, see page [1] - 21 FLANGE: the top or bottom of a girder, see page [2] - 47 FOUNDATIONS: the lowest part of the bridge which sits on or in the ground.

A-4

FUNGUS: type of plant which grows on decayed timber. GABION: wire basket filled with stones, see pages [1] 81 and [2] - 102. GALVANISING: a thin layer of zinc on steel to protect it against corrosion. GIRDER: a beam, usually made of steel, see page [2] 47. GROYNE: a wall built to change the flow of a river and protect the river bank from scour. HEADWALL: a wall at the end of a culvert to hold the soil fill above the culvert pipe. HONEYCOMBING: badly made concrete with lots of holes, see page [1] - 54. IMPACT: to hit hard, as when a vehicle hits a bridge parapet. INVERT: the bottom of a channel, usually paved. KERB: the step between the road and footpath, see page [1] 7 LAMINATIONS: thin layers, like pages of a book. LANDSLIDE: a lot of soil and rocks slipping down a mountain or hill. LUBRICATION SYSTEM: a system to supply oil or grease to a mechanical bridge bearing

A-5

MAINTAIN (as in WELL MAINTAINED): to look after carefully and repair when it is necessary. MARSDEN MATTING: a type of steel decking with holes, see page [2] - 22 MASONRY: bricks or stones set together with mortar. MASS CONCRETE: concrete without any steel in it. MOISTURE: some water or dampness. OVERBRIDGE: a bridge over a road. OVERLOADED: carrying too much weight. PANEL: a flat frame or plate. PARAPET: a wall or rail along the edge of a bridge, see page [1] -6. PIER: a support between abutments for bridges with more than one span, see page [1] - 9. PILE: a long, thin, foundation driven deep into the ground, seepage [1]-13. PILED: WALLS: walls made of long pieces of material driven into the ground PLATE (STEEL PLATE): a flat, stiff piece of steel. POINTING: the mortar between bricks or stones in masonry. PRESTRESSED (as in PRESTRESSED CONCRETE): a way of making concrete stronger with steel bars.

A-6

PTFE: a slippery white material used in bearings. REINFORCEMENT: steel bars in reinforced concrete, to make it stronger. RENDERING: a thin layer of mortar put on masonry or brickwork to protect it. RENO MATTRESS: a long thin gabion. RETAINING WALL: a wall to hold back soil, see page [1]-17. RIP-RAP: a layer of loose rocks to protect a river bed from scour, see page [2] - 104. RIVER TRAINING WORKS: constructions to stop the river changing its path, see page [2] - 39. SAFETY PINS: the small clips on a Bailey bridge, that stop the panel pins from falling out. SCOUR: holes in the river bed or bank caused by the flow of the river. SEATING : the place where one part rests on another part For example the place where a bearing is set on a pier or abutment. SERVICES: cables and pipes belonging to other authorities. SETTLEMENT: small movement downwards of a structure.

A-7

SHEET PILED WALL: a wall made from steel panels or timber boards hammered into the ground. SLAB (CONCRETE SLAB):a large piece of concrete (for example a bridge deck). SPACER PLATES: plates which are put between two other parts. SPALL: to break off in a piece. SPALLING: an area where concrete has broken away SPAN : the part of a bridge, or the distance, between the supports. SPANDREL (SPANDREL WALLS): the side walls of a masonry arch bridge, see page [1] - 12. SPREAD FOUNDATIONS: wide base to a pier or abutment, usually made of reinforced concrete. STIFFENER: piece of steel used to strengthen a steel plate or girder and make it stronger against bending, see page STONE PITCHING: stones set in cement mortar to cover a sloping bank or an invert. STRINGER: a horizontal deck member, see page [1] 10. STRUCTURAL (STRUCTURAL MEMBER): an important part of a bridge, for example a girder.

A-8

SUBSTRUCTURE. all the parts of a bridge, except for the superstructure, i.e. abutments, piers, etc. SUPERSTRUCTURE: all of the deck, including parapets, trusses, beams and running surface. SUSPENDED SPAN: the middle span of a cantilever bridge, see page [1] -25. TENSION (STRONG IN TENSION): pulling apart, see page [1] -44. TRUSS: a type of girder made from many small beams, see page [1]-1O. TRUSS BRIDGE: a bridge which has truss spans, see page [1]-1O UPSTREAM: the direction where the water in the river is coming from VIBRATION: repeated small movements caused by a heavy vehicle or perhaps an earthquake. WATER BAR: a seal to stop water, see page [1] - 24. WEB: the part of a girder that joins the flanges, see page [2]-47 WEEP HOLE: a hole to allow water to come through, see page [1]-17. WING WALLS: walls which are at the side of the abutment and part of it, see page [1] - 17.

A-9

APPENDIX B EQUIPMENT LIST This is a basic list of equipment. Each inspector should add other things that he finds are helpful. All Bridges This handbook. Waterproof clipboard for inspection reports. Waterproof pencil, pen or paint for' marking concrete or steel Flash light (torch) Straight edge (at least 2 m long). String line. Measuring Tapes (30 m and 3 or 5 m long). Bush knife or similar tool, to clear vegetation. Small shovel or trowel. Wire brush. Small paintbrush and paint, for repainting areas damaged during the inspection. Hammer (350-450gm). Set of spanners. Ranging rod or other long pole. Crack measuring gauge.

B-1

Timber bridges only Square spike. Hand drill and 5 mm diameter drill bits 5 mm wooden plugs. Saw, for cutting off wooden plugs. Squirt can of creosote or other wood preservative. Access and safety equipment A first aid box. Traffic warning signs and road cones. High visibility waistcoat. Safety helmet. Safety harness. Lifefloat. About 20 m of light rope. A ladder will be needed occasionally, and for some inspections a boat will be required.

B-2

APPENDIX C SAFETY DURING INSPECTIONS Bridge inspections can be dangerous if the inspector is not careful. These notes will help the inspector to plan his work with safety in mind. TRAFFIC During the inspection you will be walking on the bridge and its road approaches looking at details of the bridge. You might not notice oncoming traffic. Carry with you signs such as SURVEY IN PROGRESS, or DANGER or ROADWORKS, or similar signs which will warn drivers to expect people on the road. Always use the signs except on very lightly trafficked roads with clear visibility. Wear a bright coloured 'warning' jacket. To properly inspect some bridges, such as through trusses, you must climb up above the traffic. Never do this without making sure that vehicles cannot come near you. Protect the work area with cones, have assistants control the traffic or close the road for a short time, if necessary.

C-1

DANGEROUS PLACES Many small bridges over shallow water can be inspected safely by a person on his own. To properly inspect other bridges you must go into dangerous places, such as high up or near deep water. If you are going to inspect a bridge like this, take someone along (perhaps a driver) to make sure you are safe. Remember two important things: 1. Always carry a rope and safety harness with you. Use them if you have to go into a dangerous place. 2. Do not go into dangerous places without an assistant Do not miss out parts of an inspection because it is difficult to get close to a place you should look at carefully. If necessary, move to another bridge and come back to the difficult bridge on another day with an assistant to carry out the inspection that you could not do alone. Keep a look out for animals and insects which may harm you. Bridges provide lots of places where bees and wasps can build nests. Snakes and crocodiles may be found in sheltered parts under bridges. Be very careful looking at bearing shelves, especially just after a flood. If you have to go into enclosed spaces, such as box girders, make sure that the manhole covers have been off for a long time to let fresh air in before you go in. People have died going into enclosed spaces where the air is bad. Always have someone standing outside, so that you can call to him for help.

C-2

CHEMICALS Some of the chemicals used in timber treatment can be dangerous. If you make any holes during the inspection of a timber bridge, these should be filled with timber preservative. Chemicals such as Pentachlorophenol (P.C.P.) or coal-tar epoxy creosote can be used, but they are harmful to people, animals and fish. When using timber preservatives remember: take care not to get the chemical onto your skin, use gloves. wash off any accidental splashes of chemical at once. do not smoke: some of the chemicals catch fire easily. try not to spill chemicals into the water: they will kill the fish. do not leave the chemicals where children or animals can touch them. Many children die every year because chemicals are left around. do not let any chemicals, liquids, gases or dust get into your nose or mouth or eyes. If it does by accident, wash it out as quickly as you can. If anyone drinks a chemical by mistake, take him to a doctor at once. Do not wait to see if he gets ill first.

C-3

APPENDIX D EXAMPLE OF A COMPLETED REPORT FORM

The following example of a completed report form is for a bridge with: one steel pier, concrete abutments, a composite steel and concrete deck. The abutment pages are included twice; once for each abutment. The superstructure pages are also repeated because there are two spans All of the report form pages are shown here, including pages not needed for the inspection of the bridge detailed above. This is because the report form is also a detailed index for references to part [2] of this book.

BRIDGE INSPECTION REPORT FORM

CONSTRUCTION DETAILS (from the Inventory)

7a

9a

10

10a

11

12

13

13a

14

15

15a

16

17

18

19

Transport and Road Research Laboratory Overseas Unit Department of Transport Overseas Development Administration

Overseas Road Note 8

A users manual for a program to analyse dynamic cone penetrometer data

Transport and Road Research Laboratory Crowthorne Berkshire United Kingdom 1990

ACKNOWLEDGEMENTS The work described in this report was carried out in the Overseas Unit of the TRRL. The report is based on a text prepared by Howard Humphreys and Partners, Consulting Engineers. The computer program itself is based on an original mainframe program written by Neil Howitt and Dr. J. Rolt and modified by Trevor Bell of Contract Data Research Ltd. for use on microcomputers.

First published 1990

OVERSEAS ROAD NOTES Overseas Road Notes are prepared principally for road and road transport authorities in countries receiving technical assistance from the British Government. A limited number of copies is available to other organisations and to individuals with an interest in roads overseas, and may be obtained from: Transport and Road Research Laboratory Crowthorne, Berkshire, RG11 6AU England

Crown copyright 1990 Limited extracts from the text may be produced provided the source is acknowledged. For more extensive reproduction, please write to Head of Overseas Unit, Transport and Road Research Laboratory

CONTENTS Page 1 1 2 3 3 3 3 4 4 4 4 4 5 5 5 5 6 6 6 7 7 8 8 8 8 8 8 8 9 11 11 11 11 12 12 12 12 12 13 14 14 14 14 15 15 16 16 16 16 17

1. 2.

3.

INDEMNITY NOTE INTRODUCTION INSTALLATION INSTRUCTIONS PROGRAM OPERATION 2.1 Starting the Program 2.2 TRRL DCP Option List 2.3 How to Enter Data 2.4 Amend/Review Data 2.5 Do not... 2.6 What to do in the Event of a Computer Failure 2.7 Always... 2.8 Command Line 2.9 OK? Prompt 2.10 Recall File 2.11 Save File 2.12 Data Filenames PROCESSING DCP DATA 3.1 Introduction 3.2 The NEW Command 3.3 The RECALL Command 3.4 The EDIT Command 3.5 The INSERT Command 3.6 The DELETE Command 3.7 The KILL Command 3.8 The SAVE Command 3.9 The MORE Command 3.10 The OPTIONS Toggle Command 3.11 The ANALYSE Command 3.12 The VIEW Command 3.12.1 I/O: Zoom Functions 3.12.2 S/D/E/X: Panning Functions 3.12.3 G: Grid Toggle 3.12.4 V: View Point 3.12.5 P: Print or Plot 3.12.6 Esc: End Graphics 3.13 The PRINT Command 3.14 The FILE Command 3.15 The TOL Command 3.16 The Structural Number (SN) Command FILE MAINTENANCE 4.1 Introduction 4.2 Assign Data Drive 4.3 Paper Length/Beeper Control 4.4 Enter Printer Controls 4.5 Assign Graphic Devices 4.6 Test Graphics Screen 4.7 Test Graphics Printer 4.8 Specify Graphics Parameters 4.9 Batch Print 4.10 CBR Relationships

INDEMNITY NOTE
Note must be taken by all users of this program that TRRL DCP is not intended to replace normal engineering judgement. The procedures described in this manual are intended for users who already have a thorough understanding of DCP analysis and are capable of deciding which method of analysis is most appropriate for individual situations. The user must be aware of the limitations of the program and, most importantly, that incorrect data input must lead to incorrect data output. The user should be capable of assessing the accuracy of any results produced. No warranty can be given on the validity of the results and the ultimate responsibility for acceptance and subsequent use of any results lies solely with the user. TRRL DCP is the copyright of: The Overseas Unit of the Transport and Road Research Laboratory Crowthorne Berkshire United Kingdom

INTRODUCTION
TRRL DCP is a program designed to help with the interpretation and presentation of DCP test results. TRRL DCP has been written to operate under MSDOS and PCDOS compatible microcomputer operating systems. The program is totally interactive with `Help' facilities, and is highly userfriendly with rigorous error checking at data input. Throughout this manual, whenever the user is required to enter text in response to prompts from the program, the text is denoted thus: [TEXT]. Input commands in the form of single keystrokes are denoted: <COMMAND>.

1.

INSTALLATION INSTRUCTIONS

printer. In both cases a standard `test card' should appear. By <ESC>aping back through the options the user will leave the program. The software will now be fully installed and may be used at any time by typing [DCP] <RETURN>. A data file named TEST is available on disc and may be used to demonstrate the scope of the software. This file is referred to throughout the manual. Once the program has been successfully installed a back-up copy should be made and stored safely.

The TRRL DCP analysis software is contained on an installation floppy disc which MUST be installed on the host computer before it will operate. (i.e. the software will not work from the floppy disc). There are three stages of installation. Firstly the relevant files are copied automatically from the installation disc to the host computer's hard disc, then the type of computer display and graphics standard is specified and finally the data disc drive, printer controls, test graphics, etc are defined. The three stages are carried out as follows. (It is assumed that the floppy disc is in drive A and the hard disc is drive C. The appropriate characters must be substituted for A and C if other disc configurations are used.) Insert the floppy disc in drive A and type: A: [INSTALL A C] <Return>. This instruction causes the required files to be copied from drive A to a new sub directory named DCP on drive C. On completion, remove the floppy disc and store safely. The computer graphics screen must be either CGA, MDA, EGA or HERC and will be either MONO or COLOUR. To install the graphics system type the following sequence: e.g. for a Hercules system: [DCPINST HERC MONO] An unacceptable combination of graphics and screen type will result in an error message. Start the program by typing [DCP] <RETURN>. Press any key to clear the first display and choose Option 2 (File Maintenance) from the main option list. A new option list will appear. Select Option 1 to specify the data drive (i.e. where data files will be stored). Option 2 to set paper length. Options 3 and 4 should already have been preset and need not be changed. Options 5 and 6 should be used to test the graphics screen and to test graphics on the

2.
2.1

PROGRAM OPERATION
Starting the Program

The computer should be started in the normal way. TRRL DCP will be in a sub-directory of the hard disc. To start the program change to the subdirectory, and type [DCP] <RETURN>. After loading the necessary graphics devices the computer will display the following notice to the authorised user.

is also shown towards the bottom left-hand corner of the screen and will be selected if the <RETURN> key is pressed. Alternatively the user may press the number shown in front of the required option. On completion of any option, the user is returned to this main option list. If the <ESC>ape key is pressed, the user is offered the opportunity to end and to return to the computer's operating system. 2.3 How to Enter Data

Pressing a key at this point gives access to the main option list. 2.2 TRRL DCP Option List

The programs operate in an `interactive' style. Questions or requests for information are displayed within `fields' on the computer screen and responses from the operator are made through the computer keyboard. Every time an item of information has been entered through the keyboard, the <RETURN> key must be pressed for the computer to accept the data, if the data field (spaces between square brackets) has not been entirely occupied. However, if the response totally fills the data field the computer automatically accepts the data. For computers with a reverse video facility, the data field and square brackets are shown in reverse video. The computer will wait indefinitely until a response is entered to a request and, subject to the conditions above, the <RETURN> key is pressed. The program displays a series of `pages' on the computer screen. The name of the program appears in the top left corner. At the bottom lefthand side of the screen is an `error' line for the display of error messages, while in the bottom right-hand corner of the screen the user is shown which key (usually <ESC>ape) to press to end a particular operation and proceed to the next logical step in the program.

The <ESC>ape key may be used at most points in the program to leave certain pages, either when data entry is completed or the user wishes to abort. Also shown at the bottom right hand corner is the `Help' key. On most computers the <SHIFT> and [?] must be pressed simultaneously. The user may move around the option list by using the left or right arrow keys until the required option is highlighted with an asterisk. This option If help is selected a message will be displayed at the top of the screen which will give further

information or instructions on how to proceed with the program. A typical example is:

On most computers, these commands are available using the arrow keys on the keyboard. The simplest way to erase data in a complete cell is to enter a space followed by a <RETURN>. On reviewing data a <RETURN> skips to the next data field.

2.5

Do Not...

The program incorporates an error-checking system to ensure that the data entered is logical and consistent in relation to the data requested. At the bottom of the screen is an error line which will display error messages in the event of invalid data being entered. For example, if a number is being requested and an alphabetic character is entered in error, an `Invalid Data' message will appear as follows and the correct data must be re-entered.

The program is equipped with thorough error checking routines but there are still a number of operations that the user may carry out which would have disastrous consequences but which cannot be overruled by the program. (a) NEVER RESET or BOOT the computer while the program is in operation. This will re-set the whole system with the loss of data. (b) NEVER open the disc drive doors and remove or replace discs unless specifically instructed by the computer. This may result in a system failure and the loss of data. (c) NEVER press the <CTRL> and [C] keys simultaneously while the program is in operation. This may stop the program and cause the loss of data.

The invalid data message may appear for many reasons and will usually be obvious. However, the `Help' system may give some indication of why data is not accepted.

(d) NEVER turn the computer off until instructed to do so. 2.6 What to do in the Event of a Computer Failure If for any reason, there is a failure, it will be fairly obvious which data has and has not been recorded successfully. If there is any doubt, the back-up copies of the data should be used to start the data input session again. 2.7 Always . . .

2.4

Amend/Review Data

At any stage of entering or amending the data a simple editing facility is available. This involves the use of two keys which are pressed simultaneously to perform each function, one of which is the control <CTRL> key. <CTRL> + [X] move down to next batch of fields <CTRL> + [E] move back to last field <CTRL> + [D] move right on command line

There is one cardinal rule for any computerised system. ALWAYS make back-up copies of all current data after entering ANY new information. The copying procedure will vary between the different types of computer and reference should be made to the Operating System manual and Section 2.12 for details of relevant filenames. 2.8 Command Line

Throughout the program the user is presented with a command line near the bottom of the screen thus:

If your screen supports reverse video, the current option will be shown in reverse video, and will appear to the right of the word `Command'. To select options, the user may press the space bar or the left and right cursor keys, until the desired option has been highlighted. To enter the option the user presses the <RETURN> key. The user may also jump to an option by entering the first letter of the required option. To exit from the command line the user must press the <ESC>ape key. 2.9 OK? Prompt

required. A list of files already saved on disc can be displayed by typing [DIR]. 2.12 Data Filenames

Throughout the program, data files are created and stored on the data disc. [FILENAME] is a user-specified filename of up to 8 characters. A three character filename extension, [FILENAME]. DCP, is automatically given to each data file. The user is not required to enter the drive specification nor the filename extension in response to filename requests. When batch printing a file, a new file is automatically created with the same filename as the data filename but with the file extension .FIL. (See Section 4.9).

Throughout the program the user will be asked to confirm that the displayed data is OK with the following message at the bottom of the screen.

The user has three possible responses to this request: Enter [Y] for yes and continue Enter [N] for no, go back and review Enter <ESC>ape for no and quit this operation. 2.10 Recall File

The user may recall data files previously saved on disc by moving the cursor to the RECALL command and pressing <RETURN>.

Use of this option will clear any data in memory and load the recalled data. The user has only to enter the filename. Sub-directories, data drives or file extensions must not be specified. Alternatively, the user may enter [DIR] to obtain a list of those files which may be recalled. 2.11 Save File

The user may save data files on disc by moving the cursor to the SAVE command and pressing <RETURN>.

Use of this option will save the data currently displayed to the specified file. If the specified file already exists it will be overwritten. Note that it is only necessary to enter the filename and that file extensions, data drives and sub-directories are not

3.
3.1

PROCESSING DCP DATA

Introduction

Site:

This option allows the entry of new data as well as the editing and storage of existing data. Data printouts and plots of the DCP curves may also be obtained in this option. On selecting the 'Process DCP Data' option, the following screen is displayed:

Along the bottom of the screen a menu is displayed and these options are explained in turn below.

3.2

The NEW Command

The NEW command is used to clear all existing data and start the entry of new data. If data already exists, the user is required to confirm that this data is to be cleared from memory as follows:

Description of the site [20 characters] Section no. Details of the section no. [20 characters] Text: Details of the test no: [20 characters] Chainage: Details of the chainage [20 characters] Direction/lane Details of the direction and lane [20 characters] Position/offset Details of the position of offset [20 characters] Date: Date of the test [20 characters] Start layer: Description of the surface or upper layer [20 characters] Condition: Condition of the start layer [20 characters] Zero error (mm): If, for various reasons while conducting the DCP test, the reading is not zero at zero blows the reading should be entered here. Surf thick (mm): Enter the thickness of the surface layer in millimetres. This layer will generally be a very hard layer through which a hole has been chiselled or drilled. Extens @ lineHere the user should enter the line number at which any extension rod was added to the DCP apparatus. At this point the scale reading will have been recalibrated and the user will have entered two readings for the same number of blows. In the example below an extension rod of 400 mm was added after 169 blows when the reading was 466 mm. The new scale reading is 66 mm after which the data entry can continue as normal.

If data is not to be lost irretrievably it should first be saved using the 'save' command as detailed in Section 2.11. If the user enters [Y] to the 'Current data saved OK?' prompt then the data currently displayed will be cleared both from the screen and from the memory, and the cursor will move to the first field in the table ready for the input of new data.

(i) Descriptive data: As the name suggests this data describes the site and conditions and is found on the top half of the screen. (ii) Test data page 1: These are the blows and corresponding readings for the first 50 data records (iii) Test data page 2: These are the blows and corresponding readings for the second 50 data records
During data input the screen will move automatically to page 2 after page 1 has been filled with data. In order to move back to view page 1 again, it is necessary to save the file (see section 2.11) after which it is possible to select the edit facility (see section 3.4) and make any corrections required. Once the data has been entered the user should <ESC>ape in the normal way, see section 2.3 and a recalculation message will be displayed. After the program has made provision for any extension rods the above data input will appear as follows. The user is required to enter yes [Y] to confirm that the recalculation is satisfactory. 3.3 The RECALL Command

The RECALL command is used to recall and display a previously saved data file, see Section 2.11. This data file will replace any data already displayed. The user is required either to enter the filename or to enter [DIR] to produce a list of all existing filenames. If a filename has already been specified for the data shown, its name will be displayed in the data box. If the <RETURN> key is pressed, the filename will default to the display name.

After the data has been recalled from disc it is displayed. Any errors will be displayed as an error message to remind the user of invalid data. The name of the current DCP data file will be displayed in the top right hand corner of the screen, as follows:

Blows/Rdng:

Enter the number of blows in the left hand column and the corresponding reading or depth of the DCP probe in the column to the right. Once all the required data fields have been entered, (100 max) <ESC>ape from the New option.

The DCP data may then be edited, viewed, printed etc. If the user tries to recall a file which does not exist an error message will appear `File not found'.
3.4 The EDIT Command

The data in each file is split into 3 sections called:

The EDIT command is used to edit data in an existing data file. The cursor moves to the site cell and may be moved from cell to cell or field to field either by entering modified data or by

pressing the <RETURN> key. If the only data to be modified is further down the file then the down arrow key may be pressed to move down until the line to be changed is reached. The <ESC>ape key may be pressed to end an edit of the particular page as explained in section 3.2. To move back to page 1 from page 2 it is necessary to follow the procedure outlined in section 3.2. If an error exists in the data the user will be reminded by the `Invalid data' message which will appear at the bottom of the screen. Once EDIT is selected the cursor will automatically go to the line containing the error. When the user <ESC>apes from the EDIT option, the file is checked for errors and if any are found they are displayed in a message as before. The user is then prompted for yes [Y] or no [N] as to whether he wishes to leave the EDIT option. If no [N] is selected because an error message has been displayed by the error analysis, the cursor will go to the box containing the erroneous value. 3.5 The Insert (INS) Command

enters the required line number and presses <RETURN>, he is then asked to confirm this choice by yes [Y] or no [N] in answer to the question `Are you sure?' If [Y] is selected the deletion is performed. 3.7 The KILL Command

This command is used to delete an entire data file which has been recalled and is currently displayed. If the data displayed does not have a filename (i.e. has not been previously saved) then a message `No data file to kill' will appear. Once kill has been selected the user will be required to confirm the action in response to the message `Delete (filename)-Are you sure OK? If [Y] is entered the file is removed from the disc, the screen and the memory. 3.8 The SAVE Command

This command is used to add data either in the middle or at the end of an existing data set. The user is required to enter the line number at which the new data is to be inserted.

This option will store data in a user specified disc file, see section 2.11. Once a filename has been entered the file will be saved with this name. Care must be taken not to give the file the same name as one that already exists as this will overwrite the old file without any warning. If the user types [DIR] for a filename a list of all existing filenames will appear on the screen. However, if the <ESC>ape is pressed the save option will be abandoned but the data will remain on the screen, so that SAVE can be reselected if necessary. 3.9 The MORE Command

The existing data is automatically re-displayed showing blank data boxes for that line. The Edit command is automatically invoked and the user may move the cursor to the new line and enter the completion of the data entry the <ESC>ape key may be pressed and a re-calculation and error analysis will check the integrity of the data. 3.6 The Delete (DEL) Command

Once a file is recalled to the screen the first 50 sets of data on page 1 are displayed. The MORE option displays the second 50 sets of data on page 2. If the MORE option is selected again the first 50 sets of data are returned to the screen. 3.10 The OPTIONS Toggle Command This command displays a second menu as shown.

This allows the user to delete a line of data and automatically moves all the succeeding data up to fill the place of the deleted line. The user is prompted for the line number by the following:

These menu options are selected in the same manner as described in section 2 i.e. by entering the first letter of the option required or by moving the cursor and pressing the <RETURN> key. This menu also has an OPTION command which will take the user back to the first menu. 3.11 The ANALYSE Command This option calculates layer thickness and CBR values for the data entered, the processed data appearing on the screen thus:

The data will be re-displayed and re-calculated to check that the new data is valid. When the user

These options perform the following plots. (i) Blow-pen: Plots blows versus penetration, (mm) see figure 3.2. (ii) mm-blow: Plots a histogram of the penetration rate in mm/blow versus depth in mm. No boundaries are drawn to distinguish the individual layers but they may be identified by the `steps' in the plot, see figure 3.3. This plot is useful for assessing the boundary positions chosen by the program. (iii) CBR: Plots a histogram of CBR versus depth, see figure 3.4. Plots all the above on one drawing, see figure 3.5.

When the user selects ANALYSE, the program identifies layer boundaries in the material based on the rate of penetration, which is an indicator of the material strength. The sensitivity of this routine can be adjusted by changing tolerances and the EM value (see section 3.15). The results table displays the strength, CBR and thickness of each of the layers identified by the program. The layers are zones of constant penetration rate within the given tolerance and are shown graphically in Figure 3.1. Transition zones are defined between layers when the gradient of the blow/penetration curve changes gradually over a certain depth thus not permitting the positioning of a distinct layer boundary. A bracketed CBR value may appear next to the calculated CBR value for a layer when the calculated value is very high. In these cases the bracketed value is suggested as being more realistic for analysis purposes. The formula used for the CBR calculation is displayed beneath the table (see Section 4.10). 3.12 The VIEW Command The VIEW command is used to display analysed data in graphical form on the screen or to plot it on a printer or plotter. When VIEW is selected the following menu appears on the screen:

(iv) All:

The format of the initial graphics display is selected automatically. A scale is fixed so that the complete plot can appear on the computer screen with a superimposed grid. Thereafter a number of graphics functions may be used. A command line listing the various graphics display options is located at the top right hand corner of the screen. Each function is selected simply by typing the appropriate characters.

Figure 3.1

Figure 3.2 Blows versus penetration curve

Figure 3.3 Penetration rate versus depth

NOTE: Beware, on the CBR scale values are shown getting smaller as you move along the "x" axis.

Figure 3.5

Combination of Figures 3.2, 3.3, and 3.4

Figure 3.4

California Bearing Ratio versus depth

10

3.12.1

I/O: Zoom Functions

The I and O zoom functions allow a zoom-in or zoom-out on the current plot. Two rates of zooming are available. Pressing the [I] or [O] keys will zoom slowly, each step being 0.1 times the original scale, but zooming can be speeded up by pressing the <CTRL> key simultaneously which doubles the zooming rate. After using [I] or [O] the new zoom level is shown in the top left-hand corner of the screen and, in the case of [I], a dotted box will show the portion of the plot which will be displayed full screen when [V] (View) is pressed.

Panning appears as follows and the new plot may be displayed by pressing [V] as before.

3.12.3

G: Grid Toggle

[G] may be pressed to apply/remove a grid overlay: it acts as a toggle.

Normally, if a zoom-in is required, CTRL [I] will be used to zoom-in the large steps and then [I] for fine adjustment. A maximum zoom level of 500 (500 times the original plot) is the limit for zooming in, and 0.2 for zooming out. 3.12.2 S/D/E/X: Panning Functions

[S], [D], [E] and [X] may be used to pan a new plot to the left, right, up and down, respectively. (It should be noted that these are the same as the Wordstar cursor movement keys). On some computers, the arrow cursor keys will achieve the same effect. The panning functions will normally be used after zooming-in to position the window of the new plot as required. Two panning steps are available. A `fine' movement using the [S], [D], [E] or [X] alone and a `coarse' movement using <CTRL> with [S], [D], [E] or [X].

3.12.4

V: View Plot

To view a plot on screen, press [V]. This would normally be used after zooming or panning and the new plot would be of the zoomed/panned window.

11

3.12.5

P: Print or Plot

The currently displayed plot of the new window, if zooming or panning has been carried out, may be reproduced on a printer or plotter by pressing [P]. The computer screen will clear and a `Preparing Graphics Image' message will appear. On completion of printing or plotting, the current plot will be re-displayed on the screen. 3.12.6 ECS: End Graphics

The <ESC>ape key may be pressed to end graphics and to return to the option list or Command line. 3.13 The PRINT Command

This option prints the data in the following format, once the user has checked that the printer is ready for printing and confirmed this to the program.

The file name is always the same as the data file name. However, the suffix becomes [FILENAME]. FIL as explained in section 2.12. These files are used in the batch print, see section 4.9, and the final printed format is as shown above. 3.15 The Tolerance (TOL) Command

When the user selects this he will be presented with the following screen:

3.14

The FILE Command

This option writes the data to a print file in ASCII format. The format of the file is as follows:

The values E1 to E4 along the bottom of the screen are the tolerances which the program will apply to particular ranges of penetration/blow gradient during the analysis phase. These values

12

can be changed by the user. The gradient ranges to which the tolerances are applied are as follows: E1 E2 E3 E4 Tolerance for penetration gradients between 0 and 3 mm/blow Tolerance for penetration gradients between 3 and 8 mm/blow Tolerance for penetration gradients between 8 and 20 mm/blow Tolerance for penetration gradients >20 m/blow

The user must then choose whether he wishes to use the basic or adjusted structural number equation. The basic structural number is similar to the original AASHTO definition but with the addition of the subgrade contribution. The adjusted structural number takes account of the way that layer contributions decrease with depth. This option is recommended whenever the subbase is thick or multilayered and whenever the exact definition of the top of the subgrade is difficult. Once an equation has been chosen, a prompt will appear under the layer (B/S/G) column. The user must either press <RETURN>, which leaves the field blank, or enter [B] - Base [S] - Subbase or [G] - Subgrade to define the pavement layers. The program then assigns coefficients to the three layer types for use in the structural number calculation. Once layers have been defined, the user is presented with the following screen:

The EM value at the bottom right of the screen is a multiplying factor which is applied to each of the four tolerances allowing the user to manipulate the sensitivity of the layer identification routine. EM x the zone tolerance (El to E4) gives the amount by which the gradient of the penetration/blow curve is allowed to vary before the program decides that a new layer exists. Decreasing any of the zone variables (El-E4) decreases the tolerance in that zone thereby increasing the number of layers which might be detected. Changing the value of EM will affect ALL the zones as follows: Decrease in EM Increase in EM Increase in sensitivityIncrease in number of layers Decrease in sensitivityDecrease in number of layers

3.16

Structural Number (SN) Command

For the selection of this option the data file must have been analysed (Section 3.10). If this has not been done an error message `Not yet analysed' will appear. Once analysed this option will display the following:

The program now prompts for the structural number of the surface layer to be input by the user. This value, if entered, is simply added to the total for the other layers specified as base, sub-base or sub-grade. If any of these layers have not been specified the user will be prompted to input additional structural numbers. If a surface layer is not present the field can be left blank. Once values have been entered and <RETURN> pressed, the following screen is displayed:

13

4.
4.1

FILE MAINTENANCE
Introduction

The file maintenance option allows the user to define which data drive is active, the number of lines per page for the printer paper and the printer controls. It also allows him to assign the active graphics devices, to test the graphics interface and to specify the graphics parameters. The definable variables are stored in the file DCPCO and the graphics parameter file DCPGRF. If the DCPCO file is modified for colour etc, check the specified parameters. The following program menu is displayed:

Structural numbers for selected layers together with a total value for the entire pavement are given along the bottom of the screen. The suggested CBR values in brackets, see section 3.10, may change according to the designation of the particular layer.

Select the required option using cursor and <RETURN> or the option number. Otherwise press <ESC>ape to return to the TRRL DCP OPTIONS LIST. Each of the options is discussed in turn in the following sections: 4.2 Assign Data Drive

This option allows the current data drive to be reviewed with the following message:
Current Data Drive: [C]

Permissible drive letters are A to Z inclusive. Enter the new drive letter and confirm that the data is OK, or press <RETURN> to accept the displayed value. Note that sub-directories cannot be assigned to the data drive. 4.3 Paper Length/Beeper Control

This option allows the length of printer paper to be defined as lines per page. The number of lines per page is reviewed with the following message:-

14

Number of lines per Page: Beeper on

[66] [Y]

The program does not at present make use of all the printer controls and they are only included to allow for possible expansion. The control codes to be used for compressed print on/off and underline on/off will have to be obtained from the printer manual and consist of a series of ASCII values. For each control code, the operator is required to enter first the number of ASCII characters in the control sequence and then the ASCII values of each of the characters. All ASCII values are entered as decimal integer numbers. Three further controls are specified. Margin Shift: This provides a means of setting the left margin on the printout and is specified in characters, hence a inch margin at 12 cpi would require an input of 6 characters. Backspace If there was no control sequence for Underlining: underline on/off for the printer, then it should be set to [N] and if the printer is a backspacing `teletype-like' machine then backspace underlining can be used instead by typing [Y]. If there is a control sequence for underline on/off, or backspace underlining cannot be used then type [N].

Most printers use a standard 6 lines per inch, so computer paper of 11 inches in length will be 66 lines per page, while A4 paper will be 70 lines per page. Enter the new number of lines per page and confirm that the data is OK or press <RETURN> to accept the displayed value. The user may also choose whether to have the beeper on [Y] or off [N]. 4.4 Enter Printer Controls

An explanation of how the printer controls must be entered appears on the following screen:

Vertica Shift:

This permits a vertical space to be left at the top of each form/page. Enter the number of lines to be left blank at the top of each page.

Having pressed any key the following screen reveals the current control settings:

Having entered/amended the printer controls the operator is required to confirm that the displayed values are to be stored on disc. These values remain current until they are changed again. 4.5 Assign Graphics Devices

This option allows the system to be configured for screen and/or plotter/printer graphics. The following screen is presented:

Only devices which support the Digital Research GSX/GEM system may be assigned and their respective drivers must be specified and be available on the current default disc drive. If the devices of your system do not support GSX/GEM graphics then they must be set as not being available and an entry [N] must be made.

15

Otherwise, if the screen or printer/plotter supports GSX/GEM graphics, enter [Y]. Having assigned the graphics devices, the user is required to confirm that the data is `OK' before being returned to the menu. 4.6 Test Graphics Screen

The screen aspect ratio and printer aspect ratio may be specified. These are values between 0.5 and 2.0 to factor the co-ordinates on the relevant axis to accommodate for distribution. These values should normally be set to 1.0. However, should the test card show an incorrect aspect ratio, then suitable values can be entered to correct the distortion. Two further parameters can be defined. These are: (i) Graphics text size Changes the size of the text of the graphics on both the screen and printer. The required size may be obtained by experimentation. Printer Border The graphical output on dot matrix printers is often hindered by the need to plot out a border, so a flag has been introduced. By entering [Y] a printer border will be included on the plots. Otherwise entering [N] will not print out a border. Batch Print

Select this option to check the graphics interface and to produce a `test card' on the screen. The test card will contain elements used throughout the DCP options and will be similar to the following:

(ii)

4.9

Once a data file has been `filed' (see 3.14) it may be printed as a batch print. Selection of this option will display the following:

4.7

Test Graphics Printer

Select this option to check the graphics interface and to produce a `test card' on the printer/plotter. The test card will contain elements used throughout the DCP options. It should be compared with the screen image to identify any differences between the two devices. 4.8 Specify Graphics Parameters

This option allows various parameters to be modified to suit the user's system. The following screen is presented:

If the user tries to enter a data file which has not been filed an Error message filename not filed will appear at the bottom of the screen. Once the user has confirmed the printer is ready and correctly set, all the files listed will be printed one after another, each starting on a new page. It should be noted that the analysed data is not printed below the data as in the PRINT command of the Process DCP Data options (Section 3.13).

16

An example of the format of a file printed by this method is found in section 3.14. 4.10 CBR Relationships

This option allows the user to choose which relationship between layer strength and Californian Bearing Ratio to use. There is a choice of the following:-

The four options are 1. 2. 3. 4. Kleyn equation: Log10 (CBR) = 2.632-1.28 x Log10 (Strength) TRRL equation: Log10 (CBR) = 2.48-1.057 x Log10 (Strength) Expansive clay equation: Log10 (CBR) = 2.315-0.858 x Log10 (Strength) User defined equation: This option requires a user entered constant and coefficient in the equation: Log10 (CBR) = Constant-Coefficient Log10 (Strength)

Printed to the United Kingdom for The Stationery Office Dd8397622 1/97 C4 51-00 10170

17

Transport Research Laboratory Old Wokingham Road Crowthorne, Berkshire, RG45 6AU

Department for International Development 94 Victoria Street London, SW1E 5JH

Overseas Road Note 9

A design manual for small bridges

ORN 9

First Published 1992 Second edition 2000 ISSN 0951-8797 Copyright Transport Research Laboratory 2000.

This document is an output from a project funded by the UK Department for International Development DFID for the benefit of developing countries. The views expressed are not necessarily those of the DFID.

TRL is committed to optimising energy efficiency, reducing waste and promoting recycling and re-use. In support of these environmental goals, this report has been printed on recycled paper, comprising 100% post-consumer waste, manufactured using a TCF (totally chlorine free) process.

The Transport Research Laboratory and TRL are trading names of TRL Limited, a member of the Transport Research Foundation Group of Companies TRL Limited. Registered in England, Number 3142272 Registered Offices: Old Wokingham Road, Crowthorne, Berkshire, RG45 6AU.

ii

ACKNOWLEDGEMENTS The first edition was compiled by J D Parry of the Overseas Unit at TRL (Head of Unit Mr J S Yerrell) with assistance from the late Mr D M Brooks, Dr T E Jones and Mr N C Hewitt. It is based on a draft commissioned from Rendel Palmer and Tritton, Consulting Engineers of London. The numerous other sources are listed in the references. Mr P K Thomas provided an earlier text; Mr H Lewis assisted in the editing of the final version. Contributions were also made by Central Units and Bridges Division at TRL. We also acknowledge the generous help given by the following people who kindly reviewed the pre-publication draft and offered constructive comments and additions. Mr R C Petts, Intech Associates, UK Mr P Wootton, Civil Planning Partnership, Zimbabwe Mr G A Taylor, Ministry of Public Works, Kenya Major J F MacKenzie, R E Dr R J Freer-Hewish, University of Birmingham. This manual is published by the Transport Research Laboratory as part of the programme of the Department for international Development (DFID).

The second edition introduces a separate chapter on masonry as a bridge building material. This chapter is based on a draft by Mr A Beusch of Intech Associates, with further contributions from Mr J D Parry, Mr N C Hewitt and Dr A F Daly of TRL.

OVERSEAS ROAD NOTES Overseas Road Notes are prepared principally for road and transport authorities in countries receiving technical assistance from the British Government. A limited number of copies is available to other organisations and to individuals with an interest in roads overseas, and may be obtained from: International Division Transport Research Laboratory Crowthorne, Berkshire, RG45 6AU United Kingdom

Limited extracts from the text may be reproduced provided the source is acknowledged. For more extensive reproduction, please write to the address given above.

iii

iv

CONTENTS
Page 1. 2. 3. 4. 5. 6. 7. 8. 9. INTRODUCTION PLANNING SITE INVESTIGATIONS RIVER HYDRAULICS HYDRAULIC DESIGN RIVER AND SCOUR PROTECTION LOW LEVEL WATER CROSSING SUBSTRUCTURES AND FOUNDATIONS CONCRETE SUPERSTRUCTURES 1 3 13 23 33 43 55 65 115 137 155 165 193 201 215 225 229

10. STEEL/CONCRETE COMPOSITE SUPERSTRUCTURES 11. TIMBER SUPERSTRUCTURES 12. CULVERTS 13. EMERGENCY AND TEMPORARY STRUCTURES 14. BRIDGE BUILDING MATERIALS 15. MASONRY 16. DRAWINGS AND SPECIFICATIONS INDEX

vi

Introduction

This manual offers highway engineers a comprehensive set of guidelines to assist and simplify the process of designing small bridges and culverts. These structures are an essential part of every road network. They are far more common than large bridges and are simpler to design and construct. For the purposes of the manual, 'small bridges' are defined as single or multispan structures with individual spans no more than 12m long, ie taking one span to bridge a two-lane highway with shoulders or two spans to bridge a dual carriageway. The guidelines cover the entire design process, from the planning stage through site investigations and materials analysis, hydraulic design and structural design, to the final preparation of drawings and detailed specifications. There are many textbooks and other technical publications that provide excellent treatments of all these aspects of bridge design: some are listed in the manual as useful reference material for readers wishing to pursue subjects in more detail. These sources, however, are all intended for bridge engineers or students of bridge engineering. The present manual is meant to be of use in a bridge design office, but it is aimed also at the general civil engineer who is not a bridge specialist but who may nonetheless be required occasionally to construct a road that crosses a river or other obstruction. He/she may be a provincial roads engineer, extending a regional network of feeder roads with permanent bridges, an army engineer or an engineer involved in famine relief distribution, needing rapid but temporary solutions to bridging problems Because these non-specialist bridge builders have other professional responsibilities, they rarely have the time or expertise to work out all the necessary bridge design calculations from first principles. For this reason, the manual gives as much guidance as possible in the form of drawings and tables, covering two standards of traffic loading, single or multiple spans a range of bridge materials - concrete, steel, timber and masonry - and a range of in situ soils. Though the structural design of small bridges can be simplified by the use of stock solutions, the process of hydraulic design cannot be shortened in the same way. The chapters that deal with river hydraulics, hydraulic design and river works (Chapters 4 to 6) contain all the background information and procedures that the bridge designer will need in order to apply the detailed structural tables set out in subsequent chapters, but they assume the knowledge and experience of a qualified engineer as well as the availability of basic facilities for field investigations and soils analysis. Where there are several possible methods of calculating a variable - for example, allowable

bearing pressure and scour depth - the manual presents only the simplest of these methods but includes references to others. When it is thought likely to be helpful, typical calculations are worked out m the appendices to chapters.

2.
2.1

PLANNING
Site selection 2.1.1 River morphology . 2.1.2 Bridge location .. Site conditions . 2.2.1 Catchment area . 2.2.2 Water levels .. 2.2.3 Navigational and other clearance requirements Plan and sections . Design life Traffic .. Bridge width ... 2.6.1 Single lane bridges .. 2.6.2 One and a half lane bridges . 2.6.3 Two lane bridges . 2.6.4 Culverts ... 2.6.5 Low water crossings ... Paths for pedestrians and cyclists .. Design loading ... Resources .. 2.9.1 Design . 2.9.2 Trade skills .. 2.9.3 Materials . References .. 5 6 6 8 8 8 8 8 9 9 10 10 10 10 10 10 10 10 11 11 11 11 11

2.2

2.3 2.4 2.5 2.6

2.7 2.8 2.9

2.10

Planning

In this initial stage of design the highway engineer identifies a preferred location for the bridge and decides on the type, size and capacity of the structure. These decisions are made on the basis of field surveys and information about: the local terrain and site conditions; the required design life of the bridge; the likely traffic volumes; the resources he/she has available.

will influence the choice of materials and building methods. The traffic predictions enable the engineer to determine the necessary width of the bridge and the load-bearing capacity of the superstructure and substructures. Figure 2.1 shows how the information and surveys in this planning stage lead to the preparation of a general specification for the bridge, followed by further field investigations and the development of detailed designs and specifications. The data required specifically for the geotechnical analyses, site investigations and hydraulic design are discussed in Chapters 3 and 4.

2.1

Site selection

There are three initial considerations to bear in mind: a bridge site must offer appropriate vertical and horizontal alignments;

The local terrain and site conditions dictate the height, length and number of spans, and the design of the substructure foundations. The required design life and the resources available to construct the bridge

Figure 2.1 Flow diagram of the design process 5

its soils must be strong enough to ensure the stability of the structure; the bridge and its associated works should not have an adverse impact on adjoining land or buildings, or themselves be susceptible to damage from the local environment.

(Figure 2.2b). The main channels and any of the minor channels can change position and islands can disappear in the course of a single major flood. Incised rivers have relatively stable banks and are generally narrower and deeper than alluvial rivers. Some overtop their banks during flood, but the flow returns to the existing channel when the flood subsides. Steeply graded tributary streams flowing into a major river commonly exhibit abrupt changes in channel width and bed gradient where they enter the main flood plain. These changes result in the deposition of large quantifies of sediment in the form of alluvial fans (Figure 2.3). The fans consist typically of gravel to clay size debris, are usually conical in shape and have a maximum slope of about 10 per cent. The main channel across the fan can shift its position considerably in a single flood. 2.1.2 Bridge location In selecting the location for a small bridge, the engineer often has to reach a compromise between the easiest river crossing and the shortest road alignment. The choice of location then becomes an economic decision. The cheapest bridge site and the one that has potentially the longest service life is a location that: is on a straight reach of the river; is beyond the disturbing influence of larger tributaries;

For the highway engineer, rivers are the most common obstructions needing to be bridged. Occasionally he/she may be called upon to design a rail or road crossing but these are relatively simple compared to river crossings because they involve considerations only of height and span, whereas the design of a river crossing has also to take hydraulic requirements into account. 2.1.1 River morphology Rivers are classed as either alluvial or incised. Alluvial rivers erode their banks and scour their beds; they have flood plains on either side of the channel and the flow regularly overtops the channel banks to spread across the flood plain. They are continually active, scouring and depositing material on the banks and transporting quantities of sediment. Their channels are winding, and the meanders move downstream as scour and deposition of sediment takes place (Figure 2.2a). When an alluvial river is fed with larger quantities of sediment than it can transport, it deposits sediment as shoals over short lengths of the channel. These shoals deflect the flow, causing a number of minor channels to develop between islands. The river is then said to be braided

Figure 2.2 Alluvial rivers 6

Figure 2.3 Crossing an alluvial fan

.has well defined banks; has reasonably straight approach roads; permits as square a crossing as possible; has good foundation conditions.

not exceed 200 and the abutments and piers should be set parallel to the direction of flow during maximum flood (Section 2.2.2), which may not be the same as during normal flow. When crossing an alluvial fan the bridge should be located as near to the throat of the fan as possible where the tributary stream has sufficient gradient to be self-scouring (Figure 2.3). if this is not practicable and the crossing has to be located across the body of the fan, there will be a requirement for extensive training works to confine the flood flow and for the regular removal of sediment from under the bridge. hi this situation a causeway will often be a more practical solution. Bridge crossings over alluvial rivers nearly always require training works to stabilise the channel flow within the bridge waterway opening.

The site should allow the maximum gradient of the approach roads to be appropriate to the types of vehicle likely to travel on the road as well as offering vertical curves and sight distances suitable for the maximum speed of vehicles using the bridge. A bridge aligned at right angles to the river results in the shortest superstructure. A skew bridge requires more material and is more complicated to design and construct. If a skew is unavoidable the angle should

2.2

Site conditions

Once the engineer has identified a likely site for the bridge, he/she needs to obtain field information on the local terrain and river conditions in addition to the soil information and hydraulic data that are outlined in Chapters 3 and 4 The key points of field information relate to: the catchment area of the river; water levels; navigational and other clearance requirements.

flood. Even where a river is not used by regular traffic, drainage channels and other alluvial waterways periodically require dredging and river-borne equipment may need to pass the bridge. Though it is unlikely that river maintenance equipment will travel when the river is in full spate, this is the time when trees and other floating debris may be carried by floodwater. Experience from other structures on the same river, together with inquiries locally, will help to determine the required clearance between the design flood level (Chapter 5, Introduction) and the underside of the superstructure (Section 5.2). 2.3 Plan and sections The engineer should produce: a plan and longitudinal section of the river to scales of 1/1000 horizontal and 1/100 vertical; at least three cross-sections plotted to a natural scale of 1/100, one at the proposed site and one each at the upstream and downstream limits of the plan.

2.2.1 Catchment area The extent of the river catchment area determines the area to be included in plans and sections, and can be used to estimate flow volumes. If maps to an appropriate scale or aerial photographs are available, the limits of the catchment area can be marked on them and its total size calculated. Transparent squared graph paper is useful for this purpose. In the absence of suitable cartography, the size of the catchment area and its average gradient should be estimated by means of a traverse. 2.2.2 Water levels Information is needed on the highest known flood level, the ordinary flood level and the low water level at the proposed site. The highest known flood level (HFL) should be determined by local observation wherever possible, supplemented by inquiries in the locality. The silt marks that high floods generally leave on tree trunks and buildings remain visible for several years. If there are old trees in the vicinity of the site, they should be examined for the presence of small twigs left adhering to the bark at high water levels. It is usually helpful to ask people who have been living in the area for a long time about their recollections of particularly high floods, but this source of information is variable in its reliability. It is better to make such inquiries by talking to people individually rather than in groups. The ordinary flood level (OFL) is the level to which the river normally rises during the wettest part of the year. The low water level (LWL) is the level prevailing in the river during dry weather. if there is little or no flow in dry weather, the period during which the river bed remains dry should be noted. 2.2.3 Navigational and other clearance requirements The height of the bridge superstructure has to allow for the passage of any regular or occasional river craft as well as the clearance of floating debris at times of

Table 2.1 indicates the distances that should be covered by the plan and longitudinal section in relation to the size of the river catchment area. These distances may be reduced if large-scale aerial photographs are available and show a simple river channel shape. Table 2.1 Distances to be covered by site plans

Information on the terrain and other features in the vicinity of the proposed site should be marked on the plan, as well as the direction of river flow and the location of cross-sections. The longitudinal section should indicate the highest and ordinary flood levels and the low water level (Section 2.2.2). Figure 2.4 shows a typical plan and longitudinal section. Section 16.1 lists in detail the information to be included on the site plan and longitudinal section prepared as part of the final bridge drawings and specifications. Figure 2.5 shows a simple cross-section, with the relevant water levels and the channel shape indicated. The cross-section at the bridge site should show contours at close intervals and indicate any rock outcrops, scour holes and other river bed features. If a road or track already crosses the river at the proposed bridge site, the cross-section should be taken a short distance upstream, in order to obtain an undisturbed section.

Figure 2.4 Plan and section of the river at a proposed crossing

Figure 2.5 Cross section showing relevant water levels

2.4

Design life

and available funds are likely to be the strongest factors influencing the design life of the bridge. 2.5 Traffic

Unlike roads, bridges are not designed to sustain a total number of load cycles. The choice to be made is usually between a 'permanent' structure to carry specified loads with a service life of more than 40 years, and a temporary structure. The engineer's decision will be influenced by traffic predictions and by the resources available at present and in the foreseeable future. Where it is expected that future development will increase the desired capacity, the choice is between building a low-cost bridge to serve until the development occurs or building a structure that is wider, longer or stronger than initially required but which will cope with future needs. An alternative solution is to build permanent abutments and a light deck that can be upgraded or replaced when the development occurs. Service requirements can seldom be predicted with any confidence beyond a ten-year horizon. Local experience

The engineer should estimate the composition and volume of the vehicular traffic likely to use the road throughout the design life of the bridge. The volume of current traffic can be determined from a simple traffic count. The growth rate over the design life of the bridge is difficult to estimate, but the engineer should attempt to do so, taking into account the local factors which influence traffic growth, such as agricultural or industrial development, and national factors such as development planning and the general increase in gross domestic product. Traffic counting and estimating future flows are discussed in TRRL (1988). Vehicle weights can vary according to the season. Unless good quality data on vehicle weights are 9

available it is advisable to carry out an axle weighing exercise at the time of year when the heaviest loads are transported, as described in TRRL (1978).

2.6.5 Low water crossings Low water crossings are considered separately in Chapter 7.

2.6

Bridge width

2.7

Paths for pedestrians and cyclists

Apart from bridges for special applications, there are three alternative widths to be considered: single lane; one and a half lanes; two lanes.

Safety authorities recommend that segregated footways are provided for pedestrians to cross bridges, TRRL (1991). They are particularly necessary on long bridges built to minimum widths where the traffic is fast. It is possible to add a pathway for pedestrians and cyclists by means of supports cantilevered from the main deck, but the engineer has to bear in mind the effect of asymmetric loading should a large number of people congregate on the pathway. An alternative and generally more satisfactory approach is to widen the main deck by about 2m and provide a suitable barrier and parapet, as discussed in Sections 9.2.10 and 11.4.

2.6.1 Single lane bridges Single lane bridges are suitable for predicted traffic flows lower than about 200 vehicles per day. They involve only minimal disturbance to traffic flow and there is normally no safety problem, given adequate sight distance and waiting areas on the bridge approaches and clear advance signing of the width restriction. The width clearance for vehicles is usually 3.65m. Additional provision can be made for pedestrians and two-wheeled vehicles on one side of the roadway, or on both sides when the bridge is located close to a village. Footways should be a minimum of 1.5m wide. 2.6.2 One and a half lane bridges In some districts there may be a preponderance of light traffic, with only the occasional bus or heavy commercial vehicle. In this situation, the most cost effective design may be a bridge allowing two lanes of light traffic, but not wide enough for two large vehicles to pass. This solution offers economies over a full, twolane bridge in terms of both width and load carrying capacity. A carriageway width of 4.6m is sufficient for two lanes of light vehicles but restricts the loading to one lane of heavy vehicles, which are normally 2.5m wide. Adequate sight distances, waiting areas and warning signs are required at both ends of the bridge, and there is likely to be a need to make additional provision for pedestrians. Some authorities consider this width of bridge dangerous and may give preference to a wider two lane bridge. 2.6.3 Two lane bridges These should be designed to conform to the appropriate national standards in terms of load capacity, width and safety provisions. 2.6.4 Culverts Culverts occur more frequently than bridges and are not so noticeable to drivers on fast stretches of road. It is recommended that carriageway width remains constant over culverts. 10

2.8

Design loading

Most countries have some form of design loading standards for bridges, but some have not yet determined an appropriate standard for short span rural bridges which carry low traffic volumes and weights. These short span rural bridges often do not need to be designed for the heavy goods vehicles that are common in industrial areas as an appropriate loading specification for a bridge is one which caters for the heaviest predicted loads expected during the life of the structure. This manual offers standard designs that conform with two of the most commonly adopted loading standards. These are the British Standard loading for 40 tonne gross weight vehicles (BS.HA.LOAD) and the American AASHTO loading for 20 tonne gross weight vehicles (HS 20-44). These loading levels have been used in the standard designs presented in the following chapters and should be sufficient to cover the loading requirements of the majority of rural bridges. 2.8.1 BS.HA.LOAD (40 tonne maximum gross vehicle weight) The BS.HA.LOAD loading was adopted from British Standard BS 5400 (BSI 1978). The loading includes 38 tonne heavy goods vehicles as well as the new European 40 tonne, five- and six-axle trailer combinations. The revised loading specifications, now a requirement for bridges in Britain, are presented in BD 37 (Highways Agency 1988). The loading is presented in the form of a uniformly distributed load imposed on the full lane and a knife edge load placed across the lane in the most critical location and represents a fully loaded lane with the worst combination of vehicles expected in the life of the bridge. The loading includes the effect of impact

(80% on the heaviest axle), overloading (up to 40%) and a 10% contingency for future increases in traffic load and is therefore a conservative estimate of the worst expected loading. The range of vehicle configurations covered by this loading, as well as an indication of which vehicles are critical for each span, are given in BD 21 (Highways Agency 1997). 2.8.2 HS 20-44(20 tonne maximum gross vehicle weight) Loading HS 20-44 has been taken from the AASHTO standard (AASHTO 1983) to cover vehicles with a maximum gross vehicle weight of 20 tonnes. Most twoaxle medium weight commercial vehicles are loaded within the 15 tonne AASHTO limits, but when overloaded these limits may be exceeded. HS 20-44 has therefore been adopted here as a conservative standard. The vehicle configurations (axle weights, spacing, etc) covered by this loading are described in the AASHTO standard. Bridge loading specifications can vary considerably between countries. One reason for this is that standards are frequently derived from the range of legal vehicle loads in that country. When choosing a standard to use, it is important to take into account the actual vehicle loads that may use the bridge. In some cases, these can be higher than the maximum legal vehicle loads. The 40 tonne loading (BS.HA.LOAD) should be used where traffic cannot be restricted to 20 tonne vehicles. In addition, it should be used where overloaded three axle lorries, forestry or quarry vehicles and construction plant are likely to be in use. In both of the above design standards, an allowance for even heavier vehicles can be made by additional loading requirements. These are not included in the standard designs given in this manual. Where these heavier vehicles are expected, the standard designs can be checked, and modified appropriately, by a suitably qualified engineer.

use of these procedures and tables will lead to serviceable, conservatively designed structures. Engineers with experience of bridge design may wish to carry out more of their own calculations or to interpolate between recommended dimensions, and to produce designs tailored more specifically to individual requirements.

2.9.2 Trade skills The engineer should be conversant with the standards of workmanship and supervision required for structures using plain concrete, reinforced concrete, steel and timber. He/she should also be aware of the facilities he/she has available for other functions such as site investigations, and so will be the best person to judge when external skills should be engaged for processes outside the experience of his own workforce. Working alongside crews specialising in fields such as Site investigation, boring, piling or scaffolding can provide a useful opportunity to broaden the skills of technicians and labourers in the engineer's own organisation.

2.9.3 Materials All construction materials must either be found at or close to the bridge site, or must be transported there. Bringing in heavy materials in long sections such as steel beams can be difficult, especially when the bridge is being constructed in advance of the road. Steel panel bridges and steel beam bridges are often made from imported parts and may be subject to delivery delays affecting key items. They may also be the most expensive option, unless a substantial amount of the required materials can be found in the district in good second-hand condition. Although panel bridges are designed to be dismantled and used again, they are rarely used more than once. The site investigations can be extended to include a search for suitable aggregates for concrete and timber for use in the deck or as temporary support during construction. The planning stage includes making provision for processing the resources to be developed locally and storing all the materials in a form that will prevent their deterioration.

2.9 Resources There are three categories of resources to be considered: design ability; skills for the various bridge construction processes;

2.10 References available materials. AASHTO (1983). Standard specification for highway bridges. Washington: American Association of State Highway and Transportation Officials. Highways Agency (1997). BD 21/97: The assessment of highway bridges and structures. Design Manual for Roads and Bridges, Volume 3, Section 4, Part 3, Highways Agency, London. 11

2.9.1 Design It is the purpose of this manual to provide all the necessary procedural guidance, tables, dimensions and material specifications to enable a civil or mechanical engineer with some field experience to prepare appropriate designs for small bridges. The

Highways Agency (1988). BD 37/88: Loads for highway bridges. Design Manual for Roads and Bridges, Volume 1, Section 3, Highways Agency, London. TRRL (1978). A guide to the measurement of axle loads in developing countries using a portable weighbridge.

Road Note 40. Stationery Office, London. TRL (1988). A guide to road project appraisal.
Overseas Road Note ORN5. Transport Research Laboratory, Crowthorne.

TRL (1991). Towards safer roads in developing

countries - a guide for planners and engineers.

Transport Research Laboratory, Crowthorne.

12

3.
3.1

SITE INVESTIGATIONS
Methods of site investigation 15 3.1.1 3.1.2 3.1.3 3.1A 3.1.5 3.1.6 Test pits 15 Hand auger boring 16 Cable percussion boring .. 17 Rotary drilling ..17 Geophysical surveying 17 Backfilling ... 17

3.2 3.3 3.4

Extent of investigations .... 17 Sampling ... 18 Soil testing .18 3.4.1 Field tests . 18 3.4.1.1 3.4.1.2 3.4.1.3 3.4.1.4 3A.1.5 3.4.2 Density measurements 18 Shear vane tests .. 20 Penetration strength tests 20 Dynamic cone penetrometer soundings . 20 Plate bearing tests .. 20

Laboratory tests ... 20

3.5 3.6 3.7

Aggressive chemicals 20 Design review ... 21 References 21

13

14

Site investigations

The weight of the traffic, superstructure, abutments and piers will all be carried by the soils supporting the abutment and pier foundations. Th order to design appropriate foundations (Chapter 8) the engineer has to determine the nature and location of the different soil types occurring at the site of the bridge and its approaches, to depths containing strata sufficiently strong to support the bridge and embankments without significant deformation. This information is obtained by analysing samples taken from a grid of bore-holes or test pits covering the whole of the proposed site, and by testing the samples for density, shear strength, plasticity and penetration, in order to provide quantitative data for foundation design. The level and nature of the ground water also affect the foundation design and the engineer must take into account the impact of bridge construction on the ground water and hence the stability of nearby works and slopes. Since the overall behaviour of the ground is often dictated by planes or zones of weakness, it is possible to obtain a large sample of material that may not be indicative of the behaviour of the mass. For this reason, and because of the frequent need to modify the sampling technique to suit the ground conditions, strict supervision of sampling is essential. Nearby cut slopes can reveal soil and rock types and their stability characteristics, as can old excavations and quarries. There may be embankments or buildings and other structures in the vicinity of the bridge site that have a settlement history due to the presence of compressible or unstable soils. This chapter contains several extracts from Tomlinson (1986), who states, 'An engineer undertaking a site investigation may engage local labour for trial pit excavation or hand auger boring, or he/she may employ a contractor for boring and soil sampling. If laboratory testing is required, the boring contractor can send the samples to an independent testing laboratory. The engineer then undertakes the soil mechanics analysis for foundation design or he/she may ask the testing laboratory to do this analysis. Alternatively, a specialist organization offering comprehensive facilities for boring, sampling, field and laboratory testing, and soil mechanics analysis may undertake the whole investigation. This is much to be preferred to the system whereby one organization does the borings, another the testing, and yet another the analysis. A single organization has an advantage of providing the essential continuity and close relationship between field, laboratory and office work. It also permits the boring and testing programme to be readily modified in the light of information made available as the work proceeds.

Additional samples can be obtained as necessary from soil layers shown by laboratory testing to be particularly significant. In situ testing can be substituted for laboratory testing if desired. In any case, the engineer responsible for the day-to-day direction of the field and laboratory work should keep the objective of the investigation closely in mind and should make a continuous appraisal of the data in the same way as is done at the stage of preparing the report. In this way vital information is not overlooked and the significance of such features as weak soil layers, deep weathering of rock formations and sub-artesian water pressure can be studied in such greater detail as may be required, while the field work is still in progress.'

3.1

Methods of site investigation

This section outlines the following methods: test pits; hand auger boring; cable percussion boring; rotary drilling; geophysical surveying.

The descriptions are brief and intended only to remind the engineer about the uses and limitations of the tests, which should be carried out under the supervision of an experienced technician according to BSI (1981) or other accepted standard. Tomlinson (1986) and Geotechnical Control Office (1987) contain more detailed descriptions.

3.1.1 Test pits A test pit, which should be at least 1m square at the bottom, is a cheap and simple method of subsurface exploration. The pit is normally dug by hand, but a mechanical excavator may be used to remove the bulk of the material before the sides and bottom are squared and cleaned for examination. This method of investigation supplies excellent data on subsurface conditions within the depth to which the pit is excavated and enables a clear picture to be obtained of the stratification of the soils, the presence of any lenses or pockets of weaker material and the level of the water table. The maximum practical depth to which a pit can be excavated is about 3m; below a depth of about 1 .5m the sides of the pit will require support or will need to be excavated at a safe angle. Pits excavated through cohesive soils below ground water level are unlikely to need dewatering by pumps. They should be left open for some time so that seepage lines on the pit sides can indicate the ground water level.

15

In medium and fine sands it may be possible to dewater the pit by pumping. This will rarely be possible in coarse sands and gravel, where boring may therefore be necessary. Figure 3.1 reproduced from Geotechnical Control Office (1987) gives an example of how to record the data obtained from a test pit.

3.1.2 Hand auger boring The hand auger boring method uses light hand-operated equipment. The auger and drill rods are normally lifted out of the borehole without the aid of a tripod, and no borehole casing is used. Boreholes up to 200mm diameter may be made in suitable ground conditions to a depth of about 5m. The method can be used in selfsupporting ground without hard obstructions or gravelsized to boulder-sized particles. Hand auger boreholes can be used for

Figure 3.1 Trial pit log 16

ground water observations and to obtain disturbed samples and small open-tube samples.

3.2 Extent of investigations The site investigations should reveal a clear pattern of soils, rock strata and ground water over the whole site. Strength tests are required in the soils that are to be loaded by the structure (Figure 3.2 and Section 8.3.3): these tests should be taken down below foundation level, or below all deposits that are unsuitable for foundations, to a depth at least 1.5 times the expected width of the foundation, unless bed-rock is reached and confirmed by drilling with a rotary coring rig. Usually a penetration of 3m into the rock will be sufficient; in residual profiles it may be necessary to drill farther to ensure that the rock is not just a boulder or core stone. At least three boreholes should be drilled for each bridge abutment. Fewer bores may be permitted for piers if a clear picture emerges of the strata and soil properties. Each borehole and pit should be numbered and the numbers entered on a plan of the site.

3.1.3 Cable percussion boring This is an adaptation of standard well-boring methods, suitable for soil and weak rock. The sizes of borehole casings and tools are generally 150mm, 200mm, 250mm, and 300mm, giving a maximum borehole depth of about 60m in suitable strata. The drill tools, worked on a wire rope using the clutch of the winch for the percussive action, are a clay cutter for dry cohesive soils, a shell or baler for cohesionless soils and a chisel for breaking up rock and other hard layers. The clay cutter and shell bring up disturbed material for laboratory testing and identification of the strata.

3.1.4 Rotary drilling Rotary drilling rigs are available in a wide range of weights and power ratings. They require a certain expertise in operation, not least because water supplied to lubricate the drilling head can adversely affect the stability of the surrounding ground and the samples obtained from the bore. Open hole drilling, in which the drill bit cuts all the material within the diameter of the borehole, is used for more rapid progress in hard material. Better quality samples of soil and rock are obtained using core drilling, in which an annular bit fixed to the outer rotating tube of a core barrel cuts a core that is returned within the inner stationary tube of the barrel.

3.3 Sampling The choice of sampling technique depends on the purpose for which the sample is required and the character of the ground. There are four main techniques for obtaining samples: taking disturbed samples from drill tools or from excavating equipment in the course of boring or excavation drive sampling in which a tube or split tube sampler with a sharp cutting edge at its lower end is forced into the ground, either by static thrust or by dynamic impact rotary sampling, in which a tube with a cutter at its lower end is rotated into the ground, so producing a core sample taking block samples cut by hand from a trial pit, shaft or heading.

3.1.5 Geophysical surveying Geophysical tests may be helpful in supplementing the data obtained from test pits and bores, eg by tracing the boundary between two soil types, but they are rarely necessary for the planning and design of small bridges. The tests, whether sonic, magnetic or seismic, require expert handling and interpretation, and should therefore be entrusted to an organisation specialising in this work.

3.1.6 Backfilling Poorly compacted backfill will cause settlement at the ground surface and can act as a path for ground water. For boreholes in dry ground it is possible to use compacted soil as a fill material, though cement based grout is usually more successful. The back-fill of excavations can be compacted by using an excavator bucket, but hand tamping will be required at corners. Only temporary backfilling will be required where abutment or pier foundations are to be constructed. All other pits and boreholes should be properly reinstated.

Samples obtained by the last three techniques will be sufficiently intact to enable the ground structure within the sample to be examined. However, the quality of these samples can vary considerably, depending on the sampling technique and ground conditions, and most samples will exhibit some degree of disturbance. Table 3.1 indicates the mass of sample required for identification purposes, Atterburg tests, moisture content, sieve analysis and sulphate tests. Care should be taken to ensure that samples are as pure and undisturbed as possible. Before a sample is taken, the bottom of the borehole or surface of the pit must be cleared of loose or disturbed material. When

17

Table3.1 Soil sample mass required for identification Soil type Mass required kg 2 5 30

described in detail here. Relevant standards are BSI (1981) and ASTM(1985).

Clay, silt, sand Fine & medium gravel Coarse gravel

3.4.1.1 Density measurements Bulk density of soils and rock is measured by sand or water replacement methods or by nuclear methods.

taking undisturbed samples by drive sampling, it is necessary to maintain the water level in the borehole above the surrounding ground water level, so as to prevent the sample being disturbed by a flow of water into the borehole due to the differential head. Samples should be packed and labelled in appropriate containers, according to the laboratory testing programme. Block samples should be marked for orientation and protected from evaporation so far as possible until they are properly wrapped and boxed. Each sample should be labelled with a reference number for location, date, brief description and depth below ground level of the top and bottom of the sample. The sample reference numbers should be related to the borehole and pit numbers (Section 3.2)

3.4.1.2 Shear vane tests Shear vane tests are usually confined to uniform, cohesive, fully saturated soils. The presence of even small amounts of coarse particles, rootlets or thin laminations of sand may lead to unreliable results.

3.4.1.3 Penetration strength tests The strength of coarse-grained, non-cohesive soils can be assessed by the Standard Penetration test using a percussion boring rig and a split-barrel sampler. The N value obtained is used directly in the design calculations outlined in Chapter 8. The value is the number of blows required to drive the sampler 300mm into the layer under study, and it may be affected by large stones and rock. For this reason, any very high values obtained by this method should be treated with suspicion. Table 3.3 indicates an approximate correlation between N values and the relative density of granular materials.

3.4

Soil Testing 3.4.1.4 Dynamic cone penetrometer soundings The Dynamic Cone Penetrometer (DCP) is cheap and quick to use, and it causes minimal disturbance to the ground. It can be applied between boreholes or test pits to obtain a continuous profile of soil layers, or to find the boundaries of boulders. DCP tests should be made close to each borehole or test pit, to provide a correlation between soil types and penetration specific to the locality of the site, and then at small intervals between boreholes and test pits. Table 3A shows a typical correlation between DCP and SPT values.

Table 3.2 sets out the basis on which soils are classified for engineering purposes, and outlines simple field tests that help identification. Laboratory tests of soil samples and in situ field tests should be carried out according to recognised standards such as BSI (1981 and 1990). These tests include sieve analysis of non-cohesive soils, liquid limit and plastic limit tests of cohesive soils, density tests, strength tests and acidity tests. Some presumed bearing values are listed in Chapter 8, but it is recommended that wherever laboratory facilities and field equipment are available, the most appropriate of the tests outlined in Section 3.4.1 are carried out for each site.

3.4.1 Field tests These may include: density measurements; shear vane tests; penetration strength tests; dynamic cone penetrometer soundings; plate bearing tests. It is assumed that if the facilities for these tests are available, the procedures are known and need not be 18

3.4.1.5 Plate bearing tests There are a number of procedures for measuring the bearing capacity of soils and weak rocks by the use of a steel plate to which either a continuous load or a constant rate of penetration is applied, BSI (1990). If possible the plate should fill the borehole and be bedded on undisturbed material: where the diameter of the plate is significantly less than the diameter of the borehole, the results of the test are hard to interpret. Ground water should be at its natural level during the test, which may make seating of the plate in the bottom of the borehole difficult. Since the resulting bearing capacity applies only to the soil or rock immediately below the plate, a number of tests will be required to cover the surface area and depth of material to be stressed by an abutment or pier (Figure 3.2)

Table 3.2 General basis for field identification and classification of soils

19

Table 3.3

Correlation between SPT value, N, and density of granular soils N value Less than 10 1-30 3-50 Relative density Loose Medium dense Dense Very dense

3.4.2 Laboratory tests The following laboratory tests are specified in BSI (1990): liquid limit; plastic limit; density.

More than 50

The liquid and plastic limits of cohesive soils may be determined using disturbed or undisturbed samples taken from bores or test pits, and tested in their natural state. If quantities of material of suitable size can be obtained, the bulk density of soil samples of natural material may be determined by the immersion in water or water displacement methods. Sieve analyses carried out on coarse-grained, noncohesive soils also assist in their identification and classification.

Table 3.4

Typical correlation between DCP and SPT values SPT value blows/300mm 50 44 38 33 28 24 22 18 16 15 14

DCP value mm/blow 5 6 7 8 9 10 12 14 16 18 20

3.5 Aggressive chemicals The ground or ground water may contain chemicals capable of causing damage to concrete or steel. These chemicals may emanate from nearby industrial processing or may occur naturally. Measures to protect concrete or reinforcement against chemical attack are described in Section 14.1 of the manual. The principal constituents that cause concrete to deteriorate are sulphates, which are most common in clay soils and acidic waters. Total sulphate contents of more than 0.2 per cent by weight in soil and 300 parts per million in ground water are potentially aggressive (BRE, 1981).

Figure 3.2 Loading effects of small and large plates 20

Corrosion of metal is caused by electrolytic or other chemical or biological action. In industrial areas, corrosive action may result from individual waste products dumped on the site. In river and marine works, the possible corrosive action of water, sea water and other saline waters, and industrial effluents may also require investigation. In a marine environment, the most severe corrosion is found in the 'splash zone' that is only wetted occasionally. The saline concentration in ground water near the sea may approach that of sea water, particularly where land has been reclaimed. Near estuaries, there may be an adverse condition caused by alternation of water of different salinities. Laboratory tests to assess the aggressiveness of the ground and ground water against Portland cement concrete include determination of pH value and sulphate content (B SI, 1990). Since the pH value may be altered if there is a delay between sampling and testing, determinations should be made in the field whenever possible. Water sampled from boreholes may be altered by the flushing water used in drilling, or by other flushing media employed: this means that sulphate and acidity tests carried out on samples from boreholes may not be representative unless special precautions are taken.

3.6 Design review There is often difficulty in specifying ground conditions before the excavations for construction are complete. For this reason the engineer should be prepared to review his plans, both during the site investigation and during construction, if evidence is found of unexpected soil conditions.

3.7 References ASTM (1985). Annual book of ASTM standards, vol 0408. Philadelphia: American Society for Testing and Materials. BRE (1981). Concrete in sulphate bearing soils and ground waters. Digest No 250. Watford: Building Research Establishment. BSI (1981). Code of practice for site investigations. BS 5930. London: British Standards Institution. BSI (1990). Methods of test for soils for civil engineering purposes. BS 1377. London: British Standards Institution. Geotechnical Control Office (1987). Guide to site investigation , Geoguide 2. Hong Kong: Civil Engineering Services Department. Tomlinson M J (1986). Foundation design and construction, fifth edition. Singapore: Longman Singapore Publishers Pte Ltd. 21

2 1

22

4.
4.1

RIVER HYDRAULICS
Flow velocity .. 25 4.1.1 4.1.2 Direct measurement .. 25 Calculation using bed characteristics 26

4.2

Flow volume (discharge) 26 4.2.1 4.2.2 4.2.3 Area-velocity method ... 26 Orifice formula . 26 Other methods .. 29

4.3

References .. 29

Appendix 4.1 .. 30 Appendix 4.2 .. 30

23

24

River hydraulics

the geometry of waterway required at the bridge site; the backwater caused by the restriction of flow due to piers and abutments; the scour caused by the restriction; the river training works required. 4.1 Flow velocity

This chapter deals with the acquisition of the hydraulic data necessary for the efficient design of a river crossing in relation to the flow characteristics of the river (Chapter 5). The engineer has to ensure that the flow of water can pass the structure without causing damage to the bridge, the road embankment or the surrounding land. Damage can occur in a number of ways: the river may react against obstructions such as piers and abutments, and scour beneath them causing failure; the approach embankments may act as a dam during high floods, sustaining damage or causing more extensive flooding upstream; a river flowing on a shifting path may bypass a bridge and cut a new channel across the highway; a river may over-top a bridge if sufficient clearance is not provided. In order to design a structure that avoids these problems and costs no more than is necessary, the hydraulic characteristics of the river must be understood and quantified. The most economical structure is usually one which is just wide and high enough to accommodate the design flood, minimising the total cost of abutments, piers, superstructure, approach embankments, relief culverts and river training works. The hydraulic data required for the design process detailed in Chapters 5 and 6 relate to: design flood level (defined in Chapter 5), flow volume and velocity; maximum flood level, flow volume and velocity; bed characteristics - particle size, vegetation; channel shape and flood plain width; sedimentation and meander characteristics; navigational requirements and clearance of floating debris. Flow velocity measurement and estimation are treated in Section 4.1, flow volume calculation in Section 4.2. Characteristics of the river bed and navigational requirements were discussed in Chapter 2. Flood levels and channel shape are drawn on the longitudinal and cross sections described in Section 2.3. Using the hydraulic data, calculations may be made to determine:

4.L1 Direct measurement Though it may he difficult to measure flow velocities directly during a flood, the engineer should attempt to do so wherever possible, because this is the critical value and alternative methods of estimating a maximum value are less accurate. After a suitable method of depositing and retrieving a float on the river has been contrived, its travel should be timed over a distance of at least four times the channel width on a straight reach of preferably uniform section. If the shape of the channel is complex, velocities should be measured at several stations across the width. Where the channel is deep, a double float may be used to measure velocities at several depths. This device consists of a small, buoyant float attached to a large weighted object by a cord (Figure 4.l). The velocity of the small float is assumed to he that of the water at depth 'd', the length of the cord, which is varied to measure the velocity at different depths. On simple sections the mean velocity is approximately equal to the surface velocity at the centre multiplied by 0.85. Figure 4.2 illustrates typical flood velocities at different stations and depths.

Figure 4.1 Double float for measuring flow velocity

25

Figure 4.2 Typical flow velocity patterns

4.1.2 Calculation using bed characteristics. The alternative to direct measurement is to use Manning's formula to estimate the mean velocity:

level may be calculated from the cross-sectional area and velocity.

4.2.1 Area-velocity method Where there is free flow, the volume of flow may be calculated using the formula: Where V= A= velocity (m/sec) Q = A.V area of cross-section of the flooded channel (m2) length of the wetted bed across the channel (m) gradient of the surface or bed slope value of rugosity coefficient taken from Table 4.1. If the cross-section is not a simple shape, it may be divided into several parts as shown in Figures 4.3 and 4.4, and the total volume flow may be obtained by adding the flows from each calculation. This method is necessary when a river tops its banks during flood. Each part cross-section is chosen to be a simple shape and the value of V is measured for that part section, or calculated for it using Manning's formula. Appendix 4.1 sets out an example of flood discharge calculation by this method. Where Q = volume of flow (m3/sec) A = cross sectional area (m2) V = mean velocity of the water (m/sec) s= and n= (4.2)

P=

Note: When using Table 4.1 to find the rugosity coefficient (n), choose Minimum if the sides are relatively smooth and Maximum if relatively rough. For example, tree stumps on cleared land, (C)3 would have n =0.050, while a dense mass of trees would have n =0.200.

4.2 Flow volume (discharge) When a value for mean velocity of flow (V) has been obtained, the flow volume (Q) at the design flood 26

4.2.2 Orifice formula The volume of flow may also be calculated from measurements taken on an existing bridge over the

Table 4.1 Value of rugosity co-efficient, n in Mannings formula for streams and rivers up to 30m

27

Figure 4.3 River cross section divided into simple shapes

Figure 4.4 Part section I labelled for calculations

same river, by using the orifice formula: Dd= depth of water immediately downstream of the bridge measured from marks on the piers, abutments or wing walls (m) mean velocity of approach (m/sec).

V= where: Q = volume of flow (m3/sec) g = acceleration due to gravity (9.8m/sec2) L= linear waterway, ie distance between abutments minus width of piers, measured perpendicular to the flow (m) depth of water immediately upstream of the bridge measured from marks left by the river in flood (see the definitions in the introduction to Chapter 5) (m)

Co and e are coefficients to account for the effect of the structure on flow, as listed in Table 4.2 These dimensions are shown in Figures 4.5 and 4.6. The formula is claimed to give nearly correct volumes for most waterway shapes, but Q should be increased by 5 per cent when Du-Dd is greater than

Du=

Dd 4 28

Table 4.2 Values of C0 and e in the orifice formula

should be adopted as the design discharge, provided the results are not too dissimilar. if they differ considerably, the engineer has to form a judgement based on the reliability of the data on which each calculation was based.

4.2.3 Other methods The rational method of estimating flow volumes from catchment area, run-off coefficients, rainfall intensity and a time factor is a further means of checking estimates. Details of this and other methods may be found in Farraday and Charlton (1983) and Fiddes (1976). (Note: intermediate values may be obtained by interpolation) L = width of waterway as defined above W = unobstructed width of the stream Appendix 4.2 gives an example of a calculation using this formula. Whenever possible, flow volumes should be calculated by both the area-velocity and orifice formula methods. The higher of the two volumes 4.3 References For small catchment areas up to about 15 sq kms, the peak discharge is often between 1 and 2m3/sec per 25 hectares.

Farraday and Charlton (1983). Hydraulic factors in bridge design. Wallingford: Hydraulics Research Station Ltd Fiddes D (1976). The TRRL East African flood model. Laboratory Report LR706. Transport Research Laboratory, Crowthorne.

Figure 4.5 Waterway at a bridge

Figure 4.6 Backwater 29

Appendix 4
Appendix 4.1 Flood discharge calculation by the area-velocity method Calculate the discharge of a stream with clean straight banks, no rifts or deep pools and bed slope of 0.2% having a cross section, as given in Figure 4.3. a) Since the cross section is irregular, divide the channel into three sub-sections (I, II, and III) as shown in Figure 4.3. Calculate the hydraulic conditions of sub-section I as follows:

Bed slope is 0.2%, i.e. s =0.002

The flow volume for sub-section I (Q1): Q1 = A.V = 13.87 x 1.22 = 16.92m3/sec c) Similarly, the hydraulic conditions for the other sub-sections are calculated as follows:

b)

If necessary, divide the sub-section into small strips as in Figure 4.4. The area and the length of the wetted bed surface of each strip is calculated using the following expressions:

where h1 and h2 are the depths of water at chainages x1 and x2 respectively. Example calculations are shown in Table A4.1 Using Manning's formula:

Appendix 4.2 Flood discharge calculation using the orifice formula Calculate the discharge passing through a bridge with a waterway width of 18m across a stream 30m wide. In flood the average depth of flow upstream is 2.2m. Discharge at a section just upstream of the bridge, assuming a rectangular cross section: Q = A.V Q = 2.2 X 30 X V = 66. Vm3/sec (4.2) (4.4)

From Table 4.1, section (A) 1, the coefficient of rugosity (n) is 0.030.

Table A4.1 Flood discharge calculation by the area-velocity method

30

Discharge at a section just downstream of the bridge will be the same and will be given by the orifice formula:

31

32

5.
5.1 5.2 5.3 5.4

HYDRAULIC DESIGN
The hydraulic design process . 35 Bridge height .. 36 Positioning the abutments and piers ... 37 Scour ... 38 5.4.1 General scour . 38 5.4.1.1 5.4.1.2 5.4.2 The area-velocity method for predicting general scour 38 The competent velocity method for predicting general scour .. 38

Local scour 39 5.4.2.1 5.4.2.2 Local scour at piers .. 40 Local scour at abutments . 40

5.5 5.6

Backwater effect (afflux) ... 40 Culverts .. 41 5.6.1 Culvert hydraulics . 41

5.7 5.8

Use of existing abutments and piers ... 42 References .. 42

Appendix 5.1 42

33

34

5 Hydraulic design
When a bridge structure and its associated embankments encroach upon the flow of a river in flood, there is a risk to the structure, the embankments and the surrounding land. It is not economical, however, to build a bridge to clear a wide floodplain because bridge works cost more than earth embankments. Often bridges are designed to accommodate annual high flows without excessively restricting the flow of the river or incurring damage either to the structure or the surrounding land. In these cases, it is accepted that unusually high floods may over-top the superstructure and cause temporary disruption. The annual high flow is termed the design flood (DF) and the highest known flood, the high flood (HF). Precautions may be necessary to ensure that restricting the flow of the design flood does not cause either excessive backwater, resulting in flooding upstream, or scour that may damage the structure. The bridge parts should also be designed to withstand the high flood, though some damage may result to adjoining land and it may subsequently be necessary to repair scour holes, etc, in the river channel. On important roads the bridges may be designed for a design flood occurring no more than once every ten years and a high flood once every hundred years. A minor road may be served by bridges or low water crossings designed to be over4opped for a few days every year. The bridge designer is required to select a design flood level (DFL), a design discharge ie flow volume, and a design velocity, on which to base calculations of waterway geometry, foundation depth, scour protection and vertical clearance. In summary: The design flood is the maximum flow that can pass through the bridge without: causing unacceptable disruption to traffic; endangering the pier and abutment foundations with scour; damaging approach embankments; causing flood damage on the upstream side of embankments.

with a single span. The abutments are built clear of the level of the design flood and hence there is no restriction of its flow. River training works are unlikely to be required, and no backwater or additional scour is expected as a result of the presence of the bridge. Figure 5.lb illustrates a crossing of a wider flood channel. The superstructure is longer and will be very expensive if piers are not used. Both pier and abutment foundations are below DFL and will require protection from scour. The abutment walls and piers will impede the design flood. This restriction to flow will cause some backwater and additional scour of the bed, which must be taken into account in foundation design. Where a small river has a wide flood plain (Figure 51c), the most economical solution may be a relatively small bridge, approached by earth embankments. The structure will restrict the flow during the design flood but not during the normal flow. In order to prevent flood water rising too high on the upstream side, relief culverts are installed in the embankments while retaining walls and river training works channel the main flow through the bridge. The abutment foundations are designed to withstand the scour caused by the restriction to flow.

5.1 The hydraulic design process The sequence of data collection and hydraulic design is illustrated in the flow diagram, Figure 5.2, taken from Farraday and Charlton (1983). Their book, 'Hydraulic factors in bridge design', describes the process in more detail than can be shown here and is recommended for its explanations and worked examples. The general order of design is: 1 Establish the height of the superstructure, ie clearance above the DFL. Establish trial positions for the abutments according to the bed shape at the proposed crossing. Make a provisional decision regarding the number of piers that will result in the lowest overall cost of superstructure, piers and abutments. Calculate the general scour and the local scour due to abutments and piers, and draw the worst case profiles on the cross sections. Check that the backwater caused by the restriction to flow does not cause damage to surrounding land upstream of the bridge or affect the height set for the superstructure. Prepare preliminary designs of abutments and piers.

The high flood is the rarely occurring flow that it is uneconomical to include in the design flood, but which may be considered when designing the superstructure and piers of the bridge. The simplest situation is shown in Figure 5.la, where the incised river is contained within its banks at the design flood, and is narrow enough to be bridged

35

Figure 5.1 Bridges for different flood widths

Check scour and backwater effects and make adjustments as required, recalculating the effects of any changes to the waterway. Calculate the cost of the superstructure once its length is decided, and the costs of the substructures, embankments, river training works and relief culverts. In order to obtain the most economical design, or to compare the costs of structures catering for different design floods, it may be necessary to repeat the above procedure on the basis of alternative waterway conditions.

Table 5.1 Vertical clearance at DFL minimum measurements for vertical clearance between the lowest part of the superstructure and the DFL, taking into account backwater effects described in Section 5.5. These clearance measurements should be increased on rivers with a history of unusually large floating items or a navigational requirement. Table 5.1 Vertical clearance at DFL

Discharge (m3/sec) <0.3 0.3to3.0 3.0 to 30.0 30 to300 >300

Vertical clearance (mm) 150 450 600 900 1200

5.2 Bridge height The waterway below the superstructure must be designed to pass the design flood and the floating debris carried on it. Table 5.1 gives recommended

36

Figure 5.2 Flow diagram of the hydraulic design of a bridge

5.3 Positioning the abutments and piers The engineer should choose trial positions for the abutments, bearing in mind the guidelines in the introduction to this chapter and Figure 51. The requirement for piers will depend on several factors: the availability and cost of material for the structural members of the superstructure; traffic loading;

distance between the abutments; the length of timber or steel beams available; HA or HS20 loading and suspended length. A configuration of abutments and piers should be selected and the length of waterway measured, ie the distance between the abutments, measured perpendicular to the flow, minus the sum of the pier widths (Figure 4.5). 37

Using these measurements and the estimates or calculations of flow volume and velocity, outlined in Sections 4.1 and 4.2, the engineer can calculate the likely scour and backwater effects. If these prove unacceptable, a different configuration of abutments and piers should be tried.

To predict the likely general scour of a river bed between confined banks or abutments, a probable scour line is drawn on the cross-section at the bridge: the area (A) and wetted perimeter (P) below the design flood level are measured as explained in Section 4.2. Water velocity is given by Manning's formula,

5.4 Scour Scour is the erosive effect of water flow on the river bed or banks. Bridge works may alter the existing scour pattern by restricting the free flow of the stream and/or causing local changes to the current. Approximately half of all river bridge failures are due to scour alone. There are four types of scour: natural scour and channel shifting on alluvial rivers (Section 2.1.1); scour caused by changes to the river channel upstream or downstream of the bridge site; 'general' scour caused by reduction in the channel width at bridge works; 'local' scour at the base of piers, abutments and river training works where these divert the general flow.

where

n is Manning's rugosity coefficient found in Table 4.1 and s is the gradient of the water or average bed gradient.

This process is repeated, ie probable scour lines are drawn and the corresponding water velocities calculated until the product of calculated V and area A is approximately the same as the unrestricted discharge. The shape of the additional scour area should be adjusted as shown in Figure 5.3.

Examination of the river at the bridge site and at any existing structures, preferably during and soon after a time of flood, will indicate if and where river training works are required to combat channel shifting and to guide the flow through the waterway at the bridge. Any changes to the river channel should be considered as likely to affect the proposed structure and other nearby structures on the river by changing scour patterns. At the bridge site estimates are required for general scour and local scour.

5.4.1.2 The competent velocity method for predicting general scour The competent velocity of flow is the greatest velocity that will not cause scour to the river bed. In this method, the waterway at the bridge, ie the crosssectional area of the flow, is increased until the resultant mean velocity of flow is reduced to the competent velocity, Vc. In order to select an appropriate value for Vc, a probable scour depth at the bridge is estimated and the corresponding competent velocity is read from Table 5.2 for cohesive bed material, or from Figure 5.4 for non-cohesive bed material.

5.4.1 General scour There are many relationships that can be used to predict general scour. All of them assume a fairly simple river channel shape. At complex locations, particularly near the junction of two streams, these methods are known to be inaccurate.

Table 5.2 Table of competent velocities for erodible materials.

5.4.1.1 The area-velocity method for predicting general scour Despite the rise in water level on the upstream side of a bridge during flood, the flow through the structure is assumed to be equal to the unrestricted flow (Section 4.2), as calculated on the basis of velocity measurement (Section 4.1.1) or estimation (Section 4.1.2).

38

Figure 5.4 Competent velocity chart for river beds of non-cohesive soils.

The area of the waterway, A, is calculated and adjusted as shown in Figure 5.3. The mean depth used for the calculation must correspond with the area derived.

5.4.2 Local scour In addition to the general scour discussed in 5.4.1, there will be local scour of the river bed at the bridge site caused by turbulent flow around obstacles such

As piers, unless the bed is armoured. The localised turbulence also removes bed material from critical areas wherever the flow of water is diverted downwards or sideways around the ends of wing walls, abutments or river training works. Since local scour is related to bed material and water velocity, a study of scour patterns at existing structures on the same river will give an indication of conditions likely to occur at the new structure.

Figure 5.3 Adjustment of waterway to accommodate general scour 39

5.4.2.1 Local scour at piers Figure 5.5 gives a basis for estimating the local scour likely to occur at the upstream end of piers. If the piers are skewed in relation to the direction of flow, the depth of scour should be multiplied by a factor from Table 5.3. Table 5.3 Multiplying factors for local scour at skewed piers

calculated by multiplying the general scour depth by 2.25 should be used

Table 5.4 Multipliers for estimating scour depth at abutments and training works

5.4.2.2 Local scour at abutments One means of estimating local scour at abutments and river training works is to use the general scour prediction as illustrated in Figure 5.3. Blench (1969) suggests that the general scour depth should be used with a multiplier from Table 5.4. Where abutments protrude into the flow, it is suggested that the greater of the scour estimates for the piers and the level

5.5 Backwater effect (afflux) The restriction to free flow due to piers and abutments causes the head of water to rise on the upstream side of the structure. The water level downstream may also be lowered, but is usually assumed to be unaffected by the bridge. The engineer should estimate the rise on the approach side to ensure that the river bank will not be breached where floods would cause damage, and to check that the desired clearance for floating debris or river traffic is maintained below the superstructure. The magnitude of backwater can be affected by the shape of the river bed, its eccentricity, scour holes and related factors. It is also influenced by the shape

Figure 5.5 Curves for estimating depth of local scour at bridge piers 40

of the bridge works, ie by any skew, river training, embankments and relief culverts. For simple configurations, a good estimate may be made using the formula:

culverts or sometimes corrugated steel pipe culverts are used, singly or in multiples. Culvert design is the subject of Chapter 12.

where

h V g

= = =

afflux (m) average velocity of flow (m/sec) acceleration due to gravity (9.8m/sec2) unobstructed width of the stream (m) linear waterway as defined in Section 4.2.2 (m) coefficient of discharge through the bridge, taken as 0.7 for sharp entry and 0.9 for bell mouthed entry.

W = L C = =

5.6.1 Culvert hydraulics Like bridges, culverts are designed to be large enough to pass the design flood without damage to the embankment or surrounding land. In practice this usually means limiting the height of the flood on the upstream side. The required size of the culvert is found by calculating the area required to permit a flow that will maintain the upstream head of water below the critical level. The head downstream is taken to be either the design flood level before the embankment is built or the top of the culvert, whichever is the higher. The operating head H is defined as

The formula (5.2) is taken from Victor (1980). More detailed methods of calculation, including use of the orifice formula, are found in Bradley (1972) and Henderson (1966). It bears repetition that the calculations set out here and elsewhere in this chapter can produce only broad estimates of scour depth, afflux and other variables, since each river bed and bridge location has a distinct and complex set of characteristics. The bridge engineer is required to produce a conservative design, taking into account the required service life and a realistic estimate of the quality of the maintenance available to detect and repair at an early stage any damage to the river banks and bed that may affect the structure.

The head, H, can then be said to equal the sum of the losses at the culvert, ie

These losses are estimated using the velocity head:

5.6 Culverts Where a required waterway opening is less than about 15m2 and in particular, where a road crosses a stream on a relatively high embankment, it is usually cheaper to provide a culvert than a bridge. Before the hydraulic requirements of a culvert can be calculated, an assumption has to be made about the type of construction to be used. The most common forms of culvert construction are: precast concrete joined pipes; prefabricated corrugated steel pipes; precast or built in situ concrete boxes. The coefficients KE and f depend on the entry size and shape. For the concrete box culvert designs detailed in Chapter 12,

Concrete pipe culverts are commonly used for small openings up to about 2m2 and multiple pipes, with or without common headwalls, are used for larger areas. For areas greater than 2m2, reinforced concrete box 41

Hence for concrete culverts

Appendix 5
Appendix 5.1 Calculation of culvert operating head. Determine the head required to produce a discharge of 2.0m3/sec through a concrete culvert having a length of 20.0m and a diameter of 1.0m.

A typical calculation is set out in Appendix 5.1. For corrugated metal pipes projecting from the fill, Ke can he as high as 0.9. Use of head-walls can reduce this to 0.5. The f value for corrugated pipes should be 0.075. 5.7 Use of existing abutments and piers When replacing a bridge deck, the sub-structure foundations should be examined to verify that they are sufficiently deep to be unaffected by the maximum calculated scour. If they are not, scour protection as outlined in Chapter 6 will be required.

5.8 References Blench T (1969), Mobile bed fluviology. Edmonton: University of Alberta Press. Bradley J N (1972). Hydraulics of bridge waterways. Washington: US Bureau of Public Roads. Farraday and Charlton (1983). Open channel flow, Collier-Macmillan, London. Victor D J (1980). Essentials of bridge engineering. New Delhi: Oxford and IBH Publishing Co. The head required is 0.49m.

42

6
6.1

RIVER AND SCOUR PROTECTION

Protection materials 45 6.1.1 6.1.2 6.1.3 6.1.4 6.1.5 6.1.6 Rip rap 45 Gabions and Reno mattresses . 45 Filter blankets . 51 Piled walls .. 51 Vegetation .. 51 Slab concrete and stone pitching 51

6.2

Protection methods . 51 6.2.1 6.2.2 6.2.3 6.2.4 Pier and abutment foundations ... 51 Bank protection .. 51 Guide banks 52 Groynes ... 52

6.3

References .. 52

43

44

River and scour protection

grading is recommended. The stone should have a size distribution such that the ratios between the maximum and median size and between the median and minimum size are both about 2:1. Table 6.1 gives two examples of suitable stone gradings by weight. Table 6.1 Stone rip rap gradings for stream bank revetment

Scour protection is usually required where a bridge is built across a meandering stream or when some restriction to the flow of the design flood occurs at a bridge. Protection measures can take the form of: rip rap on slopes or river bed; gabion or Reno mattress aprons or revetments;

Class I Nominal 0.3m diameter or 35kg weight. Allowable local velocity up to 3m/sec. Grading: kg 140 70 35 10

groynes; piled walls; vegetation. 100% smaller than at least 20% larger than at least 50% larger than at least 80% larger than

With the exception of groynes and some forms of vegetation, the firm rule is that these protective measures must not protrude in any way into the design waterway at the crossing, since a protrusion would restrict the flow even further, causing additional scour. Experience is by far the best guide as to where river training works are required. Whether the engineer is considering measures to protect a new structure or to prevent further damage to an existing one, he will find it useful to examine protective works on other structures in the area, for their effectiveness and their durability. Observation of the currents at the site in question, preferably during a flood, and examination of plans showing how the river shape progresses, may also indicate where bank strengthening and guide walls are needed. Farraday and Charlton (1983) provide details and further references.

Class II Nominal 0.5m diameter or 180kg weight. Allowable local velocity up to 4rn/sec. Grading: 100% smaller than at least 20% larger than at least 50% larger than at least 80% larger than Note The percentages are by weight: the sizes are equivalent spherical diameters = kg 700 320 180 30

6.1

Protection materials

6.1.1 Rip rap Rip rap river bed protection consists of a carpet of loose stones, which prevents the water current from eroding the soft bed material. The stone elements must be heavy enough to resist being washed away by maximum water velocities during a flood, and they should not be installed in a manner which reduces the area of waterway designed to accommodate general scour. The main advantages of rip rap are that: it is relatively cheap; it is a flexible protection and often finds its own effective level; if appropriate lifting gear is available it is easy to install and repair.

Figure 6.1 shows recommended stone sizes for different local flow velocities. Figure 6.2 is a conversion chart to change nominal stone diameter to weight. The thickness of the stone blanket should be at least the length of the largest stones and about twice the nominal median diameter. Durable stone of a near cubic shape is best. Flaky material should he discarded.

6.1.2 Gabions and Reno mattresses The standard gabion is a rectangular basket made of steel wire mesh (Figure 6.3). It is strengthened at the comers by heavier wire and by mesh diaphragms which divide it into typically 1m compartments. The gabion is supplied as a flat pack for assembly on site and is normally filled in situ with quarried stone or rounded shingle of sufficient size to prevent the stones passing through the mesh. The gabion structure is more stable and durable if the stones are packed carefully by hand, almost as if building a stone wall. Internal tie wires at about 0.5m intervals help to retain the shape. Sections of gabions are securely wired together in position to form a continuous revetment or wall (Figures 6.7 and 6.8). Standard sizes are 2m, 3m and 4m long by 1m wide by 0.5m or 1m high.

Since stone of uniform size is unlikely to be available and does not bed down well, a judicious amount of

45

Figure 6.1 Rip rap stone size

Figure 6.2 Stone diameter/weight conversion chart

46

Figure 6.3 Gabion basket

Figure 6.4 Reno mattress

47

Figure 6.5 Timber piled bank protection

48

Figure 6.6 Area requiring scour protection around a pier base

49

Figure 6.7 Slope revetment with a launching apron

Figure 6.8 Gabion walls 50

The Reno mattress is a mattress-shaped version of the gabion, for use where the latter's height and bulk are not needed, as in lining some river banks and beds. The standard 6m length of mattress is divided into six 1m compartments or into as many as ten compartments on special order as illustrated in Figure 6.4. After assembly on site, the mattresses are wired together and filled to form a continuous blanket or lining. Details of the construction and filling of gabions and Reno mattresses are given in ILO (1986) which also contains a list of sources and material specifications. Table 6.2 indicates minimum mattress thicknesses for a range of bed soil types and water flow speeds. Table 6.2 Minimum Reno mattress thicknesses

flow through the bridge (Figure 6.5). Piled walls are less flexible than gabions or rip rap and may fail under small amounts of movement. The engineer has to make sure that the piles are driven deep enough to withstand the general and local scour, and still retain sufficient stability to withstand earth pressure on one side when the water level is low on the other. If timber is to be used, local knowledge of durability under wet and dry conditions will help selection. The shaping of the timber piles shown in Figure 6.5 contributes to achieving a tight wall. Purpose-made steel piles interlock and so make a stronger wall than lengths of W section guardrail, used railway sleepers or other improvised sections.

6.1.5 Vegetation Many plant types can be used to protect river banks against erosion. The most successful ones are almost always those found growing naturally along the river being bridged. Protection from annual floods in the form of gabion groynes (6.2.4), tethering to strong stakes or rip rap, may help tender plants to become established.

6.1.6 Slab concrete and stone pitching Where there is an erosive stream, such as a drainage ditch joining the river at a bridge, slab concrete or stones set in cement mortar may be required to armour the river bank at the point of entry. 6.1.3 Filter blankets If the river bank is a fine, non-cohesive material, it may be washed away through the voids in rip rap or gabion lining. This can be prevented by using a textile filter, or retarded by means of a gravel filter layer half as thick as the rip rap layer with a grading conforming to the following calculation:

6.2

Protection methods

6.2.1 Pier and abutment foundations Mass concrete foundations set on erodible soils need protection against local scour, though they should be unaffected by general scour if the size of the waterway is adequate. Figure 6.6 indicates the recommended position for rip rap. The top of the protective layer must not protrude above the lowest general scour level, as determined by the calculations outlined in Chapter 5. The engineer may select a suitable stone size from Figure 6.1, remembering that local velocities can be as high as 1.5 times the mean calculated velocity through the waterway opening. If large enough stone for rip rap is too expensive, gabions or Reno mattresses may be used. The appropriate mattress thickness can be read from Table 6.2.

where D15, D50 and D85 are stone sizes such that 15 per cent, 50 per cent and 85 per cent respectively of the stones are smaller by weight. 'Base' refers to the river bed material.

6.1.4 Piled walls Timber, steel or concrete piles may be driven to form a continuous wall, with the aim of either providing protection against erosion, or channelling the main

6.2.2 Bank protection The same guidelines apply to rip rap stone size for bank protection as for substructure foundations (Figure 6.1). The local velocity against the slope of a bank aligned more or less parallel to the flow may be

51

52 taken as 0.7 times mean flood velocity where the channel is straight. The local velocity at the outer bank on a severe bend may be as high as 1.5 times mean velocity. Reno mattresses, laid with their longest dimension up the bank, provide effective protection against bank erosion and scour in a similar manner to rip rap. Both methods may be used with a launching apron, as shown in Figure 6.7. Apron length la should be 1.5 times the total scour depth DS On most river banks, grasses and shrubs grow up between the stones of rip rap and mattresses, their roots protected by the revetment which they in turn help to stabilise. Steep banks are better protected by driven pile walls or gabion walls with the front face vertical or raked back at a slope of about 1 in 10 (Figure 6.8 type 1).

6.2.4 Groynes Groynes can take the form of permeable or impermeable walls and may be constructed using piles, gabions, rip rap or mass concrete. The purpose of a groyne, or series of groynes, is to divert the river flow, which is also influenced by the other features of the river bed. It follows that the length, height and location of the groynes depends not only on the required path of the river in flood but also on the river banks, bed material and any other obstructions in the vicinity, including the substructures of the bridge. Groynes should never impinge on the area of general scour. Nonetheless, they may affect the pattern of local scour at the bridge by diverting the current and may generate local scour holes, depending on the angle between the groyne and the water current. Figure 6.12 shows some typical scour effects of river groynes, but local features and experience drawn from other structures on the same river will provide the best guidance.

6.2.3 Guide banks On meandering and braided rivers, whose widths are greater than necessary to take the flood discharge, guide banks can be used: to confine the flow to a single channel to control the angle of attack on piers and prevent meanders encroaching on and eroding the approach roads.

6.3 References I L 0 (1986). Training element and technical guide for SPWP workers. Booklet No 3. International Labour Office, Geneva. Farraday and Charlton (1983). Hydraulic factors in bridge design. Hydraulic Research Station Ltd., Wallingford.

A single guide bank may be sufficient when the stream can be confined to one side of a valley possessing a natural inerodible bank (Figure 6.9). Two guide banks are necessary when the waterway opening is in the middle of a flood plain. The minimum width between guide banks is the width necessary to provide the required waterway opening through the bridge. A greater width may be preferable because it is more difficult and expensive to place an earth embankment in flowing water than to construct it on dry land or in still water. The guide banks should extend upstream by about three quarters of the waterway opening and downstream by about one quarter, unless a longer length is necessary to prevent an upstream meander working behind the guide bank and breaching the approach embankment (Figure 6.10). The plan shape of the guide bank often takes the form of a quarter section of an ellipse with a ratio of major to minor axis of 2.5 to 1, the major axis being parallel to the principal diversion of flood flow through the opening (Figure 6.11). Guide banks should be protected by revetments and by launching aprons, as shown in Figure 6.7. The top of the bank should be wide enough for maintenance vehicles and the ends of the banks should be rounded (Figure 6.11). They should extend above design HEL with a suitable freeboard allowance.

52

Figure 6.9 Single and twin guide banks

Figure 6.10 Guide banks general principles 53

Figure 6.11 Guide bank profile

Figure 6.12 Typical effects of groynes on river flow

54

7
7.1 7.2

LOW LEVEL WATER CROSSING

Road approaches 57 Fords .. 57 7.2.1 7.2.2 7.2.3 Boulders . 58 Gabions .. 58 Piles ... 58

7.3 7.4 7.5 7.6

Bed level causeways . 60 Vented causeways . 61 Submersible bridges .. 61 Signs and markers . 63 7.6.1 7.6.2 Depth gauges . 63 Guide posts 63

7.7 7.8

Designing for maintenance ... 64 References . 64

55

56

Low level water crossing

In favourable conditions, low level water crossings can provide economical and relatively simple alternatives to conventional bridges. There are two basic types of low level crossing: fords and bed-level causeways; vented causeways and submersible bridges.

7.1 Road approaches The best location for a low level crossing is similar to that recommended for a conventional bridge (Section 2.1.2), with the exception that a broad, straight stretch of river will provide slower and shallower water as well as easier road approaches. The maximum gradient for motor traffic should be 10 per cent, and for animals 5 per cent. The change of gradient should be gradual to prevent the underside of vehicles touching the road and to preserve long sight distances. A small change in horizontal alignment of the road at the crossing helps to draw the attention of drivers to a dip that may conceal an obstacle. The equal cut and fill construction of the approach roads shown in Figure 7.la requires less work than the cut and remove spoil of Figure 7.lb, but the placing of the spoil in the river channel may cause scour problems during a flood. On roads where speeds are naturally slow and where the horizontal curve on the approach side is clearly visible, steep approaches may be relieved by a diagonal descent of the river bank, as shown in Figure 7.2. Even where the road is a single track, it is advisable to make the crossing and its approaches two lanes in width, so as to allow traffic to pass any vehicle which breaks down or falls to mount the gradient. 7.2 Fords Fords are the simplest form of river crossing at places where the stream is wide, shallow and slow, the approaches gentle, and the surface firm. Improvements to the approaches are usually confined to reducing the gradient. The running surface in the

Both types are appropriate for sites where traffic volumes are low or where a reasonably short detour provides access to an all-weather bridge. For most of the year the maximum depth of water over the carriageway should be less than 150mm. The success or service life of the structure will depend considerably on its hydraulic design as outlined in Chapter 5. Fords and bed-level causeways, like conventional bridges, may be constructed so that they cause little interference with the design flood. Vented causeways and submersible bridges inevitably disrupt river flow, and so are liable to sustain damage or indirectly cause scour to the river bed or banks, which in turn may affect the road approaches to the crossing. Since fords and bed-level causeways are overtopped by any water flowing in the river channel there is no reason to raise the road surface above the stream bed. Vented causeways and submersible bridges usually present a dry carriageway for ordinary flows and are overtopped during the design flood.

Figure 7.1 Sections at a ford showing alternative road approaches

57

Figure 7.2 Plan views showing steep and shallower approaches to a ford

stream can be strengthened and made more even using stones brought in and buried just below the surface. Alternatively, if stones are carried in the flow, these may be trapped by barriers made of boulders, gabions or piles; see Bingham (1979), Hindson (1983), Metschies (1978), Leydecker (1973).

7.2.1 Boulders Large stones placed across the river bed at the downstream side of the crossing are claimed to filter the flow of water and retain gravel and sand, which eventually form a more level and even surface for vehicles. However, if the stones are too large or form too high a wall (>300mm), scour may result; if they are not heavy enough, they will be washed away at the first flood. Figure 7.3a shows a typical cross section of a ford with downstream boulders.

The gabion baskets are wired together and dropped into a prepared trench. The central gabion is filled first and, using that as an anchor, the line of gabions is pulled taut and straightened by a chain attached to a truck or winch. This tension is maintained while the remaining baskets are filled. When filling is complete, final adjustments are made to the top course of rock and the baskets are closed. Section 6.1.2 provides a more detailed account of the use of gabions. As an alternative to conventional gabions, a tube basket can be made from a roll of fencing mesh laid across the bed of the stream and filled with rock. During filling, the edges are raised and bent over at the top to form a tube and then tied. Finally a wire rope is attached as shown in Figure 7.3c and securely anchored at each end. Tube baskets need to be installed in a previously excavated trench approximately half the depth of the basket, ie 0.2 to 0.3m. After installation, sand and gravel transported by the stream is trapped behind the basket, building up a fairly level surface suitable for the passage of vehicles.

7.2.2 Gabions A more durable improvement may be made to the running surface by replacing the boulders with gabions as shown in Figure 7.3b. It is important that the gabions protrude no more than l50-300mm above the natural bed level of the river, otherwise they will act as a weir and cause heavy scour downstream of the crossing. 58

7.2.3 Piles Where gabions are unobtainable, timber piles driven into the river bed with or without stone protection make an acceptable cut-off wall (Figure 7.3d).

Figure 7.3 Ways of improving the running surface at a ford

59

To be fully effective, timber piles should have a length of about 2m, and should be placed at no more than 0.6m centres. if the river is fast flowing a continuous line of piles may be necessary. A curtain wall may be required on the upstream as well as the downstream side if the road bed is erodible.

7.3 Bed level causeways Where the type of traffic or the lack of a nearby allweather crossing justifies the expense, a pavement may be laid on the river bed. A paved ford is also called a bed-level causeway, drift, paved dip or Irish bridge (Roberts 1986). Figure 7.4 illustrates three common designs. Masonry can also be used instead of concrete: see Chapter 15 for more details.

Figure 7.4 Bed level causeways

60

To protect the pavement from scour damage, curtain walls are usually required on both sides of the roadway and these must continue up the approaches to the height of the design flood. It is recommended that curtain walls should be 1m deep on the upstream side and 2m deep on the downstream side, unless rock is reached before that depth. If the bed is not susceptible to erosion, the causeway need not have curtain walls but the bed on both the upstream and downstream sides of the crossing should be trimmed flat to reduce turbulence. Figure 7.4a shows a Section through a basic bed level causeway suitable for light traffic and maximum water flows below 2m/sec. The crossing shown in Figure 7.4b requires good concrete technology and may sustain damage to the apron that is difficult to repair. Figure 7.4c shows a design employing a good combination of concrete pavement with flexible protection. Generally, a 1:2:4 concrete, by volume, is used and slabs are jointed using crack inducers every 5m.

may be overtopped for a few days each year during the annual high flood. Since these structures present a considerable obstacle to the free flow of both an ordinary flood and the design flood, they must be built sufficiently massive to withstand water pressure and debris impact. They must also have scour protection (Chapter 6), where the bed is erodible, and marker posts. The vents are usually concrete or corrugated steel pipes from 0.6m to l.0m diameter, set in a block of concrete or masonry. Where pre-cast concrete pipes are unavailable, vaulted masonry tunnels have proved successful. Concrete or masonry retaining walls and aprons are needed to channel the flow and prevent scour at both entrance and exit. To prevent blockage of the stream by debris or silting, the pipes must be set level with the stream bed and at the same slope. No part of the vents should be narrower than the entrances. Wedge-shaped deflector walls may be required on the upstream side to guide large floating debris above the vents. Figure 7.6 shows a typical section and elevation of a vented causeway. Figure 7.7 gives dimensions for the concrete cover and reinforcement details.

7.4 Vented causeways Vented causeways are built where the river flow is too great for too many days in the year to allow the traffic to cross a ford or bed level causeway. Structures like the one depicted in Figure 7.5 are, for hydraulic purposes, multiple pipe culverts and should be designed following the guidelines in Chapter 5 and Chapter 12. However, the design flood used to calculate the vent sizes will be less than the annual flood, provided it is acceptable that the carriageway

7.5 Submersible bridges Where the traffic density justifies a dry crossing of a substantial ordinary flood and the annual high flood is much greater, a submersible bridge designed to pass the ordinary flood but to be overtopped by the high

Figure 7.5 Vented causeway 61

Figure 7.6 Vented causeway horizontal section and elevation

62

Figure 7.7 Vented caueway - vertical section

flood will be considerably cheaper than a high level, all weather bridge. Submersible bridges have the advantage of being able to pass a larger flow than the vents of a causeway of the same height, but are more susceptible to damage by the river. The overturning moment at the pier foundations becomes very large unless the piers are kept short, and the horizontal and vertical forces on the decks require solid restraint. Because of these difficulties submersible bridges are not recommended for any foundation other than rock, and even then a vented causeway or conventional bridge is likely to be a more durable alternative. However, submersible bridges have served well in some areas, and should the designer wish to adapt for this purpose the shorter standard concrete deck drawings included in Chapter 9, the only major modification required is to replace the standard fixed and free bearings by monolithic concrete with continuous reinforcement. Spans greater than 6m are unlikely to be economical.

warning of the change in vertical and horizontal alignment, and indicate the possibility of encountering water on the carriageway. In addition to advance warnings, there should be a depth gauge at each end of the crossing and guide posts on both sides for the whole length.

7.6.1 Depth gauges Depth gauges should indicate the depth of water at the lowest point of the crossing. Simple black and white markings are best with an indication of the units used. Posts should be about 300mm diameter or square, placed within easy vision of the approach but well away from possible impact damage by vehicles.

7.6 Signs and markers Like all other water crossings, low level crossings should be well signed in advance. Signs should give

7.6.2 Guideposts Guide posts should be set each side of the carriageway between 2m and 4m apart, depending on the likelihood of catching floating debris. There are two opinions about their height: a) They should be visible only when the water is shallow enough for vehicles to cross. 63

b) They should be high enough to be visible during the highest expected floods. This is the preferred method where depth gauges are well maintained, as it permits drivers to exploit the capabilities of different types of vehicle. The posts should be of durable timber, metal or concrete, depending on the materials used for the carriageway surface and the curtain walls, eg using concrete posts on concrete bases and timber posts set into gabions. An additional guide for vehicles may be provided by building a ridge down the centre of concrete causeways, as show Figure 7.4b. This ridge also offers restraint against sideways drifting of vehicles in strong currents.

each year. Figure 7.8 from BCEOM (1975) illustrates the problem.

7.8 References BCEOM (1975). Manuel d'excution de petits ouvrages routiers en Afrique. Ministre de la Cooperation, Pans. Bingliam J (1979). Low water crossings. Compendium 4. Transportation Research Board, Washington. Hindson J (1983). Earth Roads - A practical guide to earth road construction and maintenance. Intermediate Technology Publications, London. Leydecker A (1973). Use of gabions for low water crossings on primitive or secondary roads. Engineering Technical Information Service, Field Notes 5, No.6. Washington: US Forest Service. Metschies G (1978). Manual on road construction. Economic and Social Commission for Asia and the Pacific, Bangladesh. Roberts P (1986). The Irish Bridge - a low cost river crossing. University of Southampton, Department of Civil Engineering, Southampton.

7.7 Designing for maintenance Submersible crossings of all types require more maintenance than most conventional bridges. The structural design should allow for easy repair of routine damage, which is usually due to scour. On highly erodible beds it is often more successful to build gabion curtain walls and use Reno mattress aprons than to build a rigid concrete structure, recognising that some rebuilding will be required

Figure 7.8 Repair to flexible curtain walls at a ford

64

8
8.1 8.2 8.3

SUBSTRUCTURES AND FOUNDATIONS

Abutments ... 68 Piers 68 Determination of allowable bearing pressure . 68 8.3.1 8.3.2 8.3.3 8.3.4 Non-cohesive soils ... 68 Cohesive soils .. 70 Plate bearing tests 71 Presumed values .. 73

8.4

Abutment design . 73 8.4.1 8.4.2 8.4.3 8.4.4 8.4.5 8.4.6 Mass concrete abutments . 73 Mass concrete retaining walls .. 73 Reinforced concrete abutments ... 74 Reinforced concrete retaining walls 74 Raft foundations .. 74 Bearing shelves ... 74

8.5 8.6 8.7 8.8 8.9 8.10

Reinforced concrete piers ... 74 Foundations on rock 74 Earthquake restraint 74 Run-on slabs ... 75 References .. 75 Standard drawings and tables for abutments, retaining walls and piers . 76 8.10.1 8.10.2 Index of drawings 76 Index of tables ... 100

Appendix A8.1 . 114

65

66

8 Substructures and foundations

This chapter deals with the elements of a bridge that support the superstructure, ie abutments, piers and their foundations. Figure 8.1 shows some typical shapes of substructures in concrete. To assist the highway engineer by removing the necessity for complex individual calculations, the chapter includes a set of standard PC concrete abutment and pier drawings, with tables specifying dimensions, reinforcement and other details, for spans up to 12m and a range of soil conditions. The designs are conservative and take account of all the external forces from live and dead loads that substructures are normally expected to sustain. Should the engineer prefer to work from first principles or modify any of the drawings presented here, reference to Victor (1980), Pennells (1978) and Tomlinson (1986) will provide procedures and formulae.

If experience is available in structural masonry and this material is an economical alternative, it may be used as a substitute for the mass concrete detailed here. However, the engineer must be satisfied about the strength of the materials used, particularly when they are submerged in flowing water. The mass concrete abutments and retaining walls detailed in Figures 8.11 and 8.13 are conservatively designed but any changes, such as replacing the concrete stem with masonry, should be carefully evaluated, taking into account all the forces and reactions from the superstructure, the embankments and the soil under the foundations. A reinforced concrete bearing shelf for use with masonry abutments is illustrated in Figures 8.10 and 8.11, and detailed in Figures 8.20 and 8.21. Timber abutments or gabion abutments may be used

Figure 8.1 Typical substructures

67

for temporary structures but are not recommended for permanent bridges, even those with timber superstructures. Reference to Section 13.3 will provide details of timber and gabion abutments. Concrete abutments and piers as described here may be built to support a timber superstructure that will be replaced at a later date with a more permanent material. In this case the final dead weight and width, etc., must be used in the abutment design. All concrete decks must have rigid substructures like those detailed in this chapter, because uneven settlement of either abutment or pier can result in unacceptably high stresses in the materials of the decks. The positioning of the abutment and pier foundations is critically important: the alignment of the bridge and its interaction with the river are discussed extensively in Chapters 2, 4, 5 and 6. Since the most likely cause of substructure failure is scour, the engineer should pay careful attention to the estimation of general and local scour described in Chapter 5. Pier foundation depths are specified according to foundation type and protection method (Figure 8.2). There are two key points to bear in mind: the general scour area must not be obstructed or the flow will be impeded and further scour damage will result; local scour is caused by turbulence and may be contained by armouring the bed. Abutments also fail when the soil under the foundation is not strong enough to counter the combined forces from the structure and the embankment. It is recommended that spread foundations be used wherever possible, but if adequate support as defined in Section 8.3 is unavailable, a piled foundation is required. Piling techniques and materials vary from region to region. Tomlinson (1977) is a good general guide on this subject. Piling for permanent bridges is a specialist technique that should be entrusted only to experienced contractors. Where a satisfactorily strong foundation soil is found not too far below preferred foundation level, caisson support may be considered. The technique is simple if the caissons are short, but the engineer must take care that: the maximum soil reaction at the sides does not exceed the maximum passive pressure at any depth the soil pressure at the base remains compressive throughout and the maximum pressure does not exceed the allowable pressure as defined in Chapter 3. Victor (1980) contains guidance on the use of both piles and caissons. The drawings of spread foundations included in this chapter need to be amended if piles or caissons are employed. 68

8.1 Abutments In addition to supporting the dead load of the superstructure, the abutments of a bridge must: resist the vertical and horizontal live loads placed on them by vehicles and the elements; retain the approach embankments and the live loads applied to them; provide a smooth transition from the road surface to the deck running surface.

Figure 8.3 indicates the main elements of abutments. Their essential features are: a foundation slab, which transmits the weight of the abutment and a proportion of the superstructure directly to the supporting soil, or which forms a capping slab to a system of load-bearing piles; a front wall with bearing shelf that supports the superstructure and usually retains the soil of the embankment; wing walls or retaining walls may be separate from the abutments or, if they are short, may be built integrally with them. These walls retain the road embankment or river bank adjacent to the abutment and are usually built so as to bisect the angle between the road and the river bank, though they can be set at any angle to the abutments and may be built parallel to the road or perpendicular to it.

8.2 Piers Piers also perform a support function. They convey vertical and horizontal loads from the superstructure via the bearing shelf, stem and foundation slab to the supporting soil. in many instances, piers stand on saturated soils for most or all of the year: they do not retain soil embankments but are designed to withstand hydraulic pressures and impact loads. Piers are often more susceptible to scour damage than abutments and need to be orientated carefully with respect to flow direction. Their foundations should be located well below maximum scour depth.

8.3

Determination of allowable bearing Pressure

8.3.1 Non-cohesive soils The allowable bearing pressure under foundations in non-cohesive soils is governed by the permissible settlement of the structure due to consolidation of the soils under the applied loading. If standard penetration tests have been performed in boreholes (Section 3A.1.3) the values of N can be

Figure 8.2 Pier foundations in relation to scour 69

Figure 8.3 Main elements of an abutment

used to obtain allowable bearing pressures for various foundation dimensions (Figure 8.4). The allowable bearing pressure in this context is that which causes 25mm of settlement under the given breadth of foundation front to back, Br, ie measured perpendicular to the direction of water flow, on the assumption that the water table always remains at a depth of at least Br below foundation level. If the water table can be higher than this, the allowable pressures should be halved. Before applying the relationships shown in Figure 8.4, the engineer may need to correct the standard penetration values measured in the boreholes, since it has been found that the test seriously underestimates the relative densities of cohesionless soils at shallow depths. To allow for this, a correction factor, obtained from Figure 8.5, should be applied to the measured values. Appendix B.1 provides an example of the calculation procedure.

Where the N-value of a fine or silty sand below the water table is greater than 15, the density of the soil should be assumed to be equal to that of a sand having the N-value of 15 +(N-15). Very loose uniformly graded sands with N equal to 5 or less and subject to rapid changes of water level are liable to suffer large settlements under load. In these circumstances, either the sand should be dug out and thoroughly recompacted or the foundation should be supported on piles. In areas that experience seismic activity, foundations must not bear on single size material.

8.3.2 Cohesive soils Most cohesive soils at foundation level are saturated and have an angle of shearing resistance equal to zero, provided that no water is expelled from the soil as the load is applied. This is accepted as the basis for

70

calculating the ultimate bearing capacity of foundations where the load is applied relatively quickly. The ultimate bearing capacity of cohesive soils can be calculated from the following formula: Ultimate bearing capacity qf = cu Nc +p Where Cu = undrained shear strength (kN/m2) (section 3.4) Nc = bearing capacity factor P = total overburden pressure at Foundation level (kN/m2) = .D (8.1)

P where g

= density of soil above foundation level (kN/m) (submerged density where below water level) = depth of foundation level below ground surface (m).

Values of the bearing capacity factor Nc for square or circular foundations can be read from the graph in Figure 8.6. Figure 8.4 Allowable bearing pressure/penetration value relationships For rectangular foundations

Where:

Br = breadth of foundation front to back L = length of foundation

The undrained shear strength, cu of soft clays can be measured by means of field vane tests, but these results need to be corrected because the soil is sheared in a horizontal direction. The value of ca, to shear strength multiplied by the correction factor read from the graph in Figure 8.7. This factor is dependent on the plasticity index of the soil. The allowable bearing pressure is one-third of the calculated ultimate bearing capacity. 8.3.3 Plate bearing tests Ultimate bearing capacity values from plate bearing tests may be used to check the results calculated on the basis of Section 8.3.1 or 8.3.2. However, these tests are carried out with relatively small plates and stress only the soil immediately below the plate. Since pier and abutment foundations cover a larger area and their influence penetrates to greater depths Figure 3.2), the whole volume of soil affected by the structure requires examination.

Figure 8.5 Correction factors for cohesionless soils at shallow depths

71

Figure 8.6 Bearing capacity factors

Figure 8.7 Vane shear strength correction factor

72

8.3.4 Presumed values At the preliminary stage of design there may be no measured values of soil density or field strengths available. For purposes of estimation, Table 8.1 lists approximate values of allowable bearing pressures for different soil types.

Table 8.1 Presumed bearing values

8.4 Abutment design The standard designs Figures 8.8 to 8.30, set out in Section 8.10, show typical shapes and dimensions for abutments and retaining walls, built with mass concrete and reinforced concrete for heights up to 6.5m and for a range of allowable bearing pressures. Though correct in themselves, these drawings require careful adaptation, taking into account all the individual characteristics of the Site and the superstructure, eg foundation conditions, deck thickness, expansion joints etc. Specifications for the concrete and steel are given in Chapter 14, while Chapter 16 discusses the presentation of drawings and specifications.

8.4.1 Moss concrete abutments (Figures 8.9 to 8.11) The width of the top of the abutment is fixed at 1200mm to provide an adequate bearing seat for the bridge deck. Since the front face of the wall stem slopes at 1 in 10 and the rear face at 1 in 2, the width of the bottom of the wall stem is a function of the wall height H. If H is 5m, the width of the bottom of the wall stem is:

The width of the abutment foundation, which is of reinforced concrete, depends on the allowable bearing pressure of the ground, the span and the height of the abutment. The width of the toe A and the heel B and the foundation thickness D are given in Table 8.2 for a range of waterway openings from 4m to 12m, wall heights from 2m to 6.5m and presumed bearing pressures from 400kN/m2 to l00kN/m2. Table 8.3 gives details of the high yield steel reinforcement required in the foundation slab and Table 8.4 indicates the alternative mild steel reinforcement.

8.4.2

Moss concrete retaining walls (Figures &12 and 8.13) These are shown in Figure 8.10 in the common configuration bisecting the angle between river and road. The width of the bottom of the wall stem varies with the height as in the case of the abutment, but it should be noted that the slope of the front face of the stem is shown in Figures 8.12 and 8.13 as 1 in 14:

73

this is geometrically necessary for alignment with the front face of the abutment only when the wing wall is set at 45 . Widths of the toe and heel, the thickness of the wall foundation and details of the required high yield or mild steel reinforcement are set out in Tables 8.5 and 8.6.

The road approaches should be built to prevent water draining onto the bridge, but some water falling on the deck will penetrate expansion joint seals and leak through to the bearing shelves. This is particularly likely to occur when no seal at all is provided (Figure 14.2). A number of drainage configurations are available, Hambly (1979), but the two main principles to be observed are: slope horizontal surfaces to direct water away from the bearing pads; provide good access for the removal of stones, vegetation, bird nests and other debris.

8.4.3

Reinforced concrete abutments (Figures 8.14 and 8.15) Like the mass concrete abutments, these are shown with a standard width of 1200mm at the top but the front and rear faces are vertical. This structure is considerably lighter than its mass concrete counterpart and so requires different foundation details. The bearing shelf now becomes an integral part of the stem. The widths of the toe and heel, and the thickness of the foundation for various heights, span and bearing pressures are shown in Table 8.7. High yield and mild steel reinforcement for the foundations, detailed in Figure 8.17, are specified in Tables 8.8, and 8.9 respectively. Reinforcement for the wall stem is detailed in Table 8.10.

8.4.4

Reinforced concrete retaining walls (Figures &16 and &17) These match the abutments with reinforced concrete stems. Foundation dimensions and reinforcement are detailed in Tables 8.12, 8.13 and the stem reinforcement in Table 8.11. If the retaining wall is not long, the weep holes can usually be omitted. These details apply to the typical case of walls set at 450 to the abutments supporting road embankments with slopes of 1 in 2.

8.5 Reinforced concrete piers (Figures 8.22 to 825) Though piers may be built using masonry or mass concrete, reinforced concrete has several advantages, notably a more slender stem presenting less interference to flow and hence causing less induced scour. The superstructure spans detailed in Chapters 9, 10 and 11 are designed to be simply supported at the abutments and at the piers. Bach span should have one fixed and one free end. It is usual practice, though not essential, to provide one fixed bearing and one free bearing on the bearing shelf of each pier (Figure 9.2). Pier foundations are even more susceptible to damage by erosion than abutment foundations. They must be constructed on soils of well established allowable bearing pressure, on the basis of the guidelines in Chapter 3 and Section 8.3. Foundation sizes are detailed in Table 8.16 according to height, spans and allowable bearing pressures. Tables 8.17 and 8.18 detail the reinforcement sizes for the foundations and Table 8.15 for the stem.

8.4.5 Raft foundations (Figure 8.18) Where waterway openings are less than 4m wide and where the allowable bearing pressure values are too low to support standard abutment foundations, single or multiple reinforced box culverts should be used. Alternatively a continuous raft, forming a foundation to both abutments and an invert to the waterway, may be built with an articulated deck. Table 8.14 indicates the thickness of the raft and details of high yield and mild steel reinforcement. Scour protection is usually required both upstream and downstream of the raft, as in the case of vented causeways (Section 7.4).

8.6 Foundations on rock The foundation designs presented in the preceding sections are for soils readily excavated by hand or mechanical digger. Modifications may be required to suit individual site conditions, particularly when bedrock is encountered. Where foundations are set on rock at ground level or on the river bed, substantial keying will be necessary in the form of steel dowels and notching.

8.4.6 Bearing shelves (Figures 8.19 to 8.21) These designs are generally suitable for concrete, composite or timber decks, though the bearing details will be specific to the deck type. Figures 9.2 and 10.12 show the dowel and bearing pad details required for concrete and composite decks. Good drainage and the facility for removal of debris are important requirements on all bearing shelves.

8.7 Earthquake restraint (Figures 8.26 and 8.27) The bridge bearings detailed in drawings Figures 9.2, 9.3, 10.11 and 10.12 are designed to support and restrain the superstructures under normal loading from traffic, wind and temperature change, but would not be robust enough to resist the severe forces that can result from seismic activity. Bridge authorities in earthquake zones can give guidance on the precautions normally required in the area. These precautions most often take the form of stronger lateral restraint to prevent

74

sideways movement of the bridge superstructure on the abutments and/or wider bearing shelves to prevent the spans falling off their supports. Extra restraint can be provided quite easily for composite decks; a typical solution is shown in Figure 8.27. This solution is not appropriate for concrete slab decks, which are usually restrained by upstands on the bearing shelf, as shown in Figure 8.26. In areas prone to earthquakes these precautions are required at both abutments and piers.

Tomlinson M J (1977). Pile design and Construction practice. Cement and Concrete Association, London. Tomlinson M J (1986). Foundation design and construction. Longman Singapore Publishers Pte. Ltd., Singapore. Victor D J (1980). Essentials of bridge engineering. Oxford and IBH Publishing Co., New Delhi.

8.8 Run-on slabs Almost all earth embankments are subject to settlement. The amount they settle will depend on the height, the degree of compaction of the material and the strength of the subgrade. Relatively uniform settlement can be expected from most embankments until a cause of uneven compaction is met, such as a bridge or a box culvert with little fill above it (Figure 8.28a). It is difficult to compact fully the embankment material close to the bridge abutments or the culvert walls, and the result of poor compaction is more pronounced settlement. The resulting longitudinal profile is uncomfortable for road users and causes impact loads on the structure, owing to vehicle bounce. These local depressions in the carriageway close to drainage structures may be bridged using run-on slabs (Figures 8.28b, 8.29 and 8.30). They are more easily constructed at the same time as the structure, rather than afterwards as a remedial measure, and they span the fill material susceptible to settlement. One end of the slab rests on a small shelf cast onto the culvert wall or on the abutment ballast wall, while the other rests on well compacted material several metres away. Run-on slabs are usually between 3 and 6 metres long. The concrete and reinforcement details may be abstracted from the culvert detailing in Chapter 12, assuming that the slab is resting on good support for one third of its length, ie a 6m slab will have similar details to a 4m wide culvert lid. Construction details of the hinged end are shown in Figures 8.29 and 8.30. Run-on slabs are usually made wide enough to support the kerbs on the approach roads. Run-on slabs should not be required where efficient maintenance facilities are readily available (Hambley 1979).

8.9 References Hambly E C (1979). Bridge foundations and substructures. Building Research Establishment, Stationery Office, London. Pennells E (1978). Concrete bridge designer's manual. Cement and Concrete Association, London.

75

8.10 8.10.1

Standard drawings and tables for abutments, retaining walls and piers Index of drawings

Title

Figure number

Typical elevation of a single span bridge .. 8.8 Abutment - front elevation . 8.9 Abutment - plan ... 8.10 Abutment, mass concrete - vertical section .. 8.11 Retaining wall, mass concrete - plan and elevation . 8.12 Retaining wall, mass concrete - vertical section .. 8.13 Abutment, reinforced concrete - vertical section . 8.14 Abutment, reinforced concrete - side elevation 8.15 Retaining wall, reinforced concrete - plan and elevation . 8.16 Retaining wall, reinforced concrete - vertical section .. 8.17 Raft foundation . 8.18 Bearing shelf - front elevation... 8.19 Bearing shelf - details .. 8.20 Bearing shelf - reinforcement ... 8.21 Piers - general arrangement .. 8.22 Pier - plan and elevation ... 8.23 Pier - vertical section 8.24 Pier - horizontal section 8.25 Earthquake restraint - concrete superstructure . 8.26 Earthquake restraint - composite superstructure .. 8.27 Embankment settlement profiles at a culvert or bridge 8.28 Culvert run-on slab ... 8.29 Bridge run-on slab 8.30

76

Figure 8.8 Typical elevation of a single span bridge

77

78 Figure 8.9 Abutment front elevation

Figure 8.10 Abutment - plan

79

Figure 8.11 Abutment, mass concrete vertical section

80

Figure 8.12 Retaining wall, mass concrete plan and elevation

81

Figure 8.13 Retaining wall mass concrete vertical section

82

Figure 8.14 Abutment, reinforced concrete vertical section

83

Figure 8.15 Abutment, reinforced concrete side elevation

84

Figure 8.16 Retaining wall, Reinforced concrete plan and elevation

85

Figure 8.17 Retraining wall, reinforced concrete vertical section

86

Figure 8.18 Raft foundation

87

88 Figure 8.19 Bearing shelf on reinforced concrete abutment front elevation

Figure 8.20 Bearing shelf details

89

90 Figure 8.21 Bearing shelf - reinforcement

Figure 8.22 Piers general arrangement

91

Figure 8.23 Pier plan and elevation

92

Figure 8.24 Pier, reinforced concrete vertical section

93

Figure 8.25 Pier, reinforced concrete horizontal section

94

Figure 8.26 Earthquake resitraint concrete superstructures

95

Figure 8.27 Earthquake restraint composite superstructures

96

Figure 8.28 Embankment settlement profiles at a culvert or bridge

97

Figure 8.29 Culvert run-on slab

98

Figure 8.30 Bridge run-on slab

99

8.10.2

Index of tables

Title

Table number

Abutment, mass concrete foundation sixes ... 8.2 Abutment mass concrete high yield steel reinforcement in the foundation .. 8.3 Abutment mass concrete mild steel reinforcement in the foundation 8.4 Retaining wall, mass concrete foundation sizes 8.5 Retaining wall, mass concrete foundation reinforcement . 8.6 Abutment, reinforced concrete foundation sizes .. 8.7 Abutment, reinforced concrete high yield steel reinforcement in the foundation 8.8 Abutment, reinforced concrete mild steel reinforcement in the foundation . 8.9 Abutment, reinforced concrete stem reinforcement 8.10 Retaining wall, reinforced concrete stem reinforcement 8.11 Retaining wall, reinforced concrete foundation sizes . 8.12 Retraining wall, reinforced concrete foundation reinforcement . 8.13 Raft foundation depth and reinforcement 8.14 Pier stem reinforcement ..8.15 Pier foundation sizes ...... 8.16 Pier foundation high yield steel reinforcement 8.17 Pier foundation mild steel reinforcement 8.18

Note: The steel reinforcement used in these tables is either: MS Plain round mils steel bar, or HYS round deformed high tensile steel bar according to the specifications detailed in Section 14.2

100

Table 8.2 Abutment, mass concrete foundation sizes (m)

101

102

Table 8.3 Abutment, mass concrete high yield steel reinforcement in the foundation (mm)

Table 8.4 Abutment, mass concrete mild steel reinforcement in the foundation (mm)

103

104

Table 8.5 Retaining wall, mass concrete foundation sizes (m)

Table 8.6 Retaining wall, mass concrete foundation reinforcement (mm)

Table 8.7 Abutment, reinforced concrete foundation sizes (m)

105

106

Table 8.8 Abutment, reinforced concrete high yield steel reinforcement in the foundation (mm)

Table 8.9 Abutment, reinforced concrete mild steel reinforcement in the foundation (mm)

107

108

Table 8.10 Abutment, reinforced concrete stem reinforcement (mm)

Table 8.11 Retaining wall, reinforced concrete stem reinforcement (mm)

Table 8.12 Retaining wall, reinforced concrete foundation sizes (m)

Table 8.13 Retaining wall, reinforced concrete foundation reinforcement (mm)

109

110

Table 8.14 Raft foundation depth and reinforcement

Table 8.15 Pier stem reinforcement (mm)

Table 8.16 Pier foundation sizes (m)

111

112

Table 8.17 Pier foundation high yield steel reinforcement (mm)

Table 8.18 Pier foundation mild steel reinforcement (mm)

113

Appendix A8
Appendix A8.1 Example of bearing pressure calculation Determine the allowable bearing pressure of a 2.5m wide foundation, bearing on dry sand at a depth of 1.2m below ground level, Figure A8.1. The average N value measured in boreholes over a depth of 2.5m (ie foundation width) below foundation level is 17. Now the average overburden pressure (over a depth equal to the foundation width below the foundation) = Overburden density x Average depth of penetration test x Gravitational force. Assuming an Overburden density = 1.8 Mg/m3, Average depth = foundation depth + (0.5 x foundation width) = 1.2m + (0.5 x 2.5m) = 2.45m Hence the average overburden pressure = 1.8 x 2.45 x 9.8 = 43 kN/m2 From Figure 8.5 the correction factor is 2.0, so the N value to be used in Figure 8.4 is 17 x 2.0, ie 34. Reading up from the Br value, 2.5, in Figure 8.4 to an N value of 34, the allowable bearing pressure is 350 kN/m2.

Figure A8.1 Bearing pressure calculation 114

9
9.1 9.2

CONCRETE SUPERSTRUCTURES
Standard designs 117 Design details 117 9.2.1 9.2.2 9.2.3 9.2.4 9.2.5 9.2.6 9.2.7 9.2.8 9.2.9 9.2.10 Vertical profile over the bridge 117 Bearings ...117 Expansion joints .. 117 Construction joints ...118 Slab reinforcement ...118 Drainage ...118 Parapets 118 Surfacing ..118 Services carried on the bridge ..118 Provision for pedestrians and cyclists ..119

9.3 9.4

References .119 Standard drawings and tables for concrete superstructures ..120

115

116

9 Concrete superstructures
Concrete will often be the natural choice of material for the bridges on roads where a long service life and minimum maintenance are required. Not all concrete structures are trouble-free, however, and it is important for engineers and contractors to understand the factors which influence concrete durability. It is assumed here that the engineer is familiar with good concrete practice and that he and guidelines in Section 14.1. A solid concrete slab provides an economical and practical solution for spans up to 12m. It has a number of advantages over the concrete beam-and-slab form, due mainly to its simpler construction. This makes it easier to obtain both good compaction around the reinforcement and a dense surfacing which prevents the ingress of pollutants, both advantages leading to greater durability. Solid slab decks also show better load distribution qualities than beam and slab forms. This chapter gives standard designs for concrete slab decks form 4m to 12m spans for one, one and a half, and two lanes of traffic, for both BS-HA and HS20-44 loadings, using mild steel (MS) and high yield steel (HYS) reinforcement. Span lengths are measured between bearing centres as shown Figure 8.18 8.22. Traffic loadings and bridge width are discussed in Sections 2.5 and 2.6.

or navigation clearance and slab depth are known. If the deck level is required to be higher than the adjacent road, the superstructure will describe a hogging vertical curve over the bridge. This curve should be designed to the applicable highway standards, Odier et al (1974). The slab should be cast to follow this profile: thus a longitudinal section through the bridge would show a constant depth of slab and surfacing, (Figure 9.1). This detail must be made clear on the drawings, otherwise there is a risk that the contractor will cast the soffit horizontally and make up the vertical profile by varying the thickness of the surfacing, which would add excessive deadload to the deck.

9.2.2 Bearings The simplest form of bearing is made by casting the concrete slab onto the abutment bearing shelf, with only a layer of bitumen felt separating the two concrete surfaces. This may be satisfactory for very short spans, say up to 6m, but for longer spans there is a risk that thermal movement will damage both the supporting structure and the slab. For this reason, simple rubber strip bearings are shown supporting the slab in Figures 9.2 and 9.3 These bearings consist of discrete strips of black natural rubber (Section 14.6), extending over the full width of the slab soffit at the support point, with a maximum width of approximately 300mm and a maximum thickness of approximately 25mm. At the free end of the span, the bridge deck locates by friction between the rubber strip and the concrete deck, with no positive mechanical means to develop resistance against transverse loadings. At the fixed end, dowel bars passing through the pad at frequent intervals provide the necessary restraint, both longitudinally and transversely. To allow rotations to occur and the deck to expand laterally, the dowel bars are usually fitted with rubber caps where they pass into the concrete deck slab.

9.1 Standard designs The standard designs at the end of this chapter conform to either BSI (1990a) or AASHTO (1983) (Section 2.8). These designs are conservative in strength and may be modified in detail, but structural changes should be made only by a qualified civil or structural engineer.

9.2

Design details

9.2.1 Vertical profile over the bridge The vertical profile of the road over the bridge can be determined once the high flood level, floating debris

9.2.3 Expansion joints An expansion joint is required at each end of each span. Though the fixed end moves less than the free end, similar joints are usually installed at both.

Figure 9.1 Bridge with hogging vertical profile

117

Figure 9.2 shows a suitable joint between adjacent slabs of a multi-span bridge. This same type of joint can be used between the abutments and deck slab (Figure 9.3), one end of the deck being doweled and the other free. Flexible joint seals at the road surface are not recommended for bridges on gravel roads; Section 14.7 describes a more suitable alternative.

profile with a low point on the bridge, drains must be installed on the deck. The simplest way is to provide a 100mm diameter PVC pipe at the kerb line, dishing the concrete around it, and setting a metal gully grating in the surfacing (Figure 9.16). The engineer must take care to see that the discharge does not cause damage to services or sub-structures, and ensure that close attention is given to the compaction of the deck concrete around the pipe.

9.2.4 Construction joints Construction joints often provide paths for the seepage of water, leading to leaching of lime from the concrete and corrosion of the reinforcement. For this reason, it is preferable for the slab to be cast in one continuous pour. If the engineer knows that this will not be possible because the batch volume required is too much for one day's working, it is better for him to mark on the drawings a permissible location for construction joints. If a joint is unavoidable, it should be parallel to the centre line and preferably located at the high point of the transverse camber (Figure 9.16).

9.2.7 Parapets Figure 9.17 shows details of a suitable steel parapet system. These follow the general principles of the Highways Agency (1993) but circular sections have been selected for the rails since they are more readily available in most countries. Badly damaged posts or rails can be unbolted and replaced. If proprietary square section beams are available, they should be used because they cause less damage to vehicles on impact. 'W section' guardrails may also be used to good effect. Provision should be made for differential expansion and contraction between steel and concrete elements, particularly at the superstructure expansion joints. To reduce the danger of a rail penetrating a vehicle on impact, parapet rails should present a continuous face to traffic, even at expansion joints. Care should be taken with details of the holding-down fixing into the concrete plinth. As well as installing the metre-long bars through the U-bars, two extra links (reinforcement bar marked D in Figures 9.3 to 9.14) should be used at each post position, one each side close to the U-bars. Bridge parapets usually extend several metres beyond the superstructure at each end, and may be supported by the wing walls, in order to provide a barrier at the approaches.

9.2.5 Slab reinforcement The maximum length of reinforcing bar generally available is 12m. When a longer bar is needed, eg for a 12m span, two bars must be lapped. As a general rule the lap length should be at least 40 times the diameter of the bar. The main bars should not be lapped near the centre of the span, and laps should be staggered so that they do not cause a line of weakness across the slab. The lapping of bars should be avoided whenever possible so as to ensure good continuity, minimise congestion of reinforcement and so produce better compaction of concrete around the steel. The anchorage and positioning and reinforcement are particularly important where the slab sits on its bearings. For the standard designs all bars have been continued to the supports. Main bars are L-bars, marked C in Figures 9.4 to 9.15, and are positioned with the hooks at alternate ends. It is important to position the 90 degree hook so that the bend begins behind the bearing, as shown on the drawings. Accurate positioning of the reinforcement is essential in order to maintain the minimum cover of 50mm of well compacted concrete, and so avoid moisture and pollutants penetrating the slab and corroding the steel reinforcement.

9.2.6 Drainage If the bridge is single span with a hogging vertical curve and the joints between deck and abutments are sealed, surface water can be allowed to run off the bridge into the road drainage channels and no drains are required on the bridge superstructure. If the bridge is multi-span, or has a flat or sagging

9.2.8 Surfacing The standard slabs are designed to be used with 75mm of asphalt surfacing. This provides some degree of waterproofing and protection of the surface of the structural slab. In some countries it is common practice to use the upper surface of the slab as a running surface. This is not recommended, but should the engineer wish to avoid using asphalt, the minimum cover to reinforcement in the top of the slab should be increased to 75mm and kerb joint and drainage details have to be modified accordingly. Bridges serving gravel roads may need more than 75mm cover to counteract the abrasive action of stones embedded in vehicle tyres.

9.2.9 Services carried on the bridge It will sometimes be necessary to make provision for services and utilities such as electricity or telephone

118

cables to be carried on the bridge. These services can be readily accommodated in PVC ducts cast into the verges behind the kerbs, as shown in Figure 9.16. The ducts cost very little and could be included even if there were no immediate need. It steel pipes are attached to the deck, they must have provision for expansion at the fixings. 9.2.10 Provision for pedestrians and cyclists Wherever there is likely to be significant movement of pedestrians and cyclists across the bridge, a segregated pathway is recommended (Section 2.7). This pathway can be provided by an addition cantilevered out from the bridge superstructure; but as noted in Section 2.7, such additions tend to be less robust than the main structure and there is a possibility of critical asymmetric overloading of the deck, should a large number of people congregate on the pathway. A more durable, safer and often simpler solution is shown in Figure 9.18: widening the deck by 2m beyond the width required for motorised traffic, protecting the pedestrians and cyclists with the standard parapet, and installing a lighter parapet at the edge of the walkway. Comparison of the cross sections through the single, one and a half, and two lane decks shows that the

construction is identical in terms of concrete and steel. Only the width is changed. Additions can be made for pedestrians and cyclists on one or both sides by extending the same section and maintaining the spacing of the steel reinforcement. A wider superstructure will require wider abutments and piers. Should these additional features prove to be too expensive, a separate structure might be considered as an acceptable means of providing a safe crossing for non-motorised travellers. 9.3 References AASHTO (1983). Standard Specification for Highway Bridges. 13th edition. American Association of State Highway and Transportation Officials, Washington. BSI (1990a). Steel, concrete and composite bridges Part 4. Code of practice for design of concrete bridges. BS 5400 Pt 4: 1990. British Standards Institution, London. Highways Agency (1993). BD 52/93: The design of highway bridge parapets. Department of Transport, London. Odier et al (1974). Low Cost Roads design, construction and maintenance. UNESCO publication by Butterworth, London,

Figure 9.18 Segregated walkway

119

9.4

Standard drawings and tables for concrete superstructures

Title

Figure number 9.2 9.3

Pier expansion joint and bearing details -----------------------------------------------------------------------------------------Abutment expansion joint and bearing details -----------------------------------------------------------------------------------

Bridge design loading BS-HA

Bridge width 1 lane

Reinforcement MS ------------------------------------------------------------------------------------------ 9.4 HYS ---------------------------------------------------------------------------------------9.5

1 lanes

MS ------------------------------------------------------------------------------------------ 9.6 HYS ----------------------------------------------------------------------------------------- 9.7

2 lanes

MS ------------------------------------------------------------------------------------------ 9.8 HYS ----------------------------------------------------------------------------------------- 9.9

HS20-44

1 lane

MS ----------------------------------------------------------------------------------------- 9.10 HYS --------------------------------------------------------------------------------------- 9.11

1 lanes

MS ----------------------------------------------------------------------------------------- 9.12 HYS --------------------------------------------------------------------------------------- 9.13

2 lanes

MS ----------------------------------------------------------------------------------------- 9.14 HYS --------------------------------------------------------------------------------------- 9.15

Ducts, drainage and construction joints ------------------------------------------------------------------------------------------ 9.16 Parapets -----------------------------------------------------------------------------------------------------------------------------In the following drawings, the steel reinforcement is either MS - plain round mild steel bar or HYS - round deformed high tensile steel bar, according tot he specifications detailed in Section 14.2 A wearing course of asphaltic concrete 75mm thick is specified. Any change to this may require corresponding changes to the abutments. 9.17

120

Figure 9.2 Concrete superstructure, pier expansion joint and bearing details

121

Figure 9.3 Concrete superstructure, abutment expansion joint and bearing details

122

Figure 9.4 Reinforced concrete deck slab

123

Figure 9.5 Reinforced concrete deck slab

124

Figure 9.6 Reinforced concrete deck slab

125

Figure 9.7 Reinforced concrete deck slab

126

Figure 9.8 Reinforced concrete deck slab

127

Figure 9.9 Reinforced concrete deck slab

128

Figure 9.10 Reinforced concrete deck slab

129

Figure 9.11 Reinforced concrete deck slab

130

Figure 9.12 Reinforced concrete deck slab

131

Figure 9.13 Reinforced concrete deck slab

132

Figure 9.14 Reinforced concrete deck slab

133

Figure 9.15 Reinforced concrete deck slab

134

Figure 9.16 Ducts, drainage and construction joints

135

136 Figure 9.17 Parapets

10 STEEL/CONCRETE COMPOSITE SUPERSTRUCTURES

10.1 10.2 Design standards . 139 Design details .. 139 10.2.1 10.2.2 10.2.3 10.2.4 10.2.5 10.2.6 10.2.7 10.2.8 10.2.9 10.2.10 10.2.11 10.2.12 10.3 10.4 Vertical profile over the bridge . 139 Construction .. 139 Shear connectors .. 139 Protective treatment to steelwork .. 140 Bolts and nuts 140 Bearings . 140 Expansion joints 140 Construction joints 140 Slab reinforcement 140 Drainage 140 Parapets, surfacing and services 140 Provision for pedestrians and cyclists 140

References 140 Standards drawings and tables for composite superstructures 141

137

138

10

Steel/concrete composite superstructures

In countries without an indigenous steel industry, the need to use foreign exchange for structural steelwork often inhibits designers from considering steel bridges. However, there are situations in which the use of steelwork will be both technically appropriate and costeffective. Because they are easy both to erect and remove, and can be used again elsewhere, steel beams with a timber decking are often a good solution for structures intended to have a short service life (ie no more than 10 years). Their durability, however, is limited by the inability of the decking to prevent surface water and dirt from penetrating through to the steel main beams. Though regular maintenance painting can prevent deterioration of the beam webs and bottom flanges, serious corrosion can occur at the inaccessible beam top flanges, where the timber and steel meet. Where structures are intended to be permanent, the durability of the steel over a service life of 50 years or more can be achieved more readily by the use of a cast in situ concrete deck slab. Composite action of the slab and beams is secured by the use of shear connectors welded to the top flanges of the beams and cast into the concrete. As an alternative to the solid concrete slab decks described in Chapter 9, this chapter details a series of standard bridge decks constructed from steel universal beams (UB) with a composite concrete deck slab. The main beams and cross members are of standard rolled carbon steel sections (yield stress 274N/mm2, with deck slab reinforcement in either mild steel (MS) or high yield steel (HYS). The steel and concrete specifications are given in Chapter 14. Steel/concrete composite deck structures have the following advantages: The deck self weight can be less than that of an equivalent all-concrete structure. The off-site prefabrication of the main loadcarrying elements of the bridge substantially reduces the work necessary on site, resulting in more rapid construction. No temporary supports are required during of the deck slab, since the soffit shutters can be supported directly from the steel beams. This can be a particular advantage at locations with poor ground conditions, steeply sloping terrain, or with a fast stream. Steel is a reliable material which is supplied with guaranteed strength properties, enabling structures of high and consistent reliability to be produced.

10.1 Design standards The bridge decks detailed in this chapter range from 6m to 12m in length with roadway widths of one, one and a half, and two lanes. Span lengths are measured between bearing centres as shown in Figures 8.18 and 8.22: the designs are in accordance with BSI (1979) for HA loading and AASHTO (1985) for HS20-44 loading.

10.2

Design details

10.2.1 Vertical profile over the bridge The bridge should preferably be constructed either to a level profile or to a constant longitudinal grade, if this is required by the road alignment. The steel beams can be cambered to give some degree of hogging vertical curvature should the equipment be available, but the cost is unlikely to be justifiable.

10.2.2 Construction The heaviest beams required for the standard bridges detailed in Figures 10.1 to 10.11, ie those for a 12m span length, weigh less than two tonnes. They may therefore be erected using a small mobile crane or improvised lifting equipment. Alternatively, the beams may be pushed or pulled out over the gap being bridged using some form of intermediate temporary support and/or a light launching nose attached to the beams. Propping of the steel beams during deck concreting is unnecessary, since the steel section has been designed to be adequate to support the wet concrete, soffit shutters and construction loading. The engineer must appreciate that the wet concrete loading constitutes a critical design case for the steel beams. Wind loading at this stage adds significantly to the stresses in the beams. For this reason, it is necessary to ensure that the concrete slab is not poured if wind speeds of 1 8mlsec (40mph) or more are occurring or are expected within the following 48 hours.

10.2.3 Shear connectors These connectors are the only links between the concrete slab, acting in compression and bending when under load, and the steel beams, acting in tension and bending. The largest shear forces act at each end of the deck, where the shear connectors are spaced closer together. Figure 10.7 gives the dimensions and spacing for each span and load rating. Welded channel shear connectors are specified, since they can be produced from readily available material and fixed locally. The joints, however, should be of good quality welding and be protected from corrosion in storage.

139

10.2.4 Protective treatment to steelwork The degree of protection which the steelwork will require depends on the local environment. Particular care will be needed for structures in coastal locations or where there is significant atmospheric pollution. Types of paint used and surface preparation methods will depend on the local availability of materials and equipment. The engineer should try to achieve the following standard in order to ensure a reasonable life to first maintenance: l) grit blasting to remove millscale, loose rust, welding scale etc., and produce a clean surface for painting;

10.2.9 Slab reinforcement As noted in Section 9.2.5, the maximum length of reinforcing bar generally available is 12m. Where a longer bar is required, eg for a 12m span bridge, two bars must be lapped. The lap length should be at least 40 times the diameter of the lapped bars and laps should be staggered both to avoid a line of weakness and to minimise congestion of reinforcement. Main bars are Lbars, marked C in Figures 10.4 to 10.6, and are positioned with the hooks at alternate ends. Accurate positioning of the reinforcement is essential in order to maintain the minimum cover of 50mm of well compacted concrete, so that moisture and pollutants cannot penetrate the slab as far as the reinforcing bar and corrode it.

2) application of a multi-coat paint system to a total dry film thickness of 0.25mm. At least one paint coat should be applied at site after completion of construction, so that damage to paintwork incurred during transport, steel erection and concreting can be repaired. 10.2.10 Drainage Drainage should be provided if required, generally as detailed for concrete slab bridges (Section 9.2.6. and Figure 9.16). Down pipes must be of sufficient length to ensure that run-off water is discharged at least 150mm clear below the beam lower flanges.

10.2.5 Bolts and nuts Ordinary bolts, grade 8.8 to ISO (1982) together with grade 8 nuts, are specified for fixing cross members. Alternatives should match the tensile strength of 80 kgflmm2 with a minimum elongation at fracture of 12%.

10.2.11 Parapets, surfacing and services Parapet and surfacing details are the same as for concrete slab bridges (Sections 9.2.7 and 9.2.8). If ducts for services are required, they can be provided as set out in Section 9.2.9.

10.2.6 Bearings Elastomeric bearings, as detailed in Sections 9.2.2 and 14.6, are specified because they are durable, inexpensive and simple to install.

10.2.12 Provision for pedestrians and cyclists The recommendations set out in Section 9.2.10 on pathways for pedestrians and cyclists on concrete decks apply equally to composite decks. The addition of a 2m wide walkway to a composite deck of the type detailed here would also require one more I beam to match those for the road bridge.

10.2.7 Expansion joints At these relatively short deck lengths, joint movements due to temperature and live loading are small and are readily accommodated by a simple gap joint, as indicated in Figures 10.11 and 10.12. The joint is sealed by a polysulphide sealant to prevent water penetration. An alternative unsealed joint, more suitable for bridges on gravel roads, is described in Section 14.7.

10.3 References AASHTO (1985). Standard specifications for highway bridges, 13th edition. American Association of State Highway and Transportation Officials, Washington. BSI (1979). Steel, concrete and composite bridges Part 5. Code of practice for the design of composite bridges. BS 5400 Pt 5: 1979. British Standards Institution, London. ISO (1982). Fasteners - hexagon products-widths across flats. ISO 272:1982

10.2.8 Construction joints As discussed in Section 9.2.4, it is best if the slab can be cast in one continuous pour. If this will not be practicable, permissible locations for construction joints should be marked on the drawings. If a joint is unavoidable, it should be perpendicular to the centre line at a location least likely to promote corrosion in underlying steelwork.

140

10.4 Title

Standards drawings and tables for composite superstructures Figure number

Single lane dimensions . 10.1 One and a half lane dimensions 10.2 Two lane dimensions 10.3 Single lane reinforcement . 10.4 One and a half lane reinforcement 10.5 Two lane reinforcement ....10.6 Shear connector details . 10.7 Fixing details - mid span cross members ..10.8 Fixing details - end trimmers 6m and 8m spans ... 10.9 Fixing details - end trimmers l0m and 12m spans ..10.10 Pier expansion joint and bearing details . 10.11 Abutment expansion joint and bearing details 10.12 In the following drawings, the steel reinforcement is either: MS - plain round mild steel bar or HYS - round deformed high tensile steel bar, according to the specifications detailed in Section 14.2 A wearing course of asphaltic concrete 75mm thick is specified. Any change to this may require corresponding changes to the abutments.

141

Figure 10.1 Composite deck 1 lane width

142

Figure 10.2 Composite deck 1 lane width

143

Figure 10.3 Composite deck 2 lane width

144

Figure 10.4 Composite deck single lane reinforcement

145

Figure 10.5 Composite deck 1 lane reinforcement 146

Figure 10.6 Composite deck 2 lane reinforcement

147

Figure 10.7 Composite decks shear connector details

148

Figure 10.8 Composite decks fixing details of mid-span cross members

149

Figure 10.9 Composite decks fixing details of end trimmers 6m and 8m spans

150

Figure 10.10 Composite decks fixing details of end trimmers 10m and 12m spans

151

Figure 10.11 Composite decks pier expansion joint and bearing details 152

Figure 10.12 Composite decks abutment expansion joint and bearing details

153

154

11 TIMBER SUPERSTRUCTURES
11.1 11.2 11.3 11.4 11.5 11.6 11.7 Log bridges 157 Sawn beam bridges 158 Timber decking ..158 Footways 158 Abutments ..158 References . 158 Standard drawings and tables for timber superstructures .. 160

155

156

11 Timber superstructures
Timber is a viable load-bearing material for the smaller bridge decks and can offer the following potential advantages: low material cost for short spans; relatively modest requirements for skilled labour and equipment; no need to have a dry river bed for building shuttering during construction; ability to compensate for a certain amount of settlement of substructures without strain.

which have been in service for more than a hundred years, to knee-braced beams and glue-laminated arches. Though these designs are not considered appropriate to the present manual, which focuses on small bridges and simple procedures, they may be viable where specialised expertise is available. Further designs of timber beam decks, braced and trussed, may be found in Masani (1952), Federal Highway Administration (1969), Canadian Institute of Timber Construction (1970) and Ritter (1990). Where logs or sawn timber beams are unavailable or very expensive, truss-type decks may be viable. One such design, which uses multiples of a standard prefabricated frame, is described and detailed in TRADA (1985) and Parry (1981). Two other useful reference books for engineers wishing to design timber decks from first principles are Pearson, Kloot and Boyd (1962) and American Society of Civil Engineers (1975). The classification of structural timber for use in bridges is discussed in Section 14.4.2.

The limitations of timber as a bridge deck material include: susceptibility to rotting and insect attack need for frequent maintenance heavy duty or long span timber decks require a lot of material, and so may be more expensive than steel or concrete alternatives.

This chapter contains designs for single-lane timber beam decks, suitable for spans up to 12m. They are simple to construct and are particularly useful for the fast replacement of superstructures that have suffered damage (Chapter 13). There are many designs of timber decks, ranging from the famous covered truss bridges in the USA, some of

11.1 Log bridges Log bridges, such as the one with no parapet shown in Figure 11.1, are common on rural access roads in many countries. Figures 11.2 and 11.3 give details of a standard design for a single lane log bridge to carry AASHTO H520-44 loading. Whenever possible seasoned logs should be used. They should be closely matched and positioned the same way, ie with the larger diameter ends all at the same end of the deck, so that the decking planks can be fixed in contact

Figure 11.1 Timber bridge on a minor road 157

with all the logs. It is better to notch the underside of those logs which are high at the abutment than to remove a greater quantity of material along the upper surface of the logs. The maximum span is usually limited by the size of tree found locally, or by the maximum length which can be transported.

An edge beam serves as a kerb and ties the ends of the deck planks together. A pedestrian handrail is provided by extending deck planks at the post positions only, to carry an inclined post brace.

11.2 Sawn beam bridges Rectangular timber beams are also used as the main spanning members for bridges. Construction is easier with such a regular shape, since each member rests on a flat surface and fixing of the decking to the beams is more positive. Cutting allows close examination, so timber can be carefully selected to avoid serious defects. Considerable time, effort and cost, however, may be incurred in transporting logs to a saw mill, and the sawn section is not as strong as the log from which it was cut. Figure 11.4 shows standard designs for sawn beam bridges for spans up to 12m to carry AASHTO HS 2044 loading.

11.4 Footways Even though timber bridges tend to be confined to rural areas, some carry significant numbers of pedestrians. When there is motorised traffic using the bridge at the same time as high pedestrian flows, eg at times when children are going to and from school, a segregated footway should be provided on the main structure or separate from it. The footway should be at least 1.5m wide: if attached to the main structure, it may be supported on one or two additional beams as shown in Figure 11.5. A standard parapet is provided at the outer edge and another between the pedestrians and motorised traffic. The posts for this parapet are best braced below the deck in order to avoid obstructions to the footway. If all deck planks are continued across the footway, there is no need for longitudinal running strips. Alternatively, the engineer can extend only those deck planks that are required to support the outer parapet and attach continuous running strips across the full width of the footway.

11.3 Timber decking The deck is constructed by using timber boards, usually l00mm thick, set across the top of the logs. To allow water to drain easily and to avoid dirt and moisture being trapped between deck planks, an air gap of 20mm is left between the planks. In practice the logs are rarely uniform enough to fix all decking planks directly to each log. Rather than force a plank down onto a log, hardwood packing pieces should be used to make up the gap, as shown in Figure 11.3. To protect the deck timbers and encourage drivers to position their vehicles centrally on the bridge, two strips of longitudinal running boards are fixed to the decking at centres that will accommodate the wheel spacing of conventional motorised vehicles. The strips should also be sufficiently wide and spaced to cater for all common local vehicles, such as bullock carts and rickshaws as well as lorries. It is also beneficial to fix transverse threshold boards across the width of the bridge at each end of the running strips. They lessen the impact on the ends of the running strips and reduce the tendency for them to work loose. Planks for the deck and running strips should be placed with the surface that was nearest to the centre of the tree face-down, so that their edges do not curl upwards on further drying. Nailing is the least successful method of fixing the decking and running boards, since movements of the deck under traffic loads tend to pull the nails loose. Galvanised coach screws or coach bolts and nuts are far more satisfactory. The running strips in particular should be bolted to the deck planks.

11.5 Abutments The abutments can be made of masonry or concrete, as shown in Chapter 8. It is important that the abutment bearing shelf is well drained and that air is able to circulate freely around the ends of the logs. These should sit on a raised bearer, clear of moisture, soil and debris which may collect on the bearing shelf, and they may be secured in position with timber wedges fixed to the bearer. Abutments for temporary timber bridges are discussed in Section 13.3.

11.6 References American Society of Civil Engineers (1975). Wood Structures - a design guide and commentary. American Society of Civil Engineers, New York. Canadian Institute of Timber Construction (1970). Modern Timber Bridges - some standard details. Canadian Institute of Timber Construction, Ottawa. Federal Highway Administration (1969). Standard plans for highway bridges, Volume 3 Timber Bridges. Federal Highway Administration, Bureau of Public Roads, Washington. Masani N J (1952). Simple calculations in the design of forest bridges of stock spans of 152030 and 40 feet. Survey of India, Delhi.

158

Parry J D (1981). The Kenyan low cost modular timber bridge. Laboratory Report LR970. Transport Research Laboratory, Crowthorne. Pearson R G, KIoot N H and Boyd J D (1962). Timber engineering design handbook. Jacaranda Press Pty Ltd, second edition, Melbourne. Ritter M A (1990). Timber Bridges - design, construction, inspection and maintenance. U S Department of Agriculture, Forest Service, Washington. TRADA (1985). Prefabricated modular wooden bridges. Timber Research and Development Association, High Wycombe.

159

11.7 Standard drawings and tables for timber superstructures

Title

Figure number

Log or timber beam bridge - longitudinal section .11.2 Log bridge - cross section .11.3 Timber beam bridge - cross section ..11.4 Timber decks with segregated footway .11.5

160

Figure 11.2 Log or timber beam bridge longitudinal section

161

Figure 11.3 Log bridge cross section

162

Figure 11.4 Timber beam bridge cross section

163

Figure 11.5 Timber decks with segregated footway

164

21 Culverts
12.1 12.2 12.3 12A 12.5 Erosion control 167 Location and alignment ...167 Headwalls 170 Concrete pipe culverts .170 Flexible 12.5.1 12.5.2 12.5.3 12.5.4 12.6 12.7 12.8 steel culverts .. 171 Foundations 171 Backfill ...171 Design 171 Durability and corrosion protection ... 172

Reinforced concrete box culverts 172 References ... 172 Standard drawings and tables for concrete box culverts .173 12.8.1 12.8.2 Index of drawings .. 173 Index of tables 181

165

166

12 Culverts
There are many similarities between bridges and culverts, and they perform similar tasks. Bridges, however, usually accommodate longer spans: they consist of free-standing abutments and a separate articulated superstructure which carries the traffic. Culverts are often made of pre-fabricated pipes or boxes, or are cast in one or two pieces: they are usually set low in an embankment and less often bear the direct weight of traffic. Where the waterway opening is less than about 15m2, and particularly where the road crosses the waterway on a relatively high embankment, a culvert will usually be cheaper than a bridge. The hydraulic design of culverts is discussed in Section 5.6.1. Culverts may be constructed using a variety of materials from vitreous clay to glass reinforced plastics, but the most common and cheapest forms of construction are: precast concrete jointed pipes; corrugated steel pipes made of prefabricated panels; single or multi-cell reinforced concrete boxes, prefabricated or built in situ.

corrugated steel pipe culvert, well protected against corrosion by plating and bitumen coating, can be expected to have a working life in the order of 30 to 40 years in a non-aggressive environment. It is usual to design culverts to last the life of the highway. Despite the best efforts of the engineer, it must be expected that some culverts will become silted or obstructed by debris. For this reason, pipes of internal diameter less than 0.6m are not recommended since they are difficult to clean.

12.1 Erosion control The construction of a road and associated drainage structures may have the effect of taking general water run-off from a slope and channelling it. This concentration of flow is almost certain to cause erosion damage to many types of soil. Before or at the entrance to culverts it is important to control the flow velocity, particularly since an abrupt change in direction of flow may occur at these points making erosion damage more likely. Control may be achieved by reducing the gradient, installing checkdrains or by building a drop inlet. It is also important to control the velocity of flow at the outlet, because erosion damage to the road embankment or surrounding farmland occurs more frequently at the discharge end. For this reason, where gradients are steep, the arrangements shown in Figures 12.1 b and 12.1 c are preferred to that shown in Figure 12.1 a. A curtain wall is often necessary at the outlet of a culvert carrying more than a minimal flow, in order to prevent scour undermining the structure. A concrete or rip rap apron may be required to dissipate the energy of the discharge, or it may be necessary to contain the flow in a paved channel until the natural gradient becomes small enough not to induce fast flows with consequential deepening of the channel. Figure 12.2 shows the most common failure of culverts installed in soft soils, due to erosion and undermining which progress, if not checked, to the point of embankment collapse.

Single precast concrete pipe culverts are commonly used for small openings up to 2m2 while multiple concrete pipes with common headwalls or corrugated steel pipes cater for larger areas. Alternatively, reinforced concrete box culverts are used with internal box sizes up to 4m x 3.5m. Twin or multiple boxes may be required for larger waterway openings. In difficult ground conditions a flexible steel pipe has an advantage over a rigid concrete culvert through its ability to accommodate a certain amount of differential settlement over the length of the culvert without overstressing the material. A culvert made of rigid concrete sections will not be tolerant to differential settlement unless it is specifically designed for such conditions either by increasing its structural strength or by segmenting the culvert along its length to allow it to flex. In theory such a solution may appear to be satisfactory, but the provision of watertight joints to permit flexing to occur can be costly and the satisfactory performance of the joints cannot always be guaranteed. Leakage from culverts, particularly those carrying embankments, can have serious consequences, since weakening of the embankment material may lead to an embankment failure. If properly constructed, a reinforced concrete culvert is likely to have a service life in excess of 60 years and will almost certainly be more durable and require less maintenance than a steel pipe. By comparison, a

12.2 Location and alignment The proper location of each culvert is important because it affects the drainage pattern and hence the strength of the surrounding soils, the area of waterway, ditch and culvert maintenance, and erosion damage. There are four rules which, though obvious to experienced engineers, are worth listing here: Where a road crosses a valley, the lowest point requires a vent, whether there is an established stream or not.

167

Figure 12.1 Reducing water velocity through a culvert

168

Figure 12.2 Typical erosion where a culvert discharges into erodible soil

Where there is an established stream, the culvert should follow the existing alignment, unless the alignment can be improved. The gradient of the culvert should be the same as the gradient of the stream. Measures may be necessary to ensure that the watercourse does not move. This could cause severe damage and the consequent change of location of the culvert would be expensive.

the stream, there is likely to be a sediment problem. A gradient of 2 to 4 per cent is advisable where silts are carried in the flow; a minimum of 0.5 per cent is recommended for clear water. It is also important to set the culvert invert at the same level as the natural stream bed. Culverts are frequently set low to avoid humps in the road above, which results in silting and a consequent reduction in the waterway. Where the road embankment is not high enough to provide adequate clearance above the pipe, the ditch has to be lowered on both sides of the road or the embankment raised. This may necessitate widening the side slopes and demonstrates the importance of detailing drainage structures early in the design of a road. Where an established stream is met at an angle to the road alignment, it is usually better to follow the line of the stream with a skewed culvert, even though the construction cost will be increased by the greater length. An abrupt change in the direction of flow at the inlet and outlet of a square crossing may be accommodated without problem for most of the year, but when heavy rains cause maximum flow the change of direction can result in severe erosion or wash-out.

As well as venting at the lowest point, it is good practice to install culverts for cross drainage at regular intervals down a long grade. This avoids the necessity of building a large culvert at the bottom of the grade and may also provide the opportunity for safer dispersal of water in smaller mitre drains on the lower slope. The appropriate frequency of these cross drains is best decided on the basis of local experience and depends on gradient, soil characteristics, intensity of rainfall and related factors. As a general rule, there should be at least one culvert every 300m, unless the road follows a ridge. The gradient of the culvert is important because it affects future maintenance. If it is too steep, it will encourage erosion at the outlet. If it is less steep than

169

It may also be necessary to stabilise a shifting channel or even move it to improve the geometry where stream and road interact. The principal objectives here are: to avoid a bend at either end of the culvert if possible; if a bend is unavoidable, to place it at the outlet. Figure 12.4 Culvert location in a side ditch

Any change of stream channel must be constructed so that there is no possibility of the stream regaining its original course. Figure 12.3 taken from Armco (1938) illustrates these principles.

12.3 Headwalls Headwalls retain and protect the embankment at the ends of a culvert and help to counteract the dislocation of jointed pipes due to the spreading forces at the base of large embankments. They also divert debris from falling into the waterway, reduce seepage of water into the embankment and prevent small animals from burrowing alongside the pipe. Where conditions favour erosion, headwalls should incorporate deep curtain walls and side slope protection in the form of gabions or hand-pitched stone. Since large headwalls are expensive to build (Figure 12.5), it is advisable initially to make the culvert pipe long enough to reach the toe of the embankment and to monitor the effects for a year: a small headwall may be adequate, provided that side slopes are shallow enough not to need retaining, embankments are relatively low or one-piece pipes are employed, and there is no buildup of water. Headwalls are usually made of masonry or concrete. Figures 12.6 and 12.7 show standard designs for large headwalls. Small walls may be built to similar proportions, leaving out the reinforcement. Matching retaining walls, where required, can be detailed using Figures 8.12 and 8.13 or Figures 8.16 and 8.17. Headwalls, side slopes and retaining walls are illustrated in Figures 7.5, 12.8 and 12.12. Sheet metal headwalls can be purchased with sheet metal pipes.

Figure 12.3 Preferred culvert alignments

On some minor roads, ditches cross small entrances and are blocked, either deliberately or by the action of traffic. Where pipe culverts are installed, they are often placed away from the road to reduce the length. It is better for the long-term performance of the road to install pipe culverts at the outset and locate them as shown in Figure 12A. Where ditches are spaced away from the carriageway, culverts should be built from ditch to ditch, not just across the width of the pavement, so as to move the location of possible erosion damage away from the pavement. Building the culvert long enough to reach the base of the embankment also reduces the additional cost of building high headwalls.

12.4 Concrete pipe culverts On rural access earth roads, sections of concrete pipe are butted together and the embankment material is compacted around them. This is not the most permanent construction, but is cheap and often serves adequately for several years. A far better quality culvert is obtained by surrounding the pipe with concrete (Figure 12.8 and Figure 12.9). The concrete cladding resists the tensile forces that open the pipe joints and strengthens the pipe against bending. If the subgrade is likely to settle under the weight of the embankment, the concrete surround should be reinforced no matter what the depth of cover between pipe and road surface. All soft clay material and rock should be excavated

170

Figure 12.5 Illustration of headwall costs

from below the pipe to a depth of at least 200mm and replaced with good granular material, well compacted. Sections of concrete pipe may be bought locally or manufactured in-house using steel moulds. The relevant British Standard Specification is BSI (1982) or, for pipes with flexible joints, BSI (1988).

Soft foundations All soft unstable material should be excavated and backfilled to foundation level with sand/gravel mixture, crushed stone or other suitable material.

12.5 Flexible steel culverts These are made from pre-formed corrugated sections bolted together to form a circular or elliptical tube. They can accept a certain amount of longitudinal or radial deformation without sustaining damage and so are better suited to unstable conditions than rigid concrete structures. They are also easier to transport and faster to construct than the equivalent size of precast pipe culvert, but require specialist manufacture. Corrugated steel sections may also be used as permanent formwork for masonry arch structures.

Swampy foundations Where deep unstable foundations are encountered which cannot be stabilised with granular material, timber fascines can be used to spread the load.

Rock foundations Rock should be excavated to at least 250mm below the foundation level and wide enough to prevent the pipe resting directly on rock at any point. It should then be backfilled to provide a cushion for the pipe.

12.5.1Foundations Although they are to some degree flexible, for optimum performance and service life corrugated steel pipes require a firm foundation that distributes the load evenly. Where settlement is likely to occur it will be greatest under the centre of the embankment: the engineer should therefore consider the provision of a longitudinal camber to compensate for any tendency for the pipe to sag over its length. Normal practice is to adopt a camber of 1 per cent of the pipe length. To avoid the possibility of upstream ponding, the upstream half of the pipe can be laid level and all the drop provided in the downstream half.

12.5.2 Backfill The first stage is to settle the assembled pipe onto the foundation material and compact the fill under the haunches. At this stage it may be necessary to load the inside, in order to prevent the pipe being wedged upwards by the fill. Backfilling is continued in layers of approximately 200mm, loading each side evenly and fully compacting each layer before placing the next. The inside shape may need to be protected by strutting as the fill is built up on each side and eventually above the pipe. Ml the fill material should be of good quality, extending to a distance roughly equal to the pipe diameter each side and above, or up to the road base if this distance is less.

Uneven foundations When the excavation crosses soft or hard spots, the foundation should be made as uniform as possible by excavating rocks, clay pockets, etc, below the proposed foundation level and replacing good selected material.

12.5.3 Design Corrugated metal pipes are made by specialist manufacturers who will supply technical publications on the detailed design of pipes and culverts, including notes on hydraulic design. Manufacturers also offer a design service to the engineering industry. The names and addresses of two manufacturers are given in Section 12.7. 171

12.5.4 Durability and corrosion protection The degree of protection against corrosion required to extend the working life of a corrugated steel pipe structure is determined on the basis of: the importance of the structure and its required working life; the corrosive nature of the environment in which it exists.

Thorough compaction of the fill material on both sides of the concrete structure is essential to prevent settlement under traffic and to preserve the good vertical profile of the carriageway. If past experience suggests that there may be excessive settlement of the embankment on either side of a box culvert, the engineer should consider the inclusion of run-on slabs, as described in Section 8.8.

Durability studies indicate that metal loss associated with corrosion usually originates on the interior surface of the structure and progresses towards the exterior or buried face. The conclusion is that protective treatment to the exposed surfaces should generally be superior to that applied to the buried surfaces. All corrugated steel structures should be hot-dipped galvanised or treated with alu-zinc and hot coated with bitumen on the inside. Bituminous paving to the invert of the pipe will also help extend the life of the structure.

12.7 References Armco (1938). Handbook of culvert and drainage practice. Armco Ltd, London. BSI (1982). Precast concrete pipes for drainage and sewerage Part 3, specification for pipes and fittings with ogee joints. BS 5911.1982 Pt 3. British Standards Institution, London. BSI (1988). Precast concrete pipes for drainage and sewerage Part 100, unreinforced and reinforced pipes and fittings with flexible joints. BS 5911:1988 Pt 100. British Standard Institution, London. Enquiries concerning sheet metal culverts may be made to: Armco Ltd. 76 Jermyn Street, London SW1Y 6NP, England. Wells Spiral Tubes Ltd. Prospect Works, Airedale Road, Keighley, West Yorkshire BD2l 4LW, England.

12.6 Reinforced concrete box culverts Reinforced concrete box culverts may be constructed either as a monolithic box or as a U section with a simply supported top slab propping the side walls. Twin or multiple cell box culverts are used where the horizontal opening is more than about 4m. These too may be constructed with monolithic top slab and walls or with a separately cast lid. Culverts with separately constructed top slabs are simpler to construct because they require less complicated steel reinforcement and, though containing more material, they generally offer a more economical form of construction. For this reason monolithic boxes are not considered further. The side walls of the culverts detailed in this chapter are constructed monolithically with the base slab and are propped at the top by the simply supported roof slab. Owing to the reversal of the bending moment in the walls and bottom slab, reinforcement to both faces of these elements is required but the top slab of a single cell culvert requires only bottom reinforcement. For a twin box culvert with a separately cast top slab, the outer walls are again propped apart and the centre wall may be positively located to the top slab by the provision of dowel bars. The top slab of a twin box culvert is designed as a two span continuous beam spanning over the central wall and requires top reinforcement in the region of the central wall. Standard designs for single, twin and multiple box culverts are given in Figures 12.10 and 12.11, and Tables 12.1 to 12.20. Figure 12.12 shows typical details of single cell box culvert inlet and outlet.

172

12.8 Standard drawings and tables for concrete box culverts

12.8.1 Index of drawings Title Figure numbers

Mass concrete headwall for a high embankment ..12.6 Reinforced concrete headwall for a high embankment .12.7 Typical details of a pipe culvert with headwalls ...12.8 Details of concrete pipe surrounds 12.9 Single box culvert - section .12.10 Twin box culvert - section .. 12.11 Typical details of single cell box culvert inlet and outlet ...12.12

173

Figure 12.6 Mass concrete headwall for a high embankment

174

Figure 12.7 Reinforced concrete headwall for high embankment

175

Figure 12.8 Typical details of a pipe culvert with headwalls

176

Figure 12.9 Details of concrete pipe surrounds

177

Figure 12.10 Single box culvert - section

178

Figure 12.11 Twin box culvert - section

179

180 Figure 12.12 Typical details of single cell box culvert inlet and outlet

12.8.2 Index of tables

Title Single box culvert - high yield steel Single box culvert - high yield steel Single box culvert - high yield steel Single box culvert - high yield steel Single box culvert - high yield steel Single box culvert - mild steel Single box culvert - mild steel Single box culvert - mild steel Single box culvert - mild steel Single box culvert - mild steel Twin box culvert - high yield steel Twin box culvert - high yield steel Twin box culvert - high yield steel Twin box culvert - high yield steel Twin box culvert - high yield steel Twin box culvert - mild steel Twin box culvert - mild steel Twin box culvert - mild steel Twin box culvert - mild steel Twin box culvert - mild steel

Height of fill

Table number

H = 0.5m .12.1 H = 1.0m .12.2 H = 2.0m .12.3 H = 4.0m .12.4 H = 6.0m .12.5 H = 0.5m .12.6 H = l.0m ..12.7 H = 2.0m .12.8 H = 4.0m .12.9 H = 6.0m ...12.10 H = 0.5m ...12.11 H = l.0m 12.12 H = 2.0m ...12.13 H = 4.0m ...12.14 H = 6.0m ...12.15 H = 0.5m ...12.16 H = l.0m 12.17 H = 2.0m ...12.18 H = 4.0m ...12.19 H = 6.0m ...12.20

181

Table 12.1 Single box culvert high yield steel, H = 0.5m

Table 12.2 Single box culvert high steel, H = 1.0m

182

Table 12.3 Single box culvert high yield steel, H = 2.0m

Table 12.4 Single box culvert high yield steel, H = 4.0m

183

Table 12.5 Single box culvert high yield steel, H = 6.0m

Table 12.6 Single box culvert mild steel, H = 0.5m

184

Table 12.7 Single box culvert mild steel, H = 1.0m

Table 12.8 Single box culvert mild steel, H = 2.0m

185

Table 12.9 Single box culvert mild steel, H = 4.0m

Table 12.10 Single box culvert mild steel, H = 6.0m

186

Table 12.11 Twin box culvert high yield steel, H = 0.5m

Table 12.12 Twin box culvert - high yield steel, H = 1.0m

187

Table 12.13 Twin box culvert high yield steel, H = 2.0m

Table 12.14 Twin box culvert high yield steel, H = 4.0m

188

Table 12.15 Twin box culvert high yield steel, H = 6.0m

Table 12.16 Twin box culvert mild steel, H = 0.5m

189

Table 12.17 Twin box culvert mild steel, H = 1.0m

Table 12.18 Twin box culvert mild steel, H = 2.0m

190

Table 12.19 Twin box culvert mild steel, H = 4.0m

Table 12.20 Twin box culvert mild steel, H = 6.0m

191

192

13 EMERGENCY AND TEMPORARY STRUCTURES

13.1 13.2 The time factor . 195 Types of structures ... 195 13.2.1 13.2.2 13.2.3 13.2.4 13.3 13.4 13.5 Fords .. 195 Causeways . 195 Temporary beam bridges ... 195 Prefabricated bridge decks . 195

Abutments and piers .196 Protection of temporary structures .. 200 References ... 200

193

194

13

Emergency and temporary structures

structure will follow the order of increasing cost, unless specific skills or materials are unavailable. This order of preference is likely to be: fords; causeways; temporary beam bridges; prefabricated decks. 13.2.1 Fords A ford may be no more than a prepared descent to and exit from a river bed, taking into account the traction requirements of known traffic on the slopes (Sections 7.1 and 7.2).

Temporary bridge structures are intended to provide rapid solutions to essentially short-term problems of access. Decisions about their construction have to balance the high expense of the fastest solutions against the likelihood that they will yield benefits for only a limited period, and hence the need to keep their costs as low as possible. The principal reasons for requiring a temporary crossing are: damage to an existing bridge results in the need for a temporary structure on a nearby site while the disabled bridge is being repaired or replaced. (This could be an opportune time to consider upgrading the existing structure in terms of width, traffic capacity, height above the water and durability.) upgrading the road requires a wider, higher or stronger bridge and a temporary crossing is needed for a detour to carry traffic while the new road and bridge are built on the old alignment. a river crossing is required urgently and for the short term by an organisation other than a roads department, eg an aid organisation concerned with the delivery of emergency relief supplies. temporary access to a construction site is needed in advance of the construction of a permanent way.

13.2.2 Causeways A vented earth causeway can be built using pipes stocked in most road department depots, taking measures to prevent erosion of the fill around the pipes. A temporary surfacing may also be required. Sections 7.3 and 7.4 contain diagrams of the various forms of causeways. Though a ford or causeway may not be serviceable for use by general traffic at times of high water, it might be acceptable in the short term; an existing crossing of this type might even be found on a nearby older alignment.

13.1 The time factor The time factor is central to all the decisions that have to be made about the crossing, influencing in particular the type of structure to be used. For example, a washedout bridge on a strategic road will require substitution at the earliest possible moment, whereas a detour for a road upgrading project can be planned well in advance, enabling the cheapest solution to be prepared. The length of time that the temporary structure will be in service also influences its design and cost. A lowlevel structure may be adequate for a short period, but if it is likely to be needed beyond the end of the dry season, a lord or culvert may not suffice and a temporary bridge with associated bank protection may be required. A realistic estimate of the time that would be needed to build a permanent replacement is necessary, together with a worst-case view of the weather and the projected traffic loading, before a firm decision can be made about the type of temporary structure to be built.

13.2.3 Temporary beam bridges Timber may be the most readily available material in the locality. The design of timber bridge decks is detailed in Chapter 11. They do not need preservative treatment for temporary duty. if steel or concrete beams were available for short-term use, it would be necessary to ascertain their bending and shear characteristics and to employ them in a similar manner to timber beams (Section 11.2) with timber decking as outlined in Section 11.3. Chapter 10 considers only superstructures made of steel and concrete acting together in a composite manner, which is unsuitable for temporary structures. The steel beams detailed in Chapter 10 would be grossly overloaded without the composite concrete top slab.

13.2 Types of structures Generally, the order of preference for the type of

13.2.4 Prefabricated bridge decks Timber truss decks of the trestle type were popular in the nineteenth century but the skills required to build one quickly are no longer generally available. However, the Timber Research and Development Association (TRADA) in England has developed a design of bridge deck made from identical softwood timber panels, viable for spans up to about 24m. The panels may be stockpiled for use in various combinations, much as a 195

Bailey bridge is made. Figure 13.1 shows an illustration of the design. Details are available in Parry (1981) and TRADA (1985). The basic concept of steel panels, assembled to make bridges of different length, width and load capacity, was developed by Colonel Bailey of the British Army and has been used all over the world for 50 years. Other well known versions of unitary construction bridges are the Acrow Panel bridge and the Callender Hamilton designs. Suppliers of these bridges are listed at the end of this chapter. They issue detailed design and assembly instructions and provide a design and advisory service for customers. Parts of old Bailey bridges can be found in most countries, and great care should be taken to identify these parts correctly. Detail design modifications and changes to the material specifications have taken place over the years, resulting in greater load-carrying capacities. New and old panels should not be assembled in the same structure. Old components must be used only in accordance with the corresponding manuals. Prefabricated steel bridges are unlikely to offer an economical solution to the need for temporary bridging, unless their facility for being dismantled and re-used can be properly exploited. However, their component parts are so easily transported that a stock held centrally by a roads department can be made available at short notice nationwide in most countries.

13.3 Abutments and piers Most temporary bridge decks can be assembled on existing or temporary abutments. An exception is the prefabricated timber panel deck, which requires carefully made abutments at a precise distance apart, because the truss is located below the running sufface and must have a length equal to a multiple of the standard frame length. Existing abutments, if they are sound and located on an acceptable alignment, have three important advantages: they have a proven record of sustaining the dynamic and static applied loads; they have shown satisfactory resistance to attack by the river; their use avoids the time and expense of building new abutments. Temporary abutments may be made from gabions or logs. However, they are highly susceptible to scour and erosion, and should be constructed with great care using ties and anchors where possible (Figure 13.2), because they can be destroyed by a single flood. The log bridges shown in Figures 13.3 and 13A served only until the volume and velocity of the water rose and the abutments failed owing to scour. Bank seat abutments may be made in the form of log grillages as shown in Figure 13.5, if these can be built above the expected high water level and involve no interference to the waterway.

Figure 13.1 Timber panel bridge

196

Figure 13.2 Timber abutment tied back to a buried log

Figure 13.3 Temporary bridge on gabion abutments

197

Figure 13.4 Log abutments with integral wing walls

Figure 13.5 Bridge on timber grillage abutments

Where water flow is slow, timber piled abutments and piers have proved successful. The use of piers reduces the section of the road bearing beams, and a whole bridge like the one shown in Figure 13.6 can be built with timber no larger than 325 mm in diameter. Figure 13.7 shows a typical longitudinal section: its crosssection would be according to Figures 11.3 or 11.4. If the engineer is satisfied that there is strong enough material at bed level or a little below, open caisson

piers may be constructed using pre-cast concrete rings (Figure 13.8). The first ring is placed in position and excavation takes place from inside it. Rings are added as the first progresses downwards until a firm base material is reached, then more rings are added until the required deck height is achieved. Lean concrete can be used to fill the caisson and a stronger concrete is used at the top to take anchor bolts for the transoms. A height to diameter ratio of 3:1 should not be exceeded without careful calculations of toppling moments and the possible

198

Figure 13.6 Timber deck on piled piers and abutments

Figure 13.7 Longitudinal section through a bridge on timber pile piers

199

Figure 13.8 Bridge pier made of concrete rings

addition of steel reinforcement, which rather negates the concept of a temporary structure. One other serious drawback with caisson piers is the obstruction to flow presented by the relatively large surface area. If there is the possibility of fast currents, these could well cause scour at the abutments and under the caissons themselves, unless they are founded deep or on rock. The problem is discussed in detail in Chapter 5.

design is discussed at some length in Chapter 5, and Chapter 6 describes various methods of protecting structures over rivers. For most applications the two main precautions to be taken are: to allow adequate clearance between high water level and the temporary deck to build the sub-structure so that there is a minimum of interference to the flow.

13.4 Protection of temporary structures Though temporary structures, by definition, are not expected to be as robust as permanent ones, they must nonetheless offer reliable service over the required period of use. The two key aspects of durability in temporary bridging are: load carrying adequacy; protection from water damage. A temporary bridge may not be completely adequate for all vehicles that normally use the road. If the temporary structure has any limitation in load capacity, width or height, this must be clearly marked at the entrance to the road on which the bridge is located and repeated on the approach to the bridge. It may be possible to divert large vehicles to a prepared ford, while small ones are permitted to use the bridge. If possible, physical barriers should be erected to prevent drivers of large vehicles from infringing the temporary regulations in areas where drivers are known to ignore warning signs. Section 13.3 noted the influence of the river on the design of temporary crossing to be used. Hydraulic

13.5 References Parry J D (1981). The Kenyan low cost modular timber bridge. Laboratory Report LR970. Transport Research Laboratory, Crowthorne. TRADA (1985). Prefabricated modular timber bridges. Timber Research and Development Association, High Wycombe. Bailey type bridges: Mabey and Johnson Ltd. Floral Mile, Twyford, Reading RG1O 9SQ, England. The Callender Hamilton bridge: Balfour Beatty Power Construction Ltd. 7 Mayday Road, Thornton Heath CR7 7XA, England

200

14 BRIDGE BUILDING MATERIALS

14.1 14.2 14.3 14.4 Concrete 203 Steel reinforcement ... 204 Structural steel beams and steel fittings 204 Structural timber ... 204 14.4.1 14.4.2 14.5 14.6 14.7 14.8 14.9 14.10 Preservative treatments 204 Classification of timber for bridge engineering ... 206

Masonry 206 Rubber bearings 206 Joint sealant .. 208 Construction joint seals . 208 References 208 Standard drawings for unsealed expansion joints . 210

201

202

14 Bridge building materials

It is assumed that the engineer is familiar with the structural use of steel, concrete and timber. The notes on building materials in this chapter are intended only to: specify the materials and terms employed in Chapters 8 to 13; clarify certain aspects of their use; draw attention to potential problems; provide a consolidated list of the numerous references.

furnace slag with alumina content of less than 15 per cent BSI, (1996a) and pulverised fuel ash/cement mixtures with a PFA content greater than 25 per cent BSI, (1996c) also resist attack by sulfates. However, the basic concept of making a dense, impermeable concrete with adequate cover to reinforcement is still the best defence against the penetration of pollutants which attack the concrete mass or the steel reinforcement. The total chloride content of the concrete mix should not exceed 0.3 per cent of the cement, or cement plus pozzolan mixture, when steel reinforcement is used, or 0.2 per cent when using sulphate resisting Portland cement in mass or reinforced concrete. In order to avoid detrimental alkali-silica reaction, it is better to use aggregate known to be non-reactive. The alternative is to restrict the total sodium oxide content of the cement to less than 3.0kg per cubic metre of concrete. This point is discussed more fully in HA (1986) and BRE (1988b). Besides ensuring that the materials are of adequate quality, the single most important factor influencing both strength and durability of concrete is the water/ cement ratio of the mix. When specifying concrete, the maximum free-water/cement ratio by weight should generally be limited to 0.5. In calculating the free water available in the concrete, it is important that any moisture on the surface of the aggregates is included (BRE, 1988a). In marine situations or near the coast, the free-water/cement ratio should be no more than 0.42 to avoid premature deterioration due to corrosion of reinforcement or damage to mass concrete by salt scaling or sulphate attack. Careful mix design is necessary to ensure ease of placing concrete without violating the maximum free-water/ cement ratio. Cover to reinforcement and curing of concrete must also be carefully specified and supervised on site if the structure is to have an adequate service life. The cover of 50mm shown to reinforcement in the standard designs is intended as a minimum for a structure in a moderate environment. In a severe environment, such as a marine situation in a hot climate, cover may need to be increased to 75mm or even 100mm to achieve the required lifespan. To ensure that the cover is as impermeable as possible, the concrete must be moist cured with clean non-saline water as soon as surface finishing is complete for exposed surfaces, or as soon as forms can be safely stripped from formed faces (3 days for side forms, 14 days for structural supports or as advised by local engineers). This curing process should consist of covering the concrete with wet hessian and covering the hessian with polythene sheeting, which must be sealed around the edges to stop the hessian from drying out. The hessian and polythene should be kept in place until the concrete is at least 7 days old,

14.1 Concrete The designs presented in Chapters 8, 9 and 10 are based on the use of Class 30 concrete using 20mm coarse aggregate (30/20 concrete). HA (1986) defines ordinary structural concrete using ordinary Portland cement, Portland blast furnace cement, sulfate resisting Portland cement and Portland pulverised fuel ash cement. The relevant British Standard Specifications for these cement types are: BSI (1996a, 1996b, 1996c and 1996d). Class 30 concrete has a characteristic cube strength of 30N/mm2 at an age of 28 days, when tested according to BSI (1983a). This is generally interpreted as an average strength of at least 37.5N/mm2 with no sample having a strength of less than 25.5N/mm2. The equivalent American standard ASTM (1986) specifies the testing of cylindrical specimens, which return a crushing strength approximately 80% that of the cube strength to BSI (1983a). BSI (1983b) describes methods of sampling and testing concrete for strength and other characteristics; mix design is described in BRE (1988a) and the 'Man on the Job' leaflets published by C&CA (1979, 1980) are useful practical guides covering materials and work practices. Aggregates used in making concrete should conform with BSI (1992) or a similar national code. Concrete for use with steel reinforcement is more likely to achieve the standard of compaction required if the maximum aggregate size is limited to 20mm. Where there are significant amounts of sulphates in the soil or ground water, as described in Section 3.5 and BRE (1981), it is prudent to take precautions to prevent deterioration of the concrete, which is in contact with the ground. Breakdown of the concrete mass can result from reaction between sulfate and hydrates. Concrete made with less than 3.5 per cent tricalcium aluminate is considered resistant to sulfate attack BSI, (1996d). Cement mixtures including blast

203

if possible 14 days. From time to time the hessian may need to be re-dampened, taking care to reinstate the polythene afterwards. Special measures are necessary for concreting in hot weather and when there are drying winds and low humidity. Aggregates should be kept shaded, cool mix water should be used and the time between mixing and placing should be kept to a minimum. Detailed advice on how to minimise the harmful effects of hot weather on concrete is given in ACI (1991) and BRE (1971).

radius of gyration about the beam minor axis; beam overall depth. Other steel items, such as the parapets in Chapter 9, should be made from steel conforming with to BSI (1993b), the grades as specified on the drawings.

14.2 Steel reinforcement The designs included in Chapters 8, 9, 10 and 12 provide for the use of two grades of steel reinforcement, as specified in BSI (1997a), ie plain round mild steel (given the symbol R in the figures) and deformed high yield steel (given the symbol T). Table 14.1 gives their basic mechanical properties, which should be matched as closely as possible if the tables of reinforcement details given in Chapters 8, 9, 10 and 12 are not to require modification. Any deviation from the listed properties requires examination by a structural engineer. BSI (1989) specifies scheduling and bending, etc and HA (1986) gives guidance on cutting, fixing, lapping and joining. Reinforcing bars should not be welded unless the joints can be demonstrated to the satisfaction of the Resident Engineer to be acceptable in terms of location, fatigue life, durability and surface condition. Table 14.1 Grades of steel reinforcement

14.4 Structural timber The durability of a timber bridge will be conditioned by climate, biological factors and design. The most important climatic factor is moisture, since large seasonal or daily variations in the moisture content of the timber will cause it to shrink and swell, which may lead to cracking. The two most important biological factors are wood-destroying fungi and insects. Hardwoods are generally more durable than softwoods, especially against termite attack. A design life ranging from 5 years for untreated softwood to 20 years or more for hardwood heart timber can reasonably be assumed. Treatment with chemical preservatives to protect the timber against insect attack and fungi decay can extend this design life considerably. The need for treatment is determined by the type of timber, the sizes of the timber sections, and the design life required. A few species have a high natural resistance to insect attack and decay, and their heartwood may be used without treatment. However, treatment is almost always needed for framed structures, since any deterioration of their light members is likely to cause a significant loss of strength. Durability can be enhanced by good design. When detailing, the engineer should: provide for drainage of the timber superstructure to prevent retention of water; provide for drainage of the approaches to prevent water running onto the deck;

14.3 Structural steel beams and steel fittings The load-bearing beams listed in Chapter 10 conform to BSI (1993a) dimensionally and are rolled from steel conforming to BSI (1993b), grade 43c. If beams made to other specifications are to be used, the engineer must ensure that the following properties are acceptable: elastic section modulus about the beam major axis; total cross sectional area; web cross sectional area (overall depth times thickness);

keep the ends of logs, beams or trusses away from soil and water; avoid soil and debris from the deck being trapped in pockets or against the timber superstructure; space timbers and design joints to accommodate movement from shrinkage or swelling.

14.4.1 Preservative treatments To be effective in the long term, chemical treatment of timber should take place in a pressurised tank, because protection is short lived unless the preservative penetrates the timber adequately. If

204

boring or shaping of timbers takes place after treatment, the exposed surfaces should be flooded with preservative before final assembly. The active chemicals are applied diluted in water or in a spirit. Generally speaking, the hard dense heart of hardwoods does not allow the entry of preserving chemicals, nor does it need them, but where the hardwood to he used includes some sapwood, treatment should he applied. All softwoods should be treated. Newly felled timber should be given a preliminary coating of preservative to prevent early contamination by the spores of fungi, and should then be seasoned before pressure treatment. There are many commercial products available for the preservation of timber, but BSI (1997b) recommends only pressure treatment with creosote or copper-chromearsenite (CCA). Table 14.2 gives recommended quantities. Where pressure tanks for treatment are available the techniques will be known and are not detailed here. If pressure impregnation is not available, the hot and cold tank treatment using creosote is a useful alternative. The timber is immersed in a tank of cold preservative which is then heated to about 85-900C (Figure 14.1). This temperature is maintained for at least an hour and the preservative is allowed to cool down before the wood is removed. The process is described in Tack (1979), which is one of a very useful series of Overseas Building Notes, all of which are listed in this reference, and BSI (1997c). Brush or

Table 14.2 Recommended quantifies of timber preservative

spray-applied preservative treatments give only slight protection when compared to the methods referred to above and should, where possible, be used only for retreating cut or worked surfaces. Since there is a risk of corrosion of mild steel or galvanized steel in contact with salt type preservatives such as CCA, it is important not to apply metal fasteners until the fixation of the preservative within the wood is complete. This usually takes about 7 days. If the wood is treated with a preservative containing copper, aluminium fixings should not he used. All preservatives are toxic and those which are dissolved in white spirit will be highly inflammable. The specification should therefore include instructions to the effect that workers are required to use overalls, rubber gloves and eye shields.

Figure 14.1 Hot and cold timber treatment tank

205

14.4.2

Classification of timber for bridge engineering Timber is commonly classified into two basic groups hardwood and softwood. Normally hardwood is stronger, more durable and therefore more suitable for bridge building, but this is not always the case. Cutting, drilling or nailing some of the denser hardwoods may cause problems, while species such as camphor are known to have a sap which reacts with steel, corroding nails, screws and fittings. A useful guide to the strength of wood is given by its density, which in turn is related to fineness of grain and rate of growth, indicated by the distance between the annual growth rings. In general, species that have a high density are strong; but the strength of any individual piece of timber will depend on the number and size of knots, fissures and other faults within it. In this manual, to provide standard log or beam bridge designs, as shown in Figures 11.3 and 11.4, timber has been categorised into three groups: heavy hardwood, lighter hardwood, and softwood. Hardwoods with a density greater than 650kg1m3 after drying to a moisture content of 18% are considered heavy. Methods of measuring timber moisture content are given in BSI (1997c) Appendix A. The timber species set out in Table 14.3 as examples of the three groups are all considered suitable for bridge works. The lists are by no means comprehensive. The bridge engineer is expected to use knowledge of local timber and practice when judging whether a particular species is suitable for bridge construction, and which group to place it in.

Where bearings are installed before casting an in situ concrete deck, formwork around the bearings should be carefully sealed to prevent grout leakage. Before assembly onto the rubber pad, the bearing flanges of steel beams should be protected from corrosion with a thin layer of resin mortar, or a similar material.

14.7 Joint sealant The expansion gap between a concrete bridge deck and the ballast wall on the abutment is best sealed with a cold applied proprietary material, even when covered with a layer of asphalt surfacing as shown in Figures 9.3 and 10.12. The methods of application and use of joint sealants are described in BSI (1990b). Should suitable sealing material not be available, the expansion gap may be left open, but if this alternative is adopted minor variations to the abutments and deck are required Figures 14.2, 14.3 and 14.4). The most important points are to arrange good drainage for the water penetrating the joint, and to make it possible to clean the bearing shelf of small stones and other debris. In practice it is much easier to provide slopes for drainage and access for cleaning on a bearing shelf supporting a composite deck than on one supporting a reinforced concrete deck. The vertical alignment and crossfall of the approaches to all bridges should be designed to drain surface water into side ditches and not permit any water to flow onto the deck. This is even more important when the expansion gaps are not sealed.

14.5 Masonry Stone or brick masonry may be used in river bed protection, culvert headwalls, curtain walls for fords and raft foundations, bridge abutments or the barrels and spandrels of arch bridges. This is dealt with more fully in Chapter 15.

14.8 Construction joint seals When it is not possible to cast the concrete slab of a bridge deck in one continuous pass, a construction joint is made at the junction of the two parts. If suitable proprietary joint sealants for application in the horizontal and vertical plane are available, the engineer is advised to consider these for application at construction joint locations. Correctly applied, these sealants can prevent moisture seepage at construction joint interfaces and any resultant corrosion of the steel reinforcement. It is generally required that a rebate is built at the construction joint and filled with the sealant. However, should the engineer be aware that this type of joint is unlikely to be adequately maintained, he should not specify the groove or sealant, but simply call for the two concrete faces to be butted together. If proprietary internal and external water bars are available, these should also be considered for installation at construction joints to promote watertightness of the detail.

14.6 Rubber bearings Guidance on the selection of material and installation may be found in BSI (1983c). If purpose-made rubber bearings are not readily available, natural rubber sheet of hardness between 45 and 75 IRHD to BSI (1995) may be used. This is more successful when installed in discrete lengths of about 1m. Rubber bearings may be laid directly onto the bearing shelf concrete, provided the surface is flat and smooth. If a bedding pad of mortar is used, this should be kept thin enough not to affect the specified protruding length of the steel dowels. Both the bedding pad and bearing shelf should be constructed so as to drain water away from the bearing and avoid trapping debris that can harbour moisture.

206

Table 14.3 Timber species suitable for bridge beams

207

Table 14.4 Permissible short-term stresses

BSI (1983c). Steel concrete and composite bridges, Section 9.2. Specification for materials, manufacture and installation of bridge bearings. BS 5400 Section 9.2:1983. British Standards Institution, London. BSI (1989). Scheduling, dimensioning, bending and cutting of steel reinforcement for concrete. BS 4466: 1989. British Standards Institution, London. BSI (1990b). Cold applied joint sealant systems for concrete pavements Part 2. Code of practice for the application and use of joint sealants. BS 5212 Pt 2:1990. British Standards Institution, London. BSI (1992). Aggregates from natural sources for concrete. BS 882:1992. British Standards Institution, London. BSI (1993a). Structural steel sections Part 1. Specification for hot-rolled sections. BS 4:1993 Pt 1. British Standards Institution, London. BSI (1993b) Weldable structural steels. BS EN 10025: 1993. British Standards Institution, London

14.9 References ACI (1991) Hot weather concreting. Manual of concrete practice A CI3OSR-89. American Concrete Institute, Detroit. ASTM (1986). Standard test methods for compressive strength of cylindrical concrete specimens. C39. American Society for Testing of Materials, Philadelphia. BRE (1971). Problems of concrete production in arid climates. Overseas Building Notes No.139. Building Research Establishment, Watford. BRE (1975). The hot and cold open tank process of impregnating timber. Technical Note No 42. Building Research Establishment, Watford. BRE (1981). Concrete in sulphate-bearing soils and ground waters. BRE Digest 250. Building Research Establishment, Watford. BRE (1988a). Design of normal concrete mixes. Building Research Establishment, Watford. BRE (1988b). Alkali aggregate reactions in concrete. BRE Digest 330. Building Research Establishment, Watford. BSI (1983a). Methods of testing concrete Part 116. Method for determination of compressive strength of concrete cubes. BS 1881: 1983 Pt 116. British Standards Institution, London. BSI (1983b). Methods of testing concrete Parts 101 to 127. BS 1881.1983. British Standards Institution, London.

BSI (1995). Methods of testing vulcanised rubber Part A 26. Determination of hardness. BS 903: 1995 Pt A 26. British Standards Institution, London. BSI (1996a). Portland blast furnace cement Part 2. Metric units. 135146: 1996 Pt2. British Standards Institution, London. BSI (1996b). Portland cements. BS 12:1996. British Standards Institution, London. BSI (1996c). Portland pulverised fuel ash cement. BS 658& 1996. British Standards Institution, London. BSI (1996d). Sulphate resisting Portland cement. 135 4027.1996. British Standards Institution, London. BSI (1997a). Carbon steel bars for the reinforcement of concrete. 1354449:1997. British Standards Institution, London. BSI (1997b). Structural use of timber Part 5. Code of practice for the preservative treatment of structural timber. 135 526&~1 997. Pt 5. British Standards Institution, London. BSI (1997c). Wood preservation using coal tar creosotes Part 2. Methods for timber treatment. 135 144: 1997. British Standards Institution, London. C&CA (1979, 1980). Man on the job leaflets. Nos 45.101 to 45.116. Cement and Concrete Association, Slough. Highways Agency (1986). Specification for Highway Works. Part 5, series 1700 Structural concrete. Stationery Office, London. Tack C H (1979). Preservation of timber for tropical building. Overseas building notes, No 183. Building Research Establishment, Watford.

208

14.10

Standard drawings for unsealed expansion joints

Title

Figure number

Alternative expansion joint on a concrete deck 14.2 Alternative expansion joint on a composite deck . 14.3 Alternative expansion joint at a pier . 14.4

209

Figure 14.2 Alternative open expansion joint for a reinforced concrete deck and bearing shelf

210

Figure 14.3 Alternative expansion joint on a composite deck

211

Figure 14.4 Alternative open expansion joint at a pier

212

15 MASONRY
15.1 15.2 Use of stone and brick masonry for road works ... 217 Design guidelines .. 217 15.2.1 15.2.2 15.2.3 15.3 Foundations .. 217 Walls 218 Retaining walls 218

Materials and quantities 218 15.3.1 15.3.2 15.3.3 15.3.4 15.3.5 15.3.6 Mortar .. 218 Cement . 219 Sand .. 219 Water 219 Stones ... 219 Bricks ... 219

15.4

Masonry work ... 219 15.4.1 15.4.2 15.4.3 15.4.4 Joints and Pointing ... 219 Bond for stone masonry ... 220 Mortar mixing .. 220 Rules for good quality mortar-stone masonry construction: 221

15.5

Arches .. 221 15.5.1 Arched culverts 221

15.6 15.7 15.8

Dry stone masonry 222 Masonry for river bed or slope protection 222 References . 223

213

214

15 Masonry
Masonry is the term used to cover all construction work involving the laying of rough or dressed stones, bricks and cement blocks. Masonry technology usually makes use of locally available resources such as stones or bricks, sand and labour skills. Smaller masonry structures can meet the necessary design standards at a significantly lower cost than concrete ones, especially for the secondary and tertiary road network. Careful planning, however, is still required for this kind of structure and skilled labour is essential for the construction of quality masonry work. An example of a masonry structure is given in Figure 15.1.

construction of low cost structures, such as bridge abutments, retaining walls, culvert head walls and wing walls, river bed protection, drifts and the barrels and spandrels of arch bridges. if the two component materials, ie, stone / brick units and cement mortar, are both of sufficient quality to withstand load arid abrasion, then masonry may be used instead of mass concrete where it is specified elsewhere in this guide.

15.1

Use of stone and brick masonry for road works

15.2 Design guidelines These guidelines are applicable for minor works only, typically for wall heights up to 4m and spans up to 6 m. For major masonry structures detailed engineering design is required. The adoption of a simple construction method does not dispense with the need for an adequate survey and design.

History: Stone and brick masonry have been used to build walls since ancient times. Nearly all of the most famous and largest buildings and structures that have survived from those days have been built using masonry technology. Some of these structures are still in use today. Bridges for roads, waterways and railways were constructed with masonry arch methods to cross rivers or valleys of any size before concrete technology was developed.

15.2.1 Foundations Foundations for load-bearing walls (manly abutments) are usually made of reinforced concrete and must be appropriately designed (see Chapter 8: Substructures and Foundations). This requirement is essential for abutments supporting rigid superstructures. For walls up to l.5m in height, which do not support additional imposed loads, no special concrete foundation is required: however, the following guidelines should be followed: The base of the foundation should be a minimum of 400mm below ground level on uniformly firm ground.

Use of masonry today: Stone or brick masonry is very useful for the

Figure 15.1 Culvert with masonry headwall and wingwalls

215

A 50mm lean concrete layer (1:4:8, ie, cement: sand: coarse aggregate, by volume), is required as a clean base for the foundation. The first course (footing) should be laid with the largest and straightest stones, because the stability of the wall depends largely on the bearing of the stones on the ground. Where the ground has a doubtful bearing capacity, a masonry foundation is required. Chapter 8 discusses the bearing capacity of several types of soil. If there is doubt about the strength of the supporting soil then a plate bearing test can be carried out according to Section 8.3.3, or values can be used from Section 8.3.4. If a foundation is required, the width of this foundation should be double the thickness of the wall, with a minimum of 800mm. For retaining walls, foundation sizes can be taken from Table 8.5.

Attention should be given to the correct construction of the top of the wall. This is known as wall capping or the wall crown. A slope towards the outer side of the wall should be provided to allow any water to drain effectively and not to enter the wall structure. Capping should be completed as the last layer of masonry is laid.

Figure 15.3 Mortar capping

Figure 15.2 shows typical non-bearing walls and the foundations needed.

15.2.3 Retaining walls All retaining walls should be backed with a layer of freedraining material to prevent the build up of water pressure. Long walls over 2m high will also need weep holes, at least every 2m, as shown in the drawings in Chapter 8. Dimensions for retaining walls can be taken from Figures 8.12 and 8.13.

15.2.2 Walls The minimum thickness for structural walls up to a height of 1.5m built using stones of bricks is 400mm The thickness for load-bearing walls or those higher than 1.5, which also have to withstand earth pressure, should be adequately dimensioned (see Chapter 8: Substructures and Foundations, Section 8.4.1 Mass concrete abutments).

15.3

Materials and quantities

15.3.1 Mortar The mortar used for masonry structures is a mixture of cement, sand and water. Lime should not be used for road structures. The strength of the bond will

Figure 15.2 Foundations for non-load bearing walls 216

depend on factors such as: the amount of cement used; the amount of water used; the type and quality of the sand; the surface characteristics of the stones; the quality of the workmanship.

settle on the sand forming a visible layer. The height of this layer (f) and the sand layer (s) can he measured and compared. if the sand has a fines content (ie, 100f/[f+s]) of more than 5 percent, then it is not clean sand and should not be used.

It is generally recommended that the mortar should be no stronger than the bricks or blocks, so that any cracks that develop will be in the mortar. Cracking through the blocks is more difficult to repair. Recommended mixing proportions for mortar are given in Table 15.1. Table 15.1 Recommended mixing proportions for mortar

15.3.4 Water The mixing water used in the mortar must be clean. It can be taken from taps, rivers, lakes or wells. Salt water from the sea or a lake, surface run-off water and water with other chemical or organic impurities must not be used. if no other water is available, then dirty water with organic particles can be used if it is left in a drum or a similar container until the particles have settled at the bottom. Use only the clean upper part of the water.

15.3.5 Stones It is important to choose only good strong stones to build walls. The following stones should not be used: weathered stones; cracked or spalling stones; small stones (less than 200 mm diameter). The length of any stone should not be greater than three times its height. Always choose stones whose shape is as near as possible to a rectangular prism (ie, brick shaped). They must be free of dust and dirt. It is therefore advisable to wash them and, if necessary, clean them with a brush. if mortar is to be used then the stones should be wetted, but surface dry before use to ensure a good bond with the mortar.

15.3.2 Cement Ordinary Portland Cement (OPC) is recommended for use in the mortar. This type of cement should be widely available. When calculating mix proportions by volume, note that a 50kg bag will have an approximate volume of 0.036m3 (36 litres).

15.3.6 Bricks Bricks should be uniformly burnt and of similar size (maximum tolerance 5 mm). The following bricks should not be used: cracked or spalling bricks; unevenly burnt bricks; 'bent' bricks. 15.4 Masonry work The most common types of masonry wall are shown in Table 15.2, which also gives approximate material quantities. All stratified stone that has bedding planes should be laid with the natural bed as near as possible at right angles to the direction of the load. In the case of arch rings the natural bed should be radial.

15.3.3 Sand There are two main types of sand: soft sand, which has rounded particles, and sharp sand which has angular particles and is often used in concrete. For stone or brickwork mortar, soft sand is preferred because it makes a smooth, easily workable mortar. Sharp sand can be used, and it will produce a stronger mortar, but it makes a harsher, less workable mix. Only clean sand should be used and if in doubt, this can be tested with a sediment bottle test. The test consists of a jar with straight sides being half-filled with the sample. Clean water is then added until the jar is almost threequarters full, the lid is fastened and the jar is vigorously shaken. The sand should quickly settle but any silt or clay can take up to several hours to settle. Any fine material present will

217

Table 15.2 Types of masonry wall and material quantities

15.4.1 Joints and Pointing For mortar bonded masonry, it is important that no stone or brick should touch another but each one should be fully bedded into mortar. For road structures the joints are usually finished as 'flush joints'. The mortar between the stones is trowelled to a smooth surface flush with the face of the stone or brickwork using a mortar trowel or a pointing trowel. 'Ribbon' pointing, where the mortar stands out from the face of the stones, should be avoided in climates where frost is likely. Any water sitting on 'ribbon' mortar can damage the mortar if it freezes (BSI, 1976).

15.4.3 Mortar mixing The mixing of mortar for small structure works is usually carried out on site by hand. The quantity of mortar to be mixed should not be more than a mason can finish using within one hour of mixing or half an hour if it is in very hot weather or strong sun. Hand-mixed batches should not exceed 0.5 m3. The mixing should never be done on the bare ground, as this results in contamination of the mix. A mixing platform of about 4m by 4m should be built with boards, metal sheets or lean concrete. Procedure for mixing by hand 1) Measure the required amount of sand and cement using a gauge box of 36 litres (400mm x 300mm x 300mm). 2) Spread the cement and sand in alternating layers on the platform. 3) Mix the dry materials into a separate heap at least three times. This is best done with two persons, one on each side of the heap, who can shovel the heap to one side by turning the material in the process. This operation should be repeated, with the heap being thrown back to its original position and then back again, until the colour of the dry mix is a uniform grey.

15.4.2 Bond for stone masonry The bond should allow a minimum overlap of 114 length of each stone. Most of the stones are laid as stretchers, ie along the length of the wall. Header stones (also called through stones) should be laid at regular intervals across the width of the wall to bond the two faces of the wall together. The header stones should cover at least 2I3rds of the wall thickness and their overlap should not be less than 100mm.

218

Figure 15.4 Masonry bonding

4) Add water. This is best done using a watering can so that the water is spread evenly while the material is mixed again. Only the correct amount of water should be added (see guideline below). The wet mixing must be continued, turning at least three times, until the mortar is uniformly wet and has reached the required consistency. The water-cement ratio should be approximately 0.4 to 0.5, which is equal to 20 to 25 litres of water per 50kg bag of cement, or 16 to 20 litres for a 40kg bag. Trial and error is required to get the water content right, because this will depend upon factors such as the moisture content of the sand and the size of the sand particles. when the water content is right: the consistency should be such that the mortar does not flow off the trowel; the mortar can be kneaded in the hand and retains its form.

6) Use mortar of the correct mixture and consistency. 7) Ensure proper bonding and joints.

15.5 Arches Arches can be used for both culverts and bridges. Their purpose is to transmit the load above to the abutments or piers on either side of the opening. Semi-circular and semi-elliptical arches are the most commonly used shapes for road structures. They can consist of single or multiple arches. For larger structures, careful construction using wedge shaped stones or bricks is required to ensure that each stone or brick transmits load to the next stone or brick. Strong formwork will be needed to support the arch during construction. The base and foundations of any arch structure are essential for stability. The ground must be excavated down to firm material and brought back to the required level with lean concrete or good hardcore, topped with 50mm of concrete. For culverts the base should be laid to the required gradient, as discussed in Section 12.2. This gradient is usually a minimum of 0.5% for clear water and 2 to 4 % where sediment is carried in the flow. The formwork used for constructing arches can be made from strong timber, old oil drums or old car or lorry tyres laid side by side. Tyres and drums must be well matched in size and can be placed in a row on a stack of stones, bricks or a layer of compacted soil or sand to the required height. An example of arch culvert construction using wooden formwork is given in Figure 15.5. The walls are then constructed up to the level where the arch begins (springing points). When building the 219

15.4.4

Rules for good quality mortar-stone masonry construction:

l)

Construct a proper foundation.

2) Use only stones which are not cracked or weathered. 3) Clean all the stones with water and a brush. 4) Stones and bricks should be wetted before laying to ensure a good bond with the mortar. 5) Use the largest stones for the bottom layer and the corners of the wall, to ensure stability.

Figure 15.6

Rough brick arch

For larger structural openings the bricks need to be shaped as wedges and sized to appropriate dimensions, like the stones forming the arch in Figure 15.7.

Figure 15.5

Masonry arch culvert with simple wooden formwork

arch, it is important to build both sides evenly, so that the formwork does not become distorted by the weight of the stones or bricks. The arch should be left to cure for at least 2 to 3 days before the formwork is removed. The stones, bricks or soil supporting the drums or tyres should be removed first. Tyres are flexible and therefore should be easy to remove. When the formwork has been removed, the base should be cleaned of all loose material. It should then be rendered with a strong mortar screed about 50mm thick and finished slightly concave to keep small water flows away from the culvert walls. With all culverts, it is very important to ensure that the backfill down the sides and over the arch is well compacted in thin layers (150 to 200mm) and that both sides are built up and compacted to the same level at each compaction. The fill above the arch must not be less than 500mm. Figure 15.7 The example in Figure 15.6 shows a 'rough brick arch' constructed with normal bricks. The mortar joints are the wedges in this case. It is therefore essential to ensure that only good quality mortar (1: 4) is used and that all the joints are properly filled with mortar. It is also important to achieve proper bonding the longitudinal direction. Gauged stone arch

220

Stone arches should be built as 'gauged stone arches' with tapered and sized stones as shown in Figure 15.7. As with brick arches, it is important to construct both sides simultaneously to avoid deformation of the formwork, eg tyres. The joints need to be fully filled with mortar of a mix of l :4.

15.8 References BSI (1976). BS 5390: 1976: Code of practice for stone masonry. British Standards Institution. BSI (1992a). BS S62& Parts 1-3: 1992. Code of practice for use of masonry. British Standards Institution. Curtin W G, Shaw G, Beck J K and Bray WA (1987). Structural masonry designer's manual. 2nd Edition. BSP Professional Books, Oxford.

15.6 Dry stone masonry Dry stone masonry is suitable for walls which do not have to carry loads. As there is no mortar, the stones are laid to fit as tightly as possible and wedge-shaped pieces of stone are driven into the larger gaps to hold the stones firmly in place. Careful shaping, laying and bonding of the stones is essential and only skilled and experienced stonemasons should be allowed to carry out this work. Figure 15.8 shows part of a typical dry stone wall.

15.7 Masonry for river bed or slope protection River bed protection should be laid according to Section 7.3: Bed Level Causeways. As with other forms of river slope protection, any form of rigid cladding will be damaged by the water flow, unless it is laid on very firm material. if the river bed consists of alluvial material, a flexible cover of large stones (rip rap) or large stones in wire baskets (reno mattresses) will usually be more effective in preventing scour of the river bed and last longer than rigid cladding. As for bed level causeways, river bed protection beneath a bridge should be constructed with curtain walls, as shown in Figure 7.4. The top surface should be flush with the level of the natural river bed.

Figure 15.8 Part of a dry stone wall

221

222

16 DRAWINGS AND SPECIFICATIONS

16.1 Drawing number 1: site plan and longitudinal section ... 227 16.1.1 16.1.2 16.2 16.3 16.4 16.5 The site plan, drawn to a scale of about 1:500 .. 227 The longitudinal section, drawn to the same horizontal scale .. 227

Drawing number 2: bridge plan and sections . 227 Drawing number 3: substructure details . 227 Drawing number 4: superstructure main details . 227 Further detailing .. 227

223

224

16 Drawings and specifications

When the designs for foundations, substructures, superstructure and river works have been selected on the basis of the preceding chapters of this manual, and all the necessary modifications to suit local requirements and conditions have been made the engineer must prepare drawings in sufficient detail and with all necessary dimensions to enable the structure to be built by direct labour or an outside contractor. Since the drawings contain all the information required to build the structure, there should be no need for the builder to consult the designer. In practice, however, liaison between design office and construction site usually benefits both parties and is particularly helpful when unforeseen conditions are met. It is usual for the Resident Engineer, in consultation with the designer, to agree modifications with the contractor, with the aim of saving unnecessary costs, compensating for poor soils or using different materials to those specified. A complete set of drawings and calculations is also required for the bridge inventory. This set should include any modifications that may be introduced during construction and is to be updated whenever repairs or strengthening, etc take place on the structure or river works. The application of these guidelines will result in a complete set of information. Since it is helpful also to follow local practice, the following notes should be treated as advisory; their object is to provide all the necessary information in a clear and simple form.

16.1.2

The longitudinal section, drawn to the same horizontal scale

contains relevant details of the subsoil conditions obtained from bore holes and trial pits; shows the design flood level, the high flood level and the low water level with dates; specifies the vertical alignment of the road approaches and the bridge.

16.2

Drawing number 2: bridge plan and sections

This drawing contains a plan, longitudinal section and cross section of the structure. It is drawn to a scale of about 1:100 and specifies the following major dimensions: abutment and pier width, height, bearing shelf levels and foundation levels; superstructure span, width and height to the underside; wing wall lengths, heights and foundation levels; finished road surface levels over the bridge. 16.3 Drawing number 3: substructure details

16.1

Drawing number 1: site plan and longitudinal section

On this drawing the elevations, plans and sections selected from Chapter 8 give all the dimensions and levels required for the setting out and construction of the abutments, piers and wing walls. Concrete and reinforcement specifications should also be included.

16.1.1 The site plan, drawn to a scale of about 1:500 contains a north point and shows the direction of the nearest town; shows contours or spot levels of the river bed and the surrounding ground in the area of the bridge site; details the bench marks and levels established during the survey; shows the locations of the bore holes and trial pits put down during the site investigation; indicates the limits of the design flood and high flood, with direction of flow; specifies the horizontal alignment of the road approaches and the bridge. 16.4 Drawing number 4: superstructure main details

This drawing contains plans and sections detailing the bridge deck, selected from Chapters 9, 10, 11 or 13. It includes material specifications and bending schedules for reinforcement.

16.5

Further detailing

Further drawings are prepared as required in order to detail and specify materials for parapets, bearings, joints and drainage, etc, on the bridge, as well as river training, embankment protection and scour protection measures in the river.

225

226

INDEX
Term Section No.

Abutments 8.1 abutment design ... 8.4 abutments - mass concrete 8.4.1 abutments - reinforced concrete .84.3 abutments - temporary bridges .. 13.3 abutments- timber decks 11.5 afflux 5.5 aggressive chemicals ... 3.5 aprons . 6.2.1, 6.2.2 arches . 15.5 backwater . 5.5 bearing material . 14.5 bearing pressures . 8.3 bearing shelves ..8.4.6 bearings . 9.2.2, 10.26, 14.6 bed level causeways 7.3 bores - cable percussion 3.1.3 bores - hand auger .3.1.2 bores-rotary drilling .. 3.1.4 bridge height 5.2 bridges-log . 11.1 bridges - skew ... 2.1.2 bridges - submersible .. 7.5 bridges - temporary .. 13 bridges - timber beam 11.2 catchment .. 2.2.1 concrete . 14.1 construction joints .9.2.4, 10.2.8 cross-sections .. 2.3 culverts . 12 culverts - concrete box ... 12.6 culverts - concrete pipe .. 12.4 culverts - erosion control 12.1 culverts - flexible steel ... 12.5 culverts - headwalls ... 12.3 culverts - location & alignment ..12.2 curtain walls . 7.3 decks . 9,10,11 design flood .5 design life .2.4 design standards 2.8, 9.1 discharge .. 4.2 drainage .. 9.2.6, 10.2.10 drawings 16 227

Term

Section No.

earthquake restraint ..8.7 field tests .. 3.4.1. filter blankets .6.1.3 flow velocity 4.1 flow volume . 4.2 footpaths .. 2.7,9.2.10,10.2.12,11.4 fords . 7.2 foundations . 8 gabions .. 6.1.2 geophysical surveying ...3.1.5 groynes . 6.2.4 guide banks ... 6.2.3 hydraulic design .. 5 hydraulics - abutments, piers ... 5.3 Term .. Section No. hydraulics - culverts 5.6 joint sealant 14.6 joints - construction .. 9.2.4 joints - expansion . 9.2.3, 10.2.7 loading - design ... 2.8 longitudinal section 2.3, 16.1 low level crossings . 7 low water crossings, signs & markers . 7.6 masonry - Stone or brick ...15 materials, concrete, steel, etc 14 mortar ..15.4.4 navigational requirements 2.2.3. parapets .. 9.2.7, 10.2.11 pedestrians/cyclists ... 2.7, 9.2.10, 10.2.12, 11.4 penetration test .. 3.4.1 piers .. 8.2, 8.5 piled walls . 6.1.4 planning .. 2 plans .. 16 plate bearing test .. 3.4.1 prefabricated bridge decks .. 13.2.4 protection methods .. 6.2 protection methods - banks .. 6.2.2 protection methods - foundations . 6.2.1 protection methods - groynes ... 6.2.4 protection methods - guide banks . 6.2.3 protection methods - temporary structures .13.4 raff foundations .8.4.5 reinforcement 8, 9.2.5, 10.2.9 reinforcement, steel 14.2 Reno mattresses 6.1.2 Resources 2.9 228

Term

Section No.

retaining walls, mass concrete .. 8.4.2 retaining walls, reinforced concrete ..8.4.4 rip rap 6.1.1 river hydraulics ... 4 river morphology .. 2.1.1 river works .. 6 run-on slabs . 8.8 sampling .. 3.3 scour 5.4 scour protection . 6.1.5, 6.2 services . 9.2.9 shear connectors . 10.2.3 signs . 7.6 site investigations ... 3 site plan .. 16.1 soil sampling 3.3 soil testing 3.4 specifications 16 steel beams . 14.3 steel - protective treatment ..10.2.4 stone pitching 6.1.6 substructures ... 8 substructures - temporary bridges ..13.3 superstructures - composite .. 10 superstructures - concrete .. 9 superstructures - emergency . 13 superstructures - timber 11 superstructures - vertical profile ... 9.2.1, 10.2.1 surfacing ... 9.2.8 test pits ...3.1.1 timber classification 14.4.2 timber decks ... 11.3 timber - protective treatment .. 14.4.1 timber - structural .. 14.4 timber superstructure 11 traffic ... 2.5 vegetation - scour protection 6.1.5 vented causeways 7.4 water levels ... 2.2.3 width 2.6 wingwalls . 8.1

229

230

ISSN 0951-8797

OS-F

A design manual for small bridges

ORN 9

Old Wokingham Road Crowthome Berkshire RG11 6AU

Overseas Development Administration 94 Victoria Street London SW1E 5JL

OVERSEAS ROAD NOTE 10

COSTING ROAD ACCIDENTS IN DEVELOPING COUNTRIES

Main Subject area: Urbanisation and Transport Theme: Transportation

Project title: Road Safety Project reference: R5613

This document is an output from an ODA-funded research project, carried out for the benefit of developing countries.

Crown copyright 1995. The views expressed in this report are not necessarily those of the Overseas Development Administration or the Department of Transport. Overseas Centre, TRL, 1995 ISSN 0951-8987

ACKNOWLEDGEMENTS This note was drafted by Dr G D Jacobs whilst a member of the of the TRL Overseas Centre. The author gratefully acknowledges the advice and guidance provided by Professor M W Jones-Lee Newcastle University.

First published 1995

OVERSEAS ROAD NOTES Overseas Road Notes are prepared principally for road and transport authorities in countries receiving technical assistance from the British Government. A limited number of copies is available to other organisations and to individuals with an interest in roads overseas, and may be obtained from: Transport Research Laboratory Crowthorne, Berkshire, RG11 6AU United Kingdom Crown Copyright 1995 Limited extracts from the text may be reproduced provided the source is acknowledged. For more extensive reproduction, please write to Programme Director, Overseas Resource Centre, Transport Research Laboratory.

CONTENTS
Page 1. Introduction Objectives Why cost road accidents? Value of life 2. Methods that can be used to cost road accidents Classification of accidents Different approaches to accident costing/valuation Which method to use 3. Using the Gross Output Method in practice Value of the loss of output Cost of medical treatment Cost of damage to vehicles and other property Administrative and other costs Subjective costs 4. Case study - Cyprus Vehicle repair costs Estimating lost output Cost of medical treatment Police and administration costs Resource cost summary Sums to reflect pain, grief and suffering 5. 6. References Appendix: The inclusion of accident savings in highway cost-benefit analyses Background Feasibility studies in Cyprus Feasibility studies of two roads in Jordan Feasibility study in India 1 1 1 1

2 2

2 3 4 4 5

6 7 7 8 8 9 10 11 11 11 12

13 13 13 14 16

COSTING ROAD ACCIDENTS IN DEVELOPING COUNTRIES

1. INTRODUCTION
Objectives 1.1 The objective of this Note is to advise economists, planners and engineers in developing countries on a workable method that can be used to cost road accidents. There are many different methods and approaches to this particular problem; it must be stressed that no single method is ideal and that a considerable amount of data needs to be collected whatever method is used. This Road Note attempts to explain the importance of costing road accidents in developing countries and outlines in Section 2 the various methods that can be used to do this. Section 3 describes how to use the (preferred) Gross Output method and Section 4 presents a case study of its use in practice. Finally an Appendix presents results of the possible effects of including accident savings in a number of highway cost - benefit analyses.

1.2

1.3

recent road safety study undertaken in a particular country by TRL, it was shown that the annual cost of road accidents nationally was about 20 million. A series of safety improvements were outlined which, it was estimated would reduce the national cost of accidents by 5 per cent per annum (i.e. saving 1 million p.a.). These improvements (in highway design and layout, education, training and enforcement) were estimated to cost 500,000 in a programme of measures set out over a five year period (i.e. at an average annual cost of 100,000). The average First Year Rate of Return on investment was therefore about 1000 per cent and the Benefit : Cost ratio about 10:1. High rates of return such as these are fairly common in road safety appraisals and (apart from the humanitarian aspects), illustrate the economic benefits of investing in national road safety programmes. 1.7 A second need for road accident cost figures is to ensure that the best use is made of any investment and that the best (and most appropriate) safety improvements are introduced in terms of the benefits that they will generate in relation to the cost of their implementation. Failure to associate specific costs with road accidents will almost certainly result in the use of widely varying criteria in the choice of measures and the assessment of projects that affect road safety. As a consequence it is extremely unlikely that the pattern of expenditure on road safety will, in any sense be 'optimal'. In particular, if safety benefits are ignored in transport planning then there will inevitably be an underinvestment in road safety. VALUE OF LIFE 1.8 From the above it can be seen that rational decisions on the allocation of resources to road safety will require the use of cost-benefit analysis, with explicit costs of accidents and values of accident prevention. Ways in which such costs and values can be defined and estimated are described in the next chapter. However, to some people the monetary valuation of human life and safety may appear immoral and it should be stressed that at no point does this Road Note lay claim that it is possible to find a numerical sum which can be said to be the absolute "value of human life", as such. Rather, what this Note examines are the various methods that can be used to estimate the value that should be placed on various safety-improving activities (and the costs that should be associated with increases in risk) on the roads of developing countries.

Why cost road accidents? 1.4 It is now well established that many developing countries have a serious road accident problem (Jacobs and Cutting 1986). Fatality rates (per licensed vehicle) are high in comparison with those in developed countries and whilst in Europe and North America the situation is generally improving, many developing countries face a worsening situation. For example, over the period 1969 to 1986, the number of people killed in road accidents in 12 European countries combined actually fell by about 20 per cent. In 16 Third World countries combined there was, over the same time period, an increase in the number killed of over 150 per cent. Whilst these trends give cause for concern in developing countries, road safety is but one of the many problems demanding it's share of funding and other resources. Even within the boundaries of the transport and highway sector, hard decisions have to be taken on the resources that a Third World government can devote to road safety. In order to assist in this decision-making process it is essential that a method be devised to determine the cost of road accidents and the value of preventing them. So, the first need for cost figures is at the level of national resource planning to ensure that road safety is ranked equitably in terms of investment in its improvement. Fairly broad estimates are usually sufficient for this purpose, but must be compatible with the competing sectors. For example, in a

1.5

1.6

2. METHODS THAT CAN BE USED TO COST ROAD ACCIDENTS

CLASSIFICATION OF ACCIDENTS 2.1 In order to cost road accidents it is important that a country has a consistent classification method. Accidents either involve injury to a person i.e. personal-injury accidents (together with vehicle or property damage) or merely involve damage to vehicles and possibly property in which case they are termed damage-only accidents. In the UK and in many other countries, personal injury accidents are usually reported to the local police who then make a return to a central organisation (e.g. police headquarters or to a Ministry). It is standard practice for these accidents to be then classified as being either fatal, serious or slight. The definitions used in most Western European countries to define accident severity are as follows: A fatal accident is one in which one or more persons are killed as a result of the accident, provided death occurs within 30 days. A serious accident is one in which there are no deaths but one or more persons are seriously injured. A serious injury is defined in the UK as either one for which a person is detained in hospital as an "in patient", or if any one of the following injuries are sustained whether or not he or she is detained in hospital:- fractures, concussion, internal injuries, crushing, severe cuts and lacerations, or severe general shock requiring medical treatment. A slight accident is an accident in which there are no deaths or serious injuries but a person is slightly injured. This will be an injury of a minor character such as a cut, sprain or bruise. A damage-only accident is one in which no one is inured but damage to vehicles and or property is sustained. It should be noted that accident severity is defined by the most serious casualty class of any of the victims of the incident. The `cost of an accident' is therefore not the same as the `cost of casualties' resulting from that accident, at a more disaggregated level. Failure to distinguish this difference when examining the benefits of different detailed remedial measures can result in different project cost-benefit rankings. The interested reader is referred to Andreassen (1992) for a more detailed discussion of the problem. THE DIFFERENT APPROACHES TO ACCIDENT COSTING/VALUATION 2.2 In their papers on the cost of traffic accidents and evaluation of accident prevention in developing countries, Hills and Jones-Lee (1981, 1983)

identified six different methods that have been proposed for placing a cost on road accidents. All of the methods outlined were applicable to nonfatal as well as to fatal accidents but for reasons of clarity and simplicity, they concentrated on describing accidents involving one fatality only. They made the point that the appropriate method to use in any particular context may depend upon the objectives and priorities of those who intend to use the costs and values concerned (see para 2.10). A) The "gross output" (or human capital) approach 2.3 In this method, the cost of a traffic accident involving a fatality can be divided into two main categories. Firstly there are the costs that are due to a loss or diversion of current resources and secondly there are the costs that are due to a loss of future output. Included in the former will be the cost of vehicle damage, medical treatment and police/ administration costs and usually there is little disagreement as to what should be included here. Determining loss of future output of the persons) killed however is less clear cut. Usually average wage rates are used (gross of tax) to determine lost output both for the year in which death occurred and then for future years. Costs in future years that the casualty might have lived have to be discounted back to give present day values. This is not done separately for every individual killed (or inured) in a road accident; estimates are based on average (i.e. national) output or earnings data together with appropriately estimated damage, medical and police costs. In some variants of this approach, a significant sum is added to reflect the "pain, grief and suffering" of the accident victim and to those who care for him or her (see paras 3.19-3.22).

B) The "net output" approach 2.4 This differs from A) in that the discounted value of the victim's future consumption is subtracted from the gross output figure. Again, it may be difficult to visualise how an estimate can 6e derived of what a person "consumes" (m terms of food, fuel etc.) throughout his or her lifetime. When this method was used in the UK to cost road accidents (being replaced in the early 1970's by the gross output approach), the 'total consumer expenditure and the public authorities' current expenditure on goods and services" was divided by the total population. A crude estimate of "consumption per head" was thus obtained. In this approach the difference between an individuals gross output and future consumption may be regarded as a measure of the rest of society's economic interest in his continued survival. C) The "life-insurance" approach 2.5 In this method the cost of a road accident or the value of accident prevention is directly related to the sums for which `typical' individuals are willing (or even able) to insure their own jives (or limbs).

However, whilst the amount of insurance cover provided might be considered to be some estimate by the insured person of the value of his life to his dependants, it says nothing whatsoever about the value of life to the insured person himself. Thus a wealthy bachelor with no dependants may have little or no life cover, whilst a much poorer person with several children may have his or her life insured for a much greater sum. The wealthy bachelor may well, nonetheless, place a very high value indeed upon his own continued survival. Another problem with this approach is that the level of life insurance cover may be well below what it `ought' to be if the intention is to provide sufficient income on which his or her dependants are to survive. Further, to base any analysis on the insured population alone is almost certainly to choose a biased sample. This approach is of particularly limited value in developing countries where relatively few people carry life insurance. D) The "court award" approach 2.6 With this approach, the sums awarded by the courts to the surviving dependants of those killed or injured as a result of either crime or negligence are regarded as an indication of the cost that society associates with the road accident or the value that it would have placed on its prevention. In the UK, the sum awarded by the court must take into account complex issues such as degree of negligence of the defendant, whether the person killed or inured was partly to blame, whether or not the employer of the inured person is continuing to pay them any wages and whether industrial injury benefits are to be paid. (These only include private costs, i.e. not necessarily hospital costs met by the state). In addition, any sum awarded by the court will have all taxes removed. From the above it can be seen that to use court awards as implied values for the loss of life (or limb) in a road accident would be very much an imperfect solution.

in the UK at the time, it suggests that 50 could be regarded as a maximum value for life, giving from the two examples a range of less than 50 to over 20 million per life saved. These examples provide at the very least an indication that there is (or was) some misallocation of resources in life-saving activities and suggests that this would be a very imprecise method for valuing human life. F) The "value of risk change" or "willingness to pay" approach 2.8 This approach is based on the fundamental premise that decisions made in the public sector concerning the allocation of scarce resources should reflect the preferences and wishes of those individual citizens who will be affected by the decisions (Jones-Lee 1976, 1989). Accordingly, the value of a given improvement in road safety (i.e. a reduction in risk) is defined in terms of the aggregate amount that people are prepared to pay for it. Conversely the cost of a reduction in safety is defined in terms of the amount people would require in compensation for the increased risk More specifically, the value of a particular safety improvement is defined as the sum of all the amounts that people (affected by the improvement) would be willing to pay for the (usually very small) reductions in risk provided by the safety improvement. Thus the value of prevention of one accident involving one fatality is defined as the total amount that all affected individuals would pay for the very small riskreduction, both for themselves and for those they care about. Estimation of willingness-to-pay costs and values is far from straight forward. Various methods have been used and include an approach where estimates are obtained by observing situations where people actually do trade off wealth or income for physical risk. Another approach uses a complex questionnaire where samples of individuals are asked more or less directly how much money they would be willing to forfeit in order to obtain a small reduction in their own or other people's risk. For example, a detailed questionnaire might indicate that drivers were prepared to pay, on average 5 for a risk reduction of one chance in 500,000 that they would be killed on a particular journey. Then the `value of an average life' in this instance would be 5 x 500,000 i.e. 2.5 million.

2.9

E) The "implicit public sector valuation" approach 2.7 With this method an attempt is made to determine the costs and values that are implicitly placed on accident prevention in safety legislation or in public sector decisions taken either in favour of or against investment programmes that affect safety. Unfortunately, an examination of some values as derived in Britain reveals a very wide range of implied values of life not only between different sectors but also within the same sector (Mooney 1977). Thus following the partial collapse in London of a block of high-rise flats, changes were made to building regulations. According to estimates made (Sinclair et al 1972), a few lives may have thus been saved at very high cost giving an implied minimum valuation of life at over 20 million. However it was also claimed at about the same time (Heys et al 1968) that a method of preventing stillbirths could be standard practice at a cost of only 50 per life saved. Since this method was not widely practised

WHICH METHOD TO USE? 2.10 Not surprisingly, these six approaches produce substantially different costs and values for accidents involving one fatality. Typically figures derived from studies carried out in developed countries over the period 1965-1978 ranged from about 1500 to over 20 million. As stated earlier, Hills and Jones-Lee (1981, 1983) emphasise the point that the method used for costing road accidents depends on the objectives being pursued in a country by those planners and economists respon-

sible for investment planning. The reasons for costing road accidents are most likely to be either the maximisation of national output or the pursuit of social welfare objectives (such as the minimisation of injury accidents or fatalities in relation to traffic). The only accident costing/valuation methods that appear to be directly relevant to these two objectives area) the "gross output" method (well suited to the objective of maximising the wealth of a country) and b) the "willingness to pay" method (especially for social welfare maximisation and for use in costbenefit analyses) 2.11 If accident costs and values are ultimately intended for use in conventional cost-benefit analyses in order to determine the most efficient way of allocating scarce financial resources, then the most appropriate method to use by far is the willingness-to-pay approach. However, whilst this method has been adopted in countries such as UK, USA, New Zealand and Sweden, the difficulty of obtaining reliable empirical estimates has been considerable. Furthermore, whilst the willingnessto-pay approach was adopted in the UK in 1988 to cost fatal accidents, the use of the method to cost non-fatal accidents presented certain problems which have only fairly recently been resolved (Jones-Lee et al 1993, Hopkin and O'Reilly 1993). Even in the case of fatal accidents, a wide range of empirical estimates was obtained from various studies and `a considerable element of judgement' was necessary (McMahon 1991) in order to derive a value that was regarded as `a reasonable working basis for the value of a fatal casualty for use in appraising transport investments'. 2.12 The willingness-to-pay approach as used in the UK can also be criticised on the grounds that values are obtained directly for adults only (children being unable to complete the complex questionnaires used to derive values). (It was inferred that children's values would be equivalent to adults.) Children form a very high proportion of people killed or injured in developing countries (about twice that of the UK) and the willingness-to-pay approach might therefore appear, at the moment to be inappropriate. Similarly the method is used in the UK to obtain values for drivers or passengers of motor vehicles only. Again this weakens the case for its use in developing countries where significant proportions of people killed and injured are pedestrians and pedal cyclists. Lastly it may be more difficult to value changes in risk in developing countries because of the difficulty in respondents providing a monetary value where the markets do not necessarily involve money exchange, i.e a much greater use of the informal sector of the economy.

2.12 It seems unlikely therefore that reliable willingness-to-pay based costs and values will be available for use in developing countries for some time. It is therefore recommended that the gross output approach is used to cost road accidents in developing countries. However, in order to try to capture some of the `humane' considerations reflected in the willingness-to-pay approach, gross output values should be augmented by a further allowance for `pain, grief and suffering' of those involved in road accidents. This, in fact was the approach employed in the UK prior to the recent adoption of the willingness-to-pay approach. The way in which such an allowance might be added to fatal, serious and slight accidents to reflect pain, grief and suffering is discussed in the next section.

3. USING THE GROSS OUTPUT METHOD IN PRACTICE

3.1 As described in section 2, costs associated with a road accident may arise from injury to persons, damage to property and administrative procedures. Using the gross output method, these costs can be further divided into two categories, namely those that are due to a diversion of current resources and those that are due to a loss of future output. This section describes how these various costs can be derived and includes a worked example from one particular country.

VALUE OF THE LOSS OF OUTPUT 3.2 Road accidents lead to a loss of output in the year in which the accident occurs and, in the case of fatal and very serious accidents, in future years also. In the case of a fatality, the loss of a person's output is of course complete. In this situation, costs in future years have to be discounted to give present day values. The discount rate used should be that which is currently in use by economists and planners in the country concerned. In order to determine `lost output', certain assumptions have to be made. In the case of fatal accidents the number of `person years lost', is obtained by obtaining the average age of road accident fatalities and subtracting this from the average age at which a person ceases to work. In the case of serious accidents, estimates must be obtained of the average number of days that the inured person spends in hospital and then spends recovering at home from the accident. In the case of a slight accident, an estimate must be obtained of the (relatively small) number of days that the person is not working due to attending a doctor's surgery, a clinic or hospital (as an out-patient) to receive treatment for their minor injury, or being at home convalescing. Information on days lost following serious and slight road accidents can be obtained from hospital records and from information on as

many case studies as can be obtained Additional information may also be obtainable from insurance company records or employers records. Loss of output due to permanent and long term injuries depends on the number of cases, the length of absence from work and the percentage disability when work is resumed. 3.3 Having derived an estimate (and it should be stressed that it can be no more than an estimate) of the average number of days and years lost following a road accident, the value of those days and years lost must be determined. This is obtained by using figures published by government of national wage rates, before the removal of taxes. In many developing countries, a significant proportion of the population will be agricultural workers, many being self-employed and probably cultivating small plots of land For these and possibly other workers, (e.g. on short term employment), it is unlikely that published statistics of wage rates exist and estimates will have to be derived of annual incomes per capita. The most important `unpaid' workers in any country are housewives. The services rendered by house-wives are an important part of the real income of the country and the loss of these services is a loss to the country. A value must of course be placed on housewives services that are lost as a result of a road accident. In the UK this is set at the average wage of employed women and in the USA it is the value of replacing their services. In the UK, estimates of lost output are made separately for males and females, (for accidents taking place in urban and rural areas), and for road user type according to the annual distribution of accidents. If, in a particular country, separate costs are required for males and females then the way in which house-wives time is costed is important. If this subdivision is not required then, as described above, national average wage rates can be used. It is important to note that it is accidents by degree of severity that are being costed but that lost output is obtained on a person-injured' basis. The average number of persons injured per type of accident taking place must then be obtained. In the example given later there were, on average 1.45 casualties per accident in Cyprus. Ideally this information should be obtained separately for fatal, serious and slight accidents. (See also 2.1).

The total costs will be determined, apart from the number of casualties, by: the percentage of serious or fatal casualties who become in-patients, the average length of stay in hospital, the average cost per day of hospital treatment, the average number of out-patient visits, the average cost per out-patient visit, the average costs incurred by general practitioners, the costs incurred by the ambulance service.

3.4

3.5

All these factors have to be taken into account in the case of serious injuries; out-patient and general practitioners treatment can be ignored in the case of fatalities, and by definition inpatient costs cannot arise in the case of slight injuries. Some of this information may be available from sources published (usually) by the Ministry of Health. It is unlikely however that annual reports will state categorically the average cost per day of hospital treatment. Rather, it will be a case of using available statistics to produce this information. Thus it should be possible to break down total cost of all health treatment into that provided by hospitals and that provided by health centres, clinics etc. Information on the total number of hospital beds in a country is usually available from published statistics and assuming full occupancy at all times (which is not unreasonable), the average cost per bed per day can be obtained. 3.8 Efforts should be made to collect information on cost of treatment from hospitals in both urban and rural areas covering, if possible, all regions of a country. In many developing countries, hospitals often cater for patients from different income groups, religions or cultures. Information should be collected from as many classes of hospital (or private clinic) as possible. Average costs of treatment should then be weighted according to the proportion of accidents (by seventy) taking place in urban and rural areas and by the distribution of patients to the different classes of hospital. As stated in paragraph 3.6, information must be obtained on a `per accident' basis and average costs of treatment for persons killed, seriously or slightly inured must be multiplied by the average number of persons injured in the equivalent categories of accident to provide a cost of medical treatment per accident.

3.6

COST OF MEDICAL TREATMENT 3.7 The medical costs resulting from road accidents arise from hospital treatment (in-patient and outpatient), treatment by general practitioners (not included in UK), and the use of ambulances.

COST OF DAMAGE TO VEHICLES AND OTHER PROPERTY

3.9 There are three basic sources for information on cost of damage to vehicles; the insurance companies, garages and large fleet operators such as bus companies and freight operators. The best method to use will depend on local circumstances. 3.11

From the above it can be seen that returns from insurance companies need to be treated with caution. If possible information should be obtained from local garages on the cost of vehicular repair. In the UK damage - only accident values now include an estimate of unclaimed and claimed values, a departure from Dawson's method (Simpson & O'Reilly 94). Efforts should be made to ensure that information collected is representative of national accident figures. Thus proportions of the different categories of personal injury accidents, accidents in urban and rural areas and types of vehicles involved should be as close as possible to the national figures in order that the sample is not biased and is satisfactory from the point of view of coverage. If statistics on cost of vehicular repair are unavailable from insurance companies then an alternative approach is to collect information from garages, repair shops and, additionally from bus companies, freight operators etc. As with insurance company records, it is important to collect information on all classes of vehicle involved in urban and rural areas in accidents of different severity. Using this method however, the number of vehicles on which information is needed can be predetermined by collecting information on a given proportion of the accidents taking place nationally. For example, if in a particular country 10,000 vehicles were involved in road accidents in a given year than a 5 per cent sample would mean collecting information on 500 vehicles. Ideally subtotals within this figure should reflect the proportion of accidents involved in the reported number of fatal, serious and slight accidents; those occurring in urban and rural areas; and finally the different classes of vehicles involved. Information must, of course also be collected on vehicles involved in damage-only accidents - see 3.14. Information on the cost of private car repair is obtainable from garages and body repair shops. Information on buses and goods vehicles is best obtained from bus companies and freight operators. These can be both private and public sector companies and efforts should be made to collect data from both. In obtaining average values of the cost of repair of buses and trucks, values obtained from the investigation should be weighted according to the actual number of vehicles operated within the private and public sectors. An estimate will have to be obtained of the total number of damage - only accidents taking place. In most countries these do not have to be reported to the police and accurate statistics are therefore likely to be unavailable. It may be possible to obtain an estimate from insurance records which can indicate the number of vehicles involved in damage accidents per vehicle involved in personal injury accidents. In some countries insurance records may not be available. Indeed, in some countries insurance may not even be compulsory. In these

3.10 If the large majority of cars carry comprehensive insurance in a country (as opposed to Third Party cover only) and if the cooperation of insurance companies is available, then making use of information held by insurance companies may be the best approach. A specially designed questionnaire should be sent to as many insurance companies as possible. (In the original study carried out in the UK by Dawson, 4410 replies were received from 15 different insurance companies). The questionnaire should seek to establish: (i) background information such as age and sex of persons injured, locality, severity of accident, degree of personal injury (if any), number of casualties and numbers of vehicles, etc. type of insurance: comprehensive, third party, fire and theft or third party only. the payment for damage to the insured vehicle and for damage to vehicles and other property belonging to third parties.

3.12

(ii)

(iii)

There are a number of factors that should be taken into account when using information provided by insurance companies (i) public service vehicles and some large fleets of commercial vehicles may not be insured. many vehicles are not comprehensively insured. If information is collected on comprehensively insured vehicles only, this may not be a truly representative cross-section of all vehicles.

(ii)

3.13

(iii) many policies involve the insurer paying part of the cost. Therefore some payments will be far less than the true cost, leading to an underestimate of the average cost of damage per vehicle. (iv) claims for small amounts will not be submitted if it means the insurer losing his or her 'no claim's bonus'. (v) claims to insurance companies may be overestimates of the real cost of damage incurred. (vi) usually the cost of damage to a vehicle is paid for by one insurance company. With comprehensively insured cars they may, under `knock for knock' agreements, be paid by the company with which the damaged car was insured. In some cases however the cost to one vehicle may be shared between two companies and this should be identified. 3.14

circumstances ratios of non-injury accidents derived in other countries may have to be used. In the UK it has been estimated that there are at least 6 non-injury accidents taking place in urban areas and a ratio of 4 5 in rural areas for each injury accident. 3.15 Having collected information on the average cost of repair of vehicles involved in fatal, serious, slight and damage-only accidents, the average number of vehicles involved in these classes of accident needs to be determined from national accident statistics. By multiplying cost per vehicle by number of vehicles involved, the average cost of vehicular repair per accident (by degree of severity) is obtained. In the example given in the next section, a cruder method of determining cost of vehicular repair is given which makes use of relative costs of spare parts and labour in UK and the other country. As a last resort, this sort of approach can be used but efforts should, if possible, be made to obtain information at least from insurance companies. In a road accident, damage may also occur to movable property such as goods or personal effects carried by vehicles or to fixed property such as walls, lamp standards, signs etc. Information on claims for damage to property of third parties could again be obtained from insurance companies. Information on the cost of damage to street furniture can best be obtained from local authorities.

economy of the country. However, there are, as pointed out in section 2, other important issues to consider, such as suffering and bereavement, that fall upon individuals. Although these are difficult to express in monetary terms their existence is very real to the persons concerned. Moreover they are costs which the community would usually be prepared to meet in order to avoid the misery involved. If the costs given in this Note are to be used in the economic assessments of road improvements, then it is important that they should reflect the value that the community places on the saving of life and the avoidance of suffering. 3.20 It would therefore appear to be necessary to try to estimate the value that the community places on the avoidance of loss of human life. As stated earlier, this ought ideally to be done using the willingness-to-pay approach to the valuation of safety and the costing of risk. However, implementation of such an approach in a developing country will be no easy matter and as outlined in section 2, the ideal willingness-to-pay based costs and values might be approximated by adding an allowance for "pain, grief and suffering" to gross output figures. Early attempts to cost "pain, grief and suffering" in road accidents in the UK made use of awards made in courts in relation to people killed and injured in accidents. This provided some insights into possible values that could be used, but the figure that was finally adopted for fatal accidents was based on the fact that in 1967, the net output (and not gross output) method was in use. Using this method, `average consumption' was subtracted from `average output' with the result that an elderly non-productive person had a negative output and such a person's life would therefore be accorded a negative value. A figure was therefore added (5000 in 1967) to make the value positive for all age and sex groups. The figure of 5000 was thus taken to reflect the minimum value accorded by society to the avoidance of the pain, grief and suffering associated with premature death.

3.16

3.17

3.21

ADMINISTRATIVE AND OTHER COSTS 3.18 Other costs that arise as a result of road accidents include those associated with the administration of insurance, the police and court proceedings and possibly with the delays caused to other vehicles at the scene of the accident. None of these costs are particularly easy to determine. In the UK, 17% of the total cost of all insurance costs is set against administrative expenses. Half of this sum is connected with paying out of claims, the remainder (8.5%) being concerned with the handling of claims. Knowing the total sum spent on premiums paid in the UK, an estimate (8.5% of this total) can be allocated to insurance administrative costs. Specific support of the police may be required in order to obtain an estimate of the time spent by police in dealing with accidents of differing severity. Compared with other costs involved, administrative costs are likely to be low and it is probably not worth spending much time and effort in producing detailed estimates. In the example given later, percentages derived in the UK were used and ranged from 0.2% of cost of all other resource costs for fatal accidents to 14% in the case of slight accidents. SUBJECTIVE COSTS 3.19 The last section brought together the costs of accidents which directly or indirectly affect the

3.22 Fairly arbitrary values of 500 were later added to the cost of serious accidents and 15 to slight accidents. These sums represented, at the time, additions to the total resource costs derived for fatal, serious and slight accidents of 46%, 100% and 8% respectively. By the early 1980's the additional value added as non-resource costs was amended slightly to 38% for fatal accidents with values added to serious and slight accidents remaining as above. It should be stressed that these values are more or less arbitrary but neverthe less, a case can be made for their inclusion see paragraph 2.3. In the absence of more detailed research targeted at developing country societies and economics, it is suggested that an additional 38%, 100% and 8% of the resource (i.e. quantifiable) costs derived in each country are added to reflect pain, grief and suffering. 7

4.
4.1

CASE STUDY - CYPRUS

Section 3 explained how the Gross Output method might be used to cost road accidents in developing countries. In this section a case study is presented of the application of this method to cost road accidents on major inter-urban highways in Cyprus. This, it is hoped, provides further insight into the practical use of this method to cost road accidents. In 1984 the then Overseas Unit TRRL was asked (in support of a project being undertaken in Cyprus by British transport consultants), to cost road accidents taking place on a number of major inter-urban highways. Although the time available was little more than two weeks, the appraisal illustrates the approach that can be used to cost road accidents in a developing country and how the information given in sections 3.1-3.22 can be used in practice. Where information could not be collected in the time available, realistic estimates were derived, using in some instances information from the accident cost procedure used in the UK at that time. In this analysis, the 'gross output' or 'human capital' approach was used and the costs included were as follows: (a) (b) vehicle repair costs lost output due to death or injury. This was calculated as the present value of the expected loss of earnings plus any non-wage payments paid by the employer. cost of hospital treatment police and administrative costs costs of pain, grief and suffering to the casualty, relatives and friends.

limited information from insurance companies. In the time available this was the only approach possible and even this was completed on limited information from only two companies. Only vehicles which were comprehensively insured were included even though this may not have been a typical cross section. In addition, an attempt was made to include sums of money that insurees pay out themselves in the cost of repair, since policies in Cyprus usually state that the insurer pays approximately the first 50 of the cost. 4.6 In Cyprus, in 1984 there were 16,737 vehicles involved in 7,734 accidents, giving an average ratio of 2.16 vehicles per accident. This relatively large number of vehicles per accident (the equivalent ratio in the UK being 1.33) obviously had a significant effect on the overall average cost of vehicle damage per accident. The figures collected from insurance companies suggested that damage costs in injury accidents are about twice those incurred in non-injury accidents and were as follows: (a) Average cost of damage in injury accidents in Cyprus in 1984 = 1130 Average cost of damage in damage-only accidents in Cyprus in 1984 = 530

4.2

4.3

(b)

(c) (d) (e)

In this limited study, these values for average cost of repair applied to all injury accidents taking place (irrespective of severity) and an estimate needed to be made of cost of repair (separately) for fatal, serious and slight accidents. The cost of repairs in fatal and serious accidents tends to be greater than in non-fatal accidents, and adjustment figures derived in the UK were used as follows: Ratio Repair costs in slight accidents Repair costs of average personal injury accident Repair costs in serious accidents Repair costs in fatal accidents 1.0 1.11 1.4 1.8

4.4

Not all the costs incurred in road accidents in Cyprus could be readily quantified in monetary terms. For example it was virtually impossible to obtain data on items such as time delays to vehicles following an accident or out of pocket expenses to the casualty or to relatives and friends. Consequently these were excluded. The principle of not over-stating costs was followed and wherever alternative values presented themselves, the minimum value was taken in order to avoid the over-estimation of road accident benefits following highway improvements in Cyprus. Any upward revisions to the values derived would thus have the effect of increasing benefits from accident savings following the various road improvements. In this exercise, all costs (unless otherwise stated) are given in Cyprus pounds (1984 prices) with C1 approximately = 1.27

Applying these weighting ratios to the overall average cost of a personal injury accident in Cyprus gave the following values: Repair costs in fatal accidents Repair costs in serious accidents Repair costs in slight accidents Repair costs in damage-only accidents 4.7 1780 1430 990 530

VEHICLE REPAIR COSTS 4.5 The cost of repair of vehicles involved in road accidents in Cyprus was obtained by collecting

If costs of vehicular repair had not been available from insurance companies (or alternatively, garages specialising in repair work) then an alternative approach would have been to use the relative costs of spare parts and labour in the UK and the

country in question, in this case Cyprus, and to adjust costs derived in the UK using these ratios. To use this (albeit crude) approach, average costs of typical cars and spare parts used in both countries were required. In Cyprus, vehicles commonly in use were the 1600cc Mazda 626 family saloon and the 1300cc Mazda 323 small saloon. Common vehicles used in the UK were the B.L. Maestro, family saloon and the B.L. Metro, small saloon. Comparable costs are shown below. The purchase price of the vehicles in Cyprus and the UK were very close. The overall ratio of spare parts, Cyprus: UK, however, showed costs to be about twice as high in Cyprus as the UK as shown below. The cost of labour in UK garages was in 1984 about 12.00 per hour and the cost of labour per hour in Cyprus was 7.00 (taxes excluded). In the UK the total cost of vehicular repair was typically 70 per cent labour costs and 30 per cent replacement of spare parts. Thus the breakdown of 100 spent in a garage in the UK might be used to derive costs in Cyprus adjusted as follows: 4.9

An expenditure of 100 sterling in UK on vehicular repair might thus equate to 113 sterling in Cyprus. In other words the cost of repair to a vehicle involved in a road accident was probably about 10 per cent higher in Cyprus than in UK. Comparing repair costs shown earlier for vehicles involved in accidents of varying severity in Cyprus with those derived in the UK indicates that costs were about 5-8 per cent greater in Cyprus. Thus using relative costs of spare parts and labour in any country to adjust vehicular repair costs derived in the UK may be a crude but acceptable method if information is unavailable from insurance companies or garages. As a word of caution it should be pointed out that it was difficult to use this method on a later study. In this case there appeared to be no consistent pattern of ratios of spare parts in this country to costs in UK. This may well be the case elsewhere.

ESTIMATING LOST OUTPUT 4.10 The `Gross Output' method requires an estimate of current average wage rates. In the case of fatal accidents current wage rates were multiplied by the number of years lost' due to the road accident, and in the case of serious and slight accidents, days 'lost' were multiplied by the daily wage rate. With fatal accidents, the sums lost in future years were discounted back to a present value by using an appropriate discount rate. For Cyprus this was 9 per cent. 4.11 The average age of a person killed in a road accident was obtained from information provided in "Statistics of Motor Vehicles and Road Accidents" published by the Cyprus Department of Statistics and Research, Ministry of Finance. (see Table 3.1).

In Cyprus Mazda 626 (1600cc) or BL Maestro (1500cc) Cost of spare part bumper wing windscreen 102 71 78 3250

In UK

Ratio of cost & spares*

3307 42 30 38 2.42 2.36 2 00

In Cyprus Mazda 323 (1300cc) or BL Metro (1300cc) Purchase price Cost of spare part bumper wing windscreen 70 56 76 2600

In UK

Ratio of cost & spares*

2680 25 28 30 2.80 2.00 2.53

The overall average ratio is 2.4 9

TABLE 3.1
Average age of fatality in Cyprus (1984) Age Groups 0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80+ Total Weighted Average Average Age 5 15 25 35 45 55 65 75 85 Number Total Years 5 165 750 315 585 550 1170 675 340 4555 (43 years)

However, examination of accident statistics on the main inter-urban routes (relevant in this particular study) showed a higher ratio of 1.83, as follows: Route Casualties 71 91 Accidents 38 48

Larnaca-Kophinou 1 11 30 9 13 10 18 9 4 105 Larnaca-Dhah

New Nicosia-Limassol Road (Oct 84-Jan 85) TOTAL 58 220 34 120

Applying this ratio of 1.83, costs per accident became: Lost Output in fatal accident = 77,775 Lost Output in serious accident = 1,015 Lost Output in slight accident = 55 COST OF MEDICAL TREATMENT 4.15 The mayor hospital in Nicosia provided the following information: (a) first class bed = 10 + 4 treatment (minimum) + cost of medicine + cost of operation second class bed = 6 + 4 treatment (minimum) + cost of medicine + cost of operation (c) third class bed = 3 + 50% cost of operation

4.12 In Cyprus, persons in government salaried employment retire at the age of 60, persons of non-salary status retire at 65 years and self employed persons retire at ages greater than these. The weighted average of retirement age in Cyprus was estimated to be 65.5 years. The average number of years of lost output following a fatal road accident was thus about 23. The average wage in Cyprus in 1984 was 4100 per annum per person. This value, plus a further 8 per cent national insurance contribution was applied to the 23 years lost due to a road accident and discounted at 9 per cent. The total discounted lost output was estimated to be 42,500 per person. 4.13 The average number of days spent in hospital in Nicosia for all cases, not only road accident patients, was 8.5 days. However an analysis of medical records showed that for road accident patients the average length of stay was longer, approximately 13 days. Information collected from persons injured suggested that a further 24 days, on average, were spent recovering at home from a serious road accident. Thus the average lost output for a serious road accident casualty was estimated to be 37 days. With an average daily wage rate of 15/day the cost of lost output for serious accidents was thus 555 per casualty. Lost output from slight injuries was small and on average 2 days were lost. Therefore the lost output following a slight road accident was estimated to be 30 per casualty. 4.14 The above figures were calculated on a `per casualty' bass. In order to obtain costs `per accident' the costs per casualty had to be multiplied by the number of casualties per accident. Over the preceding 10 years there had been on average some 1.45 casualties per accident in Cyprus.

(b)

These values (which were in fact charges made to patients) did not reflect the true cost to the hospital because of subsidies received by in-patients from the government. The economic or resource costs were therefore considerably greater. Information collected from the Ministry of Health suggested that the overall average cost for one days in-patient treatment (including staff costs, cost of medicines, operations and overheads etc.) was 53. Using this estimate, costs of medical treatment for road accident casualties were derived as follows: a) Cost of fatal casualty

Assuming 4 days spent in hospital before dying at a daily cost of 53 together with: capital cost of ambulance capital cost of hospital hospital administration costs = approx. 235 b) Cost of serious casualty Assuming 13 days spent in hospital at a daily cost of 53 together with four outpatient visits at 14

} overhead of 23

Estimated average

10

one ambulance attendance on average and capital cost ambulance capital cost hospital hospital administration cost) = approx. 770 Estimated average overhead of 23

(c)

Cost of slight accident Damage to vehicle Lost output Medical 990 55 65

Police and administration (14%) 150 Total (d) Cost of damage-only accident Damage to vehicle Police and administration (10%) Total * Totals rounded down SUMS TO REFLECT PAIN, GRIEF AND SUFFERING 4.17 As stated earlier, the Gross Output approach includes a sum to reflect pain, grief and suffering. These sums are `notional' and are not based on detailed or complex calculations. In the estimation of lost output for persons killed in road accidents in Cyprus the average age of a fatality at 43 years was subtracted from the average age of retirement, which was 65.5 years. The average life expectancy in Cyprus, however, was as follows: Age Additional Years 39 42-46 Average life expectancy 74-78 78-82 530 50 580 1.260

c)

Cost of slight casualty Assuming 50 per cent of persons injured make an outpatient visit to the hospital and that one ambulance attendance was needed for 50 per cent of those injured. Also assumed that capital and administration cost of hospital includes one GP visit for 50 per cent injured = average estimated hospital cost for slight injury = 35. Taking into account the fact that there were 1.83 casualties per accident on major inter-urban roads in Cyprus, then
(a) (b) (c) cost of medical treatment in fatal accidents = 430 cost of medical treatment in serious accidents = 1410 cost of medal treatment in slight accidents = 65

POLICE AND ADMINISTRATION COSTS 4.16 In the time available it was not possible to obtain detailed information on the average cost of police and administration costs for the different types of accidents taking place. Instead, values based on those derived in the UK were used which suggest that police administration costs represent about 0.2 per cent of the total resource cost of fatal accidents, 4.0 per cent of serious accidents, 14.0 per cent of slight accidents and 10.0 per cent of damage-only accidents. RESOURCE COST SUMMARY (a) Cost of a fatal accident Damage to vehicle Lost output Medical Police and administration (0.2%) Total (b) Cost of serious accident Damage to vehicle Lost output Medical Police and administration (4%) Total 1,780 77,775 430 160 80,140* 4.18

Male Female

35-39 35-39

Thus the average life expectancy in Cyprus was about 79 years for males and females combined. In other words, people in Cyprus could hope to live another 14 years or so after retiring. This, however, was not included in the years of lost output and if no account is taken of this it implies that people's lives after they retire are of `little value'. This is clearly not so in any country and the addition of sums for pain, grief and suffering to some extent compensate for this. The percentage of resource costs used to reflect pain, grief and suffering were based on those values used in the UK at the time of the appraisal, and were as follows: Fatal accidents 38%

1,430 1,015 1,410 150 4,000*

Serious accidents 100% Slight accidents 8%

Using these percentages, the estimated total costs of accidents on major inter-urban roads in Cyprus were as follows:

(a)

fatal accident Resource costs Non-resource costs Total

80,140 30,450 110,600 4,000 4,000 8,000

HEYS, R F et al, 1968. Practicability and cost of oestriol assays for saving babies in a maternity hospital. Lancet, 331-332. HILLS, P J and M W JONES-LEE, 1981. The Costs of traffic accidents and evaluation of accident prevention in developing countries. In: PTRC. SummerAnnual Meeting. University of Warwick, 1316 July 1981. London: PTRC Education and Research Services. HOPKIN, J M and D M O'REILLY, 1993. Re-valuation of the cost of road accident casualties in Great Britain 1992 revision. TRL Research Report 378, Crowthorne. Transport Research Laboratory. HILLS, P J and M W JONES-LEE, 1983. The role of safety in highway investment appraisal for developing countries. Accident Analysis and Prevention, 15, 35569. JACOBS, G D, 1976. A study of accident rates on rural roads in developing countries, TRRL Laboratory Report 732, Crowthorne: Transport and Road Research Laboratory. JACOBS, G D and C A CUTTING, 1986. Further research on accident rates in developing countries. Accident Analysis and Prevention, 18(2), 119-127. JONES-LEE, M W, 1976. The Value of Life: An Economic analysis. London, Martin Robertson. JONES-LEE, M W, 1989. The Economics of Safety and Physical Risk, Oxford, Basil Blackwell. JONES-LEE, M W et al 1993. The Value of Preventing Non-Fatal Road Injuries: Findings of a Willingness-to-Pay National Sample Survey". TRL Contractor Report CR330. Crowthorne, Transport Research Laboratory. McMAHON, K, 1991. The valuation of safety. PIARC Conference September 1991. Marrakesh, Morocco. MOONEY, G H, 1977. The Value of human life. London: The Macmillan Press Ltd, SIMPSON HF and D O'Reilly, 1994. Revaluation of the accident related costs or road accidents TRL Project Report 56, Transport Research Laboratory. SIMPSON, HF, and D O'Reilly, 1994. Revaluation of the accident related costs of road accidents. TRL Project Report PR56, Crowthorne, Transport Research Laboratory. SINCLAIR, T C et al, 1972. Human life and safety in relation to technical change. Brighton: Science Policy Research Unit, University of Sussex. TRRL, OVERSEAS UNIT, 1988. A guide to road project appraisal. Overseas Road Note 5, Crowthorne: Transport Research Laboratory.

(b)

serious accident Resource costs Non-resource costs Total

(c)

slight accident Resource costs Non-resource costs Total 1,260 100 1,360

d)

damage-only accident Resource costs only Total 580 580

4.19 The above illustrates how the Gross Output (or Human Capital) approach was used to cost accidents in Cyprus. With limited time available, certain estimates and assumptions needed to be made but the methodology indicates this and above all acts as a reminder of the many factors that need to be taken into account in costing road accidents. The Appendix illustrates how costs derived were used in a number of highway cost-benefit analyses under-taken in Cyprus. These in turn showed that the inclusion of accident savings could have an effect on both the ranking of projects and on the magnitude of net benefits generated by the specific highway improvements.

5. REFERENCES
ANDREASSEN, D, 1992. Horations on the use of accident cost data. Traffic Engineering and Control, May 1992 BERGSTROM, T C, 1982. When is a man's life worth more than his human capital? in Jones-Lee, M W. (Ed). The Value of Life and Safety: Proceedings of a Conference held by the Geneva Association, Amsterdam, North Holland. DAWSON, R F F, 1976. Cost of road accidents in Great Britain, TRRL Laboratory Report 79, Crowthorne: Transport and Road Research Laboratory.

12

6.

APPENDIX: THE INCLUSION OF ACCIDENT SAVINGS IN HIGHWAY COST BENEFIT ANALYSES

BACKGROUND
6.1 In order to appraise a project, estimates need to be made of the costs associated with the project and the benefits that are expected to occur. In Western countries, benefits associated with highway engineering improvements are usually (i) direct savings on the costs of operating vehicles,

following a specific highway improvement (or improvements) (see below). With respect to point (iii) above, results are presented in this Appendix which indicate that economic benefits from reduced accidents following a highway improvement may add significantly to net present values or rates of return derived. It also shows that ranking of alternative schemes may change with the inclusion of accident savings. 6.3 In order to illustrate the effects of including the benefits of accident prevention in highway costbenefit appraisal, use is made of data collected by TRL on projects undertaken in Cyprus and Jordan over the period 1982-84. As part of these studies it was necessary to cost road accidents either nationally or on specific roads under investigation. In all these studies the `gross output' or `human capital' method was used to cost accidents. In most cases costs were derived both with and without sums added to reflect pain, grief and suffering. The information collected as part of these studies has been used to illustrate the possible effects of including the value of accident prevention in proposed highway improvement schemes. As stated above, it is difficult to assess with any degree of accuracy the likely effects on accidents of specific highway improvements. Consequently results are presented such that economic benefits from reduced accidents are assessed over a range of possible percentage reductions in accidents. Information is also presented of a more `hypothetical' exercise carved out using data obtained from a project in India.

(ii) economies in road maintenance

(iii) time savings by travellers (iv) reduction in road accidents (v) wider effects on the economic development of the region. Historically, highway cost-benefit analyses carried out on projects in Third World countries have tended to be based on operating cost savings only, although in recent years time savings have become more common. Three reasons are usually put forward for excluding savings based on reductions in road accidents. These are: (i) road accident costs (and values of prevention) are difficult to determine

(ii) changes in road accident rates following a specific road improvement are difficult to predict (iii) even if accident benefits were to be included, their effect on the economic appraisal would be minimal. 6.2 Clearly it is the object of this Road Note to assist with (i) above. It is also true that relatively few studies have been made of factors affecting accident rates in developing countries. A recent review by TRL showed that only five such studies have been undertaken in developing countries over the last twenty years or so. No consensus was achieved between these studies, nor even between the dependent variables used or the methods of analysis. Differences in traffic composition, road user behaviour and road geometry suggest that results from similar studies in developed countries (where even here, success in model development has been limited) cannot be used on projects in the developing world with any degree of confidence. Insufficient information is available at present to enable accident savings to be incorporated into specific cost benefit analyses of highway improvements in developing countries. However, studies undertaken (Jacobs 1976) etc. can be used in a theoretical exercise to provide an indication of the range of the possible changes in accident rates

FEASIBILITY STUDIES IN CYPRUS 6.4 Over the period 1982 to 1986 the British transport consultants Hughes Economic Planning carried out a number of feasibility studies in Cyprus. These included a study of a proposed road improvement from Limassol, the main port of Cyprus, to Paphos, the centre of a thriving tourist industry (see Fig 6.1), and an appraisal of the likely benefits which would result from an improved road link between the new dual carriageway running from Nicosia to Limassol and the busy port of Larnaca. At the time of this study there were three roads all of relatively poor alignment linking the new road and Larnaca (see Fig 6.1). The options were to either improve routes A and C or routes B and C. The TRL was asked to assist in these studies by providing estimates of road accident costs on each of the routes and the benefits that might result from reduced accident rates following the proposed road improvements. Having derived the above, the effect of including accident savings on the net present value (NPV) assessed by the consultants for the Limassol Paphos Road was obtained. (NPV being defined as the total discount net benefits estimated over the life of the project minus total discounted costs). Results are given in Fig. 6.2. The proposed improvements to the highway included road widening

6.5

13

Fig 6.1 Roads studied in Cyprus

from an average of 6 metres to 7.5 metres, a reduction in the number of junctions per kilometre from 1.7 to 0.5, improved sight distances and reduced road roughness. Earlier research work carried out by TRL suggests that the combined effect of these improvements might be to reduce accidents by some 20 - 35 per cent. Thus the effect of including accident savings in the appraisal (with benefits covering the period 1984 2004) might be to increase the NPV, assessed at about 14.5 million at 1984 prices by 12 - 20 per cent if accident costs include sums to reflect pain, grief and suffering. 6.6 In the appraisal carried out in 1985 the alternatives were to improve the alignment of existing routes with the road being widened from 6 metres to either 7 metres or to dual carriageway standard. Results are given in Table 6.1. It can be seen that by including possible accident cost savings in the appraisals, the NPV's might be increased by 10 to 25 per cent depending on either the standard of improvement or the routes selected. In these analyses, therefore, it would appear that the inclusion of possible accident savings significantly increases Net Present Values derived and that accident savings, when measured against the more traditional benefits from reduced vehicle operating costs are by no means insignificant.

6.7

Fig. 6.2 Feasibility study Limassol Paphos Road, Cyprus 1982

Percentage increase in net present value by including accident cost savings

FEASIBILITY STUDIES OF TWO ROADS IN JORDAN 6.8 Over the period 1980 - 1981 the British transport consultants Halcrow - Fox were involved in a range

14

TABLE 6.1
Larnaca Link Study Cyprus Option Description Per cent increase in NPV by including accident

savings
A+C A+C Both 7m A = Dual carriageway C=7m A+C Both dual 20 10 15

of transport projects in Jordan including feasibility studies of proposed major road improvements. Amongst these were proposed improvements to the roads from Salt to Suweilih and Zarqa to Rusaeifa (see Fig. 6.3). The former involved the upgrading of 8.4 km. of road from single to dual two - lane carriageway and the latter the upgrading of 2.2 km. of road from single to dual two - lane carriageway. The consultants estimated the first year rates of return of the proposed improvements using 1990 as the first year of full benefits. 6.9 Whilst the studies by Halcrow - Fox were drawing to a close, a joint British - German consortium were undertaking a broad review of the transport sector in Jordan and asked the TRL to advise on road safety issues including the costing of road accidents. Using the results of this study it was thus possible to include possible accident cost savings in two of the feasibility studies undertaken by Halcrow - Fox. Results are given in figures 6.4 and 6.5. Work carved out by TRL suggests that the upgrading of a single carriageway to dual carriageway may have the effect of reducing accidents by 30 to 50 per cent. (It should however be pointed out that results from other studies showed different results

B+C B+C

Both 7m B = Dual carriageway C=7m

11 16

B+C

Both dual

25

6.10

Fig 6.3 Roads studied in Jordan

15

Fig. 6.5 Zarqa - Rusaeifa Road, Jordan Fig. 6.4 Salt - Suweilih Road, Jordan
Percentage increase in first year rate of return by the inclusion of savings in accident costs Percentage increase in first year rate of return by the inclusion of savings in accident costs

improvement decision to form a legitimate basis for the analysis. The example was as follows. of the effects of road widening on accident rates.) On the Salt to Suweilih road the effect of including accident savings might be to increase the estimated first year rate of return by 10 to 16 per cent. (assuming sums are included to reflect pain, grief and suffering). On the Zarqa to Rusaeifa road, however, possible measures range from 40 to 60 per cent. In both cases (unlike those in Cyprus) benefits estimated by the consultants included time savings as well as changes in vehicle operating costs. Even so, possible savings from reduced accident rates are by no means insignificant; in the case of the Zarqa to Rusaeifa road they represent almost 50 per cent of benefits from time and operating cost savings. This road in fact is particularly dangerous with, on average, over 20 accidents taking place per kilometre of road per annum. In these circumstances it may well be that a project which appears not to be feasible (i.e. with a first year rate of return below the discount rate) may become so by the inclusion of accident savings. FEASIBILITY STUDY IN INDIA 6.11 In order to determine the effects of including road accident savings on the ranking of protects, Hills and Jones -Lee (1981, 1983) used data from a study in India. Whilst the protects were in the strictest sense `hypothetical', they were sufficiently typical of a Third World road 6.12 Two cities in India, with populations of 1 million and 400,000 respectively, about 190 kms apart, were connected by a two-lane stabilised gravel road. The highway authorities were considering a number of mutually-exclusive schemes to improve conditions for traffic between the two cities.

These were as follows: Scheme A - the "invest-nothing" case, with continued maintenance of the existing grave! road; Scheme B - paving the existing gravel road, with minor improvements to alignment and minimal local widening; Scheme C - paving the existing gravel road, together with substantial improvements to the width and alignment of the right of way, reconstruction of bridges, drainage culverts etc. and Scheme D - the construction of an entirely new and shorter road remaining in use, mainly for local traffic. The new highway will have some restrictions on access and a design speed of 100 km per hour throughout it's length.

In all the schemes, 1969 was used as the baseyear with a time horizon of 1990: the monetary units used for costs were the Indian Rupee and Paisa (R1 - 100 Paisa) at 1969 prices and, where appropriate, take into account the foreign exchange costs (using a shadow weighting of 1.75). All are net of taxes and subsidies. A 12% per annum discount rate was used throughout. 6.13 As far as capital costs, vehicle flow, vehicle operating costs etc. were concerned, assumptions (which were not based on the results of specific studies) were adopted that were realistic and representative of such schemes in developing countries. Following discussions with members of the World Bank who had direct experience of such schemes, it was assumed that Scheme B would raise accident-rates by 30%, Scheme C would have no effect on accident-rates, while Scheme D which was purpose-designed on a new alignment would incorporate a number of safety features which together would serve to reduce accidentrates by 30%. The authors then examined the effect upon protectrankings of varying the cost of a fatal accident from 0 - 400,000 Rupees and of varying the ratio of fatal to non-fatal accident-costs from five to twenty. For all ratios of fatal to non-fatal accident costs, variation in the cost of fatal accidents had a significant impact upon the net present value of those projects that alter accident rates and, more significantly, had a substantial effect upon projectrankings. The results for the intermediate fatal/nonfatal accident cost ratio are summansed in Fig. 6 6 and it can be seen that an increase in the cost of a fatal accident from 0 to say 100,000 Rupees serves to raise Scheme D from third to first place in the protect-ranking. In summary, the results of this sensitivity exercise and the results from Cyprus and Jordan indicate quite clearly that, far from being a matter of subsidiary importance, the size of accident costs or values of accident prevention could have a marked effect both on the ranking of transport projects, in terms of net present value within mutually-exclusive groups, and on the magnitude of net benefits generated by any given project. In short it would appear that the issue of the "appropriate" cost to associate with particular types of accidents, or values to place upon their avoidance, is not one that can legitimately be ignored on the grounds that accident costs have little overall importance in project-appraisal. The message of these sensitivity tests is that such costs may be potentially very important indeed. However before such benefits can be assessed with any degree of certainty and incorporated into specific feasibility studies, more research is needed on the effects of highway improvements on accident rates.

6.14

Proposed road improvement, India

Fig. 6.6 The effect on overall net present value of accident-prevention

(where a fatality is assumed to have a value 10 times that of an average injury)

6.15

Printed in the United Kingdom for HMSO DdP4885 6/96 C6 G5100 10170

Costing road accidents in developing countries

ORN 10

Old Wokingham Road Crowthorne Berkshire RG11 6AU

Overseas Development Administration 94 Victoria Street London SW1E 5JL

OVERSEAS ROAD NOTE 11

URBAN ROAD TRAFFIC SURVEYS

Main subject area: Urbanisation/Transport, energy efficiency Theme: Urban Transport Project title: Urban Road Traffic Surveys Project reference: ODA/2.3

This document is an output from an ODA-funded research project, carried out for the benefit of developing countries.

Crown copyright 1993. The views expressed in this report are not necessarily those of the Overseas Development Administration or the Department of Transport. Overseas Centre, TRL, 1993 ISSN 0951-8987

ACKNOWLEDGEMENTS
This note is based on a draft commissioned by the Overseas Centre (Programme Director: Dr. J. Rolt) and prepared by the Institute for Transport Studies, Leeds University. The authors were Mr N W Marler and Mr F O Montgomery, with the assistance of Prof. A D May and Mr P W Bonsall. Mr G Gardner, TRL was Project Officer. It has been edited for inclusion in the Overseas Road Note series by Mr P Disney, an independent consultant, and Mr G Gardner, Dr J C Rutter and Mr P R Fouracre. The material is based on experience gained by this team working with many overseas institutions, notably the Institute of Technology, and the Institute of Road Engineering, Bandung, Indonesia, the Training College of the Department of Land Transport, Jakarta, Indonesia, the Asian Institute of Technology and Chulalongkorn University, Bangkok, Thailand.

OVERSEAS ROAD NOTES

Overseas Road Notes are prepared principally for road and transport authorities in countries receiving technical assistance from the British Government. A limited number of copies is available to other organisations and to individuals with an interest in roads overseas. Enquiries should be made to Transport Research Laboratory Crowthorne, Berkshire RG11 6AU Crown Copyright 1993 Limited extracts from the text may be reproduced, provided the source is acknowledged. For more extensive reproduction please write to Head of Overseas Resource Centre, Transport Research Laboratory.

CONTENTS
Page 1. Introduction Scope and objectives The need for surveys 2. Survey planning General considerations Road classification system Vehicle classification system Traffic variation and sampling Fieldwork preparation Human resources Equipment Survey forms and instructions Pilot survey Liaison with other agencies Surveyor safety Data handling Processing and error checking Analysis 3. 4. Road inventory surveys Parking surveys Parking inventory survey Parking patrol survey Output 5. Origin and destination surveys: Registration number method Method Output 6. Traffic volume and classification surveys Manual traffic counts Automatic traffic counts Equipment Combatting equipment errors 7. Spot speed surveys Radar speed guns Manual short-base method Output 8. Network speeds and delay surveys Method Number of runs, routes and timing points Output 29 32 17 17 18 18 18 22 22 25 25 25 27 27 29 29 1 1 1 3 3 3 3 3 3 3 7 7 7 7 7 7 9 9 Appendix C: 10 10 13 13 13 LIST OF SURVEY FORMS (see Appendix C) A B C D E F G H J K L M N P Supervisors log Link inventory Junction inventory survey Parking beat survey Parking survey summary Registration number plate survey Traffic flow survey Traffic data summary form Speed survey along measured baseline Journey time survey from moving vehicle Delay survey by counting stopped vehicles Delay survey from vantage point Saturation flow profile survey Saturated period count Appendix B: Appendix A: Summary statistics and sample size General checklist for survey supervisors Instructions for surveyors and bank survey forms 46 45 43 11. References 41 10. Saturation flow Flow profile method Saturation period method PCU values 35 38 38 41 9. Junction delay surveys Stopped vehicle count method Method Output Elevated observer method Method Output Page 32 32 32 35 35 35 35

1 INTRODUCTION
SCOPE AND OBJECTIVES
1.1 This manual provides traffic engineers with a guide to simple, reliable traffic survey techniques, suitable for urban road traffic in developing countries. It covers surveys for urban traffic engineering, but does not cover road safety or transport planning. Public transport surveys are described in Overseas Road Note No 4 (TRRL, 1987) and road safety issues in TRRL (1990). 1.2 The manual consists of three parts: in chapter 2 general issues concerning traffic surveys are discussed; chapters 3 to 10 each describe a specific survey, and the appendices contain detailed instructions and survey forms for supervisors and surveyors. While the manual is written for the left-hand rule of the road, it can be used in countries which drive on the right, though the diagrams will need to be transposed and the accompanying text interpreted accordingly. 1.3 The surveys described in the manual are considered sufficiently robust for use in most urban situations, though it may be necessary for some small modifications to meet specific local needs and conditions. Account must be taken of any existing government guidelines and legal requirements which concern, or could have an impact on, the implementation of traffic surveys.

system, in terms of supply, demand, and performance, and identify existing problems; to determine the timing and sample rate of further surveys; as a base for predicting future demand and performance, in order to plan improvements. 1.7 Detailed data collection is required for the investigation of specific problems, the design of improvements, and for 'before and after' evaluation of the impact of changes. 'Before and after' surveys are a special type of monitoring to gauge the effects of a specific action. `After' surveys should be carried out long enough after implementation of the scheme to allow new traffic patterns to become fully established (typically 2-4 weeks), but not so long as to allow underlying traffic trends to affect the outcome. It is also desirable to carry out simultaneous 'control' surveys in a different part of the city, to reveal any underlying changes to traffic conditions.

THE NEED FOR SURVEYS

1.4 Surveys are required for both national and strategic traffic issues, and for local traffic planning, engineering and management purposes. Traffic data are important to a wide range of decision-making processes in the planning, construction, operating and maintenance of the transport system. Information is needed to support not only the case for transport investment (infrastructure planning, design and construction), but also to indicate how to make best use of existing road facilities (traffic management and maintenance policy). 1.5 The traffic data required from surveys is of transport supply (inventories and characteristics of the vehicles and infrastructure which comprise the system), transport demand (the amount the system is used and the patterns of movement) or performance (how well the system accommodates the demand placed upon it, measured in terms of traffic speeds, journey times and delay). Table 1.1 lists the main types of survey that are further described in subsequent chapters. The collection of traffic data is organised either as part of general background monitoring of traffic developments or as part of a detailed traffic investigation. 1.6 Traffic monitoring is the consistent and regular (periodic or continuous) surveying of traffic data, either nationally or locally, for a range of planning purposes: to calculate historical trends, in supply, demand and performance; to calculate hourly, daily, and seasonal variation factors; to provide a summary of the existing

TABLE 1.1
Survey types

Type (Chapter)

Information

Method

Output

Road inventory (3)

Road network Characteristics

Observation

Geometry, Land-use, Road-furniture provision Available parking space Types of parking

Parking inventory (4)

Parking supply

Observation

Parking use (4)

Demand for Parking space

Parking patrol survey

Occupancy times, Usage of space

Origin-destination (5)

Demand forecasting

Registration Number method

Route choice Through-traffic Travel times

Traffic volumes (6)

Demand

Manual counts, Automatic counts

Vehicle flows on links, Junction movements, Passenger flows, Traffic variability, Peak-hour factors, AADT Vehicle speeds on links, Speed flow measurements Network speeds, Link speeds, Network delay, Congestion points Total delays, Average arm delays, Distribution of delay times by turning movement, Delay causes Saturation flow, Junction capacity

Spot speeds (7)

Vehicle performance on links Route network performance

Short-base method, Radar observation Floating car method

Network speeds and delays (8)

Junction delay (9)

Junction performance

Stopped vehicle count, Elevated observer method

Saturation flows (10)

Junction capacity

Flow profile method, Saturated period count

2 SURVEY PLANNING
GENERAL CONSIDERATIONS
Road classification system
2.1 Roads have two basic, but possibly conflicting, functions: to move traffic smoothly and without interruption, and to provide access. The provision of access will often necessitate interruption to traffic flow, hence the conflict. The function of any specific road, therefore, is a balance between these two extremes, and is defined by a road hierarchy (see Fig 2.1). The concept of road hierarchy is fundamental to safe and efficient traffic operations and planning. The classification of a road's function within the hierarchy should also have a significant effect on its design standards (geometry, structure, etc.). Therefore it is important in any study of the traffic network to describe the road links in terms of their classification within the hierarchy. Fig 2.2 shows a possible classification system which should be amended to suit local conditions (Institution of Highways and Transportation and the Department of Transport, 1987).

Separate measurements may be required for seasonal rainy and dry periods. A particular problem in developing cities is that interruptions to traffic flows can be so common as to be considered part of the normal traffic scene. In this case the influence of the disruption should be noted, along with the traffic performance values being evaluated. 2.5 The aim of any survey should be to collect only as much data as is required to give an estimate at the desired level of accuracy. Appendix A gives detailed guidance on the use of simple statistical procedures to help choose the sample size.

FIELDWORK PREPARATION
Human resources
2.6 A traffic engineer develops the general requirements into a work plan and programme. He must choose the appropriate survey method for the task and conditions, and decide on locations and times. He must also ensure the correct preparation and training of both Supervisor and Surveyors. 2.7 Survey Supervisors undertake the surveys with teams of Surveyors. There should be no more than 10 Surveyors per Supervisor; larger teams are difficult to manage. The Supervisor must ensure the accuracy of observations and recording, timekeeping and completion of survey form headings. He must also support the survey team in general, by maintaining the supplies of forms, materials, and equipment. He also records changes in traffic conditions (for example, accidents, weather, traffic controls) using a log (Appendix B). 2.8 Shift lengths depend upon the intensity and rate of work, as well as any legal constraints or government recommendations on working hours. In surveys with a high work-rate (for example recording vehicle registration numbers) attention spans are relatively short, and shifts should be broken up into sessions of half an hour to one hour, with short breaks in between. In general, shifts should be too short rather than too long; no Surveyor should be asked to do more than is comfortably possible. 2.9 One Surveyor, however competent, should never be required to record more than twelve items (for example, three classes of vehicle and four turning movements). Two-way traffic is regarded as two items, even if not recorded separately. As a rule of thumb, one experienced surveyor using a pencil and appropriate form can simultaneously record: twelve items, up to a flow of 300 veh/hr. four items, up to 600 veh/hr. two items, up to 1200 veh/hr. one item up to 1800 veh/hr

Vehicle classification system

2.2 A fully comprehensive classified count may identify up to 20 different vehicle types It is rare that such detail will be required, however, and in order to minimise survey difficulties, five groups will often be sufficient based upon the sub-groups shown in Fig 2.3. The group or category number is based on the number of tyres on the vehicle. 2.3 Sub-divisions should be chosen to include groups of common interest. For example, in a transport planning study, divisions may be according to vehicle occupancy; for a road damage study, they may be by vehicle weight; and for traffic signals studies they may be according to their passenger car unit values (see Chapter 10). Each class must be distinguished from the others by a unique characteristic which can be seen easily in a moving traffic stream on a busy street. For example, methods which involve the counting of the number of tyres are more reliable than those which require estimation of length or weight. For clarification, a sketch or photograph of vehicle types should always be given to survey staff.

Traffic variation and sampling

2.4 Traffic variations are usually cyclical, and may be hourly, daily, or seasonal. The most appropriate days and times of survey depend on survey objectives (for example, whether average values or peak values are required). Surveys should not be conducted when traffic flow is affected by abnormal conditions, such as accidents, roadworks, public holidays, public processions, and severe weather conditions (particularly heavy rain).

Figure 2.1 Example of a road hierachy

Figure 2.2 Road Categories and Function

6 Figure 2.3 Simplified Vehicle Classification

Rates for inexperienced Surveyors should be about two-thirds of these values. For high volume flows in excess of 1800 veh/hr the use of tally counters will help. Other, less satisfactory techniques include counting in groups of ten, and creating sub-divisions of items for separate Surveyors to monitor (e.g. counting by lane or subdividing the vehicle categories). 2.10 Survey duration should be sub-divided so that anomalies can be identified by comparing different short periods, and discussed with the Surveyors involved. For example, even if the required data from a survey is the 1 hour total volume, it should be divided into 15 minute periods.

figure zero (not a dash), and suitable abbreviations are used for `not available' (n/av), or `not applicable' (n/app). Appendix C contains instructions and blank forms for each of the surveys described in the following chapters.

Pilot survey
2.15 Pilot surveys are full field tests of a survey method, preferably at the location of the main survey itself. Though they are sometimes omitted for reasons of economy, experience has shown that pilot surreys are a vital part of ensuring acceptable data quality. They can also help plan sample size and survey duration, thus ensuring the most effective use of available time and money. If the pilot is well prepared, and proves to be successful, the pilot survey data may become part of the main data set.

Equipment
2.11 Lists of equipment and materials are included with the individual survey instructions. The Supervisor must specify the equipment needed by each Surveyor; again, space is provided for this purpose on the Supervisor's field logsheet (see Appendix B). 2.12 Watches are used extensively in traffic surveys. It is recommended that digital chronometer watches which can display hour, minute, and second simultaneously be used for all timing tasks. (Stopwatches are not recommended as they are prone to operator error.) All watches should be individually numbered and each should be checked against a reliably accurate standard from time to time.

Liaison with other agencies

2.16 The approval of the police may be required for any activity on the highway, and their permanent presence may be necessary for some types of survey. However, as far as possible, the police must be made aware that there should be no unusual police presence or activity in the survey area which could affect the traffic characteristics being measured in the survey (for example, additional enforcement of speed limits).

Surveyor safety
2.17 Any work close to moving traffic has potential dangers. All survey staff should have suitable insurance cover, and each individual should be given a verbal briefing on traffic safety precautions as well as a copy of the safety card (fig 2.4) or similar. There may be legal obligations on the part of the survey organisers in respect of safety; these must be established and adhered to.

Survey forms and instructions

2.13 Survey form design should be simple, with enough space to record the data easily even under busy field conditions. Margins at left, right, and bottom of the form should be large enough to allow it to be clipped to a board, and filing holes to be punched, without obscuring information. Heading information should include the project title, survey type, and blank spaces for the Surveyor to enter the exact location (e.g. road name), a sketch plan showing the site and location of the surveyor (with measurements to an accuracy of 50m), date and time, Surveyor's name, weather conditions during the survey, and any other information concerning unusual traffic conditions. 2.14 Each Surveyor and Supervisor must have clear, detailed written instructions and training for each survey form and each survey task. This will include a detailed description of each individual task and how the survey form(s) should be filled in, with examples. It should be impressed upon them that every single item of header information must be completed on every sheet. To avoid ambiguity it should also be emphasised that no box or column is ever left blank; value zero is shown by the

DATA HANDLING
2.18 Data processing converts raw field data into a standard format, from which summary statistics, tables, and graphs can be prepared. Analysis is the process of drawing conclusions from this data. Both processing and analysis should begin as soon as possible, because potential errors are more easily identified and corrected if the surveyor concerned can be questioned. Also, additional surveys are more easily organised, if required. The main types of mistake are incorrect application of random or systematic sampling, measurement errors (misreading or misrecording, usually arising from Surveyors being required to record more information than can comfortably be achieved, use of ambiguous definitions or lack of adequate Surveyor training) and blunders (the result of direct human error). Measurement errors are often

Figure 2.4 Safety card

difficult to detect later. Serious blunders are usually detectable if proper data checking procedures are followed. However, every attempt should be made to avoid measurement errors and blunders from the start, as many errors cannot be identified once the data are collected. Processing and error checking 2.19 Data processing usually involves the transfer of raw data from the field sheets onto summary sheets or directly into computer files. This transcription process itself can be a major source of error and must be carried out with care. The transcriber should also remain alert to the possibility that the data contain errors, and report any observations which appear to be in error. With computer records it is easy to make specific error checks at the first stage of data processing. The most obvious check is for data to be entered twice, independently, and compared for differences. 2.20 The traffic engineer should look for unusual values or patterns in lists, tables, diagrams, and graphs. Discussion of unusual results with the Supervisors is often very productive. For example, a traffic volume histogram may show a large fluctuation from one 15 minute period to the next. There are no fixed rules, but an investigation could be carried out as follows: consider what reasons could have caused the fluctuation, for example, an effect such as that shown in the log. consider whether the Surveyor forgot to change time periods every 15 minutes, resulting in periods being longer or shorter than 15 minutes. The Surveyor should be questioned about this and the original survey forms examined. If this was the cause, the effect could be reduced by combining 15 minute periods into, say, one hour periods instead. compare existing data from other surreys, to see if the effect was present at adjacent locations at the same time; at the same location, at the same time; at the same location and time on different days. if no other data exist, carry out an additional (short)check survey at the same location and time of day. if no clear conclusion emerges, a judgement must be made on whether the data should be included in the analysis.

thorough, and include a visual assessment of a histogram of all the data. 2.21 Data processing resource needs (excluding analysis) vary considerably between different types of surveys. As a rule of thumb, processing time can take about twice the time of field data collection. Computers do not necessarily speed up procedures as much as might be expected, though direct entry of observations into a hand-held computer clearly has the promise of much reduced processing time. Analysis 2.22 Analysis seeks to draw conclusions relevant to the study objectives from the data characteristics and trends established in the surreys. Statistical techniques can be used to indicate, for example, whether there is a relationship between variables under examination or whether a significant difference has resulted from a particular road treatment. However, the analysis cannot, of itself, say anything about causes; why does one variable respond to a change in another or why a particular treatment has had some success? It is the job of the traffic engineer to design surveys which will yield data that can be used to test and quantify a hypothesis. To demonstrate, for example, that higher speeds result from a road widening the survey must be designed to screen out all possible alternative factors (different traffic flows, weather conditions, etc.). The use of control surrey data may be a powerful tool in this analytical process.

When an error is suspected in an individual measurement, it should be corrected or discarded. It should be corrected only if the correct value is known with reasonable certainty. It is important not to throw away data lust because it might be in error: stronger evidence is needed. In most studies a few extreme values will not materially affect the results. However, in studies interested mainly in extreme values (for example, observance of speed limits) the treatment of extreme values needs to be rather more

3 ROAD INVENTORY SURVEYS

3.1 The road infrastructure consists of links, junctions, parking spaces and terminals. The physical characteristics which will influence its use include its current geometrics and pavement structure, its traffic controls (signs, signals, road markings, and parking restrictions), sidewalks, shoulders, adjacent land use, service provision (for example, gas, water, electricity, telephones) and the intensity of non-traffic activities which encroach upon road space (for example, hawkers, builder's materials, market stalls etc.). The purpose of the inventory survey is to record this information. 3.2 The detail of an inventory survey can be varied to suit needs. Form B is adaptable to meet different levels of detail. An approximate, but simple survey method involves Surveyors walking or driving along a link, and locating objects or points by the distance (chainage) along the link and the off-set from the centre line. Chainage can be taken from a vehicle odometer or measured by pacing, or measuring wheel. As far as possible, inventory Surveyors should be restricted to observation and measurement, as opposed to making any judgements. Fig 3.1 shows an example of a completed survey form demonstrating the level of detail which can be recorded. 3.3 Junctions generally require a more detailed approach than links with, for example, measurements of corner radii and turning restrictions. Form C can be used, supplemented by additional material in sections X and Y. Traffic signal information should normally be collected in a form consistent with the equipment manufacturer's specifications. 3.4 The rate of progress of an inventory survey obviously depends upon the data items to be noted and the accuracy of measurements. However a simple link surrey for traffic planning purposes should progress at approximately 0.5 - 1.0 km/h. Link surreys can be carried out by pairs of Surveyors, one surveying to the left of the centreline, and the other to the right. 3.5 The field sheets themselves are the main output from the survey. They should be filed and referenced to a master map showing the area covered by each field sheet. Individual road links and junctions can be related to a network, by a system of link and node numbering. Existing link/node numbers should be used if these exist (for example for accident reporting or signal maintenance) as this will permit ease of cross-referencing with other survey material. Fig 3.2 shows an example in which the system chosen has 2-digit numbers for nodes, and 3 digits for links; this allows for extra nodes (Nos 51 & 53 for example) to be added if required.

PARKING SURVEYS

4.1 Parking surveys provide the data upon which the parking policy for an area can be decided. The provision of parking is obviously a major factor, primarily for private cars, in the accessibility of an area. Parking management is also a most effective low-cost traffic policy instrument 4.2 Car parking spaces can be classified into: on-street or offstreet; public (i.e. available to the public) or private; formal (i.e. marked and controlled spaces) or informal. Parking and stopping spaces must also be provided for commercial vehicles (primarily delivering and collecting freight) and public transport vehicles (for picking up and setting down passengers). Other characteristics of parking are: dimensions and layout (including access roads); time controls; charges and costs; banned and restricted locations.

4.3 Parking demand characteristics include: accumulation: the number of parked vehicles in an area, at any given moment. A graph showing the variation of accumulation in a city centre during the day, can be compared with the parking supply to show when there is over- or under-provision of parking space. parking duration: the time one vehicle remains parked in one place. parking load: the total demand on an area over a period of time, measured in vehicle-hours. It is the sum total of all vehicle durations, (equal to the area under the accumulation graph). parking volume: total number of vehicles using the parking facilities over a period of time (usually one day). turnover: rate of use of parking spaces, calculated by dividing the parking volume by the number of spaces. arrival and departure rates: which affect the design of entry and exit facilities, particularly for off-street car parks.

4.4 Two surveys are described here: parking inventory surveys, which determine the existing supply, by recording the number and location of spaces; parking patrol surreys which monitor demand and are usually for on-street (kerbside) parking but can be used for off-street facilities. The two are usually undertaken together.

10

Figure 3.1 An example of a completed link inventory form

11

Figure 3.2 Coding a Network (from Leonard and Gower 1982)

12

PARKING INVENTORY SURVEY

4.5 This survey requires a base map, upon which the surveyor marks the location and number of parking spaces. The map should be approximately 1:500 scale, depending upon the amount of detailed information to be recorded. A sketch map is perfectly adequate (Fig 4.1, for example), but officially published maps may be preferred. The parking inventory may also be incorporated into the road link and junction inventory maps. Where more than one form is required for a study area, a master reference map of the study area will be necessary, showing the area covered by each sketch sheet. 4.6 The location and number of formal parking spaces is usually determined easily because the spaces are marked. The number of informal parking spaces requires some judgement. Kerbside car parking spaces can be assumed to be 6 metres long and 2.5 metres wide. For informal areas, it is necessary to count the actual number of parked vehicles at a time of peak demand.

4.11 Routes should be defined to end at the start point, so that the surveyor wastes no time in returning to the starting point. A simple circuit may be defined by a Surveyor walking up one side of a street and back on the other. Routes are sub-divided into sections, typically of 20 spaces, to allow the identification of sections with different characteristics (for example duration, turnover). Section boundaries should occur where characteristics are expected to change, for example with change of adjacent land use from offices to shops. Data is then summarised by section.

Output
4.12 Vehicle arrivals, departures and accumulation can be calculated and entered into the boxes on the survey form itself and summarised on Form E. Parking duration is estimated by noting the patrol time at which each registration number was first seen, and the time at which it was last seen. The same result is obtained by counting the number of observations of a particular vehicle registration, and multiplying by the patrol interval. Note, however, that if the vehicle is present on the first or last patrol only a minimum duration can be calculated. Different vehicle types (or parking types) can be dealt with separately. Figs 4.2 and 4.3 show a worked example of a parking survey. 4.13 Bar charts can be drawn to summarise the parking duration pattern, if necessary by vehicle type. If examination of these indicates this is warranted, bar charts for adjacent sections can be combined.

PARKING PATROL SURVEY

4.7 Surveyors patrol a predetermined route, with a predetermined number of parking spaces. The patrol is divided into a number of sections. Surveyors record the location of all parked vehicles, their registration number, and vehicle type using Form D which can be modified to allow the type of parking to be recorded on the form, by providing an additional column alongside the vehicle registration number. 4.8 The area to be surveyed is decided from study objectives and the parking spaces within that area are identified from a parking inventory surrey. In addition to public spaces, the survey can include private, informal (e.g. waste ground), and illegal spaces (footpaths, no-parking areas), as required. 4.9 The patrol interval depends on whether parking at that location is short or long stay, but is typically 15 minutes. With low turnover, the patrol interval can be as long as 45 - 60 minutes; with high turnover 15 - 20 minutes is more appropriate. Preliminary data can be obtained from a pilot survey. The patrol interval should include an allowance of 5 minutes for rest and contingencies. 4.10 The length of route is determined by the patrol interval (less 5 minutes), together with the Surveyor's walk speed. Typical walk speed is 20 metres/minute, although it depends on the amount of data the surveyor has to record; high turnover means more data and a slower speed. If parking accumulation is required without associated duration, only the total number of each type of vehicle parked in each section needs to be recorded on each patrol. Therefore less data is collected, walk speeds are increased, and considerably longer routes are possible. Alternatively, data can be collected from a car, although a driver will be required.

13

Figure 4.1 An example of a Parking Inventory Survey

(From Wells, Traffic Engineering - An Introduction)

14

Figure 4.2 Parking patrol: example

15

Figure 4.3 Parking Patrol Survey form: Example

16

ORIGIN - DESTINATION SURVEYS: REGISTRATION NUMBER METHOD

number of Surveyors needed and to minimise the amount of data processing. The study area is defined by a cordon which should be drawn in such a way that vehicles entering or leaving the area cross the cordon only once. 5.7 Usually it is not possible to record the registration numbers of all vehicles, so in practice a subset of the total flow is chosen. It is important that the subset is chosen such that Surveyors at all points can record 100 per cent of the subset; it is much better to record 100 per cent of a small subset than, say, 85 per cent of a larger one. The proportion which the subset forms of the total flow must be known. 5.8 The most common method of choosing a sub-set is to use the last digit of the vehicle registration number. If any number between 0 and 9 is equally likely to occur, then the choice of any number automatically results in a 10 per cent subset. If two numbers are chosen, then a 20 per cent subset is selected. The last digit is chosen because it is most likely to be a random number. The method is simple, but requires every number plate to be scanned before deciding whether it has to be recorded. 5.9 Alternatively, vehicle colour can be used to distinguish the subset provided that the colour is not ambiguous; red is a poor choice, ranging from maroon to orange. Grey and white are often sufficiently unambiguous. (The pilot study should be used to check Surveyors' recognition of the colour.) Selection by colour can bias the results; for example, black is unambiguous, but there are few black cars in most cities and many of these may be official vehicles. Two (or more) colours may be used to provide a sufficient sample. If vehicle colour is used, the simultaneous volume count must also be classified by colour (as well as vehicle type) for expansion purposes. 5.10 To avoid any cause for doubt, it is recommended that the whole registration number (letters and numbers) should be recorded in full. Form F is used to record the information. It is also possible to record data on either a tape recorder or handheld microcomputer. For tape recorders the information is recorded verbally, doubling the rate of data acquisition because the surveyor does not take his eyes off the traffic. However, transcription can take longer than the survey itself, and other problems arise from equipment failure etc. Hand held microcomputers record time automatically, which if required makes recording twice as fast as using pencil and paper. If time is not to be recorded, then they do not have this advantage and are also subject to equipment failure problems. 5.11 An experienced Surveyor can record up to 300 registration numbers (of four numbers and three letters) per hour, using pencil and paper. The rate varies according to local conditions, the type of registration number, the size of the plate and numerals, etc. The rate will be lower with inexperienced personnel and if other information is recorded; for example, time of observation reduces the rate by more than 50 per cent and vehicle type, reduces the rate by about 20 per cent. A pilot survey will help establish work rates.

5.1 Origin-destination (O-D) data is used to analyze the effect of proposals for change (for example, a new traffic management scheme, or a new road) on travel through a study area. The O-D data is kept constant, but routes and journey times can be changed; the impact of the proposal on individual trips and cumulative volumes, travel times, and costs can be assessed. O-D data is usually presented as a matrix of trip volumes between each origin and destination. The origin and destination points are the start and end zones of a trip, but can also be the entry and exit points on the study area cordon. 5.2 O-D information can be derived from home or roadside interviews, postcard and sticker surveys and by registration number plate surveys. For long term transport planning studies, and large urban areas, where it is important to establish the social and economic characteristics of transport users, interview surveys are best. Registration number plate surveys obtain less information, but are easier to implement and are suitable for determining existing patterns of movement in small study areas. 5.3 For this manual, surveys dealing only with local area traffic schemes are considered. The Registration Number Method is an observation survey which matches the number plates of vehicles which enter and exit the defined study area, in order to accumulate an O-D matrix. If observers are scattered throughout the study area, as well as at the cordon, the method can also be used to obtain large samples of travel time data between observation points and to determine routing of individual vehicles through the study area. 5.4 The main problems with registration number surveys are that many reliable surveyors are needed, survey planning and coordination is complex and data processing and analysis are complicated and time-consuming (computerisation of the registration-matching process is almost essential).

Method 5.5 Surveyors record the registration numbers of vehicles inbound and out-bound to the study area, together with the time of observation and class of vehicle, if required. Two Surveyors are required at each observation point, one for each direction of traffic flow. Additional Surveyors can be located at intermediate points inside the study area, providing further information on routing and travel time. Each sighting of a vehicle provides more information on its travel time and routing, and possible stops inside the study area. 5.6 The study area should be as small as possible consistent with study objectives, in order to minimise the

17

Output 5.12 The survey begins at the same time at all locations, so that just after the start of the survey vehicles will be leaving the area which were not recorded entering it. Furthermore, just before the finish, vehicles enter the area which will not be recorded leaving it These vehicles should be ignored to obtain a clear picture of the proportion of through traffic The easiest way to remove most of them from the data set is to estimate the average time (`n' minutes) it would take to cross the survey area without stopping and to discard all registration numbers of vehicles which left or entered the area within `n' minutes of the start or finish of the survey. This is done before matching takes place. 5.13 Computer processing is recommended for matching all registration numbers between each entry and exit point pair, which have a positive travel time. The output file should list the entry point, exit point, registration number, (if used) vehicle type, entry time, and journey time. Assuming no errors, vehicles entering the study area will have been matched to an exit point, except for vehicles which either entered during the survey and stayed in the area until after the survey was finished or which were in the area when the survey started but left during the survey. These vehicles form part of the "stopping" traffic discussed below. 5.14 To identify stopping vehicles a histogram (Fig 5.1), is drawn of the travel times between each entry and exit point. The vehicles with the shorter travel times represent the non-stopping traffic, and the "tail" of longer travel times are for those vehicles which stopped in the area for some reason. It is necessary to decide the boundary between the two types. A simple rule is best employed, and applied to all cases: for example, using the first gap in the histogram longer than one minute. 5.15 For the whole area, non-stopping traffic can be expressed as a proportion of total traffic in the survey period, total traffic being all (non-stopping and stopping) matched vehicles plus the unmatched vehicles discussed. Knowing the sample proportion, it is simple to factor the sample data to give an estimate of the total flows of non-stopping traffic.

6.2 By definition AADT can be known definitively only from a continuous count over a full year. However, factors for modifying short period counts to estimate AADT can be developed using long period counts for a limited number of sites which are chosen to represent the main types of road in the network. The annual counts at the sample sites will indicate seasonal, daily and hourly variation and hence the factors which relate traffic volumes (measured at any specific time) to the AADT, for that class of road. 6.3 In urban areas peaks in traffic demand arise primarily because of home-to-work trips, occurring typically at the beginning and end of the normal working day. The `peak hour factor' expresses peak hour volume on a route as a proportion of the AADT. The peak hour factor for a particular route is often a consideration in classifying routes by type. As a rule of thumb, on urban roads the peak hour traffic is usually around 10 per cent of AADT (Note that the US Highway Capacity Manual defines peak hour factor in a different way - TRB, 1985). 6.4 Examples of presenting traffic volume data are shown in Figs 6.1 and 6.2. These include: volume maps, with a geographical background overlaid by bandwidths representing traffic volumes (see Fig 6.1 a); desire line diagrams, more often used for Transport Planning Studies, but which can also be useful in the planning stages of local area traffic surveys (Fig. 6.1 b); junction turning movements (see Fig 6.2). 6.5 Traffic volume surveys are carried out by either manual or automatic traffic counts. Manual counts are particularly useful for vehicle classification, checking automatic counter accuracy, and surveying vehicle occupancy. Automatic traffic counts, using traffic counter equipment, are normally used only on links, and are particularly suitable for long-term data collection, and analysis of seasonal, daily, and hourly variations. In most studies, a combination of automatic and manual counts is needed.

MANUAL TRAFFIC COUNTS

6.6 In a manual count a surveyor stands by the roadside, counting and classifying the vehicles as they pass, dividing the survey into fixed time periods. It is normal for the surveyor to record only one direction of flow. 6.7 Link counts should be located on straight sections of road for good visibility. Duration can be from a few minutes to several days, depending on purpose. Most counts are carried out for one day, starting before the morning peak hour, and extending for 12, 14 or 16 hours. Count periods are usually 15 minutes, with results summarised hourly. Shorter periods may be used for special purposes. Even when hourly counts are the most detailed data required, 15-minute periods should be used as errors are more easily identified (particularly start-up and closedown errors). 6.8 Turning movement counts are carried out in the same manner as link counts, except that the turning movement of each vehicle is recorded, and the vehicle

6 TRAFFIC VOLUME AND CLASSIFICATION SURVEYS

6.1 The purpose of these surveys is to collect data on the number and types of vehicles passing a specified point on a link (link counts), or making specified movements at a junction (turning counts). The occupancy of vehicles may also be recorded to provide data on the volumes of people using the roadspace. Volume of traffic is expressed as a rate of flow, usually either as vehicles per hour (veh/h), in particular the peak hour demand on the road, or vehicles per day (veh/day), often converted into the value "AADT" (Annual Average Daily Traffic).

18

Figure 5.1 Example of a Registration Plate Matching Survey Data Reduction

19

Figure 6.1 Presentation of Network Traffic Data

20

Figure 6.2 Presentation of Junction Count Surveys

21

classification system is simplified (to compensate for the extra demands on Surveyors). These surveys are primarily concerned with the performance of the junction during peak periods, and survey duration is often confined to the morning and afternoon peak periods, typically between 2 and 4 hours each. Count periods are usually 15 minutes. However, at signalised junctions short-term volumes are determined by signal operations, and cycle times in excess of 2 minutes, or count periods less than 15 minutes, may cause apparent fluctuations in flow. In such cases it is best to record flow for each cycle, using the cycle time as the counting period.

AUTOMATIC TRAFFIC COUNTS

Equipment 6.12. Automatic traffic counter equipment (Fig 6.4) consists of a detector, to detect the vehicle, and a counter to record the information. The most common types of detector used in traffic counting are a hollow rubber pneumatic tube, held to the road surface by special clips, or an inductance loop (several turns of wire laid on or in the road surface in a rectangular or diamond shape). 6.13 Pneumatic tubes detect the movement of a vehicle as the tube is depressed by a tyre; this creates a pulse of air which closes a diaphragm in the recorder adding one axle (usually half a vehicle) to the counter total. Tubes should be installed in accordance with the following guidelines: -the count site must not be located where braking or acceleration occurs, or where overtaking is common, such as near junctions, bus stops, or other parking locations. the count site must not be located near areas where children play or walk to school, as children are tempted to tamper with the counter or tube. at the count site the road should be straight, level, and free from flooding. the tube must be fixed at right angles to the direction of traffic flow. the tube must be straight when fixed, and it should be stretched some 10 per cent to ensure it remains straight. the tube must be clamped firmly to the road, with at least one clamp per lane of traffic, plus one near each kerb. the counter should be securely locked to a pole or tree, but the tube must not cross a footpath. the tube and counter must be checked frequently for damage or malfunction.

Survey forms 6.9 Survey forms must be designed for the specific needs of the study, but Form G is a good base. Typical examples of completed field forms are shown in Fig 6.3 which also illustrates the three common pencil-and-paper methods of recording and classifying each vehicle. The 'five-bar gate' method is the most versatile and is applicable to both link and turning counts. It records data quickly (without the need for extreme neatness), is the most economical in use of space on the survey form and can be summarised quickly. The 'initial letter' method is appropriate for surveys with short periods of high flow, when field manpower is limited. It has the fastest rate of data recording where one surveyor must record several items and is very fast for counts in only one direction, as the Surveyor's need to look down at the survey form is limited. It does require longer processing time than other methods, neatness to distinguish the symbols and more processing manpower. The 'crossing out numbers' method is best with low or medium continuous flow. It is the least flexible of the methods and has the slowest recording rate. However, there are no problems of Surveyors 'losing count', and totalling is immediate. 6.10 Alternatively, hand tallies can be used to record specific vehicle classes. They are faster and more accurate than pencil and paper methods, because only the total is recorded, at the end of each count period, and the surveyor is looking at the traffic flow almost continuously. However, vehicles can be missed while recording the cumulative total at the end of each count period. Note that tally counters should not be reset at the end of each time period as this wastes too much time, rather the cumulative total should be recorded at the end of each period. Tallies are also subject to mechanical failure and should be checked regularly; a common fault is not registering every vehicle. 6.11 Errors with manual counts usually arise from one of three sources: failure to define vehicle classification unambiguously, which can lead to undetected errors; failure to observe time periods accurately; surveyors having to count vehicles at a faster rate than they are capable. Particular attention should be given to these problems during the pilot survey.

Pneumatic tube detectors record axles, and the counter assumes that two axles equals one vehicle. This overestimates the number of vehicles, and a correction factor should be applied, calculated from classified count data (short sample count if necessary, but including both peak and off-peak data), as follows:
Correction Factor = Number of vehicles

x Number of axles on those vehicles A correction factor is calculated for a specific site, but various sites can be averaged to produce factors for an area, or different classifications of roads.

22

Figure 6.3 Alternative Methods of Traffic Counts

23

Figure 6.4 Automatic traffic counters

24

6.14 A loop detector detects the presence, rather than the movement of a vehicle. The counter passes an alternating electric current through the loop, creating a magnetic field above the loop. The presence of a metal vehicle in the magnetic field causes a change in the electrical inductance, which is recorded by a sensor in the counter. Non-metallic vehicles cannot be detected and pedal cycles are very difficult to detect because of their shape. Loops are usually permanent installations, laid in a shallow slot cut into the road surface, and sealed in by epoxy resin and bitumen. However, temporary loops can be fixed with sticky tape to the road surface. For loop wire specifications and installation, refer to the counter manual. 6.15 Counters are either non-storage or storage. The former accumulate a running total, which must be read manually on site but do not record data by time period. They have limited applications for measuring total volumes over a week, month, or longer. Storage counters store count data for specified time periods, for future retrieval. They are complex and expensive, but highly desirable in urban areas where hourly variations in traffic are of concern. Modern counters record data by electronic means, either on magnetic cassette tapes, or in solid-state memory. Data can be transcribed manually, but is more efficiently transcribed and summarised by computer. 6.16 A counter should be checked with at least 50 vehicles, including all major vehicle classes, over a range of speeds. This is necessary to ensure that all vehicle types which should be counted are being counted, and those not required (for example pedal cycles) are not being counted.

7 SPOT SPEED SURVEYS

7.1 The speed of vehicles can be measured instantaneously (spot speed), or averaged over distance or time. The spot speed of a vehicle varies continuously, as the vehicle accelerates or brakes. Spot speed data is used to: determine observance of, and suitability of, existing speed limits. establish suitable new speed limits. determine a suitable design speed for geometric design of the highway. provide information for use in road safety and enforcement programmes. assist the location of certain traffic signs. determine speed-flow relationships and traffic densities (May, 1990).

Combatting equipment errors

Spot speed surveys can also be used to establish trends (monitoring), for example in before and after studies. Spot speeds are usually measured on links (not at junctions) and are surveyed separately for each direction, with the surveyor normally positioned on the side of the road of the direction being surveyed. Spot-speed data is collected by either a radar speed gun (which gives automatic direct measurement) or short-base methods: timing a vehicle over a known short distance, either manually with a stopwatch or automatically using modern loop or twin-tube devices. 7.2 Radar speed guns are suited to relatively narrow roads at low or medium flows, when vehicles travel past the observer individually. They are not suited to heavy traffic volumes, congestion or multi-lane roads. Furthermore, they are complex to use, require significant training of survey staff, and are expensive. Methods where vehicles are timed over a short base line are suitable for almost all traffic conditions and types of road. They require only simple and inexpensive equipment, and are less obtrusive; the main problem is overcoming parallax error. This is reduced if a high vantage point is available (Fig 7.1). 7.3 The presence of surveyors, equipment, or unusual markings on the road surface can affect driver behaviour. The need to make the surveyors inconspicuous can affect choice of survey method and location.

6.17 Automatic traffic counters require a continuing programme of checking and maintenance, by trained technicians with access to a supply of spare parts. More sophisticated machines need more specialised maintenance without which they may prove to be of limited use. Poorly adjusted equipment gives rise to consistent over-or under-counting, while the sensitivity of the counter may deteriorate over time due, for example, to a partly blocked detector tube.

RADAR SPEED GUNS

7.4 The location of the survey, sampling of vehicles and recording of results, are exactly the same as for the manual short base method described below. The main requirements of the radar speed gun are that the operator is fully trained on the accurate use of the equipment and that the speed gun, and its operator, are concealed from

25

Figure 7.1 Speed Survey using Short-Base method

26

drivers. Measurements can be made from inside a parked car, but the car should not be parked in any location which affects the speed of the vehicles surveyed.

MANUAL SHORT-BASE METHOD

7.5 The survey location is usually at the middle of a road link. A specific point is chosen on the link, determined if appropriate by the study objectives (for example at a pedestrian crossing, to investigate an accident problem). A short-base length is created, over which vehicles can be timed. The length will depend on speeds on the road, with longer bases needed for higher speeds. Table 7.1 relates approximate lengths to average speeds.

This is the "sample vehicle". The Surveyor starts the stopwatch as the sample vehicle crosses the upstream start line, and stops it as the same vehicle crosses the downstream line. The time is recorded on the survey form, together with vehicle type and whether or not it was a following vehicle in a platoon. The procedure is repeated for the next vehicle, and so on through the survey period. Both timing and recording can be completed by one surveyor using Form J. 7.8 Unlike other types of survey, spot speed surveys are usually concerned with the non-peak periods of traffic flow, when speeds are higher. For example, where free-flow speeds are needed for setting speed limits, periods of low volume and good weather are specifically chosen. 7.9 Definition of the target population is particularly important for spot-speed surveys, and may be: all vehicles in the traffic-stream. all vehicles with some choice of speed: for example vehicles at the head of a platoon or single vehicles, on a fairly busy road. all vehicles with a free choice of speed: this would be at low flows, when a complete choice is available.

TABLE 7.1

Another approximate guide to length is that no vehicle in the traffic stream should take less than 2 seconds to traverse the short-base, in the traffic conditions prevailing during the survey. 7.6 The ends of the short-base length are marked on the road surface with paint, chalk, or tape lines; the lines should be as inconspicuous as possible to drivers. Alternatively, the downstream line can be defined by the surveyor standing directly opposite a roadside object (for example, a power pole or tree) on the opposite kerb. The Surveyor must always be at the downstream end. The short-base length must be measured accurately, preferably with a metal tape-measure rather than a measuring wheel. In addition a "sampling line" is marked upstream of the start line. The sampling line is needed so the surveyor can select the sample vehicle before he starts to record its travel time. The Surveyor must be able to see the sampling line and both timing lines, for all lanes of traffic; the pilot survey should determine whether a high vantage point is required. 7.7 Sample vehicles are selected at the "sampling line". The survey supervisor should define which vehicles are to be surveyed. This might be every nth vehicle or according to some other method to ensure a unbiased sample (For example, as the surveyor looks up he notes the first vehicle in any lane to cross the sampling line and selects the next vehicle in any lane to cross the sampling line.

7.10 Survey forms should allow space for a description of the chosen sampling method. It is not always necessary to include all vehicle classes; for example, cars alone are often the target population. If more than one class of vehicle is included, the class of each vehicle should be marked on the survey form. Output 7.11 Vehicle speeds are calculated from the times and known short-base distance. Results may be presented numerically (Table 7.2) or graphically. The most common graphical outputs are histograms (similar to Fig 5.1) and cumulative distributions of speeds which allow the extremes of the speed range to be seen. 7.12 Numerical results can be: mean speeds; the range of speeds; the proportion of vehicles above or below a certain speed (for road safety and enforcement). The 85th percentile is commonly used to describe speeds. This excludes extremely fast drivers (and gross measuring errors) and gives an estimate of what the majority of drivers consider a top limit.

27

TABLE 7.2
Example of a frequency distribution (for spot-speed data)

Class Boundary (km/h) lower upper middle

Frequency number relative

Cumulative Frequency number relative

27.5 29.5 31.5 33.5 35.5 37.5 39.5 41.5 43.5 45.5 47.5 49.5 51.5 53.5 Total =

29.5 31.5 33.5 35.5 37.5 39.5 41.5 43.5 45.5 47.5 49.5 51.5 53.5 55.5

28.5 20.5 32.5 34.5 36.5 38.5 40.5 42.5 44.5 46.5 48.5 50.5 52.5 54.5

0 1 2 14 7 20 38 29 35 15 12 9 4 0 186

0.000 0.005 0.011 0.075 0.038 0.108 0.204 0.156 0.188 0.081 0.065 0.048 0.022 0.000 1.000

0 1 3 17 24 44 82 111 146 161 173 182 186 186

0.000 0.005 0.016 0.092 0.129 0.237 0.441 0.597 0.785 0.866 0.930 0.979 1.000 1.000

28

NETWORK SPEEDS AND DELAY SURVEYS: THE FLOATING CAR METHOD

Number of runs, routes and timing points 8.6 To account for different traffic conditions throughout the day, each survey run should be related to a particular time period. Typically these periods are: morning and afternoon peaks, daytime off-peak, evening post-peak. A minimum of three runs is recommended for each time period. 8.7 Maximum route length capable of being covered in one run can be estimated from the duration of each time period being studied, the number of runs required and typical assumed speeds. For example, in a 2-hour period requiring 4 runs, the maximum route travel time would be 30 minutes; allowing (say) 10 minutes for turnaround and unexpected problems leaving 20 minutes travel time; route length at an assumed 15 km/h would be 5 km. 8.8 Route selection then depends upon covering the chosen network in the most efficient way. Survey links should be identified on a road network map of the study area, together with one-way streets and banned turns. The following points must be considered: most links are 2-way, and must be surveyed in both directions. circular routes are easier to operate; furthermore, U-turning vehicles can create problems of safety and delay, and it may be easier to have two vehicles circulating in opposite directions. in general, if more than one survey car is needed, it is better to use short routes with one car per route (per direction) than longer routes with more than one car per route. for before and after studies, the routes chosen for the before study must be repeated exactly, after the planned changes have been implemented.

8.1 Average network travel times and journey speeds are a major measure of road traffic performance. Not only are they an indication of existing road link and network performance, but they help identify specific congestion spots and are important as an input to traffic models and road investment appraisals. The basic method for measuring network speed and delay (and its causes) is the Floating Car Method. Other techniques which could be adapted to achieve the same purpose are the licence plate survey (see Chapter 4) and the elevated observer method (see Chapter 9). 8.2 The advantages of the floating car technique are its direct and accurate measurement of travel times and delays, and personal experience of the causes of delays. The main disadvantages are the large resources required for a comprehensive survey of a network, and problems of the survey vehicle exceeding the speed limit. The floating car method only surveys cars. Other vehicle types can be surveyed by following a selected vehicle. However, it is difficult to choose random vehicles to follow and the route of the vehicle is not known in advance. Safety may also be a problem, especially at high speed or in heavy traffic. 8.3 For a comprehensive study of a traffic network, surveyed links should include all the main road network, and various minor roads (Fig 8.1). As a monitoring technique for `before and after' surveys, specific turning movements should be identified for inclusion in the survey which represent `typical' movements through the site.

Method 8.4 The survey car is driven along a pre-determined route, at the typical speed of other cars. Surveyors in the car record the time at pre-determined timing points and the duration and cause of all stops and delays (see Fig 8.2). In addition, the distance between the timing points must be measured. If a good quality map exists (scale down to 1:2500) it can be used. Alternatively the distance must be measured on the street using a measuring wheel. 8.5 Pilot surveys should coverall survey links, at the same times of day as the full survey. This is to familiarise survey staff with the method and routes, and test the number of survey cars needed. Practice should also establish the driving style required to "float" or maintain one's position in the traffic stream, i.e. for the survey car to overtake the same number of vehicles as overtake it along the length of the route. The driver should not allow his driving style to change in response to the pressures of the survey.

8.9 Timing points should be located at every main road junction. Additional timing points should be added where there is a long distance between adjacent junctions, or where significant changes in characteristics occur (for example, road widening; a major traffic generator such as a market). There is little extra survey effort involved in adding extra timing points, and the extra information need not be processed if not required. Each timing point must be a specific, easily-identified, fixed point; a written description is required, with sketch map. At junctions, timing points will usually be the point of exit from the junction so that all delay associated with the approach to, and operation of, the junction is included with the approach link. Intermediate timing points are best located by roadside poles (for example, street light, telephone, road sign). To measure junction delay, it is best to have timing points just before a point which represents the longest queue length.

29

Figure 8.1 Typical routes for moving observer/journey time surveys

30

Figure 8.2 An example of a completed Journey Time/Moving Observer Delay Survey

31

Output 8.10 Data are usually presented by link (separately by direction), either as travel time or speed, and can then be aggregated for journeys or routes. Data from individual runs should be presented individually as they represent conditions at different times of day; simple averages can be calculated, but give biassed results, as more runs are made when speeds are high and delays few. If the average of runs within a period is needed for some special purpose, this can only be done by ensuring the despatcher has sent every car off (in that period) at equal intervals. To do this the despatching interval should be rather greater than the longest circuit time in the pilot survey at the most congested time. Travel times for each link or route can be compared by time of day, and may be compared with the corresponding traffic flow by time of day.

9.4 The elevated observer method is suitable for all kinds of junctions, where high vantage points are available, from which all approaches to the junction are visible (Fig 9.1). In addition to mean and total delay: distribution of delay can be calculated. delay can be estimated separately by turning movement. delays can be divided between the junction approach and the shared junction area. the cause of delay to each sample vehicle can be recorded (for example, junction controls, pedestrians, other stopped vehicle).

STOPPED VEHICLE COUNT SURVEY

9 JUNCTION DELAY SURVEYS

Method
9.1 Delay is extra journey time incurred when a vehicle is impeded, in excess of the journey time if not impeded. Most delays in an urban area arise at junctions or accesses. Junction delay is a measure of junction performance, usually presented in the form of average delay per vehicle. The data can be used to compare junctions and indicate those most in need of improvement (either in design or control) or to compare junction delay before and after an improvement. 9.2 There are two alternative survey techniques which are the Stopped Vehicle Count Method (counting the number of stationary queueing vehicles at fixed time intervals) and the Elevated Observer Method (timing sampled individual vehicles from a high vantage point). When comparing delay survey results (for example, in a before and after study), it is essential that the same survey method has been used. 9.5 Using Form L the number of stopped vehicles queueing on an approach to the function is counted at fixed intervals, usually 15 or 30 seconds, over a period of five or ten minutes. The total volume of traffic is counted at the same time. 9.6 At sites where the whole queue cannot be counted by one Surveyor, or where vehicle classification is required, the approach must be sub-divided. The distance between the stop line and a point at least 10 metres beyond the back of the longest queue is divided into a number of `boxes' such that each box is small enough for a Surveyor to be able to maintain a continuous count of the number of stopped vehicles in it. The front wheels determine the box the vehicle is in, except that the first and last boxes should include all vehicles in front of or behind the box, respectively (Fig 9.2). 9.7 A sketch plan should be prepared, showing for every approach: the number of approach lanes, and any special fixed features (for example, bus stops, major accesses). junction controls and signs. For signalised junctions, measure cycle time based on the average values over 5 cycles, and note whether left turn on red is allowed. whether delay is related to traffic movements on any other approach (for example, at roundabouts, and for opposed right turn vehicles at signals). the longest queue normally observed during the peak period. If queues extend back to the next major junction upstream then the `floating car method' or registration plate matching survey should be used instead.

9.3 The stopped vehicle count method: provides only total and average delay and cannot distinguish between delays for different turning movements. does not take account of delays other than on junction approaches; therefore the method is not generally suitable for junctions where a significant proportion of delay is within the junction area (for example roundabouts). relies on a clear determination of the number of stopped vehicles. Where the queue is "rolling" - as at roundabouts - the method is not reliable. However the method works well at traffic signals.

32

Figure 9.1 Delay Survey using Short-Base method

33

Figure 9.2 Delay Measurement by Stopped Vehicle method

34

9.8 Not more than three vehicle classes per Surveyor should be used, and the same classes should be used for delay and for flow. Non-motorised vehicles are not usually included. 9.9 An observation interval of between 10 and 30 seconds should be chosen (preferably an exact factor of 60 seconds), but must not be an exact factor of any regular signal cycle time occurring during the survey. The observation interval remains fixed throughout the survey and is the same for all approaches. Duration is usually 5 or 10 minutes at a time, which allows the different approaches to a junction to be surveyed in rotation. (All approaches must be surveyed simultaneously if delays on them are related.) An example of a completed survey sheet is shown in Fig 9.3. Output 9.10 Total delay should be presented for each approach and for each time period studied. A grand total for all approaches may also be given provided the data were all collected at the same time. The total delay (in vehicle hours) on the approach during the duration of the survey is calculated by multiplying the total number of queueing vehicles counted in the survey period, by the fixed time interval (converted into hours). The average delay is calculated by dividing the total delay by the traffic volume in the same period.

Output 9.13 The calculations of delay can be made separately by vehicle class, turning movement, or time period. The average undelayed travel time is calculated from the travel times of those vehicles identified as undelayed. For the vehicles identified as delayed, their individual delay time is calculated by subtracting the average undelayed time for that class, from the sample vehicle travel time. Mean delay is estimated by calculating the mean of all the sample vehicle delays. Total delay is estimated by multiplying mean delay by the traffic volume on that approach. The values for each approach can be added to give total junction delay in the survey period.

10 SATURATION FLOW
10.1 The maximum sustainable flow of traffic past a point is known as the saturation flow (or sat flow). It is a fundamental characteristic of a road and an important element of traffic engineering design studies, particularly related to traffic signal control and junction design. In order to compare flows of different vehicle mix, saturation flows are usually expressed in passenger car units (PCUs) in which vehicles are given a value equivalent to the number of cars that they displace from the traffic stream. 10.2 The measurement of saturation flow, PCU values and other signal-related issues, such as lost time, is difficult. For the purpose of this manual, two alternative simplified methods for estimating sat flows at a signalised junction are outlined. These are the Flow Profile Method and the Saturated Period Count. The flow profile method consists of dividing the signal green time into regular, short, intervals (typically 6 seconds) and counting the vehicles which cross the stopline during these intervals. More details of this method, including its application to lost time measurement, are given in RRL (1963). The Saturated Period Count requires the classified counting of vehicles passing in only one interval, the duration of this interval being chosen to ensure saturation. Examples of completed survey forms for both of these methods are given in Fig 10.1. 10.3 For both methods an inventory of the junction is completed during the pilot study, using Form C, and the movements to be counted are established. If the composition of the traffic stream is stable, and especially if more than 85 per cent are cars, the saturation flow may be measured using one `all-vehicle' category and then converted to PCUs by multiplying by the factor F, where

ELEVATED OBSERVER SURVEY

Method
9.11 An elevated location is found from where a Surveyor can see all the approaches to the junction (Fig 9.1). Individual vehicles are timed between fixed timing points on each approach. Vehicles are identified as either "delayed" or "notdelayed" and average travel times between the timing points determined for each category. The total and average delay is derived from the difference in travel times between the two sets, noted on Form M. 9.12 For each approach three points must be identified which are visible to the Surveyor. The finish timing point can be either the exit from the junction, as defined in the travel time surrey, or the stop line. The start timing point must be at least 10 metres beyond the longest anticipated queue and related to a fixed, easily identifiable object. The sampling point should be at least 20 metres in advance of the start point. A pilot survey should be used to test the procedure (which is similar in nature to the short base method for speeds-chapter 7) during the most congested time of the proposed survey period.

Where N is the number of each type of vehicle in the traffic stream, and PCU their passenger car unit value (see below).

35

Figure 9.3 An example of a completed Delay by Stopped Vehicle form

36

Figure 10.1 An example of completed Survey Sheets for Saturation Flow Surveys

37

If, as is often the case in developing cities, traffic composition is very variable from cycle to cycle, and no vehicle type is dominant, then this conversion cannot be used. Groups of vehicles having a common average PCU value should be recorded in this case. 10.4 Before the surveys, the number of signal cycles to be measured must be determined. As a general guide, 25 cycles should be enough at most sites, with up to 75 being required at key junctions in the network. For very quiet or unimportant sites, it may be sufficient to estimate saturation flow using previous studies. 10.5 Measurements should ideally be made at sites which are free of all obstructions; if any obstruction occurs within a cycle, the observations for that cycle should be discarded. If the obstruction persists (as happens in many cities) and is sufficiently common as to be a true representation of the traffic pattern at that site, the observations can be kept, but analyzed separately, recording the reasons. Some typical causes of obstruction are shown in the Supervisor's log (Form A). Long term obstructions should be indicated on the inventory form, completed during the pilot survey.

are carefully chosen and given full opportunity for familiarisation with the method. In some cases, the unorthodox behaviour of traffic at signals may require revision to the survey method, for example counting during the `red' period. 10.7 For unobstructed traffic conditions, the sat flow of each approach is simply the average flow for all saturated intervals (i.e. once the initial start-up period has been completed, and while the flow is still being supplied from a queue), as shown in Fig 10.2. This type of diagram can also help to analyse specific flow patterns, such as opposed turns. A plot of cumulative flow each cycle against time, gives a set of curves as shown in Fig 10.3. These can help to identify suspect measurements, and a representative value of sat flow (which is the gradient of these curves) can be determined by eye.

SATURATED PERIOD COUNT

10.8 This method gives less information than the flow profile method, but is included because it is simpler to use and has been found, in practice, to give results which are not very susceptible to enumerator error (Turner, 1993). This method consists of measuring the flow of traffic, during the entire period of saturation. Providing enough enumerators are available, this method is particularly suitable for the measurement of very large flows and/or very detailed classification. To allow saturation to develop, a suitable interval should pass between the start of green and the first vehicle to be counted. A lag of four vehicles is recommended for this, but a ten second gap is often easier to measure. Form N is used for recording the data. 10.9 The saturation flow is readily calculated by dividing the flow in each green period by its duration in hours.

10.6 The principle of the method is that the number of vehicles passing the stopline is counted at fixed short time intervals. Form N is used for recording. In addition to the points made above, the following apply specifically to the flow profile method. The length of time interval can be chosen to reflect local conditions: 6 seconds is commonly used, but if using a digital stopwatch, 5 seconds might be easier; 10 seconds may be used where queues are very long. A possible source of error arises because of the need to count (and sometimes classify) at a rapid rate. It is especially important for this survey that enumerators

TABLE 10.1

38

Figure 10.2 Using the Flow Profile method (from RRL, 1963)

A typical example of average discharge rate surveyed at a fully saturated junction

39

Figure 10.3 An example of a plot of Cumulative Saturation Flow

40

To allow for variations due to traffic conditions and enumerator error, all saturation flow measurements should be plotted as a frequency diagram (Fig 10.4) The choice of value to represent saturation flow is then a matter of judgement. For example, in Fig 10.4 the mean value would be 1800 PCUs, although it would appear that 1700 PCUs would be a more representative indicator of the current situation. For periods when interference by pedestrians is common a much lower value would be appropriate, whilst if pedestrians and buses could be controlled or segregated, then values of up to 2000 PCUs might be possible.

ROAD RESEARCH LABORATORY, 1963. A Method for Measuring Saturation Flow at Traffic Signals. Road Research Laboratory, Crowthorne. TRANSPORT RESEARCH BOARD, Highway Capacity Manual, Special Report No. 208, Washington, USA. TRRL OVERSEAS UNIT, 1987. Field survey techniques and analysis for urban bus operation. Overseas Road Note 4. Transport and Road Research Laboratory, Crowthorne, UK. TRRL OVERSEAS UNIT, 1990. Towards safer roads in developing countries. A guide for planning and engineers. Transport and Road Research Laboratory, Crowthorne, UK. Turner J, 1993. Simplified saturation flow data collection methods. TRL Project Report PR/OSC/003/93. Transport and Road Research Laboratory, Crowthorne, UK. Wells G, 1979. Traffic Engineering. An Introduction. Charles Griffin & Company Ltd., London & High Wycombe.

PCU VALUES
10.10 If a vehicle is introduced into a stream of ordinary passenger cars and in doing so displaces the equivalent of N cars, then that vehicle can be said to have a passenger car equivalent of N passenger car units (PCUs). This is not an easy phenomenon to measure, and results can vary according to the method used. As a result, more recently the PCU value of a vehicle has come to be better defined as the ratio of its average headway, in saturated flow conditions, to that for an average car. (Headway being the time between the passage of the rear axle of the vehicle in front and the rear axle of the vehicle of interest). 10.11 However, the headway ratio method can only be used where vehicles follow each other in lanes in an orderly fashion, and is therefore not applicable in many developing countries. 10.12 Because of differences in vehicle size and types, and driver behaviour, PCU values derived in Western studies cannot be directly used in developing cities. Indeed the practice of converting all vehicles to car equivalents in some developing cities, where cars may form less than 20 per cent of the traffic, is open to question. Results from previous studies are shown in Table 10.1, the 'suggested' values given in this table could be used in the absence of any other data, but only with caution.

11 REFERENCES
Gardner G, Jacobs GD, and PR Fouracre 1989. Traffic Management. TRRL Information Note. Transport and Road Research Laboratory, Crowthorne, UK. INSTITUTION OF HIGHWAYS AND TRANSPORTATION and THE DEPARTMENT OF TRANSPORT, 1987. Roads and Traffic in Urban Areas. HMSO. Leonard DR & P Gower 1982. User guide to CONTRAM version 4. Supplementary Report 735. Transport and Road Research Laboratory, Crowthorne, UK. May AD, 1990. Traffic Flow Fundamentals. Prentice Hall Inc., New Jersey.

41

Figure 10.4 Frequency Distribution Plot (Saturation Flow Survey)

42

APPENDIX A: SUMMARY STATISTICS AND SAMPLE SIZE

12.1 This appendix gives a brief outline of the techniques which traffic engineers need concerning the statistics of confidence limits and sample sizes. It begins with a description of the main statistics which are used to summarise data.

NOTATION
12.2 The notation used in this Appendix is:

usually "quartile") values. The values which split the set into 10 equally-sized parts are called "deciles", and the values which split the set into 100 equally sized parts are called "percentiles". Thus the median, the second quartile, the fifth decile and the fiftieth percentile are equal to each other in a given set of data. Percentiles are not widely used in all traffic analysis; but in some, particularly spot-speed studies, percentiles are calculated from the data to assist in determining speed limits. The 85'" percentile (the value separating the bottom 85 per cent from the top 15 per cent of the ordered data) is widely used in this context.

MEASURES OF THE SPREAD ("DISPERSION") OF A SET OF DATA

12.6 Most of the values in a data set may be close to the mean value (small dispersion) as in Fig 5.1, or they may be much more spread out (large dispersion) as in Fig 10.4. Thus "dispersion" means the amount the data are spread out.

The standard deviation

12.7 The standard deviation is the most commonly used measure of dispersion. It is of interest to Traffic Engineers partly because it can describe the variation of certain items of data, e.g. speed variations, but equally because the dispersion of a data set affects confidence limits and so sample sizes.

AVERAGES AND MEASURES OF CENTRAL TENDENCY

The arithmetic mean

The standard deviation ; 12.3 The arithmetic mean is sometimes called lust "the mean" or "the average". The arithmetic mean of a sample of n values This means, in words, that each data value in turn has the mean subtracted from it and the answer is squared. These n resulting values are then added. This sum is divided by (n-1) and the square root is taken of the result, to give the standard deviation.

The median
12.4 The median is the arithmetic mean of the two middle values of a set of numbers arranged in order of magnitude. For symmetrical distributions the mean is equal to the median. For distributions with a right hand tail, the mean is greater; for distributions with a left hand tail, the median is greater. The value of the median is less affected by extreme values than the mean. This makes it a useful measure of central tendency in cases were there is some doubt as to whether extreme values are valid or not. 12.8 When the standard deviation is calculated for the whole population rather than a sample, then we use n rather than n-1 as the denominator. Calculators usually give both. It is best for traffic engineers to use n-1 (as in the above equation) in all circumstances. The standard deviation has the same units as the data from which it is calculated.

The coefficient of variation

12.9 When it is necessary to compare dispersions, the standard deviation is often not suitable. An example might be in comparing the variation (dispersion) of travel time on two different routes. On route A, the standard deviation of travel time might be 5 minutes and on route B the standard deviation of travel time might be 10 minutes. But if the mean travel time of route B is 40 minutes, whereas that of route A, is only 20 minutes, this is not a fair comparison. In such circumstances a

Percentiles
12.5 If a set of data are arranged in order of magnitude, the middle value is the median. The median splits the set into two equally-sized parts. The data set can also be split in other ways. For example, the values which split the set into four equallysized parts, each containing 25 per cent of the set, are called the 25 percentile (or more

43

measure of relative dispersion, the coefficient of variation, is needed. The coefficient of variation, Thus, in the above example, if route B takes twice as long to travel on average as route A, it can be said the relative dispersion in travel times is the same on each route.

Where SEp = standard error of the proportion P =proportion, as calculated from the sample values This formula is appropriate if n is over about 30, but gives good estimates for smaller sample sizes.

CONFIDENCE LIMITS AND SAMPLE SIZES

12.10 When calculating a statistic (the mean or the standard deviation, for example) from a sample, the resulting value has some uncertainty associated with it because only part of the population has been measured. The sample value of the statistic is the best estimate of the true value for the population, but it is possible to calculate confidence limits about this sample value, between which the true (population) value can be expected to lie with a specified degree of certainty. To provide estimates of these confidence limits it is neces sary to calculate the "standard error" of the statistic. 12.11 In all cases, if the sample has been randomly selected, there is a probability of: * 68.2 per cent that the true (population) value of the statistic lies within confidence limits of plus or minus one standard error of the sample value. " 95.4 per cent that the true (population) value of the statistic lies within confidence limits of plus or minus two standard errors of the sample value. 99.7 per cent that the true (population) value of the statistic lies within confidence limits of plus or minus three standard errors of the sample value.

An example of estimating confidence limits of a sample mean

12.14 Suppose the speeds had been measured of a random sample of 100 cars. The mean speed of this sample was found to be 40km/h and its standard deviation 20km/h. To find the approximate range within which the true (population) mean speed can be expected to lie, with 95 per cent confidence, equation (1) is used to determine the standard error of the sample mean: SEm = 20/100 = 2.0km/h Thus the standard error is 2.0km/h and the sample mean is 40km/h. As there is close to a 95 per cent probability that the true mean lies within a range of plus or minus two standard errors, the confidence limits are:
mean speed = 40km/h 4 km/h with 95 per cent confidence.

An example of estimating confidence limits of a proportion

12.15 Suppose the proportion of vehicles exceeding a speed limit has been measured for a random sample of 100 vehicles. It was found from this sample that 0.3 (30 per cent) exceeded the limit. To find the range within which the true (population) proportion can be expected to lie, with 95 per cent confidence, equation (2) is used to determine the standard error of the sample estimate of the proportion: SEp (0.3)(0.7)/100 = 0.046

These probabilities are also fairly correct if the sample is not strictly random, but has been chosen with no bias.

Standard error of the mean

12.12

Thus the standard error is 0.046 (4.6 per cent) and the sample proportion is 0.3 (30 per cent). As there is approximately a 95 per cent probability that the true proportion lies within a range of plus or minus two standard errors, the confidence limits can be stated as:' The proportion of vehicles exceeding the speed limit is 0.3 0.09, with 95 per cent confidence" or "The percentage of vehicles exceeding the speed limit is 30 per cent 9 per cent, with 95 per cent confidence".

Using standard error formulae for estimating sample size

12.16 Equations (1) and (2) can be used to determine the required sample size (n) by rearranging them mathematically to isolate n. Thus if a mean value is needed:

44

and if a proportion is needed:

APPENDIX B: GENERAL CHECKLIST FOR SURVEY SUPERVISORS

13.1 The Survey Supervisor plays a key role in the organisation and performance of any traffic survey. There are tasks which are common to all types of survey, which can be set out as follows: 13.2 Before the survey:

Equations (3) and (4) can the be used to calculate n, as shown in the examples in paragraphs 12.17 and 12.18 below

An example of estimating the sample size needed to estimate a mean value with a stated degree of accuracy.
12.17 Suppose an estimate is needed of the mean spot speed of cars at a particular location, to an accuracy of plus or minus 10 km/h with about 68 per cent confidence. It is known from a pilot survey, that the standard deviation of car speeds is about 40 km/h. To determine the sample size of car speeds, equation (1) is used with (n) as the unknown:

establish survey station layout, and any necessary network reference numbers divide study area into survey areas as appropriate determine required number of Surveyors. arrange transport, pick-up times and locations, with adequate travel time to the site. ensure adequate supplies of materials and equipment. organise vehicles and drivers for use in mobile surveys. organise, where appropriate, route maps with all road names clearly marked showing survey route and any supervise picking up of Surveyors and travel to site. at the site, deploy Surveyors (duties and locations). hand out materials and equipment. synchronise watches, and synchronise the start of the survey, possibly using a whistle or alarm clock.

The estimated minimum sample size needed is therefore 16. If the mean speed was needed to the same accuracy (10 km/h) with 95 percent confidence, the 10 km/h must now equal two standard errors, so the standard error will be 5 km/h, and the calculation would be:

The estimated minimum sample size has risen to 64, because the range of acceptance now equals two standard errors; i.e. one standard error is 5 km/h.

An example of estimating the sample size needed to estimate a proportion with a stated degree of accuracy
12.18 Suppose an estimate is needed of the proportion of vehicles with a choice, which cross traffic signals when they are red, to an accuracy of 0.1 (10 per cent) with 95 per cent confidence. Previous studies have suggested the proportion is likely to be about 0.4 (40 per cent) (Note: if there is no prior information on the proportion, it is best to assume 0.5). To determine the required sample size, equation (2) is used; confidence limits at close to the 95 per cent level are set by plus or minus two standard errors and this means that 0.1 must equal 2 standard errors, so one standard error equals 0.05.

13.3 During the survey: check timekeeping. check accuracy of observations by Surveyors. synchronise time period changes, possibly using a whistle or alarm clock. collect completed survey forms, making certain that headers have been correctly filled out. ensure an adequate supply of blank forms and stationery

The estimated minimum sample size is 96 randomly selected vehicles.

13.4 At the end of the survey the Supervisor should ensure that all survey forms are collected and are properly labelled. Data checking and correction should take place as soon as possible after the survey, by the same team of Surveyors used for the survey itself. 13.5 The Supervisor should keep a record of the survey work using Form A as a log.

45

APPENDIX C: INSTRUCTIONS FOR SURVEYORS

LINK INVENTORY
14.1 Form B is used for recording link inventory data. The procedures to be adopted during the survey are as follows: Record field data sheet heading details; in particular orientate the field sheet by recording the street name, and junctions with other streets. Commence walking or driving along the link. Measure the chainage by pacing, vehicle odometer, or measuring wheel.

Horizontal curvature may be graded as: low (up to 30 deg.); medium (30 to 90 deg.); high (more than 90 deg). Gradients and vertical curvature may be similarly grouped. 14.3 Activities on the footpath or in the carriageway (for example street vendors, food stalls) can reduce the effective road width. Surveyors must measure the nominal road width and also record the area occupied by these other activities. This should be done even if the activity occurs only at certain times of day.

JUNCTION INVENTORY
14.4 Form C is used. The Surveyor should record the sheet heading details; in particular he should orientate the field sheet by sketching the junction and showing all the street names and a north point arrow. Information is measured and recorded for each approach road as for the link survey, but special note should also be made of: junction type (uncontrolled/priority/signals/rounda bout) signal characteristics (timing if fixed, phasing, controller type) width of approach kerb entry and exit radii (calculated from the target distance and deflection angle assuming they are circular curves) number of lanes, and lane widths. visibility splays

The information to be recorded will be specified by the Supervisor, but will include: Cross-section dimensions (road, footpath/hard shoulder, any height or width restrictions, right of way) Traffic management controls (1-way streets, banned turns, pedestrian facilities, bus, taxi and service vehicle bays) Street furniture (road signs and markings including lane widths, speed restrictions and details of parking controls, street lights, bus stops, crash and pedestrian barriers ) Service vehicle and private accesses

Frontage land use (residential, educational, shop, commercial, etc.) Road side activities which take up footpath or road space (area and type of activity.) -

14.5 Traffic signal equipment and settings require special attention and it is recommended that Surveyors with some knowledge of signal operations are used to undertake this work.

Crossfall or superelevation

Vertical curvature, gradients and horizontal curvature Pavement construction type and surface condition.

PARKING INVENTORY
14.6 Each surveyor is allocated a route through the survey area which covers all possible parking locations. As he walks the route he records on a map of the route all locations where parking is restricted or banned. For all other locations he estimates the parking capacity and records these also. Where parking spaces are marked out, the spaces can be simply summed. Where no formal markings exist estimates can be based on the following: kerbside parking spaces of 6.0 x 2.5 metres, excluding accesses and bus stops, etc. for informal areas, particularly off-street, allow 2530 square metres per car. This allows space for circulation and access/egress.

14.2 Some of this information may be classified into a simple rating scale, which the supervisor will explain prior to the survey. For example, a five point scale for surface condition might be: 1- vehicles forced to slow and drive around potholes. 2- some rough patches, poor wet weather grip 3- some worn patches without significant effect on traffic. 4- good surface 5- perfect flat surface with good wet weather grip.

46

PARKING PATROL SURVEYS

14.7 The Surveyor should familiarise himself with the survey start point, the frequency of patrol and the route. The route is divided into sections and a new copy of Form D is used for each section. After filling in the survey form headings, the patrol is started at the exact start time. For each vehicle occupying a parking space, the following is recorded: the full registration number vehicle type

the full registration number exact time of crossing the survey line (if journey times are also required, otherwise only approximate time is required). vehicle type.

14.11 If traffic is bunched, only the registration number should be recorded, to minimise the chances of missing vehicles in the chosen subset. The time can be estimated later by interpolation, provided a time is recorded for the first and last vehicle in a bunch. 14.12 Surveyors should be located where all registration numbers in all traffic lanes can be seen. Footbridges or other raised points can be advantageous, depending on their height, and the size and legibility of registration plate characters. Surveyors should not be located: just downstream from signals, as traffic will be bunched. just upstream from signals, as travel times will be distorted by junction delays.

For formally marked spaces which are empty, the box on the survey form can be left blank. However, for informal (unmarked) spaces this may not be practical; in this case the registration numbers of parked vehicles should be simply listed on the form, without blanks. 14.8 The Surveyor should patrol at approximately the same walk speed even when not busy recording data. It is not desirable for the Surveyor to hurry round in periods of low demand, causing a long wait before departing on his next circuit at the appointed time. At the end of the patrol, the Surveyor waits until the scheduled start time of his next patrol. 14.9 Vehicles entering parking spaces should be recorded, but not vehicles leaving parking spaces. Vehicles not clearly in a space should be treated consistently; for example, using the position of the front wheels to indicate which space is in use. Special problems may arise with some vehicle types: -large numbers of motorcycles may be parked at a particular spot and recording all registrations would be extremely time-consuming. bicycles may cause a similar problem, but they do not usually have registration numbers public transport or freight vehicles (for example bicycle rickshaws, motorcycle taxis, minibuses, pickups) waiting for custom may also take up parking space, and create problems.

14.13 It is important that none of the vehicles in the subset are missed. Surveyors who know they have missed or incorrectly recorded vehicles should be encouraged to say so; this enables the problem to be corrected in future surveys. Methods of correction may include: using a separate Surveyor for each sample digit. using a separate Surveyor for each vehicle class. having one Surveyor looking and shouting, and one recording, so traffic flow is continually scanned.

If the Surveyor has any doubts about the data he is recording, he should put a question mark in brackets.

MANUAL TRAFFIC COUNTS

14.14 Form G is used in this survey. Surveyors should adopt the following procedures: Identify a suitable marker (for example a telephone pole or road sign) on the other side of the road which he can stand or sit directly opposite Record each vehicle that crosses an imaginary line between himself and the marker, when it crosses the line (not before or after), either directly on the survey form (in the correct box), or on a tally. At the start of the survey (usually on the hour) commence a new survey form, and fill in the headings. At the end of each count period commence a new form, and file the completed form.

In the case of the first two examples, only the total number of each vehicle type within the area need be recorded. Public transport and freight vehicles can either be ignored or recorded, depending on the survey objectives.

REGISTRATION NUMBER SURVEY

14.10 This survey uses Form F. Before starting the survey, the Supervisor will confirm the sampling method to be used. After filling out the header information each surveyor scans the front registration plates of all vehicles coming towards him in one direction. When a vehicle is seen bearing a registration number ending in one of the sample numbers, the following information is entered:

47

14.15 Hand tallies and stopwatches should never be reset to zero at the end of a count period, because it is a common error to forget to write the total down, thus losing the data. Cumulative values should be recorded on Form H. 14.16 Surveyors must never 'invent' data, particularly at the busiest times. If mistakes or omissions are made, they should be noted and described.

ensures that the route inventory, chronometer, field forms, and stationery are prepared, and that the header information on each form is filled m. at the word "NOW", records the chronometer time precisely on the field survey form (to the nearest second, using the 24 hour clock) against the appropriate timing point. records durations of, and reasons for, delays called out by Surveyor 1. A separate line is used for every delay.

SPOT SPEED SURVEY

14.17 The supervisor must identify the short-base reference points; these are the sampling line, the timing start line, and the finish line. The vehicle types to be surveyed must also be identified. Form J is used for recording data. After filling out the headings on the survey form, the stopwatch should be zeroed and any vehicle which meets the selection criteria identified at the sample line (see paragraph 7.7); the stopwatch is started as the vehicle crosses the start line, and stopped as it crosses the stop line. The time is recorded, and the procedure repeated.

Driver: learns the route drives normally, basically "floating" in the traffic flow, trying to overtake as many vehicles as overtake him (cars turning off the route should be ignored), but driving safely. obeys instructions from Surveyor 1.

FLOATING CAR SURVEY

14.18 Form K is used in this survey. Each survey car contains three persons: two Surveyors and the driver. Their roles are as follows: Surveyor 1: controls the survey. In particular the driver must obey instructions concerning speed and manoeuvres. must ensure that he is familiar with the route and the timing points. determines when everyone is ready. instructs the driver to start. The car must enter the traffic stream before the start of the timed route. calls each timing point (in the manner: "timing point 7 .... NOW"). marks each timing point, by calling the moment of passing. times the delay if the car is stopped, and calls out duration (in seconds) and cause to be recorded by Surveyor 2. ends the run after the vehicle passes the last timing point.

14.19 If more than one car is used on a route, the cars should be spaced out. Timing points must be defined exactly. For example, if a power pole is a timing point, the time is called as the car crosses an imaginary line drawn from the pole, at right angles across the road. The exact time of passing the timing point can be related to the front of the car, the rear of the car or the surveyor. This must remain the same throughout a survey run. 14.20 Delay time is simply defined as any time lost when the vehicle is slowed down below the typical running speed of vehicles on that street. However such judgements require experienced surveyors. With less experienced surveyors, more precise definitions may be required; for example, if the vehicle is travelling at less than walking speed or stopped time is when the tyre treads of the vehicle in front can be seen. Sometimes it is not possible to see the cause of delay, and this should be noted as "unclear".

JUNCTION DELAY: STOPPED VEHICLE COUNT METHOD

14.21 Three surveyors are required for each junction arm and Form L is used for data recording. The role of each person is:

Surveyor 1: keeps a count of the number of stopped vehicles, by vehicle class. This is best done aloud, e.g. "7,6,1" could be seven motorcycles, six light vehicles and one heavy vehicle. Vehicles moving, however slowly, are not included in this count.

Surveyor 2: records all the field data, and controls all the equipment except the stopwatch (and the vehicle).

Surveyor 2: controls the timing by calling out the start, and then the timing intervals with a 10-second warning and a

48

five-second countdown ("10 seconds to go, 5-4-3-2-1NOW"). records the vehicle count called out by Surveyor 1 at each "NOW" point. fills in the surrey form headings.

FLOW PROFILE COUNT

14.26 Form N is used for data recording. Junction turning counts, if required separately, are collected using a variant of Form G. The survey involves three Surveyors whose tasks are as follows: Surveyor 1:

Surveyor 3: counts all vehicles, from the start to end of the survey, by vehicle class if required. separates these into "stopping" and "non-stopping" if possible.

starts his watch when the green light appears and shouts 'GO' identifies the last queuing vehicle when the lights turned green. shouts "one" at end of first interval," two" at the end of the second interval etc. shouts "last" if the last queuing vehicle passes during the green period , otherwise... shouts "stop" when green light disappears and stops watch.

14.22 Any vehicle which stops for reasons other than queuing must not be counted (for example, buses or taxis picking up passengers, parked vehicles). Also, other vehicles impeded by these vehicles must not be included. 14.23 Defining the number of queuing vehicles can be uncertain. It can be the number of stationary vehicles, or it can include, in addition, vehicles still moving but nevertheless delayed. The difficulty lies in defining the speed up to which a vehicle can travel while `queuing'. In general, only stopped vehicles are counted, although this results in total delay being underestimated. However, this is generally not important as delay measurements are usually used for comparison. 14.24 The number of stopped vehicles can change very quickly. For example, there may be many stopped vehicles just before the instant of recording, but at the exact instant all may have started to move slowly. The correct number to record is those vehicles stopped at the exact instant of recording. Surveyors, having seen the stopped vehicles, tend to record that number, even if the vehicles have all just started moving; the Supervisor must ensure that this does not happen. There are likely to be many zero observations of stopped vehicles.

Surveyor 2: as the rear wheels of each vehicle passes over the stopline he shouts the initial letter for the appropriate category. if measuring saturation flow only, he stops counting when Surveyor 1 shouts "last" or "finish" if measuring a complete flow profile ,or a full count of all vehicles, he keeps shouting for every vehicle passing.

Surveyor 3: writes down the initials of each vehicle as it is shouted out by the Surveyor 2. when Surveyor 1 shouts `one, two' etc., he starts recording in the next column on the survey form. When Surveyor 2 shouts 'last' stops recording. (If required for lost time calculations; when Surveyor 2 shouts last, he writes an elongated `L' in the appropriate column and when Surveyor 1 shouts `stop' he writes an elongated `S' in the appropriate column. See RRL, 1963 for more details) records anything which might have affected the measurement in the notes column. if vehicles pass during more than the 8 intervals allowed for on the survey form, go on to the next line, making sure to include a 'C for 'continuation' in column.

JUNCTION DELAY SURVEY: ELEVATED OBSERVER METHOD

14.25 Observations are recorded on Form M. The Surveyor selects a sample vehicle at the sampling line taking care to ensure that the selection of vehicles is done without bias, as described in paragraph 7.7; No element of personal choice should enter the sampling procedure. This procedure is repeated for the next sample vehicle and so on for the period of the survey. The vehicle is timed from the upstream timing point to the downstream timing point, using a stopwatch. The information he records is:

vehicle type. stopwatch time. whether the vehicle was delayed. time of day.

49

SATURATED PERIOD COUNT

14.27 Form P is used to record data. Junction turning counts, if required separately, are collected using Form G. The survey involves a minimum of two Surveyors and probably more. The two most numerous categories of vehicle are assigned to the initial two Surveyors; whether the remaining categories can be counted by one person, or whether more staff are required, must be judged by the Supervisor. He must also collect any special information required. In particular, the presence of parking or other activities in the kerbside lane for 100 metres upstream and 50 metres downstream of the junction should be noted for each cycle in the appropriate column. Note should be made of any event which has any influence whatsoever on traffic flow during each cycle in the 'notes' column on the survey sheet. 14.28 The tasks of the Surveyors are as follows: Surveyor 1: starts the stopwatch when the green light appears. looks to the back of the queue and identifies the last vehicle to join the queue when the lights changed. This is known as the 'last queuing vehicle'. shouts 'now' when the stopwatch shows ten seconds. shouts 'stop' when the last queuing vehicle passes the stopline (or when the green light disappears), and stops the stopwatch. records in the appropriate place on the form, the clock time (column 1), the elapsed time (col 6) and the number of vehicles in each category (col 7). makes a note of any event which may have affected traffic flow during the cycle or the timing of it, or of any comments made by any of the counters (col 7).

Surveyor 2: counts all vehicles of type C (or as instructed by the Supervisor) as their rear wheels pass over the stopline, from when Surveyor 1 shouts 'now' until he shouts 'stop'. gives the number of vehicles counted to Surveyor 1. informs the Surveyor 1 of anything which affects the traffic flow being measured. if Surveyor 2 loses count, he should stop counting and tell the Surveyor 1 There must be no guessing.

If other Surveyors are being used they will do the same as Surveyor 2, but count other vehicle types.

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

Urban Road Traffic Surveys

ORN 11

Old Wokingham Road Crowthorne Berkshire RG11 6AU

Overseas Development Administration 94 Victoria Street London SW1E 5JL

OVERSEAS ROAD NOTE 12

DESIGN GUIDELINES FOR BUSWAY TRANSIT

Main Subject area: Urbanisation/Transport, energy efficiency Theme: Urban Transport Project title: Design guidelines for busway transit Project reference: ODA 85596 This document is an output from an ODA-funded research project, carried out for the benefit of developing countries.

Crown copyright 1993. The views expressed in this report are not necessarily those of the Overseas Development Administration or the Department of Transport. Overseas Centre, TRL, 1993 ISSN 0951-8987

ACKNOWLEDGEMENTS This Note was prepared by Dr. P R Cornwell and J A Cracknell of Traffic and Transport Consultants Ltd., and G Gardner of the Overseas Resource Centre, Transport Research Laboratory (TRL). It has been edited for the ORN series by Dr. J C Rutter and P R Fouracre of the Overseas Resource Centre, TRL.

First published 1993

OVERSEAS ROAD NOTES Overseas Road Notes are prepared principally for road and transport authorities in countries receiving technical assistance from the British Government. A limited number of copies is available to other organisations and to individuals with an interest in roads overseas, and may be obtained from: Transport Research Laboratory Crowthorne, Berkshire, RG11 6AU United Kingdom Crown Copyright 1993 Limited extracts from the text may be reproduced provided the source is acknowledged. For more extensive reproduction, please write to Programme Director, Overseas Resource Centre, Transport Research Laboratory.

CONTENTS
Page 1. Introduction Purpose and scope of this note Bus priority measures Spot priorities Bus lanes Busways Busway transit Special operational measures Busway transit performance The case for busway transit 2. Planning considerations Planning context Allocating roadspace Feasibility and warrants Capacity concepts System planning parameters Regulatory framework Organisation and management 3. Track design Track options Means of separation Typical cross-sections Horizontal and vertical alignment Intersections Layouts Signal control Collector/distributor systems Guidance systems Other aspects 1 1 1 1 2 2 2 2 5 5 7 7 8 8 9 10 11 11 13 13 13 22 22 23 23 23 25 25 26 6. 5. Evaluation Definition of options Busway transit costs Capital costs Operating costs The impact of busway transit The users Transport operators Non-users Urban development Other impacts Economic evaluation References and bibliography 4. Bus stops Bus stop design Capacity Design considerations Layouts Bus stop spacing Bus stop location Page 27 27 27 27 27 34 34 34 34 34 34 35 35 35 36 36 36 37 37 37

DESIGN GUIDELINES FOR BUSWAY TRANSIT

1. INTRODUCTION
PURPOSE AND SCOPE OF THIS NOTE The purpose of these design guidelines is to assemble current ideas on busway transit and to offer transport planners and designers a source of practical information on their use and implementation. Thus this Note is primarily directed at practitioners rather than policy makers. The focus of the content is on busway transit rather than general bus priority measures. The note complements a TRL report on busway transit performance (Gardner et al, 1991) and contains examples of current practice, derived from published information and from observation of operational schemes around the world. 12 Busway transit is a true urban mass transit option, which bears comparison with the fixed rail systems of Light Rapid Transit (LRT) and metros. Busway transit, the physical segregation of bus and other traffic, offers the possibility of introducing a mass transit system at relatively low cost. It is important to distinguish busway transit from other bus priority measures which are more limited in their scope. In order to make this distinction some initial reference is made to other bus priority measures. The remainder of the Note then examines the components that make up busway transit: planning considerations, track requirements and bus stop design. The Note concludes with an outline to the appraisal of busway transit.

BUS PRIORITY MEASURES

1.3 Buses are one of the most space-efficient and cost-effective means of transporting large numbers of people. Where traffic flows are well below the capacity of the road network, buses can share roadspace with other traffic and, in general, there is little need for special priorities for buses. However, where road traffic volumes are high in relation to road capacity, buses suffer from the congestion and delays caused by other road users, and priorities are needed to release buses from traffic congestion. There are three main ways in which this can be achieved, which are: spot priorities, bus lanes and busways.

Spot priorities
1.4 Most bus delays occur at bus stops and junctions, rather than along running sections. Junctionrelated delays can be dealt with by spot priorities, examples of which are turn-ban exemptions and bus gates. Turn-ban exemptions permit buses to turn out of a particular road, where this movement is banned to other traffic. Bus gates permit buses to turn into a particular road, where this movement is banned to other traffic (Plate 1). However, while

Plate 1 A bus gate providing buses with unhindered access to main road: Hong Kong
1

spot priorities are a useful traffic management measure, they cannot by themselves improve bus performance over whole routes -

dissuade but physically permit crossing by both buses and general traffic. a_Buswav involves construction where schemes may be partially physically segregated from other traffic, for example in the vicinity of bus stops (e.g. by means of island stops) or may be fully segregated from other traffic by kerbs or fences.

Bus lanes
1.5 Bus lanes are road lanes reserved for the use of buses only. Short bus lane sections at junction approaches can allow buses to "queue-jump" and bus-activated traffic signal pre-emption can reduce delays. Bus lanes can also be used to give buses priority over long sections, provided they are respected by all road users. There are two main types of bus lane: with-flow and contraflow (Plates 2 and 3 respectively). Withflow bus lanes are employed extensively, but in environments where road user discipline is poor, with-flow lanes tend to be violated by other vehicles and are relatively ineffective. In contrast, contraflow bus lanes tend to be self-enforcing, since buses travel in the opposite direction to other vehicles. However, there are some indications that pedestrian/bus accident rates may be higher along contraflow than along withflow bus lanes. A traffic scheme may include both with-flow and contraflow lanes, as well as spot priorities. Although one lane is usually provided in each direction for buses, two lanes may be provided where bus volumes are high, at busy bus stops (to allow buses to overtake one another) or on long uphill sections (e g. as in Belo Horizonte, Brazil).

BUSWAY TRANSIT Special operational measures

1.8 A basic busway, comprising one lane for buses in each direction is essentially a traffic engineering measure. However, performance of this basic busway can be enhanced substantially by adopting various "special operational measures" in order to form a "busway transit system" (Table 1). Where passenger demands are high, the provision of facilities to permit buses to overtake one another at bus stops can increase throughput and commercial speed considerably. This is because bus congestion is reduced and buses are no longer "trapped" behind one another in a single lane (as occurs with trams or light rail vehicles). Plate 4 shows the Avenida 9 de Julho busway in Sao Paulo, which has bus overtaking facilities at all but two stops over an 8 km length and achieves high performance (Gardner et al, 1991).

1.9

1.6

Busways
1.7 The traffic violations experienced by with-flow bus lanes can be overcome by physically segregating buses from other traffic by means of studs, kerbs or fences. In this report, the distinction is made between a bus lane and a busway as follows: a Bus Lane is essentially a "paint-and-sign" scheme where buses are separated from other traffic by road markings or separators, which

1.10 Trunk-and-feeder operations also offer good performance. In this system, feeder buses collect passengers and bring them to a transfer terminal, where they transfer to line-haul buses; some systems allow transfer without payment of an additional fare. 1.11 Early work in Brazil led to the development of a high-capacity bus convoy scheme (COMONOR), in which buses were assembled at the beginning of a section in the order in which they would stop along the route (to form the on-street `train'). Although not

Table 1: Special Operational Measures

Busway Transit = Busway Infrastructure + Special Operational Measures

Special Operational Measures Include: bus overtaking facilities at stops; trunk-and-feeder operations; bus ordering (placing buses in the correct order at the beginning of the section); high-capacity buses (e.g. articulated or double-deck) off-board ticketing; traffic signal techniques to give buses priority at intersections; bus dwell time management (to eliminate excessive delays at very busy bus stops); and guidance systems (e.g. O-Bahn). 2

Plate 2 With-flow bus lane: Bangkok

Plate 3 Contra-flow dual bus lane: Bangkok

Plate 4 Busway transit stop, with overtaking facility: Sao Paulo

joined together, the group of buses started and stopped broadly in unison. COMONOR was initially successful but was found to be too difficult to sustain on a day-to-day basis. It evolved in Porto Alegre, for example, into "bus ordering" in which buses are allocated to one of three groups (A-B-C). The buses arrive in random order at the beginning of a section and are marshalled into the preferred sequence, though not into strict convoys (Figure 1). This method operates effectively and can improve commercial speeds at high levels of passenger demand.

1.12

Line-haul capacity can be enhanced by the use of high-capacity buses, whether articulated, doubledeck or with the use of bus + trailer. However, passenger transfer capacity at bus stops is often the constraint on system performance, and door configurations and ticketing arrangements are often more important than bus capacity alone.

1.13 Various traffic signal techniques are available to give priorities to buses. However, where bus flows are high, there is a "constant" call for green time by buses and selective-detection of buses may not be appropriate.

Figure 1: COMONOR - Bus Convoy Operations

(from Roads and Transport in Urban Areas, IHT/HMSO 1987)

1.14

Bus delays at bus stops can be minimised by collecting fares and issuing tickets prior to passenger boarding (i.e. off-board) - see Plate 5. At bus stops where passenger volumes are very high, excessive bus dwell times can occur when many passengers try to board incoming buses, and block the doorways such that the doors cannot be shut. Figure 2 illustrates how bus dwell times tend to increase sharply once the capacity of the bus is reached and crush-loading sets in. This problem can be minimised by assigning staff to control boarding. Finally, the provision of a guidance system may, under some circumstances, enhance performance. This aspect is dealt with in Section 3.

The case for busway transit

1.17 The main advantages of busway systems are (Cornwell and Cracknell, 1990): Flexibility. Since buses can join and leave a busway along its length, routes serving many parts of a city can use a busway over part or all of its length. Passengers from a wide catchment area can therefore benefit from improved services, without having to change vehicles (as required with a fixed-track system). Affordability. A basic at-grade busway along an existing right-of-way is likely to cost of the order of US$1 million/km (end-1989 values), depending upon the need for utility relocation and other local factors. Self-enforcement. Because a busway physically segregates buses from general traffic, busways are virtually self-enforcing and do not require a permanent police presence to be effective. Scope for incremental development. Sections of even a few hundred metres of busway can be useful (whereas rail transit needs a depot and a significant route length before it can attract passengers). Busway transit can be enhanced step-by-step (e.g. by adding grade separation at critical intersections; introducing off-bus ticketing) as and when finance permits.

1.15

Busway transit performance

1.16 Surveys were carried out by TRL in 1989-90 to measure passenger throughputs and bus commercial speeds for selected busway transit schemes in Brazil, Cote d'Ivoire and Turkey (Gardner et al, 1991). From these measurements, the practical capacity of busway transit was determined for various design characteristics These estimates are summarized in Table 2. The conclusion from this survey of performance was that well designed and efficiently run busway transit systems can achieve consistent flows of 25,000 passengers per hour per direction, and at speeds of up to 25 kmph.

Plate 5 Off-board ticket sales; Salvador

Figure 2 Bus Dwell Times and number of Boarding Passengers

Table 2: Measured and Estimated Busway Performance (predominantly boarding direction)

Description

Example

Measured Peak hour Flow Range pass/hour

Estimated Practical Capacity* pass/hour

Basic Busway No Options Trunk and Feeder Bus Ordering Overtaking + Express Services Optimum Combination of High Capacity Options

Ankara Istanbul Abidjan** Curitiba Farrapos Assis Brasil** Belo Horizonte Sao Paulo None

7300 - 19500 9900 } 17500 - 18300 } 15800 - 20300

5800 -18100 13900 - 24100 8200 -14700 14900 - 27900

NIA

30600

* ** -

Capacity estimated for different combinations of passenger demand and bus stop layout, using procedure, assumptions and capacity criteria given in Appendix E, RR329 (Gardner et al, 1991) Flows measured when systems were operating above capacity. Measured flows may not represent maxima because of limited demand

Foreign exchange. Since busways can usually be constructed with local labour and materials and, in many countries, operated with locally produced vehicles, the foreign exchange requirement is minimised. Existing experience. Busways enhance the use of buses, the predominant transport mode in most cities, and can draw upon the wealth of experience and knowledge of bus operations already available. One of the main disadvantages of busway transit, however, is that their implementation requires the active cooperation of the highway authority, the licensing authorities, the police and bus operators; such cooperation can be difficult to achieve.

2. PLANNING CONSIDERATIONS
PLANNING CONTEXT
2.1 Given the performance figures noted in Section 1, busway transit is likely to be suitable in a variety of locations, typical examples being: in the main corridors of medium-sized cities, where public transport travel demands are up to about 20-25,000 passengers/hour/direction (p/h/d). in the secondary corridors of large cities, to complement rail mass transit. in outer city suburbs, to structure newly urbanizing areas.

1.18

2.2

More than forty busways exist worldwide, though only a handful of cities have developed them in a systematic and comprehensive manner as the framework of the city's mass transit network. The best example of the widescale use of, and dependence on busway transit is in Curitiba. There, in a city of just over one million population, busways form the backbone of five `structural axes' radiating from the city centre and along which city growth is focused. In planning a system, it is important to distinguish between a basic busway as a traffic management measure, to meet short-term traffic objectives, and a bus-based mass transit system, including special operational measures, to meet mediumlong term objectives. Although the physical infrastructure in each case might be similar, the operational and organisational arrangements for busway transit are an important component of the system and require careful planning.

FEASIBILITY AND WARRANTS

2.6 Figure 3 illustrates the trade-off between traffic flow (degree of saturation) and bus flow; the chart indicates that this trade-off can be generalised into four basic scenarios. Case 1 is where passenger demand is modest and the road has spare capacity - bus priorities would not be needed in this case. In Case 2, although the road may be nearing saturation, bus priorities would be unlikely to be acceptable because of the relatively low volume of passengers and buses. Case 3 is where there are sufficient buses to make priorities worthwhile and acceptable, without significant disbenefits to other traffic (since the road has spare capacity) However, most important is Case 4: this is where bus priority is most needed but, because the road is already running at or near its capacity, the allocation of roadspace to buses would disbenefit other road users (unless additional capacity was provided). In this latter case strong political will is needed to implement bus priorities. In cases where existing roadspace is limited, the allocation of existing scarce roadspace to buses may be justified because: a bus lane or busway can carry up to about 20,000 p/hld, whereas a lane used by cars only is unlikely to carry more than 2-3,000 p/h/d at normal occupancy levels. it may be easier to divert cars rather than buses to alternative routes it may be more cost-beneficial to allocate existing roadspace to buses and to construct additional roadspace to be used by all vehicle types, rather than to construct the infrastructure required for a high-capacity rail mass transit system. Some of the most successful bus-related schemes have been where bus priorities were introduced in association with improvements to conditions for general traffic (e.g. Abidjan). 2.8 In some cities, decisions on roadspace allocation are taken on a purely political basis. In others, it is on the basis of the numbers of passengers which can be carried along a lane by buses or by private motor vehicles; elsewhere the decision might be made on economic grounds, taking into account time values and vehicle operating costs. Because of the wide variety of traffic and physical circumstances, a detailed assessment is generally needed for each scheme. Bus lanes can usually be justified where the minimum bus traffic is 20-40 per hour; most busway transit schemes will have flows in excess of 100 buses per hour.

2.3

ALLOCATING ROADSPACE
2.3 Where a busway is to be inserted into an existing right-of-way, difficult issues arise over the allocation of roadspace between the conflicting demands of different road users. In many cities, there is insufficient roadspace to meet the unconstrained demands of all road users and it is necessary to have a demand management policy to guide the allocation of roadspace. Where the policy is one of "laissez faire", buses are likely to suffer severely from the congestion caused by other road users. However, where there is a positive policy to restrain the use of private cars and to promote the use of public transport, busways give physical expression to this policy objective. Where passenger demands are high, there is no doubt that the number of passengers that can be transported along a bus lane or busway is substantially more than can be transported by private cars along the same lane Nevertheless, if roadspace is allocated to buses, the roadspace must be seen to be used reasonably effectively. Where bus flows are relatively low, the bus track is vacant for the majority of the time and this can give rise to political pressures from the motoring lobby to reallocate the space to cars. One possibility is to permit the use of the busway by other specified vehicles (e.g. high occupancy vehicles). Many bus lanes operate on a time-sharing basis (i.e. they are only restricted to bus use at certain times of day). However, this is a sophisticated operation which requires a high degree of road user discipline, unless bus flows are very high and buses effectively "occupy" the bus lane. Time-sharing is unlikely to be effective in many developing cities. 2.7

2.4

2.5

Figure 3: Feasibility of Busway Along Existing Road

CAPACITY CONCEPTS
2.9 The technical literature contains many ambiguous references to the "capacity" of alternative transit systems (metro, light rail, bus etc.). Such statistics usually refer explicitly or implicitly to "line- haul capacity" However, it is also important to consider "passenger transfer capacity" (i.e. the maximum number of passengers who can board and/or alight at a stop/station during a given time period) since bus stop/station capacity is often the limiting factor in a transit system. Maximum line-haul throughput decreases as passenger transfer demands increase (i e. brie-haul capacity is a variable and it is not possible to quote a single "capacity" figure for a bus system without reference to passenger transfer demands and bus stop capacities) - see Figure 4.

2.10 It is even difficult to apply the traditional concept of line-haul capacity to bus systems. This is because the number of "available passenger places" on buses passing along a bus track does not correspond to the maximum passenger throughput

achievable. Without a trunk-and-feeder system, passengers will only board buses serving selected routes In practice it is impossible to fill every bus to capacity because of the imbalance between the number of empty spaces on a bus and the number of passengers boarding at each stop In some cases, a bus leaving a bus stop will have empty passenger places whsle in others, the bus may be full and leave some passengers waiting at the stop. This variation in loading from one bus to another imposes practical limits on average load factors (number of passengers/available passenger places) - the available evidence suggests that without special operational measures, it is difficult to achieve an average load factor in excess of 70-80% without severe overcrowding on many buses. Where load factors of the order of 100.'o have been observed (e g. Sao Paulo), around 50-60% of buses were crush loaded (Figure 5). 2.11 Furthermore, since boarding and alighting times are substantially different, maximum line-haul passenger throughput is higher in the predominantly alighting direction than in the predominantly

Figure 4: Relationship Between Line-haul Throughput and Passenger Transfer Demand

Flows based upon estimation procedure as detailed in Gardner et al (1991) Standard stop 3-bay on-line, High capacity stop 4-bay with overtaking

boarding direction, for a given passenger transfer demand. For planning and design purposes, it will generally be critical to examine the predominantly boarding direction. 2.12 Unless bus entry to the busway is controlled, bus arrivals will be irregular and may lead to substantial peaking. TRL surveys of existing busways suggest that the peak 5-minute flow can be between 1.3 and 3.0 times the hourly rate.

of 80,000 or more p/h/d. But of course, stops are usually necessary and have a capacity substantially below this figure. 2.15 In many cases, the capacity constraint on a system will be a single bus stop. The interactions between passengers, bus and driver characteristics, and bus stop layout are complex. Again, "capacity" is variable and dependent upon passenger behaviour, arrival time patterns and many other factors. Stop/ station spacing also influences performance and Figure 7 illustrates the effect of bus stop spacing on bus commercial speed under typical operating conditions 2.16 Junctions influence the flow of buses along at-grade busways and consequently affect both capacity and speed of bus operation. Traffic signal-controlled functions (including pedestrian crossings) reduce the amount of green time available to the busway and impose delays to some buses. These delays reduce bus and passenger throughput, and average commercial speed In some cases, a particularly busy junction may be the busway bottleneck - this is the case on Farrapos, Porto Alegre, where the function adjacent to the central area "controls" the flow of buses off the busway and into the city centre during the morning peak. In most cases, however, junction capacity will be greater than that of the most critical bus stop

SYSTEM PLANNING PARAMETERS

2.13 Figure 6 lists the main factors which influence the capacity of a bus lane or busway The most important is probably the degree of segregation between buses and other traffic' The other critical components of a busway transit system which may constrain capacity include the running section (link), bus stops, junctions or the collector/distribution system. 2.14 In general, running sections are unlikely to be a bottleneck. Average bus headways of 4.5 seconds have been measured for bus platoons travelling along busways in Abidjan and Porto Alegre, corresponding to 800 buses/hour If buses could be fed onto and off a busway at this rate, the line-haul "capacity" without bus stops would be of the order

It may be argued that the apparent advantage of rail transit over conventional bus systems is largely due to differences in the degree of segregation rather than to other inherent technological differences

10

Figure 5: Relationship Between Crush Loading and Average Load Factor

2.17 As noted in paragraph 2.10, the choice of route structure between trunk-and-feeder and an "open" system affects achievable load factors and other performance indicators. Trunk-and-feeder operations permit high line-haul passenger flows to be achieved, particularly with the use of high-capacity buses, but at the cost of enforced passenger interchange at expensive transfer stations. Open systems can serve a wider direct catchment area (without interchange), but the irregularity of bus arrivals and dwell times limits sustainable performance levels

function. At present, no busways function in an entirely deregulated environment.

ORGANISATION AND MANAGEMENT

2.19 As indicated earlier, a basic busway can be used as a traffic management tool to segregate buses and other traffic However, in order to develop a busway transit system which can offer high performance, special operational measures are required. Some measures necessitate suitable organisational and management arrangements in order to be effective, for example: management of the bus track - such as the provision of a tow truck to deal promptly with breakdowns. maintenance of the bus track, bus stop facilities and traffic control devices. fare collection and ticketing, possibly including offboard ticketing and management of season tickets or travelcards.

REGULATORY FRAMEWORK
2.18 Busways function under a wide variety of regulatory arrangements. In Abidjan and Liege, for example, bus services are provided by a publiclyowned monopoly operator; in many Brazilian cities, including Sao Paulo, services are provided in a regulated environment by various private operators. In some cities, such as Curitiba, both public and private companies operate services as part of an integrated network, with a common fares policy, and with colour coding of vehicles according to

11

Figure 6: Capacity Determinants

RIGHT OF WAY CHARACTERISTICS Road cross-section Degree of physical segregation from other traffic Junction design and control Horizontal and vertical alignment Road surface characteristics BUS STOP CHARACTERISTICS Overtaking facilities? Spacing Number of loading positions (bays) Platform storage area Passenger information Platform height BUS CHARACTERISTICS Vehicle size and capacity Existance and control of doors Number, location, width and use of doorways Number and height of steps Floor height Maximum speed Acceleration and deceleration rates OPERATING CHARACTERISTICS Route structure and scheduling Driver behaviour Fare structure and ticketing Trunk-and-feeder Bus Ordering (or convoys) PASSENGER CHARACTERISTICS Passenger demand, by stop Distribution, by time of day Behaviour GENERAL TRAFFIC CHARACTERISTICS Volume and nature Road user discipline Encroachments (e.g. hawkers)

12

Figure 7: Relationship Between Bus Stop Spacing and Commercial Speed

driver training for the particular conditions associated with high-intensity operations. facilities and staff to undertake bus ordering, where appropriate provision of reliable and up-to-date passenger information. supervision to limit bus dwell times at busy stops in order to avoid excessive delays and service disruptions.

The relative advantages and disadvantages of median and lateral busways are summarised in Table 3. A common objection to a median busway, and the associated of island bus stops, is that passengers have to cross more of the road than in the case of kerbside busway or bus lane. This is untrue - for a return journey to and from any given point, a passenger has to cross the equivalent of one road width only; for a median busway, the passenger crosses half the width of the road on both the outbound and the return journey, whereas for a lateral busway, the passenger crosses the whole road width on one journey and not at all on the other (Figure 9). 33 As indicated earlier, the degree of grade separation between buses and other traffic can have a major influence on performance. The general arrangement for an elevated busway is given in Figure 10, and Plate 8 shows a section of elevated busway in Runcorn New Town. Plate 9 shows a tunnel in Belo Horizonte, Brazil, where buses have exclusive use of the lower level of the tunnel to pass under a hill between the city centre and suburbs.

2.20 Whether these functions are to be performed by several agencies or by a single busway transit agency, specific arrangements must be made. Particular weaknesses which occur in practice are the fragmentation of responsibilities between the highway authority, bus operators and the police. This tends to lead to inadequate maintenance of busway infrastructure and to lack of "track management" where there are several competing operators

3. TRACK DESIGN
TRACK OPTIONS
3.1 Busway track may be located along an existing or a new right-of-way. For an existing right-of-way, the bus track may be located in the centre of the road (median) or along the sides (lateral), Figure 8 shows the principal configurations and Plates 6 and 7 show examples of lateral and median busways.

Means of segregation
3.4 Some busways physically segregate buses and other traffic along their entire length using kerbs (e.g Liege, Belgium -Plate 10) or fences (e.g. Sao Paulo, Brazil- Plate 11), while others have segregation only at island bus stops (e.g. Nagoya, Japan-Plate 12) or at island stops but with heavy studs between stops (e.g. Sao Paulo, Brazil- Plate 13). 13

Figure 8: Principal Busway Configurations

14

Plate 6 Lateral busway using one half of a dual carriageway: Istanbul

Plate 7 Median busway using the central reserve of a dual carriageway: Sao Paulo 15

16

Table 3: Advantages and Disadvantages of Alternative Busway Layouts

Figure 9: Pedestrian Movements with Alternative Busway Types

17

Figure 10: Elevated Busway - General Arrangement

18

Plate 8 Elevated busway: Runcorn

Plate 9 Busway tunnel: Belo Horizonte

19

Plate 10 Kerb separators: Liege

Plate 11 Fenced separators: Sao Paulo

20

Plate 12 Separation at bus stops: Nagoya

Plate 13 Studs used to separate

21

3.5

Where bus flows are high, it may be desirable to install a median barrier to dissuade pedestrians from crossing the busway at unauthorized locations. The barrier may be in the form of a fence or a New Jersey-type barrier.

for the fact that the rear wheels of a bus go through a smaller radius than the front wheels at road bends. Horizontal and vertical alignment 3.7 It is recommended that curve radii and lateral banking is such that lateral accelerations do not exceed 1.0 m per sec. sq.; a more desirable level of lateral acceleration is 0.8 m per sec, sq.. Table 5 shows minimum radii for busways for different design speeds and street banking. The longitudinal incline of busways should be kept to a minimum, so as to maintain operational regularity, inflict minimum wear on vehicles, and provide travel comfort and general safety. Table 6 shows recommended maximum values for longitudinal inclines in relation to design speed. Near to intersections the longitudinal incline should be kept below 4 per cent.

Typical cross-sections 3.6 The width of a busway is dependent on design speed, vehicle width and operational characteristics. Typically, bus width is 2.5m and it is recommended that no bus lane should be less than 3 Om in width. Table 4 gives recommended lane widths for busways carrying in excess of 60 vehicles per hour, at different design speeds. In built-up areas of the city, design speeds are likely to be in the range 40-60 kmph For express ways and grade-sepa-rated busways, higher design speeds may be practical Lanes must be widened at curves to allow

Table 4: Recommended cross-section widths for median busways carrying more than 60 buses per hour

Width (m) Design speed (km/h) 100 80 60 40 Central separator (between lanes) 4.00 3.75 3.25 3.00 0.4 0.4 0.4 0.4 Outer separator (between bus lane and other traffic) 0.75 0.50 0.30 0.20

Bus lane

Complete busway 10.30 9.30 7.90 7.20

(source: RATP)

22

Table 5: Minimum Radii for Busways (metres)

Design speed (km/h) 100 80 60 40

Lateral acceleration (m/sec^2) 0.8 1.0 0.8 1.0 0.8 1.0 0.8 1.0

Banking 0% 964 771 617 493 347 278 154 123 5% Radius (m) 697 517 388 331 215 186 96 83 434 390 278 250 156 140 69 62 10%

(source: RATP) Table 6: Maximum Longitudinal Inclines for Busways

Design speed (km/h) 100 80 60 40

Open road % 4.0 4.0 4.5 5.5

On ramps or Under difficult conditions % 4.0 6.0 6.5 10.0

(source: VOV/VDA)

INTERSECTIONS
Layouts 3.8 Some typical intersection layouts for median and lateral busways are illustrated in Figure 11. Detailed layouts will depend upon local geometrics and traffic flows. 3.9 Where the busway or bus lane does not continue up to the stop line (as with the UK practice of providing set-backs in order to maintain stop line capacity), pre-signals can be used to manage the queue such that buses reach the head of the queue and traffic entering the junction can be controlled within its capacity (Figure 12).

3.10 In order to minimise disruption to busway operations, the number of roads crossing the busway is usually limited to main thoroughfares only. Kerbs or barriers may be placed to prevent traffic turning across the busway into or out of minor side roads. In such cases, side road traffic is restricted to right turn in/right turn out (right hand rule of the road) and "Q" and "G" turns are used to concentrate traffic onto a limited number of cross routes. Such arrangements have an impact on local access, which needs to be considered very carefully

Signal control
3.11 A high degree of traffic signal control is generally required in order to manage high bus and general traffic volumes, without excessive delays. Signals

23

Figure 11: Typical Intersection Layout

(i) Median busway

(ii) Lateral busway

Source: RATP

24

Figure 12: Presignals/Bus Advance Area

Bus Lane with "Traditional" Set Back

Bus Lane leading to Bus Advance Area

distributor systems should at least match the bus demand at the relevant locations. This can be difficult where one or more busways lead into a city centre - in such cases, special arrangements are needed to disperse high bus volumes into terminals or onto a circulation system composing bus roads or lanes. The use of a range of bus priority techniques in and around a city centre will usually be essential to enable a busway to function effectively.

may also be required to control pedestrian crossings. Signal control can be used to aid bus movements in the following ways. selective detection of buses to extend a green phase or to recall a green phase. demand dependent stages (which enables a bus to call a stage which would not otherwise occur). signal time biassing to favour a stream with a high proportion of high-priority vehicles "gating" in order to manage queues in favour of high priority vehicles.

GUIDANCE SYSTEMS
3.13 Considerable publicity has been devoted to "guided busways". A guided busway is simply a busway equipped with a guidance mechanism (tracks) to enable buses to travel at speed in a relatively narrow right-of-way. One form of guided busway "O-Bahn" - has been built in Essen, Germany, and in Adelaide, Australia (Figure 13); another system "Guided Light Transit (GLT)" - operates in Rochefort, Belgium.

COLLECTOR/DISTRIBUTOR SYSTEMS
3.12 It is important to organise suitable collector and distributor systems to feed buses onto the busway and to permit them to leave the busway, without undue congestion. The capacity of the collector and

25

Figure 13:

O-Bahn Track Dimensions

Source: VOV/VDA 1977

3.14 The prime advantages of a guided busway coin- , pared to a conventional busway are' the track provides a permanent physical presence, which makes the system more "visible" to politicians and public alike. where the right-of-way is severely constrained, or land are values are high, guided busway can operate between junctions in a right-of-way about 1 metre narrower than that of a conventional busway; however, this advantage is lost at functions (where capacity is usually critical) in the case of guidance systems which require an entry splay. guideways enable buses to pass in opposite directions at high speed in a reduced right-of-way. the track "occupies" the right-of-way and makes violation by other vehicles extremely unlikely.

3.15 The prime disadvantages are the additional cost compared to a conventional busway and the severance effect in urban areas. It appears that the prime locations for guided busway would be in suburban areas requiring high-speed operations. 3 16 Guided busway can offer broadly equivalent levels of service to light rail transit, but at much lower capital cost. Guided busway has the advantage over light rail transit that the vehicles can leave the track and so offer door-to-door service over a wide catchment area, without enforced passenger interchange.

OTHER ASPECTS
3.17 Particular attention needs to be given to the design and construction of the road pavement because of repeated loadings by heavy vehicles at bus stops and at traffic signal stop lines. It will often be

26

desirable to provide concrete slabs at stops, in order to avoid the sort of pavement creep illustrated in Plate 14 3.18 Where enforcement is expected to be a particular problem, due to poor road user discipline, physical and electronic measures are available to dissuade other vehicles from entering the busway. Plate 15, for example, shows a "barrier", which by its shape and dimensions inhibits the passage of vehicles other than buses.

passengers (e.g. Bogota, Colombia). 4.4 The main features of bus stop design which will affect capacity of the system are: the number of bus bays provided. the order in which buses stop (including the allocation of bus routes to bays). facilities for buses to overtake one another (and thus to avoid blockages). The number of bus bays at each stop on a busway transit system will typically vary between one and six, depending on the transfer demand. These bays can be either `on or off-line'. On-line bays allow no special overtaking facility; the buses stop on the busway track. Off-line bays have a stopping area which is separate from the main running track; as a result buses can easily overtake one another at bus stops. To reduce the level of interaction between passengers waiting for different bus services, bus bays can be ordered (as noted above). In such an ordered system bus services or routes would be designated to particular bays, for example, a four bay stop might have its first two bays for route (or group of routes) A and its remaining two for route (or group of routes) B. This is denoted AA-BB. Figure 14 shows some examples of these design options; indicative capacities, based on TRL analyses, are given in Table 7.

4.

BUS STOPS

4.5

BUS STOP DESIGN

Capacity 4.1 Traffic behaviour at bus stops is extremely complex and relatively little is known about bus stop capacity. In developing cities, up to about 6,000 passengers/hour may board or alight at a busy bus stop. The highest number of boarding passengers recorded in the TRL surveys of busways (Gardner et al, 1991) was some 3,600/hour during the morning peak at the KCR station bus stop in Hong Kong; the highest recorded number of alighting passengers was at Osmanbey, Istanbul, with 4,000/ hour during the morning peak. Based on survey results, and judgmental assessments, three bus stop categories are defined (in terms of passenger handling volume) for design purposes: Very High Volume: either boardings or alightings greater than or equal to 2,500/hour. High Volume: maximum of boardings or alightings less than 2,500, but greater than or equal to 1,000/ hour. Intermediate Volume: boardings and alightings less than 1,000/hour.

4.2

Layouts 4.6 Quite clearly, the capacity of a bus stop (and the busway system) will be affected by its layout; category one bus stops (very high volume transfer) will almost certainly need multiple off-line bays, which are ordered in some way. Category three bus stops will require less demanding conditions: probably on-line, single or double bay which may not have to be ordered. Figure 15 shows a number of bus stop designs: a typical layout of an "on-line" bus stop (i.e. without an overtaking lane), on a median busway in Curitiba, Brazil; a similar bus stop type in Porto Alegre, but with staggered bus stops to minimise right-of-way requirements. Experience in Porto Alegre suggests that the stops should be staggered such that to reach the pedes-trian crossing, alighting passengers should have to walk in the same direction as the bus. The overtaking facility required for category 2 and 3 bus stops can be achieved in various ways: for example, by the provision of two lanes at bus stops (e.g. Sao Paulo - Plate 4), off-line bays (e.g. Belo Horizonte - Plate 16 and Figure 15), or parallel bus stop islands (e.g. Singapore - Plate 17 and Figure 16)

Design considerations 4.3 Both the design of the bus stop, and the operational procedures at the bus stop, will affect both its capacity and that of the busway system. Thus, any design feature which enables buses to enter and leave the stop without bus congestion will be beneficial, as will any measure which enables passengers to board and alight rapidly. Often the design and operational measures complement one another. For example, bus stops may be "orderly", where bus bays are marked and used 9, or "disorderly", where buses stop one or more times "on-demand" along a length of kerbside. Disorderly stops can achieve high bus and passenger flows, but at the expense of passenger inconvenience and safety. Sometimes one objective of a busway scheme is to "order" the flow of buses and

4.7

27
.

Plate 14 Pavement creep along a busway

Plate 15 Busway entry barrier 28

Figure 14: Some Examples of Bus Bay Layouts used in Case Study Cities

30

Table 7: Indicative Bus Stop Capacities

Figure 15: Bus Stop Configurations

(i) On-line bus stop: Curitiba

(ii) On-line bus stop: Porto Allegre

(iii) Off-line bus stop: Belo Horizonte

31

Plate 16

Off-line bays: Belo Horizonte

Plate 17 Parallel bus bays: Singapore

32

Figure 16: Bus Stop Layout, Raffles Quay, Singapore

33

BUS STOP SPACING

4.8 Bus stop spacing should be chosen in relation to the density of passenger demands, the locations of large traffic generators, road geometrics and the level of service required. As indicated in Section 2, stop spacing has a large influence on commercial speed (Figure 7), with high performance being associated with relatively longer stop spacings

5.2

It will often be unreasonable to compare the costs and benefits of bus and rail systems over identical route lengths, because bus priorities are only required in congested areas ; elsewhere, buses can run on all purpose roads with general traffic, i e. no special facilities are required. The total length of infrastructure required will therefore be different. Bus and rail transit options also have implications for movement patterns and land development: by their nature, rail systems tend to concentrate passenger flows (especially where bus services are "integrated" with rail) whereas bus transit tends to offer a more dispersed pattern of movements Evaluation of these effects depends upon the land use-transport strategy of the particular city.

5.3

BUS STOP LOCATION

4.9 The relative locations of traffic signals and bus stops should be determined in relation to local circumstances. However, substantial bus delays can occur where a traffic signal is located immediately downstream from a bus stop This is because without signal pre-emption, it frequently occurs that a bus completes loading, but the traffic signal shows red. The bus is obliged to wait in the bus bay until the signal turns to green, even though other buses may be waiting to enter the stop The following buses therefore suffer a traffic signal delay before even reaching the bus stop and, at the intersection, may incur further delays.

BUSWAY TRANSIT COSTS

Capital costs 5.4 Out-turn cost data for existing busway schemes vary according to design standards, construction procedures, initial condition of the roadway, local inflation rates, exchange rate variations, and so on However, a typical at-grade, partially segregated busway track might cost of the order of US$ 1 million/km (1989 values), excluding vehicles and terminals, based on the following assumptions. no land acquisition would be required and existing road kerblines would not be moved. the existing median would be removed in order to allow busway construction. existing road drainage would be adequate and would not be modified. the existing road pavement would be adequate except in the bus stop area, where complete reconstruction would be required. no extensive diversion of public utilities would be required. General guidelines for the cost of an elevated busway pose even greater difficulties than those for an at-grade busway, due to the wide range of possible construction techniques, foundation conditions, bus stop treatments and other features. There is little direct experience of the cost of elevated busways since no extensive sections have been constructed, although several are under consideration (e.g. Karachi). Based on UK conditions, a representative elevated busway is estimated to cost of the order of 10 million (US$17 million) per km (1989 values). The approach being taken in several current proposals for elevated schemes is to consider elevation only where traffic capacity at selected junctions is critical. Clearly if bus stops can be accommodated at-grade, considerable cost savings are possible.

4.10 Placing a bus stop immediately downstream from a traffic signal controlled junction can also cause problems at high bus volumes, because buses queuing to enter the stop may back-up and block the junction (depending upon bus driver training and discipline). Successful design will involve combined consideration of signal design, passenger demand and bus flow in relation to bus stop location. The provision of grade-separated pedestrtan crossings can be considered, particularly where bus stops are located mid-block. However, enforced climbing of stairs may be unpopular with passengers.

5. EVALUATION
DEFINITION OF OPTIONS
5.1 Comparison of bus and rail mass transit options can present problems because of the different characteristics of the two systems. In particular rail systems require a depot (usually in a suburban location because of land requirements) and a substantial track length before they can be effective and attract passengers, whereas busway systems can be developed incrementally Furthermore, rail mass transit is a "closed" system in which all the costs of infrastructure and rolling stock can be attributed to the system, whereas busway transit is an "open" system in which buses may use the busway on only a minor part of the route length, making definition of "the system" and cost allocation difficult. -

5.5

34

5.6 The physical quality of most existing busways is not high and this is reflected in the above cost esti-mates. For example, passenger facilities (stops; shelters; crossings; passenger information sys-tems) are often extremely basic and of low quality, and consequently have a poor image. The bus track is often poor, since existing pavements are often retained in bus stops areas in order to minimise costs. In several existing busways, pavements have deteriorated due to tracking (e.g. Istanbul; Porto Alegre) and in some cases, the pavements have failed (e.g. Abidjan; Recife). The costs of improving the physical quality of an at-grade busway would not be great. The public image would be enhanced, scheme life would be prolonged and maintenance costs reduced. Even with such improvements, the per km costs would still be substantially below those of an equivalent light rail transit (LRT) system. 5.7 The costs of associated infrastructure will vary from place to place and will depend on local requirements. The cost of a footbridge might typically be the order of US$80-100,000. Where a comprehensive busway transit system is to be implemented, new depot and workshop facilities may be required, particularly if high-capacity buses are to be introduced Again, costs depend upon many local factors, but a new depot for about 200 buses could cost in the region of US$6-8 million (excluding land costs). A workshop and central stores facility might cost a similar sum, depending upon the scale of facilities required. (Cost estimates 1992 US dollars unless stated otherwise). If a trunk-and-feeder system is to be operated, transfer terminals will be needed along the main axes together with a terminal station at the end of each corridor. Costs depend upon many local factors but could be of the order of US$500,000 for a basic transfer station and US$800,000 for a basic terminal station (excluding land costs). The out-turn costs of the various high-quality terminals in Curitiba, for example, were of the order of US$ 1 million for each transfer station and US$3 million for each terminal. However, terminals may offer development opportunities and additional sources of revenue. The infrastructure costs of atgrade busways are generally low compared to those of other segregated mass transit modes: the typical busway track infrastructure costs of the order of US$1 million per km compares with estimated costs for mass transit railways (including rolling stock and other equipment) in the range US$8-27 million/km (1987 values) for at-grade construction, US$22-60 elevated and US$50-165 underground (Fouracre et al, 1990).

indicate the cost of the city's busways were of the order of US$5 million/km for the South corridor and US$10 million/km for the East corridor, compared with some US$27 million/km for the city's LRT (Kain et al, 1990). In Auckland, corridor studies compared LRT and O-Bahn options and the findings suggest that the capital cost of an O-Bahn would be about half that for a comparable LRT (Auckland Regional Authority, 1988). In Adelaide, where an O-Bahn has been implemented, preliminary design estimates suggested that for comparable LRT, busway and OBahn systems (including track, control systems, depots, rolling stock etc.) the costs of the conventional busway and O-Bahn systems would be 58% and 65% respectively of that of an LRT system (Wayte, 1988). 5.11 In developing countries, the foreign exchange requirement of a proposed investment can be an important criterion in the selection of a technology. Busway transit offers considerable scope for construction by local contractors and, where a local assembly or body-building industry exists, a substantial part of bus costs can be incurred locally. Operating costs 5.12 Estimated operating costs of busway transit are in the range 8-12 US cents per passenger km. (from Armstrong-Wright, 1986, with costs factored to 1993 prices). The key components are labour, energy and replacement materials.

5.8

THE IMPACT OF BUSWAY TRANSIT

The users 5.13 The majority of beneficiaries of busway transit in developing cities are likely to be existing public transport users; there has been no evidence of any major switching to bus from private modes, as a result of the introduction of priority measures in the industrialised world (although few cities have extensive, high quality, bus priority systems.) There are strong conceptual grounds for believing that most private vehicle users in developing cities are unlikely to be attracted to use of public transport; these travellers come mainly from high income groups, who will value comfort and convenience of personal transport very highly. 5.14 Many earlier studies have attested to the level of user benefits which result from bus priority measures. Typical time gains in European and North American cities, measured over the length of schemes, range between 20-50 per cent. Similar observations have also been noted in Singapore, Bangkok and Porto Alegre, in the latter, journey times were reduced by 29 per cent. Small improvements in regularity have also been noted. Busways can result in improvement in the quality of travel, particularly if the investment includes new rolling stock which is clean and comfortable. There is

5.9

5.10 In Ottawa, it was estimated that the capital cost of a busway system would be 68% of the cost of a comparable light rail system. Figures for Pittsburgh

35

anecdotal evidence from Curitiba and Nagoya that a high quality system will generate new passengers in addition to attracting transfer from other modes. Transport operators 5.15 Busways can help to reduce bus operating costs in several ways bus commercial speeds can be raised by removing buses from general traffic congestion and, since bus operating costs are speed-related, operating costs can be reduced. because buses can achieve higher average speeds, fewer buses are required to operate a given service, thus saving in capital costs. busway transit may generate opportunities for scale economies which might not otherwise be feasible (e.g. use of high-capacity vehicles).

conditions on the road network will quickly induce new car traffic to emerge. 5.19 However, there are reasonable grounds for supposing that busways (in common with other mass transit) could have some influence on the spread of traffic congestion. With increasing car ownership and use, city centre traffic congestion reaches what has been described as the threshold of the intolerable, it cannot get any worse, and assuming all traffic engineering measures have been exhausted, can only spread more widely, rather than more deeply. New roads to access the city centre may improve the situation, but there are limits to what can be achieved, simply because the land is not available and the resulting environmental damage is likely to be too great. A mass transit system, making the best use of the existing road system, provides the capacity needed to access the city centre, without the associated penalties of road building. In providing greater access, the mass transit system helps to reduce the spread of traffic congestion. 5.20 The environmental impacts of any particular scheme will require detailed assessment in the light of scheme characteristics and local circumstances. Busways, by their nature, provide a high-speed track in built-up areas where pedestrian activity will be intense. The resulting severance, safety, noise and air pollution effects all warrant particular attention. Severance effects can be minimised, and safety enhanced, by suitable urban design and by the provision of adequate pedestrian crossing facilities. Some busways have been designed so as to minimise the interaction of pedestrians and vehicles; but pedestrian crossing points are inevitably necessary (if only to access the busway), as is interaction with other traffic at grade junctions and along unprotected rights-ofway. 5.21 On-street noise and air pollution effects of busway transit can be minimised through the use of modern, LPG (liquid petroleum gas) or CNG (compressed natural gas) powered buses, or electric powered trolleybuses. 5.22 Because of the severance effect of any at-grade median transit system (bus or rail), local access requires particular attention. Strict parking, waiting and loading controls will be required in order to ensure adequate servicing and roadspace is available for moving traffic along the corridor. Urban development 5.23 The essence of a city centre is that it is the most accessible point from both within and without the city. This accessibility is important for many activities, and in particular for those central functions which serve a wide area and/or need a wide labour market: head offices, central government offices and legal institutions, financial institutions, media firms, theatres, department stores, etc. and all the

Non-users 5.16 It is impossible to generalise about the effects of introducing a busway into an existing road network. In a dense road network, a busway may influence traffic flows and journey times both along the busway corridor and on adjacent roads. In such cases, some form of network modelling is required to assess the possible reassignment of general traffic And in any event, some form of junction analysis will be required to estimate the effects of the new junction arrangements on junction capacity and delays (e.g. using software such as SATURN, OSCADY etc). 5.17 In some cases, the improved traffic management associated with bus priorities has led to travel time gains for both buses and general traffic (e.g. Bangkok). In Abidjan, introduction of a median busway across the city centre was perceived to have benefits to both the bus operator ("independence from general traffic") and to general traffic ("getting buses out of the way of cars"). Nevertheless, where a busway or bus lane continues up to a signals stop line, in order to give buses priority, there will be some loss of capacity for general traffic. This may be offset by gains to buses and passengers but the effect must be examined on a case by case basis in relation to person throughput, traffic flows and residual capacity. 5.18 Busways (and transit systems in general) are often promoted on the basis that they can contribute to relief of city centre traffic congestion through encouraging a modal switch from private to public transport. The evidence for success in this objective is, unhappily, not strong, most users of the busway will not have switched modes, but will be using a bus which has simply switched from an unreserved to a reserved track within the same right-of-way. Even where there may have been a switch from private to public transport, the improved traffic

36

supporting organisations (catering, hotels, etc.) that exist to serve these central functions. The fortunes of the city centre are at risk if the public transport system proves inadequate in supporting these central functions. In most developing cities, the majority of commuters to/from the centre depend almost exclusively on road-based public transport. If the city centre becomes congested (because too much traffic is occupying too little road space), then its relative accessibility may suffer, because the public transport system cannot perform effectively. As a result, new central functions will be discouraged from locating in the city centre and old established ones may start to drift away. Clearly, there is an intimate and vital relationship between the well-being of the city centre and its public transport system which should never be overlooked. For this reason, it is becoming increasingly apparent that urban transport development in the major cities may be reaching a stage where priorities have to be imposed, and mode choice has to managed to the advantage of public transport systems; the limited supply of road space feeding the city centre is exhausted, and the only possible relief would seem to be through the development of a mass transit system, like busway transit, which makes best use of the available road space. 5.24 Apart from promoting the performance of public transport, and thereby contributing to the healthy growth of the city, mass transit systems may have their own intrinsic developmental impact on a city Mass transit schemes have sometimes been proposed to enhance or encourage new city development and/or renewal. For example, it is reported that the development of the LRT in Manila has played a key role in shaping the urban development of the metropolis, triggering the redevelopment of the traditional centres of business and trade, and encouraging commercial growth along its route. This impact of mass transit is not fully understood and has not always worked, in particular where planning controls on urban development are weak. Generally, if a city has a buoyant economy then a mass transit system can contribute to and accentuate that condition by removing accessibility constraints; on its own, however, the mass transit system can do little. Thus ideally busways should be developed in unison with other on-going major developments within the city. 5.25 A number of mass transit schemes have managed to capture some developmental benefits for their own financial gain. This has been achieved through the commercial development of the air-space above terminals and interchanges; these revenues can contribute to both the capital cost of the structure and/or to general income. Other impacts 5.26 Public transport is often used by people who do not have access to private, motorized transport includeing children, old people and women. This means

that improvements to transit services can have important social impacts. For example, suitable bus seances can offer mobility to women who may not otherwise have access to motorized transport, and can increase their access to work opportunities, and to educational and social activities. 5.27 In developing countries, the foreign exchange requirement of a proposed investment can be an important criterion in the selection of a technology. Busway transit offers considerable scope for civil engineering construction by local contractors and, where a local assembly or body-building industry exists, a substantial part of bus costs can be incurred locally.

ECONOMIC EVALUATION
5 28 An economic analysis of any busway project should try to take account of all the impacts which have been discussed. Many of these impacts are clearly difficult to quantify. A busway scheme is likely to improve bus commercial speeds and reliability, and therefore the potential benefits are typically: journey time savings to bus passengers (including the value of increased reliability), and bus operating cost savings (including a possible reduction in fleet size). In general, the majority of benefits are likely to be associated with time savings at junctions. However, the analysis should also take into account changes in journey times and operating costs for other road users, especially if some reassignment of traffic is anticipated. Depending upon local geometry and traffic flows, introduction of a busway may increase or decrease the capacity available for general traffic, particularly at junctions, and detailed junction analyses are required to estimate these effects. 5.29 Finally, no study has examined the crucial issue of the developmental benefit to the city centre of a busway scheme. It is a very complex issue since it raises questions about the city structure and its efficient growth; these are questions which go beyond the bounds of urban transport planning, and pose major conceptual and technical problems of analysis. If the continued growth of the urban centre is an urban development objective then the busway scheme can be considered as a major positive contribution to achieving that end.

6. REFERENCES AND BIBLIOGRAPHY

ALLPORT R J and J M THOMSON, 1990. Study of mass rapid transit in developing countries; TRRL Contractor Report 188. Crowthorne : Transport Research Laboratory. ARMSTRONG-WRIGHT A, 1986. Urban transit systems: Guidelines for examining options. World Bank Technical Paper No 52 Washington DC: World Bank.

37

AUCKLAND REGIONAL AUTHORITY, 1988. Auckland Comprehensive Transportation Study Update Stage 2 Report. CHIN H C and Y TANABORIBOON, 1984. Dwell times of buses in Singapore; Proceedings of the 2nd Conference on Mass Transit in Asia. Singapore. CORNWELL P R and J A CRACKNELL, 1990. The case for busway transit; PTRC 18th Summer Annual Meeting. London: PTRC CRACKNELL J A, P R CORNWELL and G GARDNER, 1990. Study of bus priority systems in less developed countries: In CODATU V, Sao Paulo. Paris: CODATU Association. CUNDILL M A and P F WATTS, 1973. Bus boarding and alighting times. TRRL Report LR521. Crowthorne: Transport Research Laboratory. DALUORTO A, 1989. Evaluacion de pastas exclusivas para buses en Lima: In CEPAL Conference Procedings, Havana, Cuba. DEPARTMENT OF TRANSPORT, 1991. Keeping buses moving; Local Transport Note 1/91. London: HMSO. EBTU, 1982. Manual Tecnico - Tratamento preferencial ao transporte colectivo por onibus; Empresa Brasdeira dos Transportes Urbanos. Brazilia: EBTU. FOURACRE P R, R J ALLPORT and J M THOMSON, 1990. The performance and impact of rail mass transit in developing countries TRRL Research Report 278. Crowthorne: Transport Research Laboratory. GARDNER G and P R FOURACRE, 1990. Busway transit - the TRRL study; PTRC 18th Summer Annual Meeting. London: PTRC. GARDNER G, P R CORNWELL and J A CRACKNELL, 1991. Performance of high-capacity bus systems. TRRL Research Report 329. Crowthorne: Transport Research Laboratory. GIBSON J, I BAEZA and L WILLUMSEN, 1989. Bus stops, congestion and congested bus stops. Traffic Engineenng & Control, June 1989. HOUNSELL N B, 1988. Bus stop siting at road junctions. TRRL Contractor Report 89. Crowthorne: Transport Research Laboratory. KAIN J F, R GITTELL, A DANIERE, S DANIEL, T SOMMERVILLE and L ZHI, 1990. Increasing the productivity of the nation's urban transportation infrastructure: measures to increase transit use and carpooling. Draft Final Report. Washington, DC: UMTA. LINDAU L A, 1987. Bus priority systems in Brazil: from theory to practice; PTRC Summer Annual Meeting. London: PTRC.

MARLER N W, 1982. The performance of high-flow bus lanes in Bangkok. TRRL Supplementary Report SR723. Crowthorne: Transport Research Laboratory. MERCEDES-BENZ DO BRASIL, 1987. Sistema de transporte colectivo urbano por onibus, planejamento e operacao MERCER J, 1966. Rapid transit system basic design standards - 2nd Edition; Runcorn Development Corporation. NATO, 1976. Bus priority systems. CCMS Report No45; NATO Committee on the Challenges of Modern Society. Crowthorne: Transport and Road Research Laboratory. NELSON J D and P J HILLS, 1990. Innovative bus control for congested urban corridors: the application of convoying systems Traffic Engineering & Control Vol 31(5). PRETTY R L and D J RUSSELL, 1988. Bus boarding rates. Australian Road Research 18(3) pp145-152. September 1988. RATP, 1977. Autobus en site propre. Paris: Regie Autonome des Transports Pansiens. SZASZ P A, L de C MONTANS and A R GODOY. COMONOR combois de onibus ordenados nas Avenidas Rangel Pestana a Celso Garcia. CET Technical Bulletin No 22. Sao Paulo: Companhia de Engenharia de Trafego. TANABORIBOON Y, K K CHIN and H C CHIN, 1986. Performance of bus lane in Singapore - a case study. Paper presented to 13th ARRB/5th REAAA Conference TRB, 1975. Bus use of highways: planning and design guidelines. Report 155, Transportation Research Board, National Research Council. Washington, DC: Transportation Research Board. TRRL, 1987. Field survey techniques and analysis for urban bus operators. Overseas Road Note 4. Crowthorne: Transport Research Laboratory. TRRL, 1988. A guide to geometric design. Overseas Road Note 6. Crowthorne: Transport Research Laboratory. TRRL, 1990. Busway transit video. TRL Video V254. Crowthorne: Transport Research Laboratory. URBAN EDGE, 1991. Busways gain ground. The Urban Edge Vol 15(4). Washington, DC: World Bank. VINCENT R A, R E LAYFIELD and M D BARDSLEY, 1976. Runcorn busway study. TRRL Laboratory Report LR 697. Crowthorne: Transport Research Laboratory. VOVNDA, 1982. Bus transit system. Verban offentlicher Verkehrsbetnebe (VOV)Nerband der Automobdmdustne e. V. (VDA).

38

VUCHIC V R, 1981. Urban public transportation systems and technology. New Jersey. Prentice-Hall Inc.. WAYTE F A, 1988. Cost comparisons of busway, light rail and guided bus. In Proceedings of the International Seminar on Guided Bus Rapid Transit, Adelaide, October 1988. WHITE P, 1986. Public transport: its planning, management and operation. London: Hutchinson. TRANSPORT RESEARCH LABORATORY

Printed in the United Kingdom for HMSO Dd8371863 12/93 C10 65600 10170

39

Design guidelines for busway transit

ORN 12

Transport Research Laboratory Old Wokingham Road Crowthorne, Berkshire, RG45 6AU

Overseas Development Administration 94 Victoria Street London, SW1E 5JL

OVERSEAS ROAD NOTE 13

THE USE OF TRAFFIC SIGNALS IN DEVELOPING CITIES

Main Subject area: Theme: Project title: Project reference:

Urbanisation/Transport, energy efficiency Urban Transport The Use of Traffic Signals In Developing Cities ODA R6016

Copyright Transport Research Laboratory 1996. All rights reserved. This document is an output from a project funded by the UK Overseas Development Administration (ODA) for the benefit of developing countries. The views expressed are not necessarily those of the ODA.

Transport Research Foundation Group of Companies Transport Research Foundation (a company limited by guarantee) trading as Transport Research Laboratory. Registered to England, Number 3011746. TRL Limited Registered in England, Number 31.12272 Registered Offices Old Wokingham Road, Crowthorne, Berkshire, RG45 6AU.

ACKNOWLEDGEMENTS This Note was prepared by A. Cannell of Transcraft Consultants, Curitiba, Brazil and G Gardner of the Overseas Centre, Transport Research Laboratory (TRL). Useful advice and assistance was given by D. Singh and J. Cracknell. First Published 1996 ISSN 0951-8987

OVERSEAS ROAD NOTES Overseas Road Notes are prepared principally for road and transport authorities in countries receiving technical assistance from the British Government. A limited number of copies is available to other organisations and to individuals with an interest in roads overseas, and may be obtained from: Transport Research Laboratory Crowthorne, Berkshire, RG45 6AU United Kingdom

Limited extracts from the text may be reproduced provided the source is acknowledged. For more extensive reproduction, please write to: Programme Director, Overseas Centre, Transport Research Laboratory.

CONTENTS
Page Page 1. INTRODUCTION 2. CRITERIA AND WARRANTS FOR SIGNAL INSTALLATION 3. BASIC TRAFFIC COUNT SURVEYS 4. JUNCTION DESIGN AND LAYOUT Typical Layouts Siting of Signal Equipments Approaches and Lanes Signal Sequences Signal Design Techniques Right Turning Vehicles Early Cut Off Late Start Pedestrian Facilities No Pedestrian Signal Full Pedestrian Stage Parallel Pedestrians Staggered Pedestrian Facility Pedestrian Signal Displays Pedestrian Push Buttons. Audible Warnings Guard Rails Pedestrian Signal Sequences and Timings Vehicle-actuated (V.A.) Traffic Signals Stage demands Stage extension Semi-vehicle-actuated signals Traffic Signals on High Speed Roads Speed-related Green Extensions Visibility Requirements Bus Priority 5. CALCULATION OF TRAFFIC SIGNAL TIMINGS - WEBSTER'S METHOD Cycle Intergreen Period Minimum Green Period Estimation of Saturation Flow Width of Approach Gradients Traffic Composition Turning Traffic Parking, Waiting and Bus Stops 16 16 16 16 16 16 17 17 17 18 12 12 13 13 13 13 13 13 13 10. GLOSSARY 11. REFERENCES 34 36 2 4 6 6 6 6 7 9 10 10 10 10 11 11 11 11 11 12 12 12 1 Time Settings Total Lost Time per Cycle (L) Flow Factors Cycle Times Green Times Degree of Saturation Junction Capacity Analysis Traffic Signal Calculation Sheet Stage/Phase Sequence Diagram Check List for Signal Design 6. COORDINATION AND LINKING OF TRAFFIC SIGNALS Simple Progressive System (Green Wave) Mechanisms for Linking Signals Cable-Linking Cable-less Linking Fixed Time Coordinated Signals Area Traffic Control (ATC) Fixed Time ATC Systems Semi-Responsive Systems Fully Responsive Control Equipment Testing 7. THE TRANSYT PROGRAM 8. SYSTEM AND ECONOMIC ANALYSIS OF TRAFFIC CONTROL 9. SPECIFICATIONS Traffic Signal Controllers General Road Traffic Signals Inductive Loop Detectors Associated Electrical Works Traffic Signal Controller Civil Works 30 32 32 32 32 33 33 23 25 25 25 25 25 26 26 26 2 7 23 18 18 18 19 19 20 20 20 20 20

THE USE OF TRAFFIC SIGNALS IN DEVELOPING CITIES

1. INTRODUCTION
1.1 A traffic signal installation is a power-operated device which informs motorists or pedestrians when they have the right of way at a particular intersection. 1.2 The first traffic signal was installed in London in 1868 and used semaphore 'arms' together with red and green gas lamps. Unfortunately, it exploded, putting an end to this sort of control for 50 years. 1.3 However, in 1918 the first three coloured light signals were installed in New York and in 1925 they started to be used in Great Britain. 1.4 At the beginning of the 1930's an attempt at making the signals more `intelligent', or vehicle responsive, was tried in America, using microphones at the side of the road, requiring drivers to sound their horns. This was obviously not too popular and the first traffic detectors - electrical and pneumatic - were invented. 1.5 Traffic signals are now used throughout the world, using the three light signals of Green, red and amber. Also, by convention, these are normally arranged vertically with the red signal at the top and the green light at the bottom. This also helps people who are colour blind - both drivers and pedestrians - to identify the differences between the lights. 1.6 Traffic signals are used at intersections to reduce conflicts to a minimum by time sharing of right of way. This actually reduces the capacity of the intersection, but greatly enhances safety. 1.7 Conflicts at intersections are illustrated in Figure 1 which shows the potential conflict points at the junction of two roads, both with two way traffic flows, at which all crossing and merging movements are permitted. 1.8 With the provision of traffic signal control the number of potential conflicts can be reduced from 64 to zero. 1.9 The object of this report is to give traffic engineers or technicians in the cities of the emerging world a brief introduction to traffic signals, together with some practical guidelines on how to use them to obtain good and safe results.

Figure 1 Conflict points at an intersection 1.10 There is no doubt that signals are one of the most powerful tools for urban traffic control available to city authorities and their correct installation can improve both traffic flow and the safety of all road users. In comparison to other traffic improvements, signals are also relatively low capital intensive and in recent years the advancement in informatics and telecommunications has led to a new generation of low cost controllers and systems that have made modern signalling an even more attractive and powerful tool. 1.11 Essentially, traffic signals form part of the "software" of a city as opposed to the roads and bridges that are part of it's "hardware". As such they have the advantage of being cheap and often the disadvantage of being so cheap that no local lobby is interested in them, especially when city mayors fail to see the political advantages in changing an old signal for a new one. 1.12 It is thus part of the traffic engineer's task to prove to city authorities that a modern and well designed traffic signal system will bring real and visible benefits to the city.

2.

CRITERIA AND WARRANTS FOR SIGNAL INSTALLATION

where a schoolchildren crossing is present; where there is a need to maintain progressive movement of vehicles along an otherwise signalled route; and where there is a record of accidents of the type which could be reduced by the use of traffic signals.

2.1 When two or more traffic flows are competing for the same road space at a junction, some form of control - or set of rules - is needed to minimize delays and the risk of serious accidents. In some countries, a simple rule of preference states that the traffic coming from the left (or right where there is right-hand drive) has priority to enter the junction. As few people tend to know - or obey - this rule, unsignalled junctions can come under "popular control" and users have to consider that the larger vehicle, or the one that sounded the horn first, or a public transport bus, etc., may have priority. 2.2 This is obviously inefficient and dangerous, so with higher flows some form of stop or priority sign is used to inform to the user on one or more approaches that the other road has right of way. At even higher flows this form of control breaks down when the delay on the minor road becomes too high, forming queues and forcing drivers to run the risk of accepting gaps in the major road traffic that are too small for a safe crossing. At this point, time must be allocated for the rightof-way to traffic on the various approaches. 2.3 However, the introduction of traffic signals (or lights) into a city often runs the risk of these equipments being considered a panacea for all traffic problems. The engineer or technician in charge of the traffic comes under political and popular pressure to install too many signals, thus leading to the even greater risks of red-running - as the users `learn' to disrespect the red lights that they consider to be unnecessary. 2.4 To avoid this problem it is essential that the engineer or traffic department has a clear set of warrants to justify the use of signals. 2.5 If possible, these warrants should be approved by the local government bodies (elected and executive) so that requests for signals on sites that do not need them can be refused according to pre-discussed rules - and not just on the personalized decision of the head of the traffic depart-ment. 2.6 Traffic signals may be justified if, usually two, of the following criteria are present: where there is a minimum major-street/minor-street conflicting vehicle volume; where there may be need to interrupt continuous flow on the major road to allow traffic to exit from the minor road without excessive delay; where a minimum pedestrian volume conflicts with a minimum vehicle volume;

2.7

A rough and ready set of warrants might be:

traffic flows - when there is a minimum of 1000 pcu's per hour entering the junction during the peak hours. visibility - when drivers on the minor road have poor visibility for judging gaps. accidents - when three or more accidents (collisions or pedestrians) are registered per year. 2.8 Figure 2, for example, shows the relationship between major-road/minor-road flows and the type of control recommended at a junction in the UK. For a major road flow of 20,000 pcu's per day and a minor road flow of 6,000, a roundabout would be a good solution for eliminating the conflicting traffic movements - if space were available. If, however, the junction is in a built-up area, then traffic signals probably represent the best solution. 2.9 It should be stressed, however, that traffic signals if located or timed wrongly can INCREASE delays and accidents and their maintenance and electrical supply represents an ongoing cost of around US$1000 to 2000 per year. 2.10 To minimize the need for signals, the road hierarchy should try to conform to the network shown in figure 3, which offers the most efficient and safe layout.

Figure 2 UK practice for intersection control selection based on combinations of traffic flow

2.11 A method of reducing conflicts on local distributors and access roads is to physically separate traffic flows, allowing access but avoiding the pressure to install new lights.

2.12 Figures 4 and 5 show how, in some cases, conflicting flows may be avoided - provided that no economical or environmental restrictions exist. 2.13 If, however, traffic lights are to be installed, the engineer and police forces should be in agreement on how the flows are controlled. In many developing cities, the police will often take manual control, assuming that they can reduce traffic queue lengths. Research has shown that this is not true (Walker et al, 1988). Police are reluctant to stop a traffic stream even when it is no longer saturated, as shown in figure 6. It is preferable to allocate police to control illegal parking, removal of breakdowns and enforce driver behaviour.

Figure 3 Ideal urban road network

Figure 5 Eliminating the need for traffic signals -"7 esquinas", Arequipa, Peru

Figure 4 Elimination of conflicts at a junction pair

Figure 6 Typical flow/saturation relationship for police control - inefficient use of the end of the green period

3.

BASIC TRAFFIC COUNT SURVEYS

3.1 For each site where traffic signals are being contemplated it is fundamental to obtain adequate data on the traffic flows at the junction. Normally, surveys would be carried out during the peak hour periods. However, it may be important to have a broad view of the flows in the city throughout a normal working day, especially when Area Traffic Control or linked signalling are being considered. 3.2 Traffic counts are likely to be divided into two types: all day counts (normally during 16 hours of a work day) usually mid block on key roads, with the objective of defining the duration of the peak periods and general vehicle composition; and, specific junction counts carried out with the objective of providing the data for evaluation and design of the junctions.

3.3 The classification of vehicles might be cars, taxis, light vans, trucks (heavy and medium) and public service vehicles. In some cities it will be necessary to include motorcycles, cycles or other common vehicle types. The counts should be made in periods of about 15 minutes, during at least two working days. If the counts are not similar (as demonstrated in figure 8), then the counts should be repeated on another working day. A simple 16 hour survey form could look like figure 7. 3.4 Specific junction counts are aimed at providing the data for detailed evaluation and design. The peak periods can be identified from the all day (16h) counts and the junction counts should be undertaken in the peaks - including the "shoulders" just before and after the peaks. Unless a city is subject to excessive congestion, this usually means a count period of about two hours for each peak. If an ATC scheme is under consideration, counts should also be carried out at weekends. 3.5 Each surveyor can usually manage to count two independent flows. For a simple junction involving two one-way streets, two surveyors (normally temporary staff) will be needed, as shown in figure 9. 3.6 Each site should also be carefully checked to make sure that pedestrian volumes during the peak hours that might require special phases are also considered. 3.7 Counts in congested areas often suffer from the spillback of upstream queues which means that surveyors will not count the real demand of the traffic that wants to go through the junction, but only the traffic that actually manages to pass. This can lead to the classic case, in which

Figure 7 Simplified traffic survey form

a survey is made during a widespread "gridlock"; reported by the surveyors in terms of near zero flow on all approaches. 3.8 TRL ORN 11, "Traffic Surveys in Developing Cities" should be consulted for further reference. 3.9 The warrants used and/or approved by the city to justify the installation of signals are likely to include accidents. It must be stressed that an updated accident data base is essential for completing the traffic surveys.

Figure 8 16 hour traffic count on Peru Street, Mendoza, Argentinia, During two working days

Figure 9 Survey forms for a simple junction of two one-way streets

4.

JUNCTION DESIGN AND LAYOUT

right of way first. The secondary signal in this case should not be placed beyond the nearside of the junction. Approaches and lanes

4.1 The aim of any junction layout is to provide for the safe movement of traffic, both vehicular and pedestrian, without undue delay or congestion. Various alternative layouts may be considered and the ultimate choice will be governed by such factors as the nature and volume of traffic using the junction, the availability of land and the cost. 4.2 The overall capacity of a road network is limited by the capacity of individual junctions. Failure to provide the correct type of layout at one particular junction may result in accidents, congestion and delay to an extent which may impair the efficiency of the road system over a wide area.

4.7 It is essential that approaching drivers are made fully aware of the nature of the junction by adequate signing. Carriageway markings and/or channelized islands should be used to guide users on the correct path, and visibility should not be impaired. 4.8 Approaches should be marked out in lanes. Lane widths at signalled junctions should normally be between 3 and 3.6m, although 2.7m is acceptable in some instances where speeds are low and there are few large vehicles (trucks or buses). 4.9 On roads where land is available the saturation flow and capacity of an approach can be increased by widening the road to the vicinity of the junction to provide more ahead lanes. An example of this is shown in figure 10. Another option, where there are large turning movements is to divide the road space available to favour the turning lanes, as shown in figure 11. 4.10 Perhaps the most important factor affecting the capacity of a junction approach is the need to avoid obstruction to traffic flow, either temporary (a taxi or bus stopping for passengers) or permanent (a parked car). Plate 1. clearly shows the problem caused by a (very) long term parked car which has eliminated a lane of traffic. In a situation such as the example in plate 2, even the most sophisticated traffic signals will not improve the traffic flow.

TYPICAL LAYOUTS
4.3 The following descriptions of junction layout and design procedures are based mainly on UK practice. Other standards are of course possible. For example, in the UK signals are located on the kerb, at the roadside with the "primary" signal close to the stop line. In many countries overhead signals on the "far side" of the junction are the norm. Both methods have their merits, however, a country will generally have it's own standards and such standards have to be adopted in designs. The important requirement is that signals should be consistently designed, located and operated throughout the city and clear unambiguous indications given to all road users.

SITING OF SIGNAL EQUIPMENTS

4.4 The minimum requirement is one traffic signal in-stalled I m from the stopline, on the nearside of the carriage-way. If possible a second primary signal should be installed if there is a central island or divider, or more than three approach lanes. Minimum visibility distances from the primary signals are given in Table 1 TABLE 1: VISIBILITY DISTANCES

85 percentile approach speed 50 km/h 60 km/h 70 km/h 85 km/h 100 km/h

visibility distance (m) 70 95 125 165 225

Plate 1. Parked car obstructing the approach - a severe capacity loss

4.5 A secondary signal is normally installed diagonally opposite the first primary signal, as shown in figure 12. 4.6 When the signal method of control contains a special right turn phase, extreme care should be used in the siting of secondary signals for the direction of flow which loses

Figure 10 Flared junction approach TABLE 2: APPROACH LANE WIDTHS

Figure 11 Extra road space given to approaches

Approach width (m) Lane 1 3-5 5.50 6.00 8.00 8.50 10.00 11.50 3.50 2.75 3.00 4.00 3.00 3.40 3.10

Lane width (m) Lane 2 Lane 3 Lane 4

2.75 3.00 4.00 2.75 3.30 2.80

2.75 3.30 2.80

2.80

Lane 1 is nearest the kerb

SIGNAL SEQUENCES
4.11 Each signal face normally has three vertical lights with a nominal diameter of 200mrn. The height of the centre of the green lens from the surface of the carriageway (where light signals are placed at the side of the carriageway) should be not less than 2.1 metres nor more than 3.5 metres. If signals are placed over the carriageway, this distance should not be less than 5.0 metres nor more than 9 metres.

4.12 Traffic control is by means of red, amber and green signals, supplemented by additional green arrow light signals, tram signals, etc. 4.13 The signal sequence at junction traffic signals in British practice countries is red, red + amber and green, amber and red. Most Panamerican standard countries, however, use the sequence red, green, amber and red and some countries adopt other variations, eg. flashing green in place of amber.

Figure 12 Typical layout of a signalled controlled junction

steps in which the junction control is varied. The other, phase control, refers to the periods of time allocated to each traffic stream. 4.22 In UK practice a phase is used to describe a set of traffic movements which can take place simultaneously or the sequence of signal indications received by such a set of movements. A stage is that part of the cycle during which a particular set of phases receives a green indication. 4.23 In USA based practice, a phase is that part of a cycle allocated to any combination of traffic movements receiving the right-of-way simultaneously during one or more intervals. An interval is a period of time during which all signal indications remain constant. 4.24 The cycle is the complete series of stages during which all traffic movements are served in turn. The cycle time is the sum of each of the stage times.

Plate 2.

Street markets: a safety risk as well as a huge capacity restraint

SIGNAL DESIGN TECHNIQUES

4.25 Conflicts are reduced at signal controlled junctions by holding certain traffic streams stationary while others are allowed to pass. To hold all streams and release each in turn would remove all conflicts but would not be satisfactory since delays to all traffic would be high and effective capacity of the junction would be low. 4.26 The art of designing an installation is to reduce delay and increase capacity while still maintaining a high degree of safety. 4.27 Reduction in total delay and improvement in capacity can be achieved by: utilizing the lowest practicable number of stages in any signal cycle. ensuring that each approach is capable of carrying the maximum predicted traffic flow for that approach. ensuring that the time allotted to each stage is appropriate to the actual traffic flow. if appropriate, coordinating the control of adjacent junctions to maintain the flow of traffic `platoons'. allowing simultaneous non-hooking right turns. separating left turn movements with an exit lane controlled only by a "give-way" priority sign. where the degree of conflict is acceptable and movements can be executed safely with the exercise of due care, a conflicting move may be accepted (e.g. a right turn on full green). restriction of movements, e.g. banned right turns, where conflicting manoeuvres are forbidden. separation of traffic streams which conflict, assigning them to different stages.

4.14 the red light signal indicates the prohibition that vehicular traffic shall not proceed beyond the stop line provided in conjunction with the light signals, or if the stop line is not visible (or there is no stop line), beyond the light signals. 4.15 the amber light signal when shown alone, indicates the prohibition that vehicular traffic shall not proceed beyond the stop line, or if the stop line is not visible (or there is no stop line), beyond the signals, except in the case of any vehicle which when the light signal first appears is so close to the stop line or light signals that it cannot be safely stopped before passing the stop line or light signals. The time for the amber signal is normally fixed for the city or region at 3 or 4 seconds. 4.16 the red and amber light signals together indicate an imminent change from red to green. However the red light still prohibits forward movement. 4.17 the green light signal means that traffic may proceed, if safe to do so. 4.18 the green arrow signal indicates that traffic may proceed only in the direction indicated by the arrow. 4.19 a flashing amber signal in some countries means that drivers must proceed with caution. Normally displayed on all approaches with a frequency of 1 hertz (1 flash per second), this signal is sometimes used from midnight to 4 or 5 o'clock in towns with notorious night time red-running. 4.20 Pedestrian signals are red and green, either with a green walking man and a red standing man, or with "WALK/ DON'T WALK" signs. 4.21 There are two alternative concepts used in describing the control of traffic by means of light signals. One, known as stage control, is concerned with the sequential

considering different stage sequences for different times of the day. providing extra lanes for turning traffic or flares on junction approaches. combining the green periods for vehicles and pedestrians when this can be done safely. providing two separate green periods in a cycle (repeated greens) for important movements.

4.33 The approach which is permitted to flow over two stages should have a three light primary signal. The secondary signal, placed beyond the junction, should have four lights, including a right turn arrow of 300mm diameter (in addition to the full green signal) illuminated on the second stage when the opposing traffic has been signalled to stop, as shown in figure 13 and 14.

4.28 As an example of these principles, figure 12. shows a four arm junction with two stages with all movements permitted. This is a very common junction and two stage operation forms the basis of signalling techniques. Traffic on opposite arms flows simultaneously, while traffic on the other two arms is stopped. Each arm may have one or more lanes on approach but the right turning traffic may impede vehicles wishing to proceed over the junction if the road width is restricted. Where there is a relatively minor right turn flow the capacity of the junction is reduced by the road space occupied by such traffic waiting to turn right and by the time which has to be provided to this movement in the cycle. If the right turn manoeuvre is removed then reduced delay and improved capacity can be expected. An alternative route may often be indicated to traffic before the junction is reached. Usually motorists can turn left before the junction, make two right turns to appear at the junction on the left hand arm (known as a `g' turn). Alternatively motorists can pass through the junction, turn left and make two further left turns to appear at the junction on the left arm (known as a `q' turn). Such "q" and "g" turns should be carefully evaluated as there will be increased costs to set against savings injunction delay. In the case of "q" turns, the use of the junction twice by former right turn traffic may adversely affect junction capacity and thus delays and operating costs.

Figure 13 Early cut off stage sequence

RIGHT TURNING VEHICLES

4.29 The usual practice is for opposing right-turners to turn on the nearside of each other. With this arrangement locking of turning movement cannot occur but driver visibility may be restricted. 4.30 On high speed roads or where right turning movements are heavy (above 300 pcu's/h), separately signalled and segregated lanes are strongly recommended. 4.31 Another very common situation is the four arm junction with three stages. The types of control are known as either early cut-off or late start.

Figure 14 Green filter arrow for right turn

LATE START
4.34 An alternative way of dealing with right turning traffic is to delay the start of the opposing traffic by a few seconds. This method causes difficulty at the start of the following stage if the right turn flow is heavy and the opposing traffic cannot establish precedence. For this reason a late start stage is usually not recommended. 4.35 When both right turn movements are heavy, another option available is to hold both right turns with a red signal while the ahead and left turn traffic flows unhindered. All traffic is then stopped before the right turn traffic on both approaches is released together on the same stage. It is usual to separate the right turn traffic onto exclusive lanes with separate signals on each approach. This method should be employed on high speed roads.

EARLY CUT OFF

4.32 To facilitate a heavy right turn movement from one approach, the green time of the opposing approach can be cut off some seconds before the approach with the right turn.

PEDESTRIAN FACILITIES
4.36 When a traffic signal installation is being designed or modified, the nature and extent of pedestrian flow has to

10

be taken into account as well as that of vehicular traffic. The object of providing pedestrian facilities is to assist pedestrians to cross in safety, with the minimum delay to both pedestrian and vehicular traffic. 4.37 There area number of alternative methods of achieving this aim and the engineer has to consider which of these methods can be best applied to individual sites, knowing the pedestrian flow pattern, degree of saturation and site layout. 4.38 Each junction should be considered on its own merits, taking into account factors such as infirm or handicapped pedestrians, junction capacity and any available accident statistics. 4.39 If full pedestrian stages are new to the local traffic culture, great care should be taken to introduce them only when accident data and high pedestrian flows justify their need.

NO PEDESTRIAN SIGNAL
4.40 The presence of traffic signals at an intersection provides assistance to pedestrians in crossing the arms of a junction, especially where refuges are available, and in many cases no further facility is necessary. An extended all red period between two traffic stages to assist pedestrians is not recommended. This practice leads to increased delays to traffic and to driverdisobedience since the extended period will always be present even when there are no pedestrians. Figure 15 Full pedestrian stage

FULL PEDESTRIAN STAGE

4.41 With this facility, all traffic is stopped while pedestrian movement is signalled across all arms of the junction. This method will cause delay to traffic. However, the stage can be programmed only to operate during certain hours or by demand from push buttons. Where the crossing is across a dual carriageway, additional push buttons on the central reserve should also be considered. 4.42 Although pedestrians may be allowed to cross any of the approaches to an intersection there will usually be one approach upon which the pedestrian problem is most acute. The pedestrian stage should immediately follow the end of the vehicle stage on this approach. The signal sequence should be arranged to ensure that on termination of the pedestrian period, the right of way will revert to a nominated stage in the absence of other demands. This is shown in figure 15.

STAGGERED PEDESTRIAN FACILITY

4.44 Where carriageway widths permit, a large island in place of the normal refuge may be provided. Pedestrians can negotiate one half of the carriageway when traffic on that approach is held on red at the junction signals. Normal pedestrian signals are shown during this period. The other half of the road is controlled by separate signals which are located at the opposite end of the island. Normally the stagger should be at least one crossing width in order to alert pedestrians that the crossing is in two sections. A right-hand stagger may reduce junction intergreen times by placing approach stop lines closer to a junction. A left-handed stagger, as shown in figurel6, is normally preferred as pedestrians stepping on to the central refuge will turn towards the approaching traffic stream.

PEDESTRIAN SIGNAL DISPLAYS

4.45 Normally each signal face has two lights arranged vertically (the upper red standing man and the lower green walking man) of 300mm nominal diameter. An alternative size of 200mm nominal diameter may be used when specified. 4.46 The red stationary man, when illuminated by a steady light, indicates to a pedestrian that he should not cross or start to cross the carriageway at the crossing.

PARALLEL PEDESTRIAN STAGES

4.43 Where it is possible to prohibit permanently some turning movements a combination of pedestrian and vehicle stages can be installed. By virtue of banned turns, pedestrian facilities can be provided across appropriate arms. In order to reduce the possibility of vehicles turning illegally, kerb radii should be kept as low as possible.

11

tone for the green walking man period and an intermittent tone for the flashing green period. 4.51 Each proposal for use of audible signals at junctions should be considered on individual merits and carefully checked against real demand, safety aspects and potential risks, technical feasibility of the equipment or supplier, local layout and environment (these signals are not popular with nearby residential blocks of flats). 4.52 An additional benefit to the visually handicapped can be given by fixing metal plates with the street names in Braille onto traffic signal posts in the vicinity of schools or other buildings frequently used by them.

GUARD RAILS
4.53 It is desirable in some cases to restrict the crossing of pedestrians to certain approaches at an intersection and guard rails can be used to prevent pedestrians crossing at dangerous places (for example where filtering traffic may be moving at times unexpected by pedestrians). Guard rails should always be provided on large islands where staggered pedestrian movements are allowed. Normally minimum length of guard rails provided at each side of a crossing should be 15m.

PEDESTRIAN SIGNAL SEQUENCE AND TIMINGS

Figure 16 Left-handed stagger stage sequence 4.47 The green walking man signal, when illuminated by a steady light, indicates to a pedestrian that he may cross the camgeway at the crossing. 4.48 The green signal, when illuminated by an intermittent light (flashing green man) indicates that a pedestrian who is already on the crossing should proceed to complete the crossing with reasonable speed; and/or a pedestrian who is not already on the crossing should not start to cross. 4.54 Pedestrian time should be sufficient to enable pedestrians to cross the full width of the road with relative ease at normal walking speed. An assumed walking speed of 1.2 m/s for the measured crossing distance is satisfactory in determining the minimum times. A staggered crossing can be considered as two separate crossings. 4.55 Normally, minimum green periods of less than 5 seconds are considered too short and are not recommended. 4.56 Provided that the above minimum requirements are met, the green period of a parallel pedestrian stage may be determined by the predominant traffic flow running in parallel. 4.57 The vehicle clearing times before the start of all pedestrian stages should be checked to ensure that the last vehicle clears the crossing by the time the pedestrian green signal is lit. A summary of pedestrian facilities is given in table 3.

PEDESTRIAN PUSH BUTTONS

4.49 Pedestrian push buttons units mounted on signal posts may be used for calling up pedestrian stages. Additional push buttons are also necessary on wide refuges where pedestrians may be trapped at the end of the pedestrian stage. It is advisable to have push buttons located at each side of the pedestrian crossing, so that pedestrians approaching from either direction can pass a push button before reaching the crossing.

AUDIBLE WARNINGS
4.50 Audible warnings, in the form of pulsed tones, are intended for the benefit of visually handicapped pedestrians. The set up consists of a post-mounted audible device which emits different patterns of audible signal, representing different pedestrian signal indications e.g. a slow hammering tone during the red standing man period, a quicker

VEHICLE-ACTUATED (V.A.) TRAFFIC SIGNALS

4.58 With vehicle -actuated (VA) signals the duration of the green periods and the cycle time will vary in relation to the traffic flow into and through the controlled area. A vehicleactuated signal responds to demands recorded for

12

the various directions of flow. Once a green has been given to a particular direction of flow, the length of green for that direction will be extended until all the traffic has passed through the junction, or the maximum green time for that direction has been reached. 4.59 Vehicle actuated signals will be most appropriate for isolated junctions where coordination with other signals is not important and for locations with fluctuating light or medium traffic flows.

TRAFFIC SIGNALS ON HIGH SPEED ROADS

4.65 When traffic signals are installed on roads where the 85 percentile approach speed at a junction is between 60 km/h and 105 km/h on any arm, drivers have a difficult decision to make when green changes to amber: they are often faced with a choice between attempting to brake to a halt at the stop line, or continuing at the same speed through the junction and clearing it safely. 4.66 They may fail to achieve either, thus putting themselves and others at great risk. 4.67 Because of the increased braking distances required at high speeds, drivers need adequate warning that they are approaching a signalled junction. High approach speeds also result in drivers misjudging the lengths of gaps in opposing traffic when making a right turn at the junction -again leading to increased risk. 4.68 On high speed roads, the use of right turn clearance phases should be avoided. Right turning movement, across high speed flows should be channelized and controlled with a separate vehicle phase, or preferably banned.

STAGE DEMANDS
4.60 On the approach to a red signal, a green signal will be demanded on the arrival of a vehicle on that approach. This demand is stored in the controller which will serve stages in cyclic order omitting any stages for which no demand has been received. Where it is essential that one stage must always follow another, the appearance of the first stage will automatically insert a demand for the second stage. 4.61 When a stage loses right of way on a maximum green period change, then a demand is inserted for a reversion to that stage after other demands have been met.

STAGE EXTENSION
4.62 When a green signal is displayed, the period for which it is displayed may be extended by vehicles detected moving towards the signal. The purpose of this extension, or the sum of several extensions, is to permit vehicles to pass the stop line before the maximum green period is reached.

SPEED-RELATED GREEN EXTENSIONS

4.69 To assist drivers and minimize risk it is necessary to provide green extensions, the extensions being related to the 85 percentile approach speed. Normal approved vehicle detection equipment is used within 40m of the stop line on each approach and in addition approved speed discrimination or speed assessment equipment can be used. 4.70 Advance warning signs are necessary on each approach, according to local or regional standards. 4.71 When the 85 percentile approach speed on any arm exceeds 105 km/h it is recommended that traffic signals should not be installed.

SEMI-VEHICLE-ACTUATED SIGNALS
4.63 With some semi-vehicle-actuated signals, detectors are installed on the side roads only (i.e. not all approaches) and the right-of-way normally rests with the main road, being transferred immediately or at the end of a preset period to the side road when a vehicle passes over the side road detector. The green period on the side road can be extended in the normal way by successive demands up to a preset maximum. After right-of-way has been returned to the main road, it cannot be taken away from the main road until the preset period has expired. 4.64 Another modified form of V.A. signals is to operate one or more demand-dependent stages within a fixed cycle time. The demand dependent stages which may consist of vehicle phases (such as right turn traffic, minor flows) or pedestrian phases may be slapped or extended in accordance with the prevailing situation detected. The advantage of this type of control is that a fixed cycle time can be maintained for linking with surrounding controllers.

BUS PRIORITY
4.72 The great majority of passengers in the cities of the developing world travel by bus. Although these road users normally have less political influence than the more affluent car owners, the traffic engineer should consider how to improve bus flows at signalized junctions. 4.73 The simplest form of priority is to guarantee that saturation on the approaches most used by buses is kept as low as possible, even if this means additional waiting times for the other stages.

13

Figure 17 Phase and stage sequence for early cut off operation

4.74 In ATC systems, the TRANSYT program (see section 7) permits bus flows to be treated separately thus providing optimum settings for buses. 4.75 rity to buses, not necessarily within ATC systems has been achieved at traffic signals by a number of methods. These include: the selective detection of buses using on-bus transponders and detectors in the approaches to signals; the use of segregated lanes, exclusively for buses on approaches to junctions, within which detectors are installed to actuate the signals; and

- the use of pre-signals on the approaches to junctions. These enable traffic queues to be relocated upstream of the junction and control traffic and bus flows to an advance area so that all vehicles are able to clear the junction. (TRL ORN 12, 1993).

14

TABLE 3: SUMMARY OF PEDESTRIAN FACILITIES

Type of facility No pedestrian signal -

Characteristics Traffic signals, even without signals for pedestrians, can help pedestrians to cross by creating gaps in traffic streams. Especially applicable where there are refuges and on one-way streets. All traffic is stopped. Demanded from push buttons. More delay to vehicles than combined vehicle/pedestrian stages. Combined vehicle/pedestrian stage often accompanied by banned vehicle movements. Useful across one-way streets. Pedestrians cross one half of the carriageway at a time. Large storage area in the centre of the carriageway required. Stagger preferably to face on-coming traffic. For junctions close to capacity. The crossing point is situated away from the junction but within 50m. Normal staging arrangements as above apply.

Full pedestrian stage

Parallel pedestrian stage

Staggered pedestrian facility

Displaced pedestrian facility

15

5.

CALCULATION OF TRAFFIC SIGNAL TIMINGS - WEBSTER'S METHOD

distance 'x' should be determined from the position of the pedestrian crossing. Where pedestrians are losing right-of-way the start of the following stage should be delayed until the crossing area is clear.

CYCLE
5.1 A complete series of stages during which all traffic movements are served in turn is known as a cycle. The cycle time is the sum of each of the stage times.

MINIMUM GREEN PERIODS

5.8 Minimum Green Periods cannot be overridden by any demands, whether emanating from vehicles, manual control devices or received remotely from central computers or linked controllers. Such a period is built into signal controllers. The shortest minimum green period normally used for vehicle stages is six to eight seconds. 5.9 Site conditions may require a longer period where large numbers of heavy vehicles have difficulty in starting, or the approach is on a steep gradient. 5.10 Where pedestrians and traffic share the same stage, minimum green times may be governed by the time required by pedestrians to clear the crossing.

INTERGREEN PERIOD
5.2 The period between the end of the green display on one phase and the start of the green display on the next phase gaining right-of-way is known as the intergreen period. It comprises an amber display, red + amber display and may also contain a period when the red signals are shown to all approaches simultaneously. In some countries thus intergreen period is composed solely of an amber signal and an all red period. With a five second intergreen the amber and red + amber periods occur consecutively. Any period over five seconds will include a period where red signals are shown to all approaches simultaneously. (an all red period). Safety requirements may dictate a longer period to be given in the following circumstances to allow vehicles to clear the intersection when the distance across the junction is excessive. to improve safety on high speed roads.

ESTIMATION OF SATURATION FLOW WIDTH OF APPROACH

5.11 The Road Research Technical Paper No. 56 suggested that the Saturation Flow (S) be expressed in terms of passenger car units (pcu's) per hour and with no turning traffic or parked vehicles; S = 525w ,

on roads where there are insufficient numbers of right-turning traffic to justify provision of a separate stage.

where w is the width of the approach road in metres and 5.15<w <18.3m. 5.12 For widths less than 5.15 m the following values may be used: w(m) S(pcu/h) 2.70 1790 3.00 1850 4.00 1950 4.50 2250 5.15 2700

5.3 It should he noted that an intergreen period which is too short will be potentially dangerous but a period which is too long is equally unsatisfactory since it may lead to delay, frustration and lack of observation by drivers. A guide to determining the length of the intergreen period is illustrated in Table 4. 5.4 A vehicle which passes over the stop line at the start of the amber display must be clear of the potential collision point in relation to a vehicle starting at the onset of the green of the following stage, when travelling at the normal speed for the intersection. The distances AF and BF should be determined and those distances which give the highest difference used The recommended intergreen period can then be determined. 5.5 When East-West arms are losing right-of-way if AFBF is greater than CH-DH, then 'x' = AF-BF (or vice versa). 5.6 When North-South arms are losing right-of-way if DE-AE is greater than BG-CG, then x'= DE-AE (or vice versa). 5.7 When the following stage is a pedestrian stage the

5.13 Research and experience has shown that S may be more accurately reflected by the number of lanes rather than the overall width of the approach - which tends to underestimate flows in situations where narrow lanes of, say, 2.70m are used. 5.14 The correct value of S can be checked against the formula: S = Si, (approx. 1790n - 100) where n = max. no. of lanes (min 2.70m) and Si is the value of S for each lane. 5.15 In practical terms this value is about 1790 pcu/h, except for the nearside lane which is 100 units less. The higher value of 525w or 1790 - 100 should be used.

16

TABLE 4: CLEARANCE TIMES FOR INTERGREEN CALCULATION

Ahead Traffic Distance 'x' (m) Intergreen (seconds) 9 5 10-18 6 19-27 7 28-36 8 37-46 9 47-54 10 55-64 11

Turning Traffic Distance 'x' (m) Intergreen (seconds) 9 5 10-13 6 14-20 7 21-27 8 28-34 9 35-40 10 41-45 11

5.16 Provided that the junction is operating at under 90%n of full capacity, the calculated green times or TRANSYT settings will still be valid even if local factors slightly alter the real values of S. However, for key junctions that are operating at capacity there is no alternative but to measure the saturation flow according to TRL Overseas Road Note 11. GRADIENTS

urban bus also assumes a vehicle of 10-12m. For motor cycles, the lower value suggested would reflect a high proportion of smaller types of vehicles, (under 125cc). 5.20 It is worth remembering that although there has been a lot of discussion on the changes of the value of S due to smaller, faster cars, etc., these factors tend to act equally on all the approaches of a junction. TURNING TRAFFIC

5.17 For each 1 % of uphill gradient the value of S should be reduced by 3% up to a maximum of 30% reduction. For downhill gradients another 3 % should be added to the value of S (up to a max. of 15%) for every 1 % of gradient. TRAFFIC COMPOSITION-PASSENGER CAR UNITS 5.18 The effect of different vehicle types on the satura-tion flow is eliminated by transforming all vehicles into standard car units, using a conversion factor, such as the values suggested in Table 5. 5.19 The figures shown in Table 5 give an idea of the range of values for different vehicle types in a number of countries. The suggested values represent typical conditions in developing cities, where heavy goods vehicles are normally larger than in Europe or Japan. The value for an

5.21 When this traffic has an exclusive lane with no opposing flow, S may be given by the formulae: 1800 S= 1+1.515/r 3000 S= 1+1.515/r where r = av. radius of turn 5.22 In mixed traffic each left turning vehicle is equivalent to 1.25 straight ahead vehicles, and each right turning vehicle is equivalent to 1.75 straight ahead vehicles. (for two lane) (for one lane)

17

TABLE 5: SOME TYPICAL PCU VALUES AT URBAN INTERSECTIONS

vehicle type Car Mini-bus < 20pass. Motor Cycle Heavy Goods Bus>20pass. Auto Rickshaw Pedal Rickshaw Pedal Cycle Horse &Cart Bullock Cart

England 1966 1.0 1.0 0.33 1.75 2.25 -

France 1974 1.0 0.3 2.0 2.0 0.3 -

Japan 1974 1.0 0.33 1.75 1.75 0.2 -

Indonesia 1984 1.0 1.25 0.2 2.25 2.62 0.52 0.93 -

India 1.0 0.25 2.8 3.6 0.6 1.4 0.4 2.6 11.2

Cairo 1985 1.0 1.0 0.5 1.6 2.5 4.0 4.0

Chile 1984 1.0 1.26 0.64 2.23 1.52 -

Suggested Values 1.0 1.25 0.3 2.5 2.5 0.5 1.0 0.3 3.0 -

PARKING, WAITING AND BUS STOPS

5 23 A parked car on an approach causes an effective loss in width, which can be expressed as: loss = 1.68 - 0.9 (z - 7.62)m g where z (>7.62) is the distance from the stop line of the parked car and g the green time in seconds. 5.24 In practice, the handling of parked or waiting vehicles in developing cities is extremely complex and involves a lot of assumptions on the part of the engineer. Traffic signals are generally put on the busiest junctions - and these are exactly the favorite street corners for commerce and irregular bus stops. 5.25 Road signs and road markings may prohibit parking, but enforcement is normally difficult or ineffective. Inmost cases the engineer will be left with the choice of reducing the effective carriageway width by lm (for the odd parked car, a bus stop or loading), by 2m (a row of parked cars with some double parking and occasional bus stop) or even 3m (next to a school where double parking is almost the rule) - or in situations such as the street market shown in Plate 2. 5.26 One of the other main causes of reduced saturation flow is the proximity of a signal controlled junction down-stream. This may result in the need for vehicles to slow down because of poor linking or vehicles backing up from another congested signalized junction downstream. The adverse effect of adjacent signals on one another can be particularly serious during peak hours. In these circumstances site observations should be made to evaluate the actual effective green usable by various vehicle phases and to determine the necessary adjustments on signal timings i.e. green splits and offsets.

TIME SETTINGS TOTAL LOST TIME PER CYCLE (L)

5.27 With a 2 second red/amber start and 3 second amber finish, the Intergreen Period for two conflicting stages is 5 seconds plus any all red time. 5.28 The lost time for a single phase during the green and amber period is normally about 4 seconds, in other words, the effective green time is thus the green time + 1. Then, per cycle: Total lost time = 4*number of stages + all red 5.29 Full pedestrian stages should be treated as all red tunes and included for the calculations.

FLOW FACTORS
5.30 The flow factor `y' for each phase is given by the formula: design flow for an approach (q) y= saturation flow for an approach (S)

5.31 Where more than one approach is operating during a phase, the maximum `y' should be used. Where an early cut-off or late start is to be used in connection with a right turn the `y' values for the right turn and for the approach with shortened time should be added together to represent one phase, unless the straight on traffic on the same approach as the right turn has a higher value, in which case this latter figure should be taken. 5.32 The sum of these higher `y' values for all the phases is the `Y' value, which is in fact a measure of the congestion, and which will be used for calculating the optimum signal setting.

18

Figure 18 Variation with time of discharge rate of a queue in a fully saturated green period

CYCLE TIMES
5.33 For an isolated signal installation, where the mean traffic level is constant and where vehicle arrivals are at random, the U.K. Transport Research Laboratory (TRL) has shown that the optimum cycle time for minimum delay (Co) is given by: 1.5L + 5 Co = 1-Y 5.34 The cycle time which is just sufficient to pass the traffic (Cm) is given by: L Cm = 1-Y 5.35 This is the minimum possible cycle time which may be associated with excessively long delays. In designing linked signals a cycle time should be chosen which provides a margin over this minimum cycle time for the key intersection. 5.36 In practice it will be generally appropriate to choose a practical cycle time (Cp) which will allow the installation to be loaded to 90 per cent of its capacity: 0.9L Cp = 0.9 - Y 5.37 Where pedestrian crossing volumes are high it will be desirable to use as short a cycle time as practicable to minimize delays imposed on pedestrians. In these cases, recommended practice normally limits cycle times to below 90 seconds. Cycle times lower than 50 seconds, on the other hand, tend to waste too much of the cycle time in lost time. secs. secs. secs.

5.38 Even when pedestrian movement are low, a practical upper limit to cycle times is around 120 seconds.

GREEN TIMES
5.39 Signal setting for the effective green periods (g) should be in proportion to the y values on each approach, with an allowance for lost time: g1 = g2 y2 y (c L) g= Y Where: g = effective green period y = flow factor G = actual green period c = cycle time L = total lost time Y = sum of y flow factors c-L = total effective green time Therefore: y1 (c L) g1 = Y y2 (c L) g2 = Y 5.40 To simplify the calculations, for most cycle times of around or above 60 seconds, the same value can be used for y1 etc.

19

both G and g. In other words, if the total lost time for the junction is 8 seconds and the desired cycle time is 60 seconds, both G and g can be taken as 52 seconds. 5.41 If parallel pedestrian facilities are included in the junction method of control, the minimum green times for the minor movements could well be dictated by parallel pedestrian crossing green times. This could distort the green split calculation and in situations where pedestrian signals are being introduced the engineer is faced with the dilemma of either reducing the minimum pedestrian time or oversaturating the junction capacity, in turn leading to disrespect of the pedestrian signal. In many developing cities this problem still remains.

5.46 Most of the useful formula have been incorporated so that the engineer may perform the operations directly, without referring back to this guide. This particular model was composed using a spreadsheet and can be adapted according to local conditions and the program most suitable. STAGE/PHASE SEQUENCE DIAGRAM 5.47 The method of signal control should be fully illustrated by the Stage/Phase Sequence diagram, complete with the following details: diagrammatic junction layout signals operation sequence design flows in pcu/h pcu. factor (if necessary), for converting unclassified counts from veh/h to pcu/h intergreen periods required actual green times (G = g - 1 )

DEGREE OF SATURATION
5.42 Degree of saturation (X) for individual approaches may also be expressed as; qc X= gs where: q = design flow c = cycle time S = saturation flow g = green time for approach 5.43 The degree of saturation should be the same for all the predominant arms of an intersection when the signal timings are at optimum settings and is given by the equation: 2Y Xo = 1+Y

5.48 All traffic signal calculations should be regarded as a good first estimate. There is no substitute for on-site checks and the `fine tuning' that should take place once the changes or installations have been implemented. 5.49 This section has outlined the "Webster" method of traffic signal design, which, although manual, has great merit in that it is simple to apply and enables a clear understanding of the principles used. However, there have been great advances in PC based programs for signal design, such as TRANSYT, OSCADY, LINSIG, and others which enable calculations to be canned out rapidly and accurately.

CHECK LIST FOR SIGNAL DESIGN Step 1 - Identify Traffic Flow Volumes
Traffic flow volumes are identified, including turning movements.

JUNCTION CAPACITY ANALYSIS

5.44 The ultimate capacity of an intersection may be defined as the maximum flow which can pass through the intersection with the same relative flows on the various approaches and with the existing proportions of turning traffic. Capacity will normally increase as the cycle time increases, since the ratio of lost to useful time decreases. In practice, for maximum reserve capacity assessment, a maximum cycle time of 120 seconds should be adopted. However it should be noted that for new installations, the maximum operating cycle time should be limited to about 90 seconds.

Step 2 - Identify Junction Layout, Lane Geometry and Site Characteristics

The junction layout, including lane geometry and site characteristics are identified. It may be necessary, if revealed in Step 4 or Step 7, to modify the layout to cater for turning movements, pedestrians or to enhance capacity and/or safety.

TRAFFIC SIGNAL CALCULATION SHEET

5.45 A `Traffic Signal Calculation Sheet' is shown as an aid for performing traffic signal calculations.

Step 3 - Identify Signal Phasing and Method of Control

The method of control to be used for analysis is identified.

20

21

Step 4 - Check Turning Movements and Pedestrians

Adequate provision for turning movements and pedestrians should be checked. It may be identified at this stage that the assumed method of control would need adjustment before continuing. Adequate allowance in calculations for parallel pedestrian minimum green crossing times should be made.

Step 11 - Determine Offset and Other Controller Settings

Offset and other controller settings such as minimum green, maximum green ,etc. are then finalized. Offsets for linking signals may be prepared with the aid of time-distance diagrams. or programs such as TRANSYT. (see section 6).

Step 12 - Prepare Documentation

For record purposes, drawings showing the junction layout, method of control, stage/phage diagram, traffic flow, etc. need to be prepared and maintained. Standard symbols should be used wherever applicable.

Step 5 - Estimate Saturation Flows

The saturation flows for various approaches/movements are identified. In critical cases the saturation flows for important movements may have to be measured on site.

Step 6 - Compute Y, L,
The lost times, flow factors and sum of the critical flow factors are computed.

Step 0 - The Proven Need for traffic signals at the junction according to approved warrants, should always come first and be part of the documentation.

Step 7 - Compute Reserve Capacity

The maximum reserve capacity of the intersection is then calculated as a measure of operating performance. If this is not satisfactory, then it may be necessary to go back to Step 2, modify data and layouts and recalculate. A minimum provision of 25% reserve capacity should be provided wherever possible for new junctions. A lower standard may be adopted for existing junctions where further improvement is restricted by space limitations.

Step 8 - Compute Co, Cm, and Cp.

The optimum, minimum and practical cycle times for operating the junction are then computed for further analysis, if necessary.

Step 9 - Select C
It is then necessary to select a cycle time for operating the intersection. Sometimes, for reasons of linking, the selected cycle time may be different from the values calculated in the previous step.

Step 10 - Compute Green Times, Degree of Saturation

The green times of the various phases are then computed. Degree of saturation may be computed as well if detailed analysis of signal operation is required. If good linking to other junctions requires a cycle time that results in very low degree of saturation and a very high reserve capacity, consideration should be given to double cycling this junction within the linking group, i.e. running it at half the linking cycle time.

22

6.

COORDINATION AND LINKING OF TRAFFIC SIGNALS

case, a desired speed should be used to ensure that the platoon conforms to the legal limit. 6.5 The timings of the signals in a simple progressive system can be prepared with the aid of a tune-distance diagram, examples of which are shown in figures 19 and 20. 6.6 On these diagrams, distances between junctions along the route are plotted along the abscissa (y axis) and the travel times are plotted along the ordinate (x axis). The slope of diagonal lines represent the chosen speed of progression and green stages of successive junctions are offset in time. Normally the problem is one of determining, by trial and error, the optimum through-band speed and width for a fixed cycle time. For one-way roads, the green bands follow each other in sequence. The driver, having passed one intersection, will then receive right of way at the others. 6.7 When the flow of traffic is two-directional and where the intersections are not equally spaced, the situation is more complex and it may be necessary to come to a compromise on progression between the two directions. It may also be necessary to take into account other requirements such as demands from cross-street traffic. 6.8 The time-distance diagram method can be used to bias in favour of a particular direction of flow e.g. to favour a heavy inbound flow in the morning peak at the expense of increased delay to the fewer vehicles travelling in the opposite direction. The situation may be reversed for the evening peak. 6.9 Cycle time for a coordinated signal system is normally dictated by the timings of a key junction, i.e. the junction which is most heavily loaded Spare green time should be allocated as required to clear traffic turning into the main route from side roads in order not to delay the through platoons. 6.10 To minimize congestion, opportunities for leaving the system should also be greater than for entering. 6.11 In heavy city centre traffic a design `speed' of about l0m/s ( about 40km/h) usually gives good results. For suburban traffic, where traffic is lighter and signals are about 300m apart, a design velocity of 15m/s (about 60km/ h) can be used as a first estimate, provided this does not conflict with local speed restrictions. 6.12 On two-way roads, good coordination can usually be obtained by using a common cycle time equivalent to twice the average travel time between junctions. 6.13 For example, using a design speed of l0m/s on a two-way road between nodes A.B.C, with links lengths AB+ 260m and BC = 300m. The travel times are then 26 and 30 seconds with an average of 28s. A good estimate of cycle time would then be 2x28 =56seconds, or in practice a cycle time of 55 to 65 would be adequate (with the lower option possibly leading to higher speeds).

6.1 Many cities in the developing world have grown to huge dimensions in a very short period and lack the large scale road infrastructure needed for modern traffic, as well as the capital to build it. This leads to the inefficient use of the existing road network, often in the form of extensive one-way systems, in turn leading to the need for a large number of traffic signals mostly fairly close to one another. 6.2 The effect of vehicles stopping at a signalised junction is to form the vehicles into a queue behind the stop line. When this queue is released as the green is given, it will discharge initially at its maximum rate (i.e. saturation flow) and move forward as a `platoon'. If, as this platoon approaches another signal controlled intersection, its arrival is made to coincide with the start of the green period, the vehicles will experience no delay at this new junction. If the platoon has to stop, a queue may form leading to spillback, which in turn may block the upstream junction. 6.3 The objectives of coordination of signals are to: prevent the queue of vehicles at one intersection from extending back and interfering with an upstream junction. increase the capacity of the linked route. enhance driver comfort by offering less stops and smoother flows in a controlled manner. offer minimum overall delay for road users, reducing overall travel time. reduce fuel consumption - and hence pollution -within the area. impose on drivers a safer behaviour, as they normally bunch together at the speed designed for green coordination (or waves). Speeding is thus minimized, accident risk and severity are reduced as is the associated risk for pedestrians, as they can judge oncoming speeds with greater accuracy.

SIMPLE PROGRESSIVE SYSTEM (GREEN WAVE)

6.4 The most commonly-used linking system works with a cycle time common to all intersections and the signals are so timed that the `go' periods are staggered in relation to each other according to the road speed to give a 'progression' of green periods along the road in both directions. Thus road speed should be considered "reasonable" by drivers; if speeding is common before linking, then a measured speed will be too high for safe operation. In this

23

Figure 19 Co-ordinated signals for one-way traffic

Figure 20 Co-ordinated signals for two-way traffic

Red (amber omitted, only north-south phase shown Green

24

MECHANISMS FOR LINKING SIGNALS CABLE-LINKING

6.14 Local traffic signal controllers at intersections working in a linked group may be cable-linked to a master controller. The master controller generally ensures that the local controllers, or `slave controllers' as they are termed, operate in synchronization by sending control pulses or instructions down the link cable to every slave controller.

6.20 The main facilities provided by ATC Systems are Optimized signal coordination. With the aid of centralized computer control, signal settings can be optimized on an area basis to provide minimum overall delay and reduced journey times. Control flexibility. Changing traffic conditions can be catered for by vehicle actuation or predetermined multiplan operation. The time settings of traffic signals can also be altered very quickly by manual intervention at the control centre, such as modifying the existing signal timing plan or replacing it with a new plan. Fault monitoring. One of the most important facilities offered by ATC systems is the continuous monitoring of the operation of the traffic signal equipment linked to the computer. Any fault condition detected is reported to the Control Room immediately and fault repairs can be carried out quickly. Priority for emergency and public transport vehicles. For fire engines which always start from a certain fire station, special plans may be prepared for predetermined `preferred routes' and stored in the central computer. Priority arrangements can be given to public transport routes, busways, etc. Accident reduction. ATC systems improve road safety, especially in the traffic conditions of some developing cities. Table 6 shows some results of traffic accidents before and after the construction of an ATC system.

CABLE-LESS LINKING
6.15 Linking of signals may also be achieved by cable-less means such as radio, mains synchronization, etc. Originally conceived to offer the benefits of wide area coordination while avoiding the expensive costs of underground ducting, these systems are now less competitive than the simpler ATC systems, as they do not allow for supervision of faults or such basic Responsive System facilities such as queue control. -

FIXED-TIME COORDINATED SIGNALS

6.16 These systems are based on assumptions that traffic flows are repetitive over a weekly (24h, 7day) cycle and that the appropriate predetermined signal settings can be prepared to cope with predictable traffic flows. The traffic plans, or signals settings can be selected according to the day of week and time of day. These systems require controllers with synchronized internal clocks and, typically, a minimum of 6 plans to choose from. 6.17 These would normally include plans for the: Morning Peak. Off Peak (work day). Evening Peak. Off Peak (night, holiday or Sunday). Nighttime Yellow Flashing, etc.

TABLE 6: ACCIDENTS IN ATC SYSTEMS

ATC System

Accident Reduction (%) 13 24 8.5 20 14 8 18 20 24

AREA TRAFFIC CONTROL (ATC)

6.18 Area Traffic Control, (ATC) is the centralized control of traffic signals on an area-wide basis using micro-processor and computer technology. 6.19 Usually the traffic controllers on street are linked to one or more central computers in the control centre, via data transmission cables. The cable network can either be provided as a dedicated network or private circuits leased from the telephone company (or a mixture of both, according to cost). Urban Traffic Control, (UTC) involves central coordination of signals as an ATC, but will include other facilities such as car park space control and variable message signing.

Toronto Atlanta Witchita Falls Sydney Glasgow Paramatta West London Rochester Curitiba

6.21 On certain corridors these improvements may be even greater. For a well designed system on downtown streets together with suburban corridors, an accident reduction of about 20% (all types) may be expected.

25

FIXED TIME ATC SYSTEMS

6.22 These ATC systems operate on a strategy of fixed time traffic signal plans for each controller. Signal timings and plans for junctions alter in accordance with a preset timetable held in the central computer or in a microprocessor inside the controller itself. These systems work very well for traffic patterns which are predictable and which change quite slowly. Traffic signal timings are developed from historical traffic flow data collected for the junctions and the timings require updating periodically, depending on the traffic fluctuations at each site. 6.23 For large urban areas with possibly several hundred traffic signal controlled intersections, a large system, complete with a special computer, monitor rooms and possibly closed circuit television is an entirely appropriate arrangement. Usually this type of maxi system will demand (and can justify) its own dedicated team of hardware and systems engineers, traffic engineers and technicians and system operators. 6.24 For medium-sized cities in the developing world, however, the benefits accruing from the installation of a maxi system cannot be justified in terms of the initial costs and staff resources demanded to run it. Experience has revealed that fixed time systems are cheap and simple to install and maintain and recent developments in electronic technology have made the Mini ATC, or Compact ATC systems almost as efficient as the traditional 'centralized' systems. The features of these systems are fairly standard, and can be summarized as: Computer and data transmission system in a standalone unit which does not require a specially controlled environment. "off the shelf" system operating software. automatic printout of faults; can be left to run virtually unattended. usually these systems are purchased as a complete package to a standard specification. This minimizes the staff input at the outset of the scheme.

examining all possible alternatives, during the evaluation of any ATC or UTC system.

SEMI-RESPONSIVE SYSTEMS
6.27 To overcome some of the rigidities inherent in the fully fixed time strategy, whilst avoiding some of the sophistication and complexities of fully responsive strategies, some degree of traffic responsive control can be introduced into the fixed time philosophy. 6.28 This can take two main forms Using strategic detectors to detect fluctuations in traffic flow, the system automatically changes from one fixed time traffic plan to another. Basically, this is equivalent to the introduction of a flexible timetable into the system. Installing vehicle presence detectors and pedestrian push buttons at junctions and introducing a VA 'window' into the traffic signal timing plans. The difficulty with fully traffic actuated signals is that the concept of signal linking within a fixed cycle time breaks down under fully traffic-actuated control. The VA-window concept allows a certain amount of 'local discretion' in the traffic signal plan. This permits the controllers at certain times in the signal cycle to decide whether to introduce a certain stage or whether to remain on the stage running, or to control the extension of queues by allowing for green extensions, etc.

6.29 The former is very effective in overcoming difficulties in defining accurately plan change points. Also, as only a relatively small number of detectors may be required, a heavy ongoing maintenance commitment on detectors is unnecessary. And in developing cities detectors are often a problem. 6.30 The second type of system can be very effective in coping with minor fluctuations in traffic flow, queues and with the problems of light vehicle and pedestrian flows in the off peak period. Though semi-responsive systems do require more complex signal controllers on street.

6.25 Standardization usually means that the costs in-volved to the purchasing of such a system are modest. As the system requires no special architectural/environmental arrangements, and will virtually 'run itself,' the staff input on the part of the purchaser/local administration is also small. Much of the flexibility and monitoring facilities of the maxi systems are, however, maintained and signal timings can be kept up to date by a small team, possibly on a part-time basis in addition to normal traffic engineering duties. 6.26 One of the major difficulties and costs of ATC systems is in installing an adequate data transmission network. This item should be given special attention,

FULLY RESPONSIVE CONTROL

6.31 The logical extension of computerised traffic control philosophy is the fully automated system, where the computer automatically calculates and adjusts signal timings to suit actual traffic conditions on street. 6.32 The SCAT (Sydney Coordinated Adaptive Traffic) System is based on a limited distributive intelligence system with a main computer performing certain monitoring functions and several satellite computers controlling a group of controllers on street. 6.33 All intersections within a subsystem operate on a common cycle length and are equipped with inductive loop vehicle detectors on all approaches. These are located in

26

each lane immediately in advance of the stopline and perform the dual functions of providing traffic flow data for strategic control and local or `tactical' vehicle actuation. 6.34 The SCOOT (Split Cycle Offset Optimization Technique) System consists a number of SCOOT cells or computers, each cell being able to control up to 60 junctions, handling input data from up to 256 vehicle counting detectors on street. Unlike the SCAT system, the SCOOT detectors are placed as far upstream from the approach to the junction as possible and are then calibrated to strike a balance between flow and occupancy. 6.35 In normal operation SCOOT estimates whether any advantage is to be gained by altering the timings. If an advantage is predicted then one or more of the timings are changed by small amounts. By this means frequent, but small, changes allow the signal timings to match fluctuations in traffic demand. There are no large and abrupt changes in signal timings, although overtime, major changes in splits, cycle time and offset can occur. 6.36 It must be remembered that in order to operate any fully responsive system at least 80% of detectors need to be fully operational. This requires constant maintenance and adequate continuous funding --exactly the areas which are often a problem for transport projects in the developing world. DUET DIAL-UP EQUIPMENT TESTING 6.37 It is arguable whether the greater benefit obtained from an ATC system is derived from signal coordination or from the ability to monitor continuously the performance of signal equipment. For signals in remote locations or where the cost of a dedicated ATC telecommunication network is not justified, DUET may provide a compromise solution. With the aid of this technology the equipment can be interrogated over public telephone lines as and when necessary. 6.38 Each local controller is connected to the public telephone system. Communication is achieved by dialing a telephone number from the Central Office, which connects the local controller via modem to equipment in the Central Office for testing/fault reporting purposes. 6.39 This arrangement avoids the need for a dedicated private circuit but provides only a limited fault monitoring facility as equipment faults will only be detected when the interrogation takes place, which may be some time after the fault has occurred. 6.40 An alternative is for the local controller to have the ability to `ring-up' the Central Office when a fault is detected. At the same time, it should be possible to change and load plans from the Central Control Room via a DUET setup.

6.41 The curve shown in figure 21 represents the cost-benefit relationship for the various options of control out-lined in this section. It should be noted that the principal gains are obtained with less complex measures and that high-tech solutions although extremely useful options in big cities - are normally only recommendable as a final stage in the ongoing process of traffic signal improvements.

Figure 21

Cost/benefit relationship for different types of signal co-ordination

27

7. THE TRANSYT PROGRAM

7.1 The TRANSYT program was developed by the U.K. Transport and Road Research Laboratory in a series of experimental applications in Glasgow, Scotland, where an ATC System had been installed. 7.2 The network being modelled in TRANSYT is represented by `nodes' inter-connected with `links'. Each signalised junction is represented by a node; each distinct one-way traffic stream leading to a node is represented by a link. TRANSYT takes the flows on each link and models traffic behaviour from this. 7.3 Figure 22 represents a small network of three junctions together with the traffic flows for one of the peak hours.

as the common cycle time for the network and the cost of delays to vehicles. Signal optimisation data allows the user to specify what type of optimisation is to be carried out. Link data requires the user to specify information such as the flow and saturation flow on a link, as well as during which stages the link receives a green signal. The node data requires the user to specify the staging of the signals at each node, minimum greens and intergreens. 7.6 This data is resumed in the input table as follows:

7.7 TRANSYT will produce an optimised set of signal timings for each node within a network. The signal timings are calculated by a method which works by varying both the signal offsets and the duration of the individual stage green times. Together, these signal timings result in delay being reduced to a minimum within the network. 7.8 The program can also be used to advise on the best cycle time for a network. This area of TRANSYT will also advise the user about the benefits to be gained from doublecycling any of the nodes. 7.9 For the optimised signal timings calculated by TRANSYT, a table of results, relating to performance and delay on each link, is given. From this table the user can deduce how saturated each link is, the potential size of any queues, the cruise time for vehicles, the amount of delay each vehicle can expect to suffer, as well as a series of cost related indices in terms of fuel consumption and delays. 7.10 This program has been immensely successful and is recognized as one of the most important traffic signal tool available to engineers. Versions are now available for use in P.C.'s and these also allow increased emphasis to be attached to key links in the network and the whole process can, if desired, be biased either towards minimum delay or minimum stops to vehicles. The average speed within the network and total fuel consumption are also given and these provide the basic data for evaluating the operating benefits to be gained from a new plan or system. Special routines are also available to model bus movements separately and aid in the selection of optimum cycletimes. 7.11 In practice, the TRANSYT Performance Index may be improved in "Grid" type networks by building into the initial offsets simple green waves on the main streets of the network. The output is then normally a much improved version of the desired coordination. If left to optimize from an all zero setup, the program may opt, when signals are close together, for a solution of the `open all lights north-south, then east-west'. This leads to speeding as drivers note that all the lights along the road go green at the same time.

Figure 22 Simple road network and traffic flows 7.4 This is then transformed into a "TRANSYT Diagram" such as figure 23; note that turning movements are given separate link numbers.

Figure 23 Transyt diagram of the network 7.5 TRANSYT requires the user to input a variety of information about the network. The data required is basically of four types; network data, signal optimisation data, link data, and node data. Network data includes such things

28

TABLE 7: TRANSYT OUTPUT FOR A SIMPLE NETWORK

1 LINK

NUMBER

FLOW INTO LINK

SAT FLOW

DEGREE OF SAT (%) 87 81 81 87 86 86 76 67 67 15 82 73 73 10

(PCU/H) (PCU/H) 10 11 12 20 21 22 30 31 32 39 40 41 42 49 TOTAL DISTANCE TRAVELLED (PCU-KH/H) 843.0 1200 550 400 1150 1101 199 500 1000 200 100 800 550 150 100 2400 35005 11L 2600 38005 21L 1600 36005 31L 1600 1600 3200S 41L 1600 TOTAL TIME SPENT (PCU-H/H) 79.8

MEAR TIMES PER PCU CRUISE DELAY (SEC) (SEC) 25 31 24 29 31 39 34 11 5 20 23 27 16 9

--------------DELAY--------------UNIFORM RANDOM+ COST OVERSAT OF (U+R+O.HEAN Q) DELAY (PCU-H/H) ($/H) 5.3 + 3.6 + 1.7 + 6.2 + 7.2 + 1.7 + 3.2 + 2.2 + 0.1 + 0.5 + 3.0 + 3.1 + 0.4 + 0.2 + TOTAL UNIFORM DELAY

--------STOPS-------MEAN COST STOPS OF /PCU STOPS (%) ($/H) 86 98 94 91 87 108 92 38 22 61 80 90 81 40 ( 9.9) ( 5.2) ( 3.6) ( 10.0) ( 9.1) ( 2.1) ( 4.4) ( 3.7) ( 0.4) ( 0.6) ( 6.1) 4.73 ( 1.2) ( 0.4) TOTAL COST OF STOPS (S/H) 61.2)

-------QUEUE-------MEAN MAX. AVERAGE EXCESS (PCU) (PCU) 27 + 24 24 28 + 31 31 12 13 13 2 18 + 17 17 1 PENALTY FOR EXCESS QUEUES ($/H) ( 0.0)

PERFORMANCE INDEX. WEIGHTED SUM OF ( ) VALUES ($/H) 77.9 43.8 24.9 84.5 86.4 19.4 42.2 27.8 2.8 5.0 48.1 37.8 6.5 2.5 TOTAL PERFORMANCE INDEX (S/H) $09.8

EXIT NODE

GREEN TINES START START END ENE 1ST 2ND (SECONDS) 38 4 4 42 2 2 12 53 53 12 63 32 32 63 89 33 33 87 37 37 48 7 7 48 27 58 58 27

10 8 8 10 12 12 10 13 13 10 10 10 10 10 MEAN JOURNEY SPEED (EM/H) 10.6

3.1 ( 68.1) 1.2 ( 38.7) 0.9 ( 21.3) 3.1 ( 74.6) 2.4 ( 77.3) 0.4 ( 17.4) 1.6 ( 37.9) 0.8 ( 24.1) 0.2 ( 2.4) 0.1 ( 4.4) 2.2 ( 42.0) 1.0 ; 33.1) 0.3 ( 5.3) 0.1 ( 2.1) TOTAL TOTAL RANDOM+ COST OVERSAT OF DELAY DELAY (PCU-H/H)(PCU-H/H) ($/H) 38.2 17.5 ( 448.6) + ( DELAY LITRES PER HOUR 78.0

1 1 1 2 2 2 3 3 3 3 4 4 4 4

TOTALS

******************************************************************************************************************
FUEL CONSUMPTION PREDICTIONS CRUISE LITRES PER HOUR 59.7 + + STOPS LITRES PER HOUR 60.7 TOTALS LITRESPER HOUR = 198.3

NO. OF ENTRIES TO SUBPT = NO. OF LINKS RECALCULATED=

10 115

29

8.

SYSTEM AND ECONOMIC ANALYSIS OF TRAFFIC CONTROL

8.4 An analysis of existing traffic signals, with or without coordination, plus those sites where counts or other data suggest the installation of signals, using the above criteria will tend to show the areas or sub-areas that would probably benefit most from a modern traffic signal scheme, such as those indicated in figure 34. 8.5 Obviously, if the city has a population of several mullion with a large and fast growing fleet of cars and buses, then some form of responsive control, such as SCOOT should be considered. In the city centre, important isolated junctions can also benefit from the application of an intelligent vehicle responsive system, such as MOVA (TRL. 1993.). Most cities in the developing world, however, will obtain the highest benefits from updating their traffic control system for a simpler model using prefixed timings. 8.6 In this case, TRANSYT is a good tool for evaluating a proposal. 8.7 Having arrived at the study area (or areas) a TRANSYT simulation for each area should be carried out, using real traffic data for each junction. If this data is not available, then new counts have to be made. 8.8 If the number of junctions in the initial proposal leads to a cost that is unacceptable to the city authorities, then the difficult task of reducing the proposal has to be undertaken, eliminating (or transferring to a later stage) those junctions or sub-areas that show least benefits - providing that neither the safety nor the basic concept of the system is impaired. 8.9 TRANSYT simulations of these new areas, scaled to meet budget requirements, will then provide the existing operating costs and travel delays (using existing flows and timings) for each peak period

8.1 In many cities throughout the developing world there is enormous scope for introducing modern traffic controllers capable of offering linked signal timings compatible with the different traffic volumes found during the day. The TRANSYT Program can now be run on a 486 PC and similar advances in micro processors have given local controllers the "intelligence" that used to occupy large main frame computers, thus simplifying the need for expensive telecommunication networks. 8.2 ATC systems improve road safety, they no longer require expensive "command centres", with a large public service staff and they normally have exceptionally high Benefit/Cost ratios that make them potentially attractive to international lending agencies and banks. 8.3 Warrants for an ATC System vary from city to city and there are no hard and fast rules that can be followed, however, some basic guidelines would cover: the city centre, where there is a high density of signals the streets used by public transport vehicles one-way suburban roads signals are up to 600m apart the main corridors and on the side streets up to 300m from the corridor the access roads to the city from major highways isolated junctions (using DUET or similar) that are overloaded during peak hours.

City centre

Figure 24 Definition of ATC system sub-areas

30

optimized operating costs and delays (using the optimizing process), for each peak period existing operating costs and delays for the off peaks optimized operating costs and delays for the off peaks.

8.15 DUET systems require the cooperation of the local telephone service - normally by contracting the lines, including their maintenance and installation costs. This requires a written proposal providing annual costs per junction. 8.16 Software, engineering and system maintenance costs vary greatly from city to city according to the size and specifications of the proposed system. However, an additional 30-40%o of the total would give a rough estimate. 8.17 Considering the speed of advances in electronics and telecommunications a useful life of 10 years can be expected for ATC/UTC projects, allowing for a residual value of 20% of the total project cost after year 10. 8.18 For two projects in Latin America, both nonresponsive using prefixed timings, the economic data for the systems can be summarized as: 8.19 Project a), Cost = US$ 5,794,000. Benefit/Cost Ratio = 2.41. Internal Rate of Return = 39.88 8.20 Project b) Cost = US$ 5,000,000. Annual time savings = US$ 2.000.000. Annual fuel savings (excluding national and local taxes) = US$ 3.709.548. Annual accident cost savings = US$ 2.640.000 8.21 In both cases it is clear that projects of this kind have an extremely high economic return. In fact, when a city proposes to upgrade its traffic signal system from isolated junctions with just one cycle time and plan to a simple ATC network it is no exaggeration to claim that the scheme will pay for itself in a question of months.

8.10 This same process can be repeated for a three, five and ten year horizon, using the relevant traffic growth factors (based on recent traffic or vehicle growth data). The yearly time and operating benefits of the proposed system can then be estimated over the project's useful life. 8.11 According to Table 6, a 20% reduction on the accident costs within the proposed area(s) could also be included. 8.12 In developing cities it should be remembered that bus (public transport) normally plays a major role and so these travel benefits should be calculated, if possible, separately. The value of time savings is always a delicate point - especially as these are aggregated in small amounts for a very large number of people. However, it is standard practice to allocate a reasonable value to these for economic evaluation. 8.13 On major public transport trunk routes, the extra fleet capacity gained during the peak hours may also be relevant and could be included as a real benefit if this means the postponement of additional fleet capacity during the project's lifespan. 8.14 Table 8 below shows the costs of installing a new traffic signal layout to different parts of the world. A rough estimate of junction cost for a modern, simple ATC System would be around US$20.000 - excluding the costs of data transmission and any detectors.

TABLE 8: ESTIMATES OF TRAFFIC SIGNAL INSTALLATION COSTS PER JUNCTION

city

pop. (106)

junctions with signals

cost/simple junction (US$)

cost/complex junction (US$)

So Paulo Santiago Curitiba Nairobi Harare Bombay Istanbul Islamabad

12.0 4.5 1.6 1.5 1.0 11.0 4.0 0.4

4000 980 350 27 152 305 300 41

4.150 14.300 12.000 18.000 8.000 12.000 14.000 10.450

36.000 28.600 18.000 26.500 16.000 18.000 55.000 13.550

31

9. SPECIFICATIONS
9.1 Most countries have developed their own national specifications for traffic signal materials, based either on local experience and production or on international standards used in the larger markets of North America, Japan, European Community, etc. Detailed specifications of equipment and services are beyond the scope of thus report, however, some basic requirements can be resumed as follows

capability of incorporating pedestrian push buttons, pedestrian wait indicators and pedestrian audible signals.

9.10 Most controllers offer a minimum setup of 4 stages and/or 8 phases, allowing for expansion to 8 stages and/or 16 phases, without the need to change the cabinet or internal architecture of the equipment.

GENERAL ROAD TRAFFIC SIGNALS

9.11 Vehicular signals should contain three optical systems arranged vertically each having a nominal diameter of 200mm. 9.12 Pedestrian signals should contain two optical systems arranged vertically and shall normally each have a nominal diameter of 300mm - an alternative size of 200imm nominal diameter may also be used where necessary. 9.13 Where an optical system incorporating a green arrow is used, it should normally have a nominal diameter of 300nim an alternative size of 200mm nominal diameter may also be used if necessary. 9.14 Signal heads should have adequate mechanical strength and durability to withstand the conditions of installation and normal use operations. It should be dust proof and weatherproof against corrosion and action of direct sunlight without significant deterioration. The signal head assembly together with the attachment of a backing board should be capable of withstanding wind velocities up to 160 Km/hour in any direction and temperatures over a range of -25C to 70C. 9.15 Normally, the signal heads will be made from polycarbonate or cast aluminium. 9.16 High intensity lamps should be used for enhancing visibility.

TRAFFIC SIGNAL CONTROLLERS

9.2 The increasing demand for better traffic control mechanisms and the advancements in technology have introduced the microprocessor-based traffic signal controller now a fairly standard, cheap and reliable form of control. 9.3 A major difference arising from the use of microprocessors is that the operating parameters such as timings are stored in memory rather than by switches or potentiometers. Changes in operation are accomplished by the use of hand-held terminals, although safety requirements are met by storing the basic site configuration data and control program on programmable read only memories (PROMS), alterable only by signal maintenance personnel. High reliability is also achieved by removing as many moving parts as possible and by using inbuilt self-monitoring and automatic fault reporting capabilities. 9.4 Essential features of this type of controllers are: ` intelligence' and flexibility inherent to microprocessor systems offers the possibility of easy modification as requirements change. capability of operating in any of the following modes:

9.5 Hurry Call. This mode maybe requested by the local detector or by switch input to allow priority to be given to a particular movement for the passage of emergency vehicles such as fire engines or rapid transit vehicles. 9.6 Manual. This mode will allow the controller to operate under emergency manual control. 9.7 Local Coordinate Mode. Multi-plan operation in this mode shall be possible in accordance with plans and timetables stored in the memory. 9.8 Local Isolated Vehicle Actuated Mode. The con-troller in this mode may operate with a mixture of fixed time, demand dependent or vehicle actuated phases. 9.9 Amber flashing on all approaches. capability of linking to a pedestrian controller for coordinated control.

INDUCTIVE LOOP DETECTORS

9.17 The basic system consists of a loop of wire (typically 2 or 3 turns) buried approximately 5mm below the road surface. The ends of the loop are returned to the vehicle detector, usually housed some distance away in the controller cabinet. A small electric current is passed through the loop and this causes an electric field to be built up. 9.18 A change in the inductance of the loop occurs when a vehicle is positioned directly over it - or is passing over the loop. The change in inductance is sensed by the unit and a signal change indicates the presence of a vehicle. 9.19 Provided that loop detectors are properly installed, tested and maintained, they will work reliably and offer efficient and accurate detection as well as providing congestion, counting and vehicle classification information.

32

ASSOCIATED ELECTRICAL WORKS 9.20 These works include Laying of aspect cables, linking cables (or telecommunications cables in an ATC area). necessary cable jointing work. signal wiring and connection work.

TRAFFIC SIGNAL CONTROLLER CIVIL WORKS

9.21 Again, the specifications of civil works vary from country to country, however, it is recommended that all cabling should be underground using ducts as shown in figure 25. Inspection boxes, normally 60cm on each side should be built every 50m (maximum). In developing cities the pavement is often dug up for new services, so to avoid the severance of cables, the ducts should be protected by a plastic strip in bright colours (normally yellow with black identification).

Figure 25 Cross section of cable duct

Plate 3.

Microprocessor controller suitable for pole installation at simple junctions

33

10. GLOSSARY
A short glossary of terms used in this volume in alphabetical order is given below: All red period Period when red signals are shown to all approaches simultaneously, usually of short duration to allow vehicles to clear the intersection. Audible signals

Early cutoff overlap Condition in which one or more traffic streams are permitted to move after the stoppage of one or more other traffic streams, which during the preceding stage had been permitted to move with them. Effective green period The period in the green and amber periods throughout which flow could take place at saturation flow levels. Extension

Signals in the form of pulsed tones provided for the benefit of visually handicapped pedestrians. Area Traffic Control (ATC) Also called Urban Traffic Control (UTC), this is the centralized control of traffic signals on an area wide basis by means of computer. Backing board The Background board is usually coloured black with a white border which is placed behind signal lanterns to make them more visible against a bright sky or other street or shop lights. Cycle time A cycle is a complete series of stages during which all traffic movements are served in turn. The cycle time is the sum of the stage times. Degree of saturation The degree of saturation at an approach is the ratio of the design flow to the actual capacity of a particular approach, weighted by the amount of green the approach receives in a cycle. Delay

A request for the continuation of the green signal to a predetermined maximum, made by a vehicle which, when the request is made, has the right of way. Fixed-time traffic signals Traffic signalling equipment in which the stages and their duration in each cycle are preset to suit predictable traffic conditions. Flow factor The flow factor or `y' value of an approach is the ratio of the design flow to the saturation flow of the particular approach. Green filter arrow An additional green arrow mounted by the side of the three light display to indicate early movement in the direction of the arrow and is terminated by a full green light signal. It is normally used for early discharge of left turners ahead of other movements at the same approach. Green split The ratio of green time allocated to each of the conflicting phases in a signal sequence. Grid lock

Traffic Delay is the lost time by vehicles due to traffic `friction', congestion or control devices. Demand A request for right of way for traffic on a phase which has no right of way when the request is made. The demands normally being stored in the controller and served in a prearranged order. Demand - dependent stage/phase A demand-dependent stage or phase is one which appears only on demand from a vehicle detector or pedestrian pushbutton. i.e. it is skippable. Detector A device to detect the presence or passage of a vehicle in the roadway.

The state in which downstream traffic completely blocks the junction, which then forms queues that blocks other junctions, and so on until the whole area is blocked with stopped traffic. Indicative green arrow An additional green arrow mounted on the right of the three light display of the secondary signal only, to indicate the early cutoff of an opposing flow. Intergreen period The period between the end of the green display on one stage and the start of the green display on the next stage is known as the intergreen period. Late start overlap Condition in which one or more traffic streams are permitted to move before the start of one or more of the traffic streams which, during the subsequent stage are permitted to all run together.

34

Lost time Lost time in the green and amber periods is the wastage time during which no flow takes place. Total lost time per cycle is the sum of these lost times for the critical phases plus other lost times due to red-amber periods, all red periods and pedestrian green and flashing green times. Maximum green running period The maximum time that a green signal can run after a demand has been made by traffic on another phase. Minimum cycle time The minimum cycle time which is just sufficient to pass the traffic. Minimum green running period The duration of the green signal following the extinction of a red/amber signal during which no change of signal lights can occur. Offset The time difference or interval in seconds between the start of the green indication at one intersection as related to the start of the green interval at another intersection from a synchronized system time base. Optimum cycle time

Red-running The act of disobeying, consciously or not, the red signal requiring vehicles on a determined approach to stop and remain stationary. Right of way The condition which applies when a green signal is displayed to traffic at the approach thereby permitting that traffic to proceed. Saturation flow The maximum flow which could be obtained if 100 percent green time was awarded to a particular approach. Semi-vehicle-actuated signals Modified vehicle-actuated signals whereby detectors are installed on the side roads only. Some semi-vehicle-actuated signals also operate one or more demand-dependent stages within a fixed cycle time. Stage A condition of the signal lights which permits a particular movement of traffic. Stages usually, but not always contain a green period. They are arranged to follow each other in a predetermined order but .stages can be skipped, if not demanded, to reduce delay. Spillback

The theoretical cycle time for attaining minimum vehicle delay. Passenger car units Passenger car units for a given type of vehicle are expressed in terms of the number of moving passenger cars it is equivalent to, based on headway and delay characteristics. PCU factor An average pcu value derived for the convenience of signal calculation to convert unclassified (by type) vehicle counts from vehicles per hour units to pcu per hour units. Phase Vehicle-actuated (V.A.) signals The sequence of conditions applied to one or more streams of traffic which during the cycle, receive identical signal light conditions. Two or more phases may overlap in time. A series of phases is usually arranged in a predetermined order but some phases can be omitted if required. Practical cycle time The cycle time at which the traffic signal installation will be loaded to 90 per cent of its capacity. Reserve capacity A measure of the spare capacity of a signal controlled junction, expressed in terms of the percentage of the current total flow factor value (Y) which will be available for further increase of traffic flows. An abbreviation of `Traffic Network Study Tool', a program developed by the Transport and Road Research Laboratory for optimizing signal timings. A vehicle extension period is the additional duration of the green signal which is secured by the actuation of a detector. Transyt Traffic signalling equipment in which the duration of the green time and cycle length varies in relation to the traffic flow on its approaches. Vehicle extension period Queue from a downstream junction which affects the traffic flow at the junction being examined. Traffic signals A system of different coloured lights, including arrow-shaped lights, for controlling conflicting streams of traffic and pedestrians. Traffic signal controller The electronic control equipment which activates the signal phases at an intersection.

35

11. REFERENCES
Burrow, I.J. OSCADY: A Computer Program to Model Capacities, Queues and Delays at Isolated Junctions. TRRL Research Report 105. Transport and Research Laboratory. 1987. Cannell, A. E. R. and Kaestner, C. Some Aspects of Area Traffic Control in Semi-Developed Countries. Traffic Engineering acrd Control. 1983. Companhia de Engenharia de Transito -CET. Semaforos, Aspectos de Seguranqa. Sao Paulo. 1990. Department of Transport. Junction Layout for Control by Traffic Signals. Highway, Safety and Traffic Advice Note TA 18/81. Department of Transport. 1992. Fouracre, P. R. and Gardner,G. Traffic Signals in Developing Cities, Overseas Unit Working Paper. Transport and Road Research Laboratory. 1990. Gardner, G, Fouracre, P.R. and Jacobs, G.D. Traffic Management. Overseas Unit Information Note. Transport and Road research Laboratory. 1990. Hunt, P.B, Robertson, D.I, Bretherton, R.D. and Winton, R.I. SCOOT: A Traffic Responsive Method of Coordinating Signals. TRRL Report 1014. Transport and Road Research Laboratory. 1981. Hurdle, V.F. Signalized Intersection Delay Models- A Primer for the Uninitiated. Transportation Research Record 971. Transportation Research Board. Washington, DC. 1984. Institution of Highways and Transportation/ Department of Transport. Roads and Traffic in Urban Areas. HMSO, London. 1987. Jourdain, S. Urban Intersection Control. The Book Guild Ltd. Lewes, Sussex. 1992. Robertson, D.I. and Hunt, P.B. A Method of Estimating the Benefits of Co-ordinating Signals by TRANSYT and SCOOT. Traffic Engineering acrd Control 1982. Robertson, D. I. Transyt: A Traffic Network Study Tool. RRL Report 253. Road Research Laboratory, Crowthorne.1969. Robertson, D 1, A Method of Estimating the Benefits of Coordinating Signals by TRANSYT and SCOOT. Traffic Engineering and Control. London. 1982. TRL. Urban Road Trafic Surveys. Overseas Road Note 11. Overseas Centre, Transport Research Laboratory. 1993. TRL. Design Guidelines for Busway Transit. Overseas Road Note 12. Overseas Centre, Transport Research Laboratory. 1993.

Vincent, R.A. and Pierce, J.C. MOVA: Traffic Responsive, Self-Optimising Signal Control for Isolated Intersections. TRRL Research Report 170. Transport and Road Research Laboratory. 1988. Vincent, R.A, Mitche, A.I. and Robertson, D.I. User Guide to TRANSYT Version 8. TRRL Report LR888. Transport and Road Research Laboratory. 1980. Walker, J.S, Jacobs, G.D. Gardner, G.and Kunchit Phiu-Nual. The Development of Traffic Management Policies in Bangkok. Paper presented to CODATU Conference. Jakarta. 1988. Webster and Cobbe. Traffic Signals. Road Research Technical Paper No. 56. H.M.S.O. London, 1966. Willumsen, L.G. and Coeymans, J.E. Research into the Value of Area Traffic Control Techniques in a Developing Contry. TRRL Contractor Report 99. Transport and Road Research Laboratory. 1988.

36

ISSN 0951-8987

OS-E

The use of traffic signals in developing cities

ORN13

Transport Research Laboratory Old Wokingham Road Crowthorne, Berkshire, RG45 6AU

Overseas Development Administration 94 Victoria Street London, SWIE 5JL

OVERSEAS ROAD NOTE 14

HYDROLOGICAL DESIGN MANUAL FOR SLOPE STABILITY IN THE TROPICS

Subsector: Theme: Project title: Project reference:

Geoscience G5 Terrain surveys and site investigations R5606

Copyright Transport Research Laboratory 1997. All rights reserved. This document is an output from a project funded by the UK Overseas Development Administration (ODA) for the benefit of developing countries. The views expressed are not necessarily those of the ODA.

Transport Research Foundation Group of Companies Transport Research Foundation (a company limited by guarantee) trading as Transport Research Laboratory. Registered in England, Number 3011746. TRL Limited. Registered in England, Number 3142272. Registered Offices: Old Wokingham Road, Crowthorne, Berkshire, RG45 6AU

ACKNOWLEDGEMENTS This manual was commissioned by the Overseas Centre (Programme Director: Dr J Rolt) and prepared at the University of Bristol. It was written by Professor M G Anderson, Dr D M Lloyd and Dr M J Kemp and is based on ODA funded research previously carried out at Bristol, and a final ODA project report submitted in 1989. The co-operation from the Research Institute of the Malaysian Public Works Department is also gratefully acknowledged.

First Published 1997 ISSN 0951-8797

OVERSEAS ROAD NOTES Overseas Road Notes are prepared principally for road and transport authorities in countries receiving technical assistance from the British Government. A limited number of copies is available to other organisations and to individuals with an interest in roads overseas, and may be obtained from: Transport Research Laboratory Crowthorne, Berkshire, RG11 6AU United Kingdom

Limited extracts from the text may be reproduced provided the source is acknowledged. For more extensive reproduction, please write to: Programme Director, International Division, Transport Research Laboratory.

CONTENTS
Page 1. INTRODUCTION 2. THE COMBINED SLOPE HYDROLOGY/ STABILITY MODEL THE HYDROLOGY MODEL THE STABILITY MODEL MODEL VALIDATION MODEL APPLICATION ASSUMPTIONS AND LIMITATIONS OF THE COMBINED MODEL 3. INSTRUCTION 1 THE USE OF SLOPE STABILITY DESIGN CHARTS IN STABILITY ASSESSMENT INSTRUCTION 1.1 When to use stability design charts INSTRUCTION 1.2 How to use the developed slope hydrology/stability charts 4. INSTRUCTION 2 INSTRUCTIONS RELATING TO SUCTION STRENGTH RELATIONSHIPS The use of the resistance envelope procedure INSTRUCTION 2.1 Procedure for the construction of a resistance envelope for a slope INSTRUCTION 2.2 Using the resistance envelope summary charts INSTRUCTION 2.3 Using the resistance envelope to determine the threshold soil water conditions for slope stability INSTRUCTION 2.4 Calculation of the likely average depth of the slope failure zone INSTRUCTION 2.5 Implications of the resistance envelope method on slope instrumentation, stability analysis and remedial action INSTRUCTION 2.6 The use of the resistance envelope procedure in the choice of stability analysis method 18 18 17 17 14 14 14 10 14 9 9 7 9 4 4 4 5 6 1 5. INSTRUCTION 3 INSTRUCTIONS RELATING TO THE MEASUREMENT OF PERMEABILITY INSTRUCTION 3.1 Determination of soil permeability for conditions below ground water level using piezometers INSTRUCTION 3.2 Determination of the permeability of soil above the groundwater level INSTRUCTION 3.3 Field measurement of the saturated infiltration capacity of soils INSTRUCTION 3.4 Laboratory permeability tests 6. INSTRUCTION 4 INSTRUCTIONS FOR PIEZOMETER MONITORING INSTRUCTION 4.1 Piezometer selection based on ground water conditions INSTRUCTION 4.2 Calculation of the time lag, T, Equalisation ratio, E, and intake factor, F, for a piezometer INSTRUCTION 4.3 Choice of procedures available for improving the response characteristics of piezometers (if appropriate) INSTRUCTION 4.4 Selection of the monitoring and data recording equipment for the piezometer system 7. INSTRUCTION 5 INSTRUCTIONS FOR THE MEASUREMENT OF SOIL SUCTIONS INSTRUCTION 5.1 Selection of instrument type based on monitoring range INSTRUCTION 5.2 Selection of tensiometer equipment INSTRUCTION 5.3 Tensiometer response characteristics 41 39 38 38 35 38 35 29 26 26 23 26 23 22 22 22 Page 22

Page INSTRUCTION 5.4 Tensiometer installation, monitoring and maintenance INSTRUCTION 5.5 The measurement of suctions greater than 80kPa 8. INSTRUCTION 6 INSTRUCTIONS CONCERNING DATA LOGGER SYSTEMS INSTRUCTION 6.1 Selection of the data logger system INSTRUCTION 6.2 Choice of sensor INSTRUCTION 6.3 Transducer selection INSTRUCTION 6.4 Calculation of tensiometer and piezometer readings using a calibrated transducer INSTRUCTION 6.5 Selection of data loggers with solid state memory design INSTRUCTION 6.6 Other data logger system considerations: power consumption, reliability and security APPENDIX A: Design Charts LIST OF REFERENCES 51 53 57 51 50 49 49 49 48 44 48 44

HYDROLOGICAL DESIGN MANUAL FOR SLOPE STABILITY IN THE TROPICS

1. INTRODUCTION
1.7 A new approach to slope stability analysis is described which considers the hydrological and hence stability system as dynamic, changing through time in response to rainfall infiltration. A combined slope hydrology stability model has been developed which is used to simulate the changes in soil moisture conditions within slopes in response to individual rainfall events. The model generated dynamic hydrological conditions are used as input soil moisture conditions for stability analyses using effective stress conditions. The stability analysis accounts for both the detrimental effects of positive pore pressures and the beneficial effects of soil suctions within the unsaturated zone of the profile as they change with time. 1.8 The combined model has undergone validation and assessment using an instrumented field site in Malaysia and also by application and comparison of model generated results to a large sample of known stable and failed slopes in Hong Kong. The model characteristics, development and validation are discussed to the following section of this manual. 1.9 The scheme has been used to produce design charts which provide a summary of the overall minimum factors of safety that result in response to particular storm events for a range of different slopes and antecedent conditions. These design charts are included in Appendix A of the manual. The charts allow a rapid assessment of the stability of a wide variety of potential slope conditions that accounts for the hydrological response of a slope to rainfall infiltration. They therefore represent a considerable advance on conventional stability charts in which hydrological conditions are treated as static. In addition these design charts are particularly useful in that they graphically high-light the importance of both antecedent moisture conditions and slope material permeability in controlling slope stability response to individual rainfall events - controls which are argued as important in controlling stability as material strength which is customarily used as the criterion for slope design. 1.10 The purpose of this manual is to familiarise the geotechnical engineer with the importance of the dynamic hydrological control on stability for the tropical slope condition, the modelling procedure that has been developed to investigate it, the use of model generated design charts in slope design, and those procedures and measurement techniques that should be adopted in addition to standard procedures to allow a better appreciation and understanding of the stability of tropical residual soil slopes.

1.1 The development of this manual arises from the perceived need to alter the decision process involved in the design, construction, analysis and maintenance of engineered slopes and embankments in the tropics. It therefore aims to be a source of reference and a guide to the practising geotechnical engineer with the purpose of improving awareness and account of the dynamic hydrological condition in slope design and stability assessment in the tropics. 1.2 Standard approaches to slope design for slopes within the tropics have proved inadequate as evidenced by their failure, implying the need to account for some further factors in slope design - factors that obviously exert a controlling influence on stability conditions but which have been ignored in standard methods of analysis. 1.3 In tropical residual soils most failures are caused by rainfall. Residual soil slopes are often unsaturated with the water table at depth due to the high permeability of the soil. Suctions can be expected to exist in the profile and these enhance the stability of the slope. With the infiltration of rainfall such suctions are reduced or even removed and can lead to the development of positive pressures in a perched water table zone. As a result there is a decrease in soil shear strength and in the overall mass stability of the slope. 1.4 It is, therefore, critical in the analysis of slope stability in the tropics to account for the pore pressure distribution in cut slopes and embankments and most importantly, the dynamic variation of this distribution with time. 1.5 Despite this fact the major emphasis of recent research has been concerned with mass strength and stability analysis procedures. Table 1 shows an assessment of the state of knowledge of the various aspects of slope stability predictions for Hong Kong as compiled by Hencher et al., (1984). From this it can be seen that there is a significant deficiency in knowledge relating to groundwater and pore water pressure conditions and predictive capabilities. This is in spite of work that has emphasised the role of relationships between soil moisture deficit and rainfall to slope failures. 1.6 It is the relative lack of success in moving from such relationships (successful though they have proved to be for landslide warning systems) to predictive capabilities for pore pressure distributions in hillslopes that lies behind the position regarding the state of knowledge of groundwater and pore pressures in the assessment of slope stability identified in Table 1.

Table 1: Assessment of the state-of-knowledge of the various aspects of slope stability predictions for Hong Kong conditions (updated from Hencher et al, 1984)

Aspect

Current State-of-Knowledge for Hong Kong Conditions

Overall Rating of Knowledge Very good

Methods of Stability Analysis

Janbu (1954, 1973) method of analysis for non-linear surfaces thought satisfactory. Recommended factors of safety of 1.2 to 1.4 are satisfactory (GCO, 1984). Computational data is often poorly handled (Lumsdaine & Tang, 1982).

Geometry of Failure

Post-failure geometry is easily defined. Often difficult to decide critical potential failure surface for design, especially where geology is complex (Hencher et al, 1984; Hencher & Martin, 1982; Hudson & Hencher, 1984). Site investigation procedures are adequate, but descriptions often poor. Complex weathering profiles are difficult to describe (Hencher & Martin, 1982). Understanding of influence of geological details on hydrogeology is poor.

Good to very good

Geology

Fair

Shear Strength

Mass strength as distinct from sample strength is poorly understood. Laboratory tests are commonly used to determine saturated strengths of samples in terms of effective stress, but doubt exists about applicability of test results (Brand, 1982). Limited amount of insitu strength testing carried out (Brand et al, 1983). Weakening effect of relict joints recognised (Koo, 1982). Effects of boulder and corestone content unknown (Hencher & Martin, 1982).

Fair to Poor

Groundwater and Pore Pressures

Useful correlations available between landslides and rainfall (Lumb, 1975; Brand et al, 1984). Rapid changes in pore pressure with rainfall are very difficult to predict for design (Anderson et al, 1983). Only limited attempts made to model groundwater (Leach & Herbert, 1982). Extrapolation of insitu measurements seems best design approach (Koo & Lumb, 1981; Endicott, 1982). Some progress made with field instrumentation (Pope et al, 1982; Brand et al, 1983). Erosion pipes are important in transmitting water (Nash & Dale, 1983; Brand et al, 1986).

Poor

1.11 The manual divides into two distinct sections. The first section provides a summary of the development and a description of the combined slope hydrology and stability modelling scheme. Direct application of the developed combined model is generally impractical due to the demands made on data and computational hardware and so the purpose of this section is to familiarise the reader with the assumptions and limitations of the developed model. The importance of an understanding of these assumptions cannot be over stressed as sensible application of the results generated by the scheme can only be undertaken with a full knowledge of those conditions that are, and are not, represented by the combined model. 1.12 Results of multiple applications of the model are summarised in the form of design charts and a methodology for their application for rapid assessment of slope stability that accounts for the influence of dynamic slope hydrological conditions provided. The charts emphasise the important control the dynamic hydrological condition has on slope stability. It is therefore necessary that standard approaches to slope design and site investigation are augmented to take account of this fact. The second section of this manual provides a set of instructions that should be adopted to achieve this end. 1.13 Instruction 1 details the scope and methodology of application of the developed design charts. The charts graphically illustrate the importance of the hydrological controls on slope stability - especially the effect on stability of slope antecedent conditions and material permeability. It is important to take account to these hydrological factors if suitable applications of the design charts are to be undertaken. Subsequent instructions therefore relate specifically to these points. Instruction 2 describes the Resistance envelope procedure. This approach allows rapid assessment of the form of the hydrological controls on slope stability for any existing slope. Such information is of particular use in the choice and location of instrumentation, the form of stability analysis required, and assessment of the remedial action required if a slope has experienced failure. Instruction 3 provides procedures for the measurement of slope material permeability. This has shown strong controls on slope hydrological conditions and hence stability. Determination of material permeability should be considered an essential part of any ground investigation concerning the construction of cut slopes, embankments, or works resulting in any change to hydrological conditions in existing slopes (both natural and man made). Similarly, accurate determination of antecedent conditions must also be considered essential in any such ground investigation. Instructions 4, 5, and 6 therefore relate specifically to the monitoring of soil moisture conditions. Instruction 4 concerns the use of piezometer systems, Instruction 5 the measurement of soil moisture conditions in the unsaturated zone of the profile and Instruction 6 the use of data logging systems. These instructions emphasise the factors that might otherwise not be taken into account, but which will

significantly affect the value of data obtained from any such monitoring schemes. 1.14 These instructions should be used to complement standard site investigation procedures and stability analyses currently employed to allow improved assessment of slope stability conditions in the tropics.

2.

THE COMBINED SLOPE HYDROLOGY/STABILITY MODEL

2.1 In view of the need to consider the short term over-storm response in the analysis of slope stability in the tropics the coupled slope hydrology/stability model has been developed. This scheme allows analysis of the effect of storm events on the slope hydrological conditions in both the saturated and unsaturated sections of the profile, and the resultant effect this has on the stability conditions. The scheme simulates the dynamic stability conditions allowing identification of the minimum factor of safety, the characteristics of the failure, and time of occurrence for any particular initial slope condition and rainfall event. The salient model characteristics are summarised below. 1) The scheme consists of a dynamic, 2-dimensional hillslope hydrology model directly coupled to a 2dimensional slope stability analysis procedure. 2) The hydrology model predicts both positive and negative soil water pressure conditions, at each iteration step (model time increment), as they change in response to rainfall events. On each hour of the simulation, the current hydrological conditions are used as the hydrological input data for the stability analysis. 3) The stability model is capable of accounting for the influence on material strength of both positive and negative (suctions) soil water pressures in the analysis. 4) Both individual models are able to accommodate potential uncertainty in the input parameter values. THE HYDROLOGY MODEL 2.2 The procedure adopted in the modelling of the slope hydrological system is a forward difference explicit block centred finite difference scheme. The hydrology model considers only two dimensions, modelling the soil moisture flow within a unit cross section of any given slope - a cross section that is assumed to be representative of the conditions within the slope. The scheme requires the model user to subdivide the slope into a series of columns which are themselves divided into cells, the centre point of which is assumed representative of the whole cell. The number of cells per column defines the surface topography of the slope and, as there is specified a maximum one cell difference between columns, it is the dimensions of the cells that affects the topographical resolution of the scheme. 2.3 Having defined the surface topography and the profile depth to be modelled, the slope can be divided into two hydrologically distinct layers - this being achieved by

defining the depth of the top layer from the surface. This top layer is modelled as a band of uniform depth throughout the slope length. The hydrology model simulates the effect of infiltration, evaporation and surface detention. Inflow from the backslope and a drainage level at the base of the modelled profile can also be accommodated. Flow within the unsaturated zone is assumed vertical and to obey Darcy's law with the unsaturated hydraulic conductivity defined by the Millington and Quirk (1959) procedure. Lateral flow is modelled at the top of the saturated zone in each column. Flow is assume to obey Darcy's law -hydraulic conductivity being equal to the saturated value and the hydraulic gradient equal to the gradient of the water table at that point. 2.4 After cessation of rainfall, water is lost from the surface due to evaporation. A maximum rate is specified and the model applies a sine based function during daylight hours. Between 1800 and 0600 hours (assumed night time) evaporation is assumed to be a hundredth of its daily maximum. 2.5 In instances where the rainfall results in the infiltration and surface detention capacities being exceeded, overland flow is initiated. The equations governing all these processes are given in Figure 1. THE STABILITY MODEL 2.6 On each hour of the simulation of the slope hydrology a stability analysis model is evoked. In the combined model the Bishop's method of analysis is used to determine the stability of the slope. The soil moisture conditions predicted by the hydrology model are converted to piezometric head values before transfer to the stability model routine. 2.7 The stability model accommodates the effects of both positive and negative soil water pressures on soil strength in the analysis. Like positive pressure, soil suctions are directly input into the Mohr Coulomb equation for soil shear strength with overestimates of shear strength being minimised by specifying a maximum negative head of 20kPa in the analysis. 2.8 As soil moisture conditions change through time in response to rainfall infiltration, so the minimum factor of safety slip surface position can be expected to change. Within the stability analysis a search procedure for the minimum factor of safety slip surface is incorporated by specifying a grid of circle centres and a circle radius incrementing procedure. This allows analysis of a whole suite of potential slip surfaces, the minimum factor of safety slip surface obtained being assumed the overall minimum condition for that time period. Thus a picture of the dynamic slope stability response to changing soil hydrological conditions is developed.

Figure 1

Two dimensional soil water model: computational points and governing equations

MODEL VALIDATION 2.9 1) Model validation can be divided into three categories:

possible to falsify the model. Two separate but compatible approaches can be taken: 1) The comparison of measured soil water conditions with those predicted by the hydrology model. This approach is currently being undertaken. For this purpose a new instrumentation scheme has been designed, developed and installed on a cut slope on the Kuala Lumpur to Karak highway, West Malaysia. Data from this program has been used to compare observed against model predicted soil moisture conditions. An example comparison is given in Figure 2. The resolution of the modelling mesh is such that the predicted soil moisture response is for a depth of 50cm, whilst the observed value is at a depth of 15cm. The lagged and damped response of the predicted values in comparison to the observed is therefore expected in this example comparison. The application of model predicted conditions (stable/failed) to a large sample of cut slopes, some of which have been observed to fail, others that are apparently stable. This approach has the major advantage that, at the current stage of model development, it provides information to evaluate the performance of the model and also determine the scope of design charts required for engineering applications. This has been successfully undertaken in Hong Kong where some 40 slopes with detailed site investigation information were analysed. Design charts (summaries of results of simulations for a range of antecedent and rainfall conditions) generated from application of the combined model

Mathematical validation: an assessment of mathematical assumptions employed in simplification of the physical system

the the

2)

Computational verification: the checking of the computer code after translation from mathematical equations Operational validation: an assessment of the degree to which the physically based model compares with reality

3)

2.10 Mathematical validation of a model is a subjective process as it involves the justification of the assumptions used in the simplification of the physical system. It also serves to familiarise the potential user with assumptions employed in the development of the model, preventing the use of the model in conditions whereby these assumptions may be violated. 2.11 Computational validation of the scheme has included tests on the computer code and both deterministic and stochastic sensitivity analyses. 2.12 Full operational validation of such physically based models is impossible due to the inherent variability of the natural system (no one condition is the same as another -full operational validation would require consideration of an infinite number of cases and parameters). However, it is

2)

Figure 2

An example comparison of observed soil water suctions and model predicted values

correctly classified 77% of the field failed slopes and 68% of the field stable slopes. This represents a considerable improvement over other currently available slope stability charts which gave a percentage correct classification no greater than 50%. MODEL APPLICATION 2.13 There are eight parameter groups that are necessary to run the combined hydrology/stability scheme. These comprise: 1) Evaporation 2) Rainfall 3) Initial near surface soil moisture conditions 4) Initial ground water table 5) Slope height 6) Slope angle 7) Soil permeability 8) Soil strength 2.14 Evaporation is defined by a daily maximum value to which the model applies a sine based function. Rainfall is most usefully defined as a return period event (design storm)- a 24 hour 10 year return period event for example. The remaining parameters need to be determined from site investigation. Where such information is not available then reasonable estimates based on experience and available information need to be made. 2.15 In addition to the assessment of the dynamic stability conditions of a cut slope in response to rainfall infiltration and drainage, the coupled scheme has the

facility to allow investigation of the effect parameter uncertainty has on the predicted result. Any measured parameter that is input into the scheme suffers from errors introduced by the sampling and measurement procedure. Input parameters can be randomly selected from distributions defined by their mean and associated standard deviations, and a distribution of output results developed from multiple simulations of the model. If the input and output parameter distributions are normal then probabilities can be assigned to particular factor of safety values and slip surfaces for any given modelled condition. Though computationally intensive this technique allows for a much greater appreciation of the overall stability conditions and likely failure characteristics of any modelled slope, see Figure 3. 2.16 It is often necessary to be able to form a rapid assessment of stability conditions for a particular slope (or large number of slopes) and individual application of the model to enable this would be impractical. Consequently the combined model has been applied to a range of slope conditions (both slope form and antecedent conditions) with varying characteristics (soil strength and hydrological properties), and the results summarised in the form of `design charts'. Such design charts provide a quick reference data source that allows rapid assessment of the minimum factor of safety condition in response to specified rainfall events for a range of potential slope conditions. Any slope can be matched to its nearest idealised modelled slope, and an assessment made of the stability which accounts for the effect of storm rainfall infiltration. This form of analysis is described in Instruction 1 of this manual.

Figure 3

Probability (%) of the failure surface being above illustrated surfaces, with assumed input parameter variability for a 35 slope and a 1000 year storm event 2) The resolution of the scheme. Temporal resolution of the stability analysis is every hour of the simulated hydrological conditions. It is assumed that between hour variation in the stability conditions is not of a magnitude significant enough to affect identification of, or anything other than insignificant error in, the minimum stability conditions for any modelled rainfall event. Over the range of conditions considered by the design charts this assumption is considered valid. For hillslopes consisting of highly porous, coarse, sandy soils (Ksat>10-4ms-1) this will not be a valid assumption However, it is unlikely that rainfall intensities, even in the tropics, will approach these permeability rates and a net loss of soil moisture will occur. This will result to improved stability conditions and so it is unlikely that the combined model, as a result of temporal resolution, will overestimate the factor of safety conditions in this circumstance. 3) The form of the stability analysis. The analysis of the stability conditions is achieved using the Bishop's method assuming a circular failure surface. The majority of observed failures in steep tropical residual soil slopes are shallow and near circular in cross section. Where failure surface location is influenced by material property variation such as relict discontinuities (resulting in wedge and block failure) this form of analysis is unsuitable and the design charts would not provide accurate assessment of the slope factor of safety conditions.

ASSUMPTIONS AND LIMITATIONS OF THE COMBINED MODEL 2.17 In the development of a model of a physical system assumptions need to be made to simplify the problem to one of practical proportions. It is important to be aware of such simplifying assumptions so that applications of model generated results are not made to conditions whereby the modelling assumptions are violated. Hence, it is the aim of the following section to familiarise the potential design chart user with the assumptions made in the development of the combined slope hydrology slope stability model so that sensible applications of the resultant design chart product are made. 2.18 Important modelling assumptions include: 1) Uniformity of material. Throughout the hydrology and stability modelling process, homogeneity of both soil layers in the hydrology model and any defined zone of material to the stability model is assumed. Factors that may influence soil water infiltration and redistribution such as preferential flow paths (macropores, root channels), seepage, and the soil mass strength such as relict discontinuities are not considered. This assumption is considered reasonable for most tropical soil slope conditions. In conditions where such features may extend along significant sections of slope profiles then model predicted conditions cannot be considered wholly representative (being likely to overestimate the factor of safety) of actual stability conditions.

4)

Shear strength envelope. Whilst effective stress conditions are represented by the combined model it is assumed that the shear strength envelope is linear. At low stress levels, such as in very shallow failures, this may not be representative of the actual conditions (see for example Crabb and Atkinson, 1991; Perry, 1994). It is therefore important that the appropriate soil strength (or range of soil strength) conditions are considered when making an assessment of stability using the design charts.

3. INSTRUCTION 1
THE USE OF SLOPE STABILITY DESIGN CHARTS IN STABILITY ASSESSMENTS
3.1 An important part of most slope analyses is the determination of the stability of the slope usually in terms of its factor of safety. This is conventionally achieved by carrying out a full stability analysis usually using a computer package which can be a time consuming process. Rather than carrying out detailed analysis of a slope it is often more practical to utilise `design charts'. These summarise the results of a large number of analyses of stability for `typical' slope forms and ranges of potential slope conditions. By the use of such design charts rapid assessment of stability is possible for a range of potential conditions (strength and hydrological). There is, therefore, great utility in providing easy to use look-up charts. 3.2 Existing stability charts serve a useful purpose but are limited by their poor representation of the soil water conditions. Generally, soil suctions are ignored in such analyses and water table representation is poor. Most importantly, all commonly used stability charts assume only static hydrological conditions, no account being made of the effect rainfall has on the hydrological condition and consequently stability. This factor is considered essential for the improved analysis of slope stability in the tropics. 3.3 Application of the developed combined hydrology slope stability model to any standard engineering analysis of stability, though preferential, is limited by the model's demand on data, computational hardware requirement, and the need for system familiarity. This, in combination with the need for rapid assessment of stability, emphasises the utility of design charts that summarise results of model simulations. 3.4 The developed model has been applied to provide design charts that summarise the dynamic response in the factor of safety (in terms of the overall minimum factor of safety) to a specified return period rainfall event for a representative set of initial slope conditions. An example of the developed slope stability design chart is illustrated in Figure -1. 3.5 The suite of potential input conditions covered by the design charts are summarised in Figure 5. The summary of design charts are given in Appendix A. 3.6 Rainfall - Previous studies of slope instability in the tropics have shown correlation between landslide occurrence and the 24 hour rainfall total (Brand et al, 1984). Four 24 hour rainfall events have been used in the development of the design charts, these being 250mm, 350mm, 450mm and 550mm. By matching the 24 hour design rainfall, for any particular return period, to those used for the development of the design charts, the factor of safety (for any particular slope condition) can be determined.

3.7 Slope form- The charts have been specifically developed for application to engineered cut slope/embankment design and hence summarise slope stability response for the four slope angles of 2:1 (63), 1:1 (45), 1:1.5 (34), 1:2 (27). Slope heights range from 6 meters to 36 metres in 6 metre height increments. 3.8 Soil characteristics - The range of soil strengths is designed to envelope those expected in the tropical soil conditions (see Figure 5). Similarly the soil permeability values used in the design charts aim to represent the range of values that can be expected in the majority of cases in the tropical condition (1x10-7ms-1 - 1x10-5ms-1). 3.9 Antecedent conditions - The initial surface soil moisture condition for the sake of simplicity is limited to consideration of a 1 metre suction equivalent and a uniform suction gradient to the water table is assumed. Four initial water table conditions are represented in the charts - the water table is assumed to pass through the slope toe and to extend upslope to a level expressed as a percentage of the slope height at the top of the slope (0%, 25%, 50%, & 75%). 3.10 The important relationship between hydrology and stability is apparent from comprehensive analysis of the stability response to the 24 hour rainfall events provided by the design chart factor of safety data base. Initial water table height and soil permeability exert strong controls on the stability of slopes, the influence of which is argued as important as that of the strength of the soil (c', '). 3.11 These results serve to emphasise the need for thorough investigation of slope hydrological conditions in the analysis of slope stability in the tropics.

INSTRUCTION 1.1
When to use the slope stability design charts
3.12 Standard engineering slope stability charts are used in the assessment of stability conditions in the absence of a full analytical analysis. The accuracy of the prediction depends on the closeness of fit between the actual slope characteristics and the chart assumed conditions. Differences between the two approaches result from the resolution of the stability charts in terms of the profile, soil material strength and soil water conditions. However upper and lower band estimates can be established and an indication of the likely range of values obtained. If this range is large then a more detailed analysis will need to be undertaken. 3.13 Standard slope stability charts such as those of Bishop and Morgenstern, and Janbu should be considered for: 1) 2) Initial approximate stability predictions Slope stability analysis in the absence of a more detailed stability assessment Highlighting those slopes that require more detailed analysis.

3)

Figure 4 Example design chart for a 250mm 24 hour rainfall event and slope cut at 2:1

3.14 However such charts only provide a summary of conventional slope stability analysis methods. As such they represent static analyses with defined static soil moisture conditions - no account is made of the dynamic hydrological conditions. As a result, analyses usually take the form of assessment of the stability for an assumed worst case condition - that is a condition hydrologically that is considered the worst the slope is likely to experience. Estimation of such worst case conditions in the tropics is however difficult. By assuming a water table at the surface (as a theoretical worst case condition) over conservative designs will be derived with their associated unnecessary construction costs. Underestimation of the potential worst case soil moisture condition (maybe by a `design by precedent' approach) can result in slope failure. 3.15 By using the model generated design charts which summarise the slope stability response to known recurrence interval rainfalls with `realistic' initial water table conditions an improvement on both standard engineering charts and standard slope stability assessment methods is achieved. For this reason it is recommended that the developed charts are used as part of any analysis of stability for the tropical residual soil condition.

3.16 However, while the developed charts are considered to provide a more accurate and realistic assessment of tropical slope stability conditions to conventional forms of analyses (in that they account for the effect of rainfall on slope hydrology and stability) the limitations of a design chart procedure must be appreciated. All slope charts are restricted to consider slopes that have an approximately straight cross section in profile, simplified hydrological conditions (e.g. a specific groundwater level), homogeneous soil material of known shear strength, and a potential failure surface controlled by the stress conditions rather than by geological features.

INSTRUCTION 1.2:
How to use the developed slope hydrologystability charts
3.17 To use the design charts it is necessary to select a number of options so as to match the modelled condition to the slope of interest. 3.18 Figure 5 diagrammatically illustrates the modelled conditions covered by the design charts. These are

10

Figure 5 Summary of the range of conditions covered by the developed design charts

rearranged in Figure 6 to give structured choice diagram. The result of making these choices allows the most appropriate chart to be selected. These are displayed as separate charts for each slope angle and each design rainfall, giving 16 charts. Figure 4 gives an example design chart whilst the complete set of design charts are listed in Appendix A of this manual. To fully utilise the developed charts the following input information is required: 1) 24 hour rainfall for a specified return period 2) Slope height 3) Slope angle 4) Soil permeability 5) Water table height 6) Soil strength 3.19 The use of the developed charts is best illustrated by example. Consider a slope which has the following (approximate) characteristics: Soil Permeability Soil strength Groundwater level 1x10-6ms-1 c' = 5kPa, ' = 35 50% of the slope height at the top of the slope

Slope angle Slope height Rainfall for area

1:1 15 metres 450mm (total for a 24 hour 100 year event)

3.20 Select appropriate chart from Appendix A, on the basis of rainfall (450 mm 24 hour event) and slope angle (1:1). Select the central section of the chart (Ksat = 1x10-6ms-1) and, as there is no specific 15 metre slope, obtain the factor of safety values for slope heights that envelope this value in the 50% water table height category for the relevant soil strength (c' = 5kPa, '=35). For the 12 metre slope the factor of safety is between 1.2 and 1.3 and for the 18 metre slope the factor of safety is between 1.1 and 1.2. It is therefore reasonable to assume that the factor of safety for the 15 metre slope is likely to be about 1.2. It is useful to note at this point the important influence that soil permeability has on the hydrological conditions and hence slope stability by alteration of the effective soil strength. For soils of the same material strength but permeabilities greater than 1 x 10-6ms-1 this slope design would be unstable. This example serves to stress the importance of determining slope material permeability in site investigation (see Instruction 3).

11

Figure 6

Decision structure for design chart selection

12

SUMMARY DECISION STRUCTURE FOR INSTRUCTION 1

13

4. INSTRUCTION 2
INSTRUCTIONS RELATING TO SUCTION/ STRENGTH RELATIONSHIPS The use of the resistance envelope procedures
4.1 In this section the resistance envelope procedure is illustrated in terms of its utility in predicting the expected average soil water conditions which will initiate slope instability. Additional features allow for the rapid analysis of a number of slopes in terms of; 1) the hydrological controls on the stability 2) depth of the likely failure surface 3) an estimate of the existing factor of safety 4.2 By using this procedure it is possible to advise on the type and installation of soil water instruments, the type of laboratory tests and the method of analysis that should be employed in the assessment of slope stability. 4.3 Results from using the resistance envelope procedure and back analyses have shown that some slopes display a degree of suction controlled stability. This is shown by the fact that some slope failures occur even though soil suctions within the material still persist. If this is found to be the case, then validation may be necessary in the form of suction controlled laboratory tests such as that described by Anderson and Kemp (1987). 4.4 In instances where slope failures occur in partially saturated soil then the analysis and design of slopes should be by a methodology consistent with the actual slope conditions. The use of the effective stress theory for saturated soils is tried and tested and gives realistic predictions. This is not the case for partially saturated conditions in which suctions may be important. A number of attempts have been made to incorporate suction into an effective stress analysis directly. However, no straight forward solutions can be applied, making it necessary to clarify the most appropriate methodology. 4.5 The resistance envelope procedure can be used to graphically indicate the average equilibrium normal stresses within a slope. Envelopes have been constructed from a large number of previous analyses which have been organised in the form of a dimensionless chart - the envelopes being independent of slope material strength, Figure 7a. By using this chant the required strength for slope stability can be assessed from knowledge of only the slope profile and material density.

INSTRUCTION 2.1
Procedure for the construction of a resistance envelope for a slope
4.6 1) Obtain and draw the slope cross sectional profile. A number of profiles may be necessary to account for variations in slope angle. 2) Obtain the slope angle (), slope height (H, metres) and soil material density (, kNm-3). 3) Given the slope characteristics in (2) select a range of normal stress values () over which the envelope is required (for example, 0 - 150 kPa) 4) Determine the dimensionless normal stress component ( / H) by dividing the normal stress by the material density and slope height. 5) Using the dimensionless chart, Figure 7a, obtain the dimensionless shear strength component ( / H) for the corresponding normal stress component and the given slope angle. 6) Obtain the shear strength () by multiplying the shear strength component ( / H) by the material density () and slope height (H). 7) The resistance envelope for the slope can now be drawn by plotting the values of normal stress () against the values of shear strength () on an equal axis diagram. An example of this procedure is given in Table 2.

INSTRUCTION 2.2
Using the resistance envelope summary charts
4.7 To simplify the above process, resistance envelope design charts for slopes cut a 1:1 (figure 7b) and 2:1 (figure 7c) have been constructed for a range of slope heights (10, 20, 30, 40. 50 metres). Each envelope has been constructed using the procedure described above assuming a soil density typical of a residual soil, 18 kNm-3. To use these charts select the appropriate slope angle and trace the curve for the applicable slope height. For slopes that have a height between those illustrated interpolation between the nearest two slope heights should be undertaken.

14

Table 2: Procedure for the construction of a resistance envelope (kPa) 0 5 10 20 30 40 50 70 90 / H 0 0.01 0.02 0.04 0.06 0.08 0.10 0.14 0.18 / H 0 0.011 0.022 0.043 0.063 0.076 0.097 0.125 0.140 (kPa) 0 5.5 11.0 21.5 31.5 38.0 48.5 62.5 70.0

Slope Height - 25m Slope angle - 50 degrees Material density - 20 KNm-3 (a) Select a range of normal stress values over which the envelope is required (e.g. = 0 to 150 kPa). (b) Determine the dimensionless normal stress value by dividing the normal stress by the material density and slope height to give 6/H). (c) Using the dimensionless chart, figure 7a, obtain dimensionless shear values (/H) for the corresponding normal stress values for the given slope angle. (d) Determine the shear strength () by multiplying the values in (c) by the material density and slope height.

Figure 7a

Dimensionless resistance envelopes from which a resistance envelope for any slope angle (80 > > 20) and height (H) can be generated

15

Figure 7b&c

Resistance envelopes for slopes of varying heights and angles 1:1 (b) and 2:1 (c)

16

INSTRUCTION 2.3
Using the resistance envelope to determine the threshold soil water conditions for slope stability.
4.8 The resistance envelope constructed by following Instruction 2.1 or 2.2 represents the stability equilibrium conditions for a dry slope. By superimposing this envelope on the saturated shear strength envelope obtained from shear box (BS 1377 : 1990, part 7) or triaxial tests (BS 1377: 1990, part 8) of the slope material it is possible to identify the threshold soil moisture conditions required for stability equilibrium over a range of different normal stress levels. The resistance envelope approximates to the required shear strength to maintain stability and so, assuming stability is maintained, any differences in shear strength for a given normal stress when compared to the saturated envelope (the actual available shear strength) can be considered a function of the effective stress conditions. The normal stress condition which exhibits the greatest difference in shear stress when comparing the saturated envelope to the resistance envelope indicates the soil water conditions (in terms of negative pressures) necessary to maintain slope stability. This match can be obtained by a horizontal transformation of the saturated envelope by adjustment of the pore water value, u, as illustrated in Figure 8c. Three type conditions can thus be envisaged: 1) Figure 8a. The strength envelope always lies above the resistance envelope. The available strength is always greater than the required strength and instability will only be induced by the development of positive pore water pressures. In such circumstances a ground water rise or the development of a perched water table will control stability. 2) Figure 8b. The envelopes coincide at a particular normal stress giving stability equilibrium while at all other points the shear strength is greater than that defined by the resistance envelope. Instability at this common stress level may occur through the dissipation of suctions by infiltration or rise in ground water. 3) Figure 8c. A part of the strength envelope lies below the resistance envelope. For this stress range suction must be maintained to effect slope stability. The threshold suction required can be estimated from the lateral displacement of the two envelopes for the particular shear stress value.

INSTRUCTION 2.4:
Calculation of the likely average depth of the slope failure zone.
4.9 The comparison of resistance envelope to saturated strength envelope procedure described above can be used to provide an indication of the probable most critical normal stress condition - i.e. the normal stress condition at Figure 8

'''

Three resistance envelope and shear strength relationships showing the equilibrium soil water conditions

17

which there is the greatest difference to shear stress. An approximation of the likely average failure depth can be obtained by dividing this normal stress value (a) by the material density (7) such that: d = where: d= = = average depth of failure zone (m) critical normal stress (kNm-2) material density (kNm-3) /

4)

An estimate of the minimum factor of safety can be obtained (see Figure 8) by dividing the saturated shear stress value (S 1) by the resistance envelope value (S2) for the normal stress condition that exhibits the greatest difference in shear stress when comparing both envelopes. The type and suitability of remedial works can be alluded to once the critical hydrological conditions have been established. Two type cases can be identified: slopes for which stability is not suction controlled. These will benefit most from remedial measures designed to reduce the ground water level (such as horizontal draining). slopes for which stability is controlled by suction. These will benefit from remedial techniques that prevent infiltration. Slope covers such as chunam and gunite (provided they are well maintained) can increase soil suctions by reducing infiltration and this in turn enhances slope stability. Additionally, any proposed change to slope vegetation should be carefully considered to ensure that surface infiltration is not increased and that sufficient evapotranspiration occurs to maintain soil suctions.

5)

4.10 If a circular failure surface is assumed then it is likely the maximum failure depth will slightly exceed this value. The assumption of material homogeneity is made in this procedure. Where stability is controlled by discontinuities this technique cannot be expected to provide an accurate assessment of the likely failure depth.

INSTRUCTION 2.5:
Implications of the resistance envelope method on slope instrumentation, stability analysis and remedial action
4.11 A summary of the hydrological and slope stability implications of the results of the resistance envelope procedure is provided in Table 3. This table should be considered in conjunction with Figure 8. A number of recommendations can be made: 1) The technique can be used to help decide on the type of instrumentation necessary once the importance of positive and/or negative pressures on stability conditions has been established. The type of instrumentation program required, in relation to potential failure mechanisms, is described in Table 3. 2) The depth and position of instrumentation schemes can be identified in order to monitor conditions in the critical stress range, see Table 3. 3) The envelope characteristics can form the basis for advising on the type of slope stability analysis, in particular: consideration of either deep or shallow slip circles the form of treatment of the effective stress conditions and decisions made as whether to use the Fredlund et al (1978) approach to unsaturated soil stability analysis. assess whether there is a need to obtain actual monitored data (for example when a suction control is identified) or whether simple ground water assumptions can be made.

INSTRUCTION 2.6:
The use of the resistance envelope procedure in choice of stability analysis method.
4.12 When suction control is identified as a major component of stability equilibrium then it is realistic to include suction in the detailed analysis of slope stability. Standard analysis of stability uses the Mohr Coulomb relationship with effective stress conditions. In this circumstance the effective normal stress is represented by ' = u w where: ' = effective normal stress = normal stress uw = pore water pressure 4.13 For low suction conditions (<20 kPa) then suction can be considered directly in the Mohr-Coulomb relationship, so that the effective normal stress is given by: ' = u suction where: u suction = soil suction (kPa)

18

Table 3: Summary interpretation of the selected resistance envelope and shear strength relationships

19

4.14 If suctions Greater than 20 kPa are likely then direct incorporation of suction into the effective stress equation will lead to an over estimate of the suction effect and the analysis will be unrealistic. In these circumstances it is necessary to choose one of the following options 1) to specify a lower more conservative suction 2) define a maximum suction value of 20 kPa 3) adopt the Fredlund et al (1978) approach. 4.15 Of these three approaches that of Fredlund et al (1978) is the only one to account for all stress state variable combinations within the analysis - the other two techniques simplifying consideration by ignoring the air/water phase, employing a conservative simplification to the analysis. The equation for analysis of slopes with matrix suctions greater than 20 kPa requires an additional independent variable ( b) which relates to the matrix suction (u a u w) such that: = c' + ( - ua)tan (ua uw)tan b 4.16 The b has to be determined using sophisticated laboratory tests in which both the pore air and water pressures are controlled. This analysis can be simplified by assuming typical values of b and a rearrangement of the Fredlund et al (1978) formula. It can be shown that the air water interaction term can be simply incorporated into the standard Mohr Coulomb equation such that: = c new + ( u w ) tan ' where: c new = c' + ( u a u w ) (tanb -tan') that is, by modification to the cohesion component, analysis can be made of the actual shear stress conditions using the standard method of analysis. 4.17 An average b value for a decomposed granite soil is given as 15.3 whilst for a decomposed rhyolite a value of 13.8 is suggested (Ho and Fredlund, 1982). Typically the b value has a standard deviation of 5. For such analyses it is assumed that the u a component has a value of zero.

20

SUMMARY DECISION STRUCTURE FOR INSTRUCTION 2

21

5. INSTRUCTION 3
INSTRUCTIONS RELATING TO THE MEASUREMENT OF PERMEABILITY
5.1 The importance of permeability in controlling slope stability to the tropics has been shown in Instruction 1. It is therefore necessary, along with an assessment of the slope antecedent hydrological conditions, to identify the saturated permeability (saturated hydraulic conductivity) of the material forming the slope. It is the purpose of the following Instruction to detail the methods available for the measurement of permeability. This should be considered as a necessary procedure during either site investigation or laboratory analysis of sample material, complementary to standard site investigation. Standard site investigation procedures are reported elsewhere (BS 1377, BS5930, GCO, 1984) and are therefore not discussed here. 5.2 The determination of soil permeability can be achieved either in the field or in the laboratory and the method by which it is assessed is dependent on whether the soil being tested is saturated or unsaturated (above or below the groundwater level).

2) Rising or falling head test 5.5 In this test the change in hydraulic head over time is used to calculate the soil permeability. Saturated permeability is then defined by: A.loge (H1 / H2) k= F(t2 t1) where: k= F= A= Saturated permeability (ms-1) The piezometer shape factor (see Instruction 4.2) cross sectional area of the standpipe casing (m2)

(If the casing is not vertical then the horizontal area should be used) H1 = H2 = Water head (m) at time t1 (s) Water head (m) at time t2 (s)

INSTRUCTION 3.2
To determine the permeability of soils above the groundwater
5.6 The method described above cannot be used for the determination of soil permeability above the groundwater level as the assumption that saturated flow is maintained is broken. Steady state conditions can only be used if the test becomes very long (as t approaches infinity) and then the permeability can be determined using a constant head test. To approximate thus situation through extrapolation it is necessary to plot the flow rate, q, against the reciprocal of the square root of time (1/ t). When plotted in this form the test data should provide a linear relationship from which the flow rate at infinity can be calculated (by extending the line back to 1/ t = 0 allowing determination of q (infinity)). Once this value has been obtained the permeability can be calculated from the constant head test equation such that: k= where: k= Saturated permeability (ms-1) The piezometer shape factor (see Instruction 4.2) Head measured from the centre of the piezometer ceramic to the level maintained in the standpipe for the test (m) q (infinity) /(F.H)

INSTRUCTION 3.1
Determination of soil permeability for conditions below ground water level using piezometers
5.3 Two test methods using the Hvorslev (1951) equations can be adopted for the measurement of permeability of soils using piezometers below the ground water level. These are 1) Constant head test 2) Rising or falling head test 1) Constant head test 5.4 In this procedure the water level is maintained at a constant height in the piezometer standpipe. The volume of inflow or outflow required to do this can be used to assess the saturated permeability of the slope material at the piezometer installed depth. The permeability is defined by: k= where: k= q= F= H= Saturated permeability (ms-1) Inflow or outflow rate required to maintain Hc (m3s-1) The piezometer shape factor (see Instruction 4.2) Water head above or below the standing groundwater level maintained at constant head (m) q/(F.Hc)

F= H=

5.7 The methods described have detailed techniques for the measurement of soil permeability at depth (standpipe installed depth). It is also possible to determine the surface saturated permeability or saturated infiltration capacity.

22

This is defined as the maximum rate at which water will enter the soil surface by infiltration when the soil is saturated.

INSTRUCTION 3.3:
Field measurement of the saturated infiltration capacity of soils
5.8 The saturated infiltration capacity of a soil can be measured using a ring infiltrometer, Figure 9. The principle of the test involves the flooding of a known surface area of soil and then measuring the volume of water lost through the soil over a known time period. The infiltration rate is defined as: IR = where: IR = q = A = 5.9 Infiltration rate (ms ) Volume of water lost (m3) in time t (s) Flooded surface area (m2)
-1

the boundary effects. This is achieved by installing a second ring that has a diameter twice that of the inner ring. The depth of the outer ring is most effective when it is below that of the soil wetting front during the test. As for Instruction 3.2 the flow rate, q, is plotted against the reciprocal of the square root of time to allow extrapolation to t= and hence the saturated infiltration rate.

INSTRUCTION 3.4:
Laboratory permeability tests
5.10 As with field tests there are two techniques available for the determination of soil permeability: 1) The constant head test 5.11 Figure 10a illustrates the apparatus required for the determination of permeability in the laboratory using the constant head method. Saturated permeability is calculated from the following equation: k = V.L/(A.t.H)

q/(A.t)

The accuracy of the test can be improved by reducing where: k V A L H = = = = = saturated permeability (ms-1) volume of flow (m3) in time t (s) horizontal surface area of the sample (m2) sample thickness (m) hydraulic head (m)

2) The rising or falling head test 5.12 Figure l0b illustrates the apparatus required for the determination of permeability in the laboratory using the rising or falling head approach. The saturated permeability of the sample is defined as follows: k = where: k rt rc L H1 & H2 = = = = = saturated permeability (ms-1) radius of the head tube (m) radius of the sample tube (m) sample thickness (m) water heads at time t1 (s) and t2 (s) respectively (m) (rt2.L/(rc2.[t2 t1])).loge (H1/H2)

Figure 9 Surface ring infiltrometer

23

Figure 10

Laboratory equipment for the measurement of hydraulic conductivity a) constant head b) falling head

24

SUMMARY DECISION STRUCTURE FOR INSTRUCTION 3

25

6.

INSTRUCTION 4

Instructions for piezometer monitoring

6.1 Piezometers are used to monitor the pore water pressure conditions at a particular- location and time beneath the ground surface. In this context it is a method by which the level and movement of the groundwater level can be established for input into a stability analysis. As a result of the variable nature of field groundwater and soil conditions a number of piezometer systems have been developed each with their own specific characteristics. For realistic and accurate measurement of conditions it is necessary to match the instrument specification to the likely groundwater response. To optimise performance the design most suited to the application required has to be determined. 6.2 Figure I1 allows illustration of the decisions that need to be made when selecting a piezometer system. In most cases the piezometer selection can be made using Instruction 4.1 but the performance, design and installation need evaluation for particular applications so that the suitability of system to its application can be judged. 6.3 The performance of a piezometer is represented by its time lag characteristics which can be determined experimentally in the field or by an empirical approximation. A tong time lag will result in poor piezometer performance manifested by a delayed response to groundwater change and a reduction in the peak recorded value in comparison to actual values. The type, shape and installation characteris-

tics of piezometers affect the time lag and are factors that must be addressed in the choice of system for installation. Broadly, piezometers can be divided into two categories 1) Open systems in which the piezometer is vented to the atmosphere (high volume systems). Closed systems in which the piezometer is sealed from the atmosphere (low volume systems).

2)

6.4 As well as considering the performance of the piezometer so must the operational aspects of groundwater monitoring be addressed. Many piezometer systems come with an integral data logging facility - when this is not the case others can be adapted (Instruction 6). The choice of either an automatic or manual system should be assessed in terms of the value of the data, the equipment reliability and system security.

INSTRUCTION 4.1:
Piezometer selection based on groundwater conditions
6.5 Figure 11 illustrates the piezometer selection procedure for four typical (observed or anticipated) pore pressure responses. Table 4 details the four response curves and the basic piezometer requirements whilst Table 5 summarises the characteristics of commonly used piezometer systems. A summary of piezometer selection based on anticipated response can be expressed as follows (with reference to Figure 11):

Figure 11 Four typical piezometer response characteristics

26

Table 4: Summary details of the piezometer responses given in Figure 11

Groundwater Response

Approximate Average Rate of Change 1m / month

Pore Pressure Range

Piezometer system

Positive

Open standpipe system. Low air entry piezometer tip. Manual monitoring (dip meter) Casagrande piezometer, filter design to be calculated from piezometer basic time lag and required shape factor. Consider automatic monitoring - bubbler or acoustic systems Rapid response, low volume factor system required closed hydraulic or transducer. If always 0 kPa then use low air entry piezometer tip. Use a data logger system to control monitoring cycle and store frequent data Main specifications as C above. Must use a high air entry ceramic to avoid de-airing of the system. For near surface monitoring may consider automatic tensiometer system, see - Instruction 5.2

1m /week

Positive

1m / day

Positive

1m / day

Positive and negative

CURVE A:

Little variation in the pore pressure conditions. Only approximate groundwater location is required. Consider using an open hydraulic stand-pipe system with a low air entry piezometer tip. Manual monitoring (dip metre) can be used. If a large number of piezometers are required and/or installations are in remote locations, consider using automatic systems (e.g. data logger with either a down stand-pipe transducer system, or vibrating wire piezometers). Fluctuating positive pore pressure conditions. For most cases a Casagrande standpipe piezometer is suitable with either manual or automatic monitoring. For soils with permeabilities that are less than 1x10-9ms-1 consider a low volume system (minimise the internal diameter of the stand-pipe) to avoid a lagged response. The instrument suitability and filter design can be assessed by calculation, see Instruction 4.2. Rapid changes in positive pore pressures as a response to individual storm events. To monitor the worst soil pore pressure conditions it is essential to use a low volume

system. Closed hydraulic piezometer systems or vibrating wire piezometers are therefore preferable. Further enhancement in response can be achieved by using a low air entry ceramic, provided the tip always remains saturated. To ensure maximum efficiency it is recommended that an automatic monitoring system is employed. CURVE D: Rapid pore pressure changes that may be both positive or negative. This response is typical of that experienced in steep tropical residual soil slopes when piezometer tips are located at shallow depths and within zones where perched water tables develop. It is essential to match equipment to the range and rapidity of change in conditions likely to be experienced. A closed system is preferable to ensure a low volume factor and hence achieve rapid response. The system must be capable of being de-aired in situ and it is necessary to use a high air entry ceramic tip. Automatic monitoring is essential - a transducer based system provides maximum flexibility. Tensiometers should also be installed to provide additional negative pore pressure information, Instruction 5.2.

CURVE B:

CURVE C:

27

Table 5: Summary of Piezometer characteristics

Type Open hydraulic (Casagrande) Low air entry tip Range Atmospheric pressure to top of standpipe Response Slow due to high volume factor of the standpipe. Design filter size to obtain suitable factor F. Response improved by decreasing standpipe diameter Deairing Self de-airing as long as standpipe diameter is >12mm Monitoring Manual - dip meter Automatic Bubbler system Acoustic system Transducer within standpipe Comments Manual system is cheap, simple and easy to read making it the first choice for the measurement of positive pore water pressures. For reasonable to good response the material permeability should be less than 1 x 10-' CMS -' Automatic systems increase the overall flexibility but not its response. In general, suitable for majority of applications unless fast response required. Only satisfactory if the tip is always saturated. Tip can be installed during construction and hydraulic lines can run in any direction as long as the absolute vertical distance between the tip and sensor is known. Multiple system can be used, however. Separation distance restricted by elevation changes and hydraulic resistance in the tubing. For large separations use electrical system. High air entry tip must be used when the material around the tip may become partly unsaturated. This system also capable of measuring suctions. Other considerations are the same as above. Transducer/tip can be located at any position and is connected to a data logger using electrical cabling. Large piezometer tip separations (up to 500m) can be achieved using a small amplifier circuit. Electrical system is extremely flexible and adaptable However, it is the most expensive system.

Closed hydraulic Low air entry tip

Any positive. May be restricted by the head difference between the tip and monitoring sensor

Moderate to rapid depends on the volume factor of the monitoring device

Can be de-aired

Manual Mercury Manometer Bourdon Gauge Automatic Pressure transducer (multiple systems require either a number of transducers or a fluid scanning switch)

Closed hydraulic High air entry

-1 atmosphere to any positive pressure. Same restrictions as above

Similar to above Slight restriction due to lower permeability tip

Can be de-aired

As above

Electrical Transducer located within a high air entry tip

-1 atmosphere to any positive depending on the transducer specification

Rapid

Not insitu without modification

Transducer output is in millivolts which can be manually measured or automatically using a data logger

28

INSTRUCTION 4.2
Calculation of the time lag, T, equalisation ratio, E, and intake factor, F, for a piezometer.
6.6 Closed piezometer systems that employ transducers, for most practical purposes, can be regarded as having an instantaneous response, though response time will be lengthened by long distances between the monitoring device and the piezometer tip. 6.7 For open systems the piezometer response can be represented by the system time lag, that is the time taken for the system to respond to an external change in pore water pressure. This can be affected by the shape and type of piezometer used, the response of the soil to instrument installation, both the consolidation and swelling of the soil in response to changes in the soil moisture content, and the erosion and/or build up of soil fines in and around the piezometer tip. 6.8 A number of important concepts need consideration. For any change in external pore pressures there will be an associated response from the piezometer. With an ideal system this will be instantaneous and of the same magnitude as the external change. However, due to the factors described above the response is characteristically delayed and of a smaller magnitude to that in reality. The difference between the measured and actual response introduces the terns equalisation and equalisation ratio. Equalisation is the matching of the change in external pressure to that recorded in the piezometer. The equalisation ratio is the ratio between the actual change and that recorded in the piezometer. The time taken for flow to or from a piezometer until equalisation (or a desired level of equalisation) is achieved is called the hydrostatic time lag. In addition to the hydrostatic time lag there is the stress adjustment time lag. This is the time lag introduced into the system as a result of changes in the soil stress conditions both from initial disturbance of the soil from installation and transient consolidation and swelling of the soil with changes in the soil moisture conditions. Stress adjustment time lag is minimised by having a small well point (filter and ceramic) and large volume factor (open system). This is in direct opposition to the requirements to minimise hydrostatic time lag. 6.9 For tropical residual soils and for circumstances where observations are to be extended over time it is the hydrostatic time lag that is the dominant factor in controlling the overall system time lag. Emphasis should therefore be placed on maximising the well point and minimising the piezometer volume factor. Calculation of the time lag of a piezometer system 6.10 The hydrostatic time lag of a piezometer system can be calculated using the Hvorslev (1951) basic time lag function. This formula allows calculation of the lag time

(T), for the equalisation of an initial pressure difference when the original flow rate (q) is maintained, see Figure 12. Given that the flow rate, q, can be calculated by: q= where q= F= Ho = flow rate (m3s-1) piezometer shape or intake factor (m) Hydraulic head at time zero (m) FkHo equation 4.1

and that the volume of the piezometer, V is: V= where V= A= piezometer volume (m3) Piezometer cross sectional area (m2) A.Ho equation 4.2

Ho =

Hydraulic head at time zero (m)

then the basic time lag, T, can be calculated by: T= V/q = AHo /FkHo = A/Fk equation 4.3

6.11 Figure 12a and b illustrate that the basic time lag corresponds to an equalisation ratio of 0.63. Figure 13 summarises a number of basic time lag responses, from which the head changes and equalisation ratios can be calculated for specific time periods. 6.12 From equation 4.3 it can be seen that piezometer system design should seek to minimise the volume factor, V, and maximise the intake factor, F, to provide monitoring with a minimum time lag. Calculation of the equalisation ratio, E 6.13 The equalisation ratio of the system, E, is defined by: E = 1-H/Ho = 1 e-( t/T) where H = Hydraulic head at time t (m) Ho = Hydraulic head at time zero (m) t = time interval from the initial change (s) T = the basic time lag of the system (s) From this equation it can be seen that the equalisation ratio is dependent on the basic time lag of the system. 6.14 Figure 14 illustrates the effect that different equalisation ratios have on the piezometer response to a particular groundwater change. Two points to note are: equation 4.4

29

1)

The time of the piezometer peak is determined by the equalisation ratio of the system The magnitude of the peak is controlled by the equalisation ratio.

2)

6.15 An equalisation ratio of 0.9 can be considered adequate for most practical cases and this corresponds to a time lag of 2.3 x basic time lag: T90 = Where T90 = T = Lag time for 90% equalisation Basic lag time of system 2.3T equation -1.5

Calculation of the intake factor, F 6.16 From equation 4.3 it can be seen necessary to calculate the intake factor of the piezometer system if calculation of the time lag and equalisation ratio is to be achieved. Figure 15 summarises the equations for calculation of the intake factors for a variety of piezometer constructions (Hvorslev, 1951). For a standard Casagrande type piezometer installation the intake factor, F, is given by: 2L F= Figure 12a Basic geometry of the Hvorslev (1951) solution to response times of open piezometer systems 1n Where: L= D= Length of the piezometer tip (m) Diameter of the piezometer tip (m)

[ L / D + 1 1 + [L / D] )]
2

equation 4.6

6.17 Subsequent verification of this formula by Brand and Premchitt (1980) suggest the following modification: 2.4L F= 1n

[ 1.2L / D + 1 (1 + [1.2L /D] )]

2

equation 4.7 The difference between the two formulas however is small enough to make either procedure appropriate. 6.18 Worked example for the calculation of the basic time lag for a stand-pipe piezometer Calculate the time lag for a piezometer installation given the following data: Filter length Filter diameter Figure 12b Analytical solution to the Hvorslev (1951) response times = = 21 cm 10 cm 1.92cm

Stand-pipe tube diameter =

Soil permeability

1x10-5cms-1

30

Figure 13 Summary diagram for calculating piezometer basic time lag and equalization ratios

Figure 14 Dampening characteristics of various piezometer specifications

31

Figure 15 Shape factors (F) for various piezometer constructions according to Hvorslev (1951)

32

1) Calculate the intake factor for the system using equation 4.6. Given that:

2)

If the ceramic permeabilty is less than 50 tunes that of the soil whilst the filter permeability is greater that 50 tittles that of the soil (Kf /Ks > 50 and Kc /Ks < 50) then the overall system intake factor, F*, needs to be calculated. This requires the calculation of the intake factor, F (FS), and the ceramic intake factor, Fc. This allows determination of the influence factor, R or R' (given in Figure 17), which is used to calculate the overall system intake factor, F*. An example solution is given below.

Given the following conditions determine the overall system intake factor, F* Piezometer and soil properties: 2) Calculate the basic time lag of the system using equation 4.3. Given that: Filter length and diameter Ceramic tip dimensions 100 x 10 cm 15x3.3cm (internal) and 4.85cm (external diameter) 1 x l0-4 and 1 x l0-3cms-1

where:

Soil and tip permeability

By using equation 4.6 the intake factor for the soil/filter, Fs, can be calculated:

3) Calculate the time lag to monitor 90 % of the actual peak value. Given that: The intake factor, Fc, for the ceramic can be calculated from the following equation: where:

then

where 6.19 Not only can the shape of the piezometer filter affect the intake factor, F, but also the permeability of the ceramic, see Figure 16. Kemp et al (1989) have carried out finite difference modelling of the ceramic/filter/soil system and stress the need to consider a combined filter and ceramic intake factor, F*. The results of this study provide the following recommendations: 1) If the filter and ceramic permeabilities are more than 50 times that of the soil (Kf /Ks > 50 and Kc /Ks > 50) then the combined intake factor, F*, approximates to the Hvorslev intake factor, F, and no correction factor need be applied. 1 d1 d2 hence: = = = length of the ceramic internal diameter of the ceramic external diameter of the ceramic

33

Figure 16

Schematic representation of the piezometer filter zone

Figure 17 Calculation of the influence factor on intake shape factor

34

Using the chart in Figure 17 and locating the positions for Fs/Fc = 0.855 and Kc/Ks = 10, a value for R of 0.90 can be determined. The intake factor for the system therefore can be calculated using the formula:

to give an overall intake factor value:

3) If the ceramic and filter permeabilities are both less than 50 times that of the soil (Kf /Ks < 50 and Kc/Ks. < 50) then the design of the entire piezometer system can only be accurately determined by using the finite element modelling procedure outlined in Kemp et al (1989). Solutions which ignore the effect of the ceramic filter will significantly over-estimate the overall piezometer shape factor, F*, thus underestimating the time lag of the system.

INSTRUCTION 4.3:
Choice of procedures available for improving the response characteristics of piezometers; (if appropriate)
6.20 The basic procedure for response improvements is to adopt systems or improvements that reduce the basic time lag, T, of the system. This can be achieved by either reducing the volume factor or increasing the intake factor, F (or F*). Methods to achieve this for open stand-pipe piezometers are: 1) Reduction of the volume factor by decreasing the stand-pipe diameter. For example, a reduction from 19 cm to 14 cm will halve the basic time lag of a piezometer. Care must betaken however as diameters less than 121mm prevent de-airing, so should only be used in ground that is permanently saturated. 2) Increasing the filter intake factor, F, by increasing the dimensions of the filter. 3) Conversion of an open hydraulic system to a closed system by inserting a pressure transducer at the tip and sealing the tube using an inflatable packer (see Figure 18) For closed hydraulic systems there are a number of methods available to improve the response characteristics: 1) Reduction of the monitoring instrument volume factor by adopting a lower volume monitoring procedure, e.g. changing from a manometer or bourdon gauge to a pressure transducer. 2) Increasing the system stiffness by a reduction in the length of tubing by locating the sensor as close as possible to the tip. Figure 18 Inflatable packer used for converting an open standpipe piezometer to a closed piezometer system

3) A small improvement can be gained by increasing the intake factor. If the tip is always saturated then a low air entry tip can be specified. 6.21 Of the number of methods available to improve piezometer performance, that which is most effective is the change of system itself. In particular, the change from open to closed system gives by far the greatest improvement in response time and equalisation ratio.

INSTRUCTION 4.4
Selection of the monitoring and data recording equipment for the piezometer system.
6.22 Two approaches to piezometer monitoring are possible, the choice of either manual or automatic systems.

35

1)

Manual

With manual monitoring data acquisition, and hence data resolution, is totally dependent on the number of site visits. It is therefore possible to miss the worst conditions if rapid temporal changes occur. Open stand-pipe piezometers can be manually read using a dip meter, whilst worst conditions can be approximated using a Halcrow bucket system. This consists of a series of cups lowered down a stand-pipe to known depths. On returning to the site, the uppermost full cup locates the approximate position of the highest recorded water level. All manual operations require manual data transfer prior to processing. For multiple systems the work load involved in this process becomes significant. 2) Automatic (see also Instruction 6)

(ii) The storage space available in the data logger is matched to the number of instruments and monitoring resolution. The maximum period of time between site visits should be calculated so that no data is lost due to lack of system memory. (iii) The system demand on power should be minimised (low energy systems are preferred) so that the system can run unattended for suitable lengths of time without battery failure. 6.23 In general, the advantages offered by an automatic monitoring facility in terms of increased data resolution and response times (in the case of closed transducer based systems) would advocate their use for conditions in the tropics, especially for the assessment of worst case conditions for input to slope stability assessment. Such systems are however expensive. For short term monitoring projects a manually read piezometer system is the only economical choice. However, if monitoring of groundwater is to be made over a period of time then the high initial cost of such equipment and low running cost may be balanced out by the high running cost of frequent manual readings (labour cost), see Figure 19. In all cases it is necessary to calculate the difference in overall cost between manual or automatic based systems and balance that against the data quality demands. 6.24 When installing any equipment, care must be taken to ensure system security - both against theft and vandalism. This is especially the case for high cost automatic based systems.

Automatic monitoring systems allow a large amount of data to be accumulated between site visits. In addition such systems usually offer an operator controlled reading frequency, providing far greater resolution of monitoring. Consequently, such systems are advisable when rapid fluctuations in groundwater are likely if worst conditions are to be identified. With the importance of system reliability there has been a move to the use of solid state electronic data logging facilities. When considering such systems there are a number of points that should be addressed: (i) That on site monitoring devices are compatible with the logging facility.

Figure 19

Comparative illustrative costs of bubbler piezometer system and manual piezometer (read daily)

36

SUMMARY DECISION STRUCTURE FOR INSTRUCTION 4

37

7. INSTRUCTION 5
INSTRUCTIONS FOR THE MEASUREMENT OF SOIL SUCTIONS
7.1 Due to the high evaporation rates, steep slope gradients and often low natural groundwater levels experienced in the tropics, significant suctions may develop beneath the soil surface. Many studies have shown that for a large number of slopes soil water suctions may persist throughout major storm events. Back analysis and monitoring of failed slopes (especially in Hong Kong) have shown the importance of suction in maintaining a factor of safety greater than 1. This has resulted in the development of stability analysis procedures which incorporate the beneficial aspect of soil suction in the assessment and monitoring to provide the input data for such calculations.

INSTRUCTION 5.1
Selection of instrument type based on monitoring range.
7.2 The selection procedure for establishing the most suitable equipment for monitoring soil water suction is outlined in Figure 20. An important distinction between the type of equipment is in the form of suction measurements. For suctions less than 80 kPa direct field measurements can be undertaken with a water to water (instruments to soil) interface. Tensiometer systems are used for the direct measurements of soil suctions. For suctions greater than 80 kPa indirect measurements systems are required. At such suctions water in tensiometer based systems cavitates leading to unreliable readings. Normally suctions greater than 80 kPa would not be included in the analysis of slope stabilty but if it is necessary to assess such suctions an indirect MCS system (Instruction 5.5) is suggested.

Figure 20 Instrument selection for soil water suction monitoring

38

7.3 The initial assessment of the suction range that exists can be provided by using a quick draw tensiometer though this is limited to depths no greater than 50 cm below ground level.

7.5 In the event of positive pore water pressures developing, then the tensiometer acts as a closed hydraulic piezometer. 7.6 Manual tensiometers are particularly useful as they are relatively cheap, reliable, robust and easy to install. Examples are detailed to Table 6 which include; 1) Jetfill tensiometers - Figure 21

INSTRUCTION 5.2:
Selection of tensiometer equipment
7.4 Tensiometers are water filled instruments with a high air entry ceramic tip. The instrument is inserted into a small bored hole in the ground (usually 21 mm diameter). The soil matrix suction is measured directly by obtaining a water equilibrium across the tip between the soil water and the water confined in the instrument. Since soluble salts are free to pass though the membrane, the measurement is free from any osmotic effects. The soil water suction is represented by the water tension within the tensiometer and may be measured using a manometer, pressure gauge or a transducer. The actual soil water tension at the instrument tip is the recorded value on the measuring device less the vertical distance from the sensor (e.g. transducer diaphragm) to the tensiometer ceramic mid point: Total suction (mH2O) = Recorded suction (mH2O) z where: z =the vertical distance from transducer to the tensiometer ceramic mid point (m)

2) Quick draw tensiometers - Figure 22 3) Small diameter tensiometers - Figure 23 7.7 In addition to manual tensiometers there area number of automatic versions, which have either the sensor central to a number of instruments (Scanivalve system) or at the instrument location (transducer system). Automatic systems are considered in Table 7 and include 1) Scanivalve system - Figure 24 and Figure 25 2) Single pressure transducer tensiometer - Figure 26 3) Multiple transducer system (Anderson et al, 1990) Consult Figure 20 and Tables 6 and 7 to determine the most suitable system for the given application.

Table 6: Manual tensiometer systems Type Jetfill Principal Use Soil sucton (<80kPa) field studies. Typical operating depth, 0-2m. Useful in extending the range and scope of automatic systems. Equipment details Simple, cheap robust and reliable. Maximum installation 4 metres. Install using hand auger, tip must fit snugly into a 21.5mm diameter hole. On-site de-airing. Maximum installation depth (<50cm). Extremely portable and has a fast equilibrium time with soil conditions, enabling many readings to be taken. In soft soils instrument can be pushed straight into the ground. For firmer soils use manufacturer's push-in auger. Same principle as the two above but connects the instrument tip to the pressure gauge using flexible tubing, so can be installed during ground preparation.

Quick-draw

Rapid reconnaissance of near surface suctions (<50cm). Useful in establishing preferable instrument locations as well as obtaining soil water conditions, while conducting other field tests.

Small diameter tensiometer

Similar a.3es as the jetfill but with additional flexibility of installation during construction.

39

Table 7: Automatic Tensiometers Type Principal Use Equipment details

Scanivalve system

Suitable for detailed soil suction studies radiating profile and spatial variations within a localized area, e.g. establishing the effective stress conditions during storm events. Operational depths up to 4m also will monitor positive pressure conditions.

A number of ceramic tips are connected via plastic tubing toa fluid scanning switch. In turn, each tensiometer tip is linked hydraulically to a pressure transducer. The system requires a data logger to activate the timing of the scanivalve switching and for data recording. This system is limited in range by the elevation of the tip relative to the transducer. Also any faults in the switching mechanism or transducer will affect all the readings. The scanivalve system has proved to be less reliable than fully electrical alternatives. This system locates the transducer transducer within the same instrument as the tensiometer tip. The tensiometer may be monitored using a single station data logger or linked to another multichannel data logger. This system can become expensive if a number of locations need to be monitored. By modifying a jetfill to accept both a bourdon gauge and a pressure transducer (using a `y' piece) a simple flexible system can be constructed. This system is fully interchangeable and does not have any elevation considerations. Instrument separation of up to 500m can be achieved using a simple amplifier circuit. This system may be integrated into existing data logger systems.

Single pressure transducer tensiometer

Specialized electronic instrument designed for automatic data logging. Suitable where a small number required for specific locality monitoring, or where spatial influence is small.

Multiple transducer system

Use is the same as for the scanivalve system, additional advantage that spatial location of instruments less critical. Suitable for detailed geotechnical studies requiring information about the near surface (0-2m depth) soil water conditions.

40

Figure 21 Jetfill tensiometers and example installations (after McFarlane, 1981)

Figure 22 `Quick draw' tensiometer

INSTRUCTION 5.3
Tensiometer response characteristics
7.8 The theory of tensiometer response is in general equivalent to that of piezometer systems in that there are two main considerations: 1) the intake properties of the instrument 2) the volume factor of the instrument. 7.9 n practice the calculation of tensiometer response times is not necessary as the volume factor is sufficiently low to give rapid equalisation times. The quick draw

tensiometer has a very rapid response time, being only a few minutes. Figure 27 illustrates the quick draw response characteristics to three suctions in which 90 % of the equalisation occurs within two minutes. 7.10 With transducer systems better response characteristics can be achieved, especially when the transducer is located at the instrument or sensor. The response of the scanivalve system, though transducer based, maybe slightly restricted by the length (and hence `stiffness') of the hydraulic line between the tip and the scanivalve.

41

Figure 23 Small diameter (8mm) tensiometer

42

Figure 24 Automatic scanivalve tensiometer system

43

Figure 25 Redesigned tensiometer - scanivalve system employing microprocessor control unit (after Anderson and Kneale, 1987)

INSTRUCTION 5.4
Tensiometer installation, monitoring and maintenance
7.11 When installing tensiometers it is critical that the ceramic tip forms a tight seal with the soil medium so that the water within the instrument is in contact with the soil water. In general, jetfill tensiometers are inserted in hand augured holes (0-2 metres deep). For deep installations the instruments can be inserted laterally from downslope dug caissons. In the case of the scanivalve system, where the tips are attached to flexible tubing, the tips may be installed by auguring or during actual slope construction. 7.12 Before installation it is advisable to de-air the tips by either boiling in water or by applying a large suction to the tip while in de-aired water. The tip should be lowered to the base of the hole taking care not to scratch the ceramic on stony material. Slight initial back saturation will speed up the sealing of the hole around the tip. 7.13 If the range and value of soil suctions maintained during and after storm events are to be assessed then an automatic form of monitoring is required. Consider using an electrical transducer based system as this offers the greatest flexibility. Care must be taken in siting the transducer in scanivalve based systems as the difference in elevation between the sensor and tensiometer tip can cause cavitation of the instrument water.

INSTRUCTION 5.5:
The measurement of suctions greater than 80 kPa
7.14 Suctions exceeding 80 kPa would not in practise be used in stability analyses. It is however useful to outline the MCS system for monitoring soil suctions. The MCS equipment consists of a porous block of low heat conductance which is inserted into the ground. By applying a heat source at a point centred within the block and measuring the temperature rise (heat dissipation) a value for the water content can be obtained. 7.15 The effectiveness of this system is in its ability to measure suctions greater than 80 kPa though the instrument is limited by its response time and calibration needs. Instrument sensitivity in the 0 - 100 kPa range is approximately 6 kPa, see Figure 28. For suctions above 200 kPa, however, the sensor results are open to question. 7.16 Although the system allows the measurement of soil suctions in the 0 - 200kPa range it is limited by the time required to obtain a stable reading and also dependent on the absorption or desorption cycle of the soil, Figure 29. Typical response times are in the order of 160 hours making it unsuitable for transient conditions. In addition care must be taken during instrument installation as entrapped air can result in considerable measurement errors.

44

Figure 27

Response times of the 'quick draw' tensiometer (after Sweeney, 1982)

Figure 26 Individual pressure transducer system

Figure 28

MCS sensor suction - output response (after Lee, 1983)

45

Figure 29

Hysteresis in the suction moisture content relationship

46

SUMMARY DECISION STRUCTURE FOR INSTRUCTION 5

47

8. INSTRUCTION 6
INSTRUCTIONS CONCERNING DATA LOGGER SYSTEMS
8.1 Data logger systems enable a number of individual instruments to be monitored remotely. Such systems may be used in conjunction with both negative and positive pore pressure monitoring equipment, though some systems are

equipment specific (e.g. acoustic piezometer system). The main purpose for opting for a data logger is for data handling (enabling frequent readings), data storage (reducing the frequency of site visits) and data transfer from the field to the office. Figure 30 illustrates the components of a data logger system and the interrelationships between field monitoring and office processing of data. An example of a complete data logger system is the bubbler system illustrated in Figure 31.

Figure 30 Field and office components of data logger systems

48

Figure 31 Bubbler data logger system

8.2 Central to the design of modern data loggers is the use of solid state electronics and electrical digital signals. Such systems have proved to be reliable and flexible. Key areas for consideration relate to the system flexibility, compatibility (field to office) and reliability. These must be compared with system cost, value of the data and the length of monitoring before selection can me made of the most appropriate system. 8.3 Data logging systems allow on site recording of information and so are specifically associated with automatic systems. If the frequency of data gathering is high or the location of the monitoring area remote then the use of a data logger should be considered. The decision should be made by evaluating the cost of getting the data against the value of the data.

INSTRUCTION 6.2: Choice of sensor

8.6 The choice of sensor is largely controlled by the monitoring instrument and the parameters of interest (for example thermocouples for temperature, transducers for pressure, and potentiometers for displacements). The signals from the sensors can be transmitted to a central multiplexor linked to a data logger. In most cases it is more convenient to have the sensor instrument specific, so that any malfunction will affect only one set of readings. If retrieval of the sensor is desirable then sensor location is important (i.e. at the surface or central to the data logger). 8.7 The sensor chosen should operate throughout the monitoring range and be sensitive enough to that overloading of the sensor does not occur or that the length of cables between instrument and sensor does not affect instrument calibration.

INSTRUCTION 6.1:
Selection of the data logger system
8.4 Two approaches can be adopted to the selection of data logging systems: either the selection of an `off the shelf complete system or the construction of a system from individual components. Details of a number of systems that have previously been used are given in Table 8. 8.5 In general, the systems that rely more on electrical than mechanical components prove the most reliable. How-ever, fully automatic electrical systems can allow equipment failure to go undetected for some time until data is processed and so it is recommended that some form of manual monitoring is also available for on site system operational verification. For data logger based systems care must be taken not to corrupt data in storage and to ensure a sufficient data back-up before information is erased from the data logger memory.

INSTRUCTION 6.3:
Transducer selection
8.8 Transducers provide a useful sensor for monitoring piezometric and tensiometric pressures. When choosing the type of transducer to be employed it is necessary to consider: 1) the working range (e.g. 0 2 bar) 2) signal output (in relation to cable length and hence signal loss; data logging compatibility) 3) transducer type

49

Table 8: Examples of reported data logger systems

Author 1. 2. 3. 4. 5. 6. 7. Burt (1978), Anderson and Kneale (1987) Jones et al. (1984) recorder and tipping bucket raingauge Bosworth (1985) Anderson and Kneale (1987) Anderson and Kneale (1987) Durham et al. (1986) Anderson et al. (1990)

System Automatic Scanivalve for a tensiometer system Cheap data logger shown operating a stage recorder Tipping bucket recommendation Piezometer bubbler system (Geotechnical instruments) Acoustic depth monitoring system Digital pulse train data logger system for piezometers, water sampler and tipping bucket Combined automatic tensiometer piezometer system

8.9 Differences between transducers relate to the pressure conditions on the reverse side of the membrane. Three common types are: 1) Absolute transducers where the diaphragm pressure is measured relative to a vacuum. Total pressure is measured and therefore it is necessary to separately monitor atmospheric pressure to allow compensation of the recorded values. 2) Sealed gauge transducers where the diaphragm pressure is measured relative to a fixed pressure (usually atmospheric when the instrument was constructed). Depending on the accuracy required it may or may not be necessary to make atmospheric compensation. These transducers are temperature sensitive and are therefore best suited to applications such as water pressure monitoring within standpipe piezometers where temperature fluctuations are small. 3) Vented gauge transducers where the membrane is vented to the atmosphere requiring no atmospheric compensation. For field use extreme care must be taken to prevent moisture or humidity entering the transducer via the vent tube. Moisture can be removed from the air by passing the vent tube through a sealed container of silicon gel. The silica gel should be regularly replaced to maintain a dry environment within the transducer.

INSTRUCTION 6.4:
To calculate tensiometer and piezometer readings using a calibrated transducer.
8.10 When a transducer is used with a piezometer or tensiometer installation it is necessary to have the following information before calculation of actual tip pressures is possible. 8.11 For a tensiometer installation r sing an absolute transducer: XDR ATM XDRC ATMC = output of transducer (mV) = output of the atmospheric transducer (mV) = output of transducer at time of calibration (mV) = output of atmospheric transducer at time of calibration (mV) XDR gain = gain of the transducer (mH2O/mV) ATM gain = gain of the atmospheric transducer (mH2O/mV) TL = vertical difference in height between transducer diaphragm and ceramic tip centre (m) Given this information the tensiometer tip pressure is given by: Tip pressure (mH2O) = TL + (XDR-XDRC) x XDR gain - (ATM-ATMC) x ATM gain

50

8.12 For a sealed gauge transducer installed within a piezometer stand-pipe the pressure at the transducer location is given by: Pressure (mH2O = DL - (XDR-XDRC) x XDR gain - (ATM-ATMC) x ATM gain where: DL= depth of the transducer from the surface (m)

INSTRUCTION 6.6:
Other data logger system considerations, power consumption, reliability and security
8.15 Many systems are designed to run on an independent power supply. If battery power is used it is essential that a minimum power consumption system is adopted when monitoring for long periods. Typically the system should have a `wake up' sequence so that the sensors are only powered during the monitoring cycle. The frequency of site visits between battery changes and the interval time of reading will be constrained by the instruments power consumption. Approximate maximum times between battery changes must be calculated before installation. 8.16 The operating reliability of solid state electronics is generally very good as long as conditions remain within tolerance levels. Particular problems occur if there is exposure to moisture as a result of the high humidity and condensation experienced in the tropics. To eliminate such problems it is advisable not to expose any electrical components in the field and ensure that adequate casings are used. It is recommended that silicon gel bags are included within the instrument casings to remove moisture vapour. Problems may also result from exposure to extreme temperatures. This is especially likely when above ground transducers are used. It is important that for such systems instruments are properly insulated. 8.17 With the use of expensive electrical data logger systems there should be considerable concern about security. The equipment is best located out of general sight and preferably within lockable steel cabinets that are concreted into the ground. Cables between instruments and data loggers should be concealed by burial.

INSTRUCTION 6.5
Selection of data logger with solid state memory design
8.13 The use of data loggers with a solid state memory is becoming more frequent due the desirability of such systems. In some systems a typical 100 k byte memory (RAM) can store up to 50,000 readings. In most cases however, the logger is designed to enable data access and processing at set intervals during the monitoring period. In meeting this demand one of three different memory access approaches are frequently used: 1) Retrieval of the field logger and down loading of the information in the office or laboratory. For continuous readings it is therefore necessary to have two data loggers. The use of a separate interrogator with its own built in memory to down load the data logger in the field and then transfer to the office/laboratory for further down loading using a standard RS 232 link. The use of data loggers with removable memories. Spare memory cartridges are required and must be powered up all the time. These systems often need a specialist memory reader for office down loading.

2)

3)

8.14 The use of an on-site field interrogator allows for immediate data and instrument checking. However it does require field operators and office down loading facilities. A cartridge system has the advantage of transferability in that the data can be easily transported (for example by post) for processing. This is especially useful when the instrumentation is at a remote location.

51

SUMMARY DECISION STRUCTURE FOR INSTRUCTION 6

52

APPENDIX A: DESIGN CHART

Variation of factor of safety for a 250mm 24 hour rainfall event

53

Variation of factor of safety for a 350mm 24 hour rainfall event

54

Variation of factor of safety for a 450mm 24 hour rainfall event

55

Variation of factor of safety for a 550mm 24 hour rainfall event

56

LIST OF REFERENCES
ANDERSON, M G and P E KNEALE, 1980. Pore water changes in a road embankment. Highway Engineer, 25, 93-5. ANDERSON, M G and D P McNICHOLL,1983. On the effect of topography in controlling soil water conditions, with specific regard to cut slope piezometric levels. Hong Kong Engineer, 11, 35-4t. ANDERSON, M G and P E KNEALE, 1987. Instrumentation of pore pressure and soil water suction. In: ANDERSON, M G and K S RICHARDS (Eds). Slope Stability: Geotechnical Engineering and Geomorphology. Chichester: John Wiley and Sons. ANDERSON, M G and M J KEMP, 1987. Suction controlled triaxial testing: laboratory procedures in relation to resistance envelope methods. Earth Surface Processes and Landforms, 12, 649-654. ANDERSON, M G, M J KEMP, D M LLOYD and A OTHMAN, 1990. A combined automatic tensiometer piezometer system. Earth Sin face Processes arid Landforms, 15, 63-71. BOSWORTH, D, 1985. An investigation into the controls of overland flow generation. Ph.D Thesis, University of Bristol (unpublished). BRAND, E W and J PREMCHITT, 1980. Shape factors of cylindrical piezometers. Geotechnique, 30, 369-384. BRAND, E W, 1982. Analysis and design in residual soils. Proceedings Conference Engineering Construction in Tropical Residual Soils, Hawaii American Society, Civil Engineers, 89143. BRAND, E W, G W BORRIE, and J M SHEN, 1983. Field measurements in Hong Kong residual soils. Proceedings International Symposium Field Measurements in Geotechnical Engineering, Zurich. BRAND, E W, J PREMCHITT and H B PHILLIPSON, 1984. Relationship between rainfall and landslides in Hong Kong. 4th International Symposium Landslides, Toronto, 1, 377-84. BRAND, E W, M J DALE AND J M NASH, 1986. Soil pipes and slope stability in Hong Kong. Quarterly Journal of Engineering Geology, 19 (3) 301-303. BSI, 1981. Code of practise for site investigations. BS 5930: 1981. London: British Standards Institution. BSI, 1990. Methods of test for soils for civil engineering purposes. BS 1377: 1990. London: British Standards Institution.

BURT, T P, 1978. An automatic fluid scanning switch tensiometer system. British Geomorphological Research Group, Technical Bulletin 21. CRABB, G I and J H ATKINSON, 1991. Determination of soil strength parameters for the analysis of highway slope failures. In: CHANDLER, R J (Ed). Slope Stability Engineering, Developments and Applications. London: Thomas Telford, 1318. DURHAM, I M, E M O'LOUGHLIN and I D MOORE, 1986. Electronic acquisition of hydrological data from an intensively instrumented hillslope. Hydrological Processes, 1, 78-87. ENDICOTT, L J, 1982. Analysis of piezometric data and rainfall records to determine groundwater conditions. Hong Kong Engineer, 10, 53-56. FREDLUND, D G, N R MORGENSTERN and R A WIDGER, 1978. The shear strength of unsaturated soils. Canadian Geotechnical Journal, 15, 313-321. GEOTECHNICAL CONTROL OFFICE 1984. Geotechnical Manual for Slopes (2nd edition), GCO, Hong Kong. HENCHER, S R and R P MARTIN, 1982. The description and classification of weathered rocks in Hong Kong for engineering purposes. Proceedings 7th South East Asian Geotechnical Conference, Hong Kong, 1, 125-142. HENCHER, S R. J B MASSEY and E W BRAND, 1984. Application of back analysis to Hong Kong. Proceedings 8th South East Asian Geotechnical Conference, Bangkok, 631-638. HO, D Y F and D G FREDLUND, 1982. Increase in strength due to suction for two Hong Kong soils. Proceedings Conference Engineering Construction Tropical Residual Soils, American Society of Civil Engineers, Hono-lulu, Hawaii, 263-295. HUDSON, R R and S R HENCHER, 1984. The delayed failure of a large cut slope in Hong Kong. Proceedings International Conference Case Histories in Engineering, St Louis, Missouri. HVORSLEV, M J, 1951. Time Lag and soil permeability in groundwater observations. US Army Corps of Engineers, Waterways Experiment Station, Bulletin 38. JANBU, N, 1954. Application of composite slope surfaces for stability analysis. European Conference Stability Earth Slopes, Stockholm, 3, 43-49. JANBU, N,1973. Slope stability calculations. In: Embankment Dam Engineering, Casagrande Volume, John Wiley and Sons, Chichester, 47-86.

57

JONES, J A A, M LAWTON, W F WAREING and F G CRANE, 1984. An economical data logging system for field experiments. British Geomorphological Research Group, Technical Bulletin, 31. KEMP, M J, D F NASH, and M G ANDERSON, 1989. Filter design for a Casagrande type piezometer system. Earth Surface Processes and Landforms, 14, 375-382. KOO, Y C, 1982. The mass strength of jointed residual soils. Canadian Geotechnical Journal, 19. 225-231. LEE, R K C, 1983. Measurement of soil suction using MCS600 sensor. MSc Thesis, University of Saskatchewan (unpublished). LEACH, B and R HERBERT, 1982. The genesis of a numerical model for the study of the hydrology of a steep hillslope in Hong Kong. Quarterly Journal of Engineering Geology, 15, 243-259. LUMB, P, 1975. Slope failures in Hong Kong. Quarterly Journal of Engineering Geology, 8, 31-65. LUMSDAINE. R W and K Y TANG, 1982. A comparison of slope stability calculations. Proceedings 7th South East Asian Geotechnical Conference, Hong Kong, 1, 31-38. McFARLANE, 1981. Suction - moisture relationships in residual and remoulded Hong Kong soils. Public Works Department, A7aterials Division, Hong Kong, Report 24. MILLINGTON, R J and J P QUIRK, 1959. Permeability of porous media. Nature, 183, 387-388. NASH, J M and M J DALE, 1983. Geology and hydrology of natural tunnel erosion in superficial deposits in Hong Kong. Proceedings Conference Geology of Superficial Deposits, Hong Kong. PERRY, J, 1994. A method for defining non-linear shear strength envelopes, and their incorporation in a slope stability method of analysis. Quarterly Journal of Engineering Geology, 27 (3). POPE, R G, R C WEEKS and P N CHIPP,1982. Automatic recording of standpipe piezometers. Proceedings of the 7th South East Asian Geotechnical Conference, Hong Kong, 77-89. SWEENEY, D J, 1982. Some in situ soil suction measurements in Hong Kong's residual soil. In: McFEAT-SMITH, I and P LUMB (Eds). Proceedings South East Asian Geotechnical Conference. Hong Kong: Institute of Engineers, 91-105.

58

ISSN 0951 - 8797

OS-E

Hydrological design manual for slope stability in the tropics

ORN 14

Transport Research Laboratory Old Wokingham Road Crowthorne, Berkshire, RG45 6AU

Department for International Development 94 Victoria Street London, SWIE 5JL

Overseas Road Note 15

Guidelines for the design and operation of road management systems

ORN 15

First Published 1998 ISSN 0951-8797 Copyright Transport Research Laboratory 1998.

Subsector: Transport Theme:; T2 Project title: Road Network Management Project reference: R6024

This document is an output from a project funded by the UK Department for International Development (DFID) for the benefit of developing countries. The views expressed are not necessarily those of the DFID

TRL is committed to optimising energy efficiency, reducing waste and promoting recycling and re-use. In support of these environmental goals, this report has been printed on recycled paper, comprising 100% post-consumer waste, manufactured using a TCF (totally chlorine free) process.

Transport Research Foundation Group of Companies Transport Research Foundation (a company limited by guarantee) trading as Transport Research Laboratory Registered in England, Number 3011746 TRL Limited. Registered in England, Number 3142272 Registered Offices Old Wokingham Road, Crowthorne, Berkshire, RG45 6AU.

ACKNOWLEDGEMENTS This Overseas Road Note provides detailed guidance on the design and operation of computer-based road management systems. It has been issued in parallel with a reference textbook published by Macmillan Press Ltd covering the wider area of road maintenance management (Road maintenance management: concepts and systems by Robinson, Danielson and Snaith). Mr C C Parkman (TRL) was project officer and editor. The original draft of the Note was prepared by Dr R Robinson in association with May Associates. Contributions were also made by Dr J Rolt, Mr T Toole, Mr P May and Mr R Abell. Mr H Lewis performed a final technical edit. The Note draws on material from a number of sources as well as experience gained by this team working with many institutions. Chapter 3 has been developed from guidelines produced by May Associates for the European Commission. The recommendations for information quality and data design are developed from work at the World Bank. Much of the document draws on the on-going research carried out by TRL in this subject. The helpful comments of Mr H S Thriscutt and Mr N Ings on earlier versions of the draft are gratefully acknowledged.

OVERSEAS ROAD NOTES Overseas Road Notes are prepared principally for road and transport authorities in countries receiving technical assistance from the British Government. A limited number of copies is available to other organisations and to individuals with an interest in roads overseas, and may be obtained from:

International Development Advisory and Information Unit Transport Research Laboratory Crowthorne, Berkshire, RG45 6AU United Kingdom

Limited extracts from the text may be reproduced provided the source is acknowledged. For more extensive reproduction, please write to the address given above.

CONTENTS
Page 1 Purpose and structure of the note Purpose Structure Part A: Principles 2 Road management Categories of work Management functions The management cycle Introducing procedures Computer-based systems 3 Approach External factors Institutional factors Stages in the design process Multiple system implementation Use of technical assistance and the private sector Part B: System design 4 System requirements Identifying objectives Cost-benefit analysis Priorities for system implementation 5 System specification Identifying the system users Confirming system outputs Identifying data and models Information quality levels Types of model Criteria for model selection Model calibration 6 7 Specification of network information systems Specification of planning systems 1 1 1 3 5 5 5 5 7 7 11 11 12 12 12 12 15 17 17 17 19 23 23 23 24 24 26 27 27 29 33

iii

Page 8 9 10 11 Specification of programming systems Specification of preparation systems Specification of operations systems Computer requirements Sources of procurement Customisation Modular approach to software Programs and databases Hardware Part C: System operation 12 Training Training needs analysis Training levels Training topics Monitoring training 13 Systems management Defining responsibilities Control of systems and data User access Data updating Data control Monitoring and feed-back Institutional issues Technical issues 14 References 39 43 49 53 53 53 53 53 54 55 57 57 57 57 60 61 61 61 61 61 62 62 62 62 65 67 73

Appendix A: Glossary of terms Appendix B: Institutional appraisal check list Appendix C: Example applications of the information quality level concept

78

iv

Purpose and structure of the note

Purpose 1.1 This Note is intended for engineers and managers in road administrations who are responsible for the specification, procurement, implementation and operation of computer-based road management systems. It offers guidance to help them reach informed decisions about the type of road management system which will best match the needs of their administration and the most effective methods to be used for operating the system. 1.2 As well as explaining the benefits which computers can bring to the task of data management, the Note points to the problems that can occur if computer-based systems are designed inappropriately or operated without a full awareness of their strengths and limitations. The aim is to alert managers to these risks so that, in planning and implementing their systems, they can take steps to prevent problems from developing. Engineers and managers do not need to be computer specialists to gain advantage from the Note; but they should be familiar with computer applications such as spread sheets and word processing programs, in addition to having a sound basis of professional experience in road management. 1.3 The Note does not discuss management methods which do not rely on computers, though manualprocesses are likely to be involved to some extentin any system. Manual management methods are described in Overseas Road Note 1 (TRRL Overseas Unit 1987) for road maintenance systems, and Overseas Road Note 7 (TRRL Overseas Unit 1988b)for bridge inspection and data systems. 1.4 While the Note is not addressed to organisations engaged in developing or marketing road management systems, or to institutions funding projects for system implementation, they too may find it a useful source of reference. Structure 1.5 The Note has three main parts, each focusing on the priorities of a different level of responsibility in the road administration. Part A is meant for senior policy and decision-makers. It outlines the principles of best practice in road management and the role of computer-based systems in supporting management procedures (Chapter 2), and defines the philosophy underlying the Note (Chapter 3) - in particular, the need to ensure, before a computer-based system is introduced, that the road administration has the institutional capability and the commitment to implement

1.6 Part B is intended for use by professional staff who have the task of recommending the type of system design to be adopted. It addresses the processes involved in system design, starting by identifying the objectives of the system and the components it needs to include (Chapter 4). It then identifies a generic approach to system specification, based on the outputs that might be delivered by the system and the data and computer models required to produce those outputs (Chapter 5). 1.7 Chapters 6 to 10 discuss in turn the development of specifications for management systemsconcerned with road network information, strategic planning, work programming, work preparation and operations, while Chapter 11 reviews the procurement of computer software and hardware. 1.8 The operation of road management systems is the subject of Part C, which is intended for staff involved in system implementation. Chapter 12 deals with the training and competence-building activities needed within an administration to ensure the successful introduction of a computer-based system and its continuing operation. Issues related to the day-to-day management of the system are covered in Chapter 13. 1.9 The Note concludes with a series of Appendices which provide reference material and background information on specific issues raised in themain body of the text.

it effectively and to sustain its use over the long term.

Part A: Principles

Road management

2.1 Road management starts from the premise that the road network is an asset which needs to be maintained and improved so as to secure the best performance and value-for-money and the maximum service life. The aims of road management are to enable the network to withstand the damage caused by wear and tear, to prevent sub-standard conditions from developing, and to ensure that traffic can continue to travel, in a manner which is safe, efficient, reliable and which causes the least damage to the environment. These aims are achieved through a series of works and activities which depend for their effective management on the maintenance of up todate information about the features and condition of the road network. Categories of work 2.2 The works and activities undertaken as part of road management are generally categorised by their frequency and the budget from which they are funded (Table 2.1). The terms used in the table are explained in the glossary which forms Appendix A. Management functions 2.3 The management of these works and activities is best viewed in terms of four main functions: Planning. Programming. Preparation. Operations.

The focus of attention is transferred from the network as a whole to the specific locations where works are being undertaken. The time horizon narrows from a span of several years to the individual budget year and then down to the current week or day. The level of management responsibility decreases. The information required for each function changes in scope from summary or sampled data about the entire network to detailed and precise data about specific road sections. Where computer systems are used to support management activities, automated processes which produce standard reports on a pre-defined basis are progressively replaced by processes in which managers work interactively with the computer. There is a transition from tasks which are conventionally viewed as client function to tasks which are increasingly amenable to being contracted out.

The management cycle 2.5 In performing each of die four road management functions set out in Table 2.1, managers need to follow a logical and clearly defined sequence of steps. This approach, termed the `management cycle', is shown diagramatically in Figure 2.1. Each successive step in the cycle requires an accurate and up-to-date supply of information if the correct decisions are to be made. For this reason, the maintenance of a database of management information is at the heart of the cycle. 2.6 Road management can be viewed as a process which integrates the cycles of activity involved in each of the management functions of planning, programming, preparation and operations. While these functions have different objectives, they draw on a shared fund of information about the road network: in other words, there needs to be a continuous flow and

2.4 Table 2.2 indicates the application of these management functions within a road administration. The progression from planning through to operations is accompanied by several changes in emphasis:

Table 2.1 Categories and examples of road management works

Table 2.2 Road management functions

Figure 2.1 The management cycle

feedback of data both within each management cycle and between successive cycles. Table 2.3, which spans the whole process of road management, shows examples of management cycles and their component tasks. Introducing procedures 2.7 To ensure consistency in its approach to each management function, a road administration should have a clearly defined policy framework. In order to adhere to the standards as set out in this framework, it is important, amongst other things, to introduce and document procedures for each activity in the road management process. A procedure will normally specify: the purpose and objective of the activity; the units within the administration to which it applies; the meanings of any terms requiring definition used in the procedure; the component tasks of the activity, shown as a logic network, flow chart or work plan; the responsibilities for fulfilling the procedure which attach to particular posts within the administration, including requirements for liaison and consultation; any special considerations of health and safety and environmental protection which may apply to the tasks covered by the procedure.

management through the use of manual techniques: operating check-lists, equipment use and maintenance cards, diagrams and wall charts have a part to play even in the most advanced management system. But effective road management requires continuous access to information about every aspect of the road network and the activities undertaken to keep it in good condition. With their power and relatively low cost, modern computer systems are ideally suited to assist in this task, particularly where large amounts of data have to be managed. Even so, it needs to be remembered that the sole purpose of computer-based systems is to support the human resources engaged in the management process, and not the other way round. 2.10 Each of the management functions inherent in the care of the road network can benefit from the power of computers, notably through the creation and maintenance of a network-wide database. It is essential to use a computer system as a means of reinforcing the effectiveness of agreed procedures, rather than allowing it to dictate the way road management is to be performed. In the past, computer systems have sometimes been brought into operation without being matched correctly to the priorities and procedures of the administration. As a result, they rapidly lose credibility. The situation becomes even worse when the operational procedures of the road administration are expected to change to reflect the particular requirements of a proprietary computer system. 2.11 Two types of computer-based systems are used in road management: network information systems, which correspond to the core of the management cycle (Figure 2.1) and are used to assemble, organise and store data about road network decision-support systems, which are used in each stage of road management to assist in the tasks that form the perimeter of the management cycle planning, programming, preparation and operations; they process network data as a basis for decisions about road management activities and almost always require a fully functioning network information system.

2.8 If a procedure is to work successfully, the staff who will be involved in managing and implementing it need to understand what it is intended to achieve, and it has to be accepted as a logical and practical course of action. For this reason, it is important to consult them when the procedure is being drafted and to give them an opportunity of contributing to its detailed formulation. The aim is to arrive at a procedure which staff regard as helping rather than hindering their work, and to which they can feel a sense of professional commitment. The quality of road management practice is largely dependent on the degree to which appropriate and clearly documented procedures are applied within a road administration. So far as developing a computer-based road management system is concerned, it will be difficult to achieve worthwhile results until sound and effective management procedures are in place. Computer-based systems 2.9 Once an administration is equipped with these procedures, it makes sense to consider introducing a computer-based system to assist in the process of road management. There is, of course, much that can be achieved in road

2.12 Terms such as `maintenance management system' and `pavement management system' can easily cause confusion, since systems described as such and produced by different vendors often possess quite different characteristics. Table 2.4 gives examples of the way systems offered in support of road management functions are commonly described.

Table 2.3 Examples of management cycles for road management functions

Table 2.4 Examples of different system descriptions

Related management function Planning

System description Strategic analysts system Network planning system Pavement management system Programme analysts system Pavement management system Budgeting system Project analysis system Pavement management system Bridge management system Pavement/overlay design system Contract procurement system Project management system Maintenance management system Equipment management system Financial management/accounting system

Programming

Preparation

Operations

Approach

3.1 The guidance set out in this Note reflects a structured approach to the design and operation of computer-based systems. Its key principle is the need to ensure that a road administration has the institutional capability to specify a computer-based system which will match its requirements and priorities correctly, and that it has the competence to use the system efficiently before investing resources in the procurement of hardware and software. In the past, there has been a tendency for road administrations to review the existing range of commercial products and then simply choose the one which appeared to have the most functionality within the available procurement budget. This course of action, which omits a rigorous and comprehensive assessment of a road administration's abilities and requirements, has usually proved unsuccessful, with the result that resources have been wasted and administrations left with systems which created more problems than they solved. 3.2 A computer-based system is a technical response to the challenge of managing and applying road management data. For this response to deliver its expected benefits, two prior conditions have to be fulfilled: first, the social, economic and regulatory context in which the road administration functions the set of external factors which govern its activities- needs to be conducive to effective performance; and second, the road administration itself has to develop the institutional competence to support the introduction of a computer-based system. The relationship between technical, institutional and external factors is shown in Figure 3.1. Simply implanting a computer system in an organisation that is ill-equipped to use it will be a waste of resources; without the underlying base of institutional capability and external support, the pyramid will collapse. 3.3 This principle holds good whether the organisation responsible for managing the road system is in the public or the private sector.

Organisations will have differing strategic objectives for example, a public sector administration may be concerned to maximise overall social benefits from the road network, whereas for a private toll road operator maximising revenue may be a corporate priority - but in both cases the feasibility of a computer-based system depends on the capability to sustain it. Experience from road administrations in industrialised economies shows that the implementation of a new management system, with all its institutional changes, takes a period of between five and ten years and requires a considerable investment of financial and human resources. The process should not be expected to require less time or investment in developing and emerging economies. 3.4 The external and institutional factors that determine the feasibility of a computer-based system need to be addressed before any start is made on designing the system. External factors 3.5 By definition, external circumstances such as the state of the economy, the legal and regulatory framework and, to an extent, the budgetary limits within which the road administration is required to operate are outside its own control; but it is essential for any administration setting out on the path of computerisation to try to secure the most favourable climate for change. Measures which need to be explored include: obtaining from government a commitment to view road management as an economic priority and to provide the necessary :Funding to launch management systems successfully and sustain their operation over the long term; gaining government commitment to the establishment of effective legal, regulatory and administrative mechanisms to reinforce the responsibilities and powers of the road administration;

Figure 3.1 Hierarchy of factors affecting road management (the `Brooks pyramid') 10

examining the potential for innovative methods of financing and possible partnerships with the private sector, to see whether these may offer a more dependable continuity of funding; promoting the concept of efficient road management both in government circles and in the wider public forum (for example, by publicising the economic and political consequences of not maintaining the road network to the required standards); communicating effectively the message that neglected roads mean higher operating costs for all road users.

3.10 The physical design and implementation stage focuses on the technical aspects of the system. It covers the procurement and commissioning of the appropriate hardware and software and the operational, training and management measures that bring the system into implementation. 3.11 Figure 3.2 shows the overall sequence of stages and activities forming this design process. The sequence is fully consistent with the structured approach to systems design applied in the information technology sector. 3.12 The approach may appear unduly complicated for a small system, but its applicability needs to be viewed against the potential costs of implementing and operating the system. These costs are normally dominated by the expense of data collection. In the absence of a rigorous approach to system design, there is a considerable risk that data requirements will be specified inappropriately, with the result that the costs of acquiring and maintaining data far outweigh the practical benefits of the system. Multiple system implementation 3.13 An institutional appraisal may sometimes indicate a requirement for several different types of computerbased system. In such cases the process of implementation has to be undertaken carefully and needs to be phased in gradually as part of along-term development plan. The temptation to try to do everything in a single bout of procurement should be avoided; experience suggests that it is much better to move forward step by step. In this way each step is fully assimilated within the administration before moving on to the next. This type of approach should enable the administration's longer-term objectives to retain flexibility for change and adaptation, while allowing each successive stage of implementation to benefit from the build-up of experience and competence. Use of technical assistance and the private sector 3.14 Systems do not necessarily need to be developed in-house by the road administration. Box 3.1 indicates parts of the process where the use of consultants and other sources of technical assistance from the private sector may offer advantages.

Institutional factors 3.6 As a further preliminary step, the road administration needs to undergo an institutional appraisal that will assess its readiness and suitability for the introduction of a computer- based system, identify management characteristics which may constrain the effective implementation of the system, and provide the basis of a programme of action to resolve constraints and improve and strengthen the technical capability of the administration. An example of the factors which have to be examined in this appraisal are listed in Appendix B. 3.7 One important point to confirm as part of this appraisal is the degree to which senior management staff of the administration understand the implications of computerisation and share an intention to meet the institutional and technical challenges which it presents. Stages in the design process 3.8 The steps which result from the institutional appraisal may be categorised as the logical and physical stages of the design process. The first step in the logical design phase, that of obtaining commitment to proceed with the development of any system, is of prime importance and its need cannot be overstated. Without a firm commitment from an organisation, both with regard to external factors (3.5) and also institutional factors (3.7), there will be limited potential for a system achieving its objectives. 3.9 The logical design stage starts from a commitment on the part of the administration to introduce the system and make it work. The administration has then to confirm the objectives of the system and decide its requirements in terms of the components to be included. This leads to the preparation of a detailed specification for the system, which identifies users and their required outputs and hence defines the required data and analytical processes.

11

Figure 3.2 Overall approach to the design and operation of systems

12

Box 3.1 Use of technical assistance from the private sector Requirements During the requirement phase, some knowledge of the market may be helpful to confirm what is technically feasible. It can be useful to seek an objective outside assessment. Those involved within the roads administration may lack up-to-date knowledge of the capabilities of modern systems, or may find it difficult to identify the real business benefits of systems being considered. Similarly, few consultants have in-depth knowledge of developing and implementing a wide range of systems. Those that are experienced often have proprietary systems of their own. Such systems may be very competent and robust, but may not meet all of the requirements of the road administration. Clearly, the choice of objective outside assistance, and the terms of reference for its employment, are important. Specification During the specification phase, the investment should be justified in terms of costs and benefits. A specification can then be written that focuses on the minimum facilities that are needed to deliver those benefits. At this stage, it is vital to avoid expansion of the scope of the system (as this can affect the cost significantly) without being sure that the incremental benefits outweigh the additional costs. There have been many examples where the perception of the success, or otherwise, of a new system has been affected because of unrealistically high initial expectations. When drafting specifications, it is important that engineering requirements are translated into a suitable logical design that can be understood by information technology (IT) specialists. Procurement Available products should be reviewed, either as off-the-shelf solutions, or as a basis for local customisation. The cost of this approach can then be compared with the cost of bespoke development. The sustainability of any off-theshelf systems should always be reviewed in terms of the required on-going support that might be necessary after initial implementation. Each case will be slightly different, but it should be noted that purpose-written systems are normally considerably more expensive than customised products, irrespective of whether the work is done in-house or by outside firms. Engineers are advised to seek the advice of IT experts to write operational requirements (i.e., the functional user requirement of the system), detailed specification and contract documents, as appropriate. The procurement should be handled by someone who is familiar with computer system projects, but always with reference back to the road administration in the event of technical queries. Operation The greatest disappointments often occur during system implementation. The setting of unrealistic delivery schedules and underestimation by the road administration of the time required to collect, load and validate data, can all contribute to delays. Firm project management is required and outside assistance may be appropriate. The private sector can also be used to assist with on-going operation of the system. Tasks such as network referencing, inventory collection, traffic counts, axle load surveys, and condition surveys (carried out by manual methods and by machine) can all be undertaken in this way, often more cost-effectively than by the road administration using its own resources. These works are relatively easy to define and, in several countries, specialist contractors have evolved for these activities. Similarly, for bridge management, consultants and contractors can often bring specialist skills and equipment to the tasks of bridge inspection and assessment. More radically, consultants can be used for all aspects of system operation, including the provision and operation of all hardware and software. Use of the private sector also gives access to state-of-the-art technology without the need for investing directly in development costs. Such an approach reduces the need for institutional development within the public sector.
Source: Button, C, 1994. Computerised road rnanagement: the appliance of computer science. Highways, 62 (No 4) London: Thomas Telford, pp. 12-13.

13

14

Part B: System design

15

16

System requirements

4.1 Defining the requirements of the systems needed within the administration is a matter of identifying the objectives which each system is intended to fulfil, and determining the components needed to meet these objectives. Identifying objectives 4.2 Two approaches can be used to help identify system objectives. The first is related to the framework of strategic policies adopted by the road administration; the second uses a method known as `problem tree analysis'. 4.3 The policy framework approach derives its logic from the principle that the introduction of new technology should assist the administration in achieving its strategic objectives. It therefore makes sense for the objectives of the computer-based system to reflect the broader policy framework adopted by the administration, in terms of its overall aims and the means by which achievement of these aims is measured. Box 4.1 offers examples of the use of this approach. 4.4 The method termed problem tree analysis examines the effects of problems identified within an organisation to determine their basic causes. The results of this analysis are then used to identify the means of achieving the desired solution or objective. The method is described in more detail in a document on project cycle

management produced by the Commission of the European Communities (1993). 4.5 Problem tree analysis establishes cause and effect relationships between the negative aspects of the existing situation, and identifies `bottle- necks' that should be treated as matters of priority. The results of the analysis can be recorded in a tree diagram showing the effects of a problem and its causes. The `negative situations' of the problem tree diagram are then converted into `positive achievements' in the diagram by replacing the causes of the problem with the means of achieving the required end. The effect of the problem will generally have several impacts and causes, and several means and results will be needed to achieve the objectives by addressing all aspects of the problem. 4.6 Box 4.2 sets out an example of problem tree analysis in the field of road management. This demonstrates the conclusion that the implementation of a road management system is likely to be only one component of an overall solution. A range of measures will normally be required to help improve road management capabilities within an administration. Cost-benefit analysis 4.7 The decision to introduce a computer-based road management system has to be seen as a business decision, not as simply a technical option. This decision needs to be reached on the basis of a

Box 4.1 Use of a policy framework approach in deciding system objectives System objectives can be defined by reference to the objectives of the road administration. A typical objective in the area of road user costs might be: Arterial roads will be maintained, so far as budgets will allow, to minimise the sum of road user and road maintenance costs in the longer term. This objective would need to be met by a system designed to minimise the longer-term road user and road maintenance costs in a constrained budget situation. An appropriate planning system would be required to help achieve this objective. Decisions would be needed about which road user costs (vehicle operation, travel time, accidents, and so forth) should be included, since the phrasing of the objective does not specify this point. In addition, turning longer-term plans into actual programmes of work, with agreed budgets, and arranging for the budgeted work to be performed would require die implementation of programming and preparation systems. Similarly, a typical road administration cost objective might be: The road administration will endeavour to provide value-for-money by meeting the above objectives at minimise cost, subject to the available budget. An operations system would be needed to support this objective. The scope of the system could include the facility to develop effective costing procedures for works, and the generation of work records that could be used to monitor and audit performance against targets.

17

Box 4.2 Use of problem tree analysis A road administration may be concerned that its road maintenance activities are incurring costs which perhaps seem high by international standards, or when its overall costs per km are compared with those of neighbouring administrations. Analysis of this problem might identify several possible causes of high cost levels. These could be shown in the form of a problem tree, such as that below, which only isolates one of a number of possible causes.

The tree diagram can help identify possible solutions to the problem. The effect of the problem `high road maintenance costs' can be converted into the objective of achieving `lower road maintenance costs'. The impact of 'inefficient use of resources' can be converted into the desired result of `effective use of resources', and the cause of `lack of resource management' can be converted into the means of `implementing an operations management system', as illustrated below.

18

comparative assessment of costs and benefits. Though it is not always easy to quantify the potential benefits of the system, experience suggests there is a real risk of costs escalating out of control unless they are analysed carefully in advance. 4.8 The costs of a computer-based system will typically be incurred by: data collection and updating - the principal cost item; hardware and software; staff training and retraining.

administration, but planning systems are likely to be easier to implement, even though the immediate benefits may be less. This is because planning systems may require fewer data to be collected on a relatively infrequent basis, and will be operated by staff in the organisation who are more likely to have computer skills. Box 4.5 provides additional examples of system implementation priorities. 4.15 Funding agencies such as the World Bank, EBRD or the European Community may well attach the highest degree of priority to a planning system, since this will provide the macro-level data which funding agencies find useful when developing the sector policy for a country.

4.9 The benefits of the system are likely to result from: improved asset management; improved control of contracts and costs; better management information.

4.10 Cost-benefit analysis should follow normal procedures for an economic appraisal in the road subsector, as set out in Overseas Road Note S (TRRL Overseas Unit 1988a). 4.11 Boxes 4.3 and 4.4 offer examples respectively of costs and benefits typically associated with the introduction of a computer-based road management system. Priorities for system implementation 4.12 The order of priority for implementing systems of different types will be determined from the assessment of requirements, supported by the results of the costbenefit analysis. Each road administration will have specific needs, but in most instances the first priority will be to introduce a network information system, if one is not already in place. The reason is that a more comprehensive and reliable base of information about the road network can in itself yield significant benefits to the administration. In any event, all other types of system require a network information system as a database on which they can operate. 4.13 Issues determining priorities for implementation include the following: the level of institutional development within the road administration; the nature of the administration's existing procedures; the form of work procurement - i.e. in-house or contracted out; the competencies of staff at various levels in the organisation.

4.14 There are also other factors to be considered. Operations systems may hold out the possibility of greater benefits to the road 19

Box 4.3 Typical costs of implementing a road management (planning and programming) system Implementation costs Computer hardware Cost of new hardware or the upgrading of existing facilities to the standard of the new system, including computers, printers, storage devices, cabling, and so forth System System licence cost Other software licences Cost of licences for software purchased for, or for use with the system, including database management system, system interfaces, and so forth Data collection equipment Cost of equipment purchased for use with the system, including data capture devices (data loggers), rut depth measuring devices, distance measuring devices and safety equipment for condition surveys; this would include the cost of vehicles if purchased solely for use with the system Establishment of network All costs of setting up or integrating an existing referenced network for the system Entry of network to database Personnel and other costs of entering the network into the system database Collection of inventory Cost of item inventory collection or the upgrading of existing inventory Entry of inventory to database Personnel and other costs of entering inventory to the system database Initial staff training Cost of training staff to facilitate the introduction of the system, including training of inspectors, computing staff, engineering staff, and so forth Annual running costs Computer support staff Cost of all staff involved in providing support to computing operations Engineering staff Cost of all engineering staff including inspectors, surveyors, engineers, etc. Maintenance of hardware Cost of maintaining computers and ancillary equipment, including all necessary computer consumables Maintenance of software Cost of in-house maintenance of the system database system as well as any third party software maintenance and upgrade costs Maintenance of fretwork Cost of keeping the network referencing information up to date, including any surveying and data entry Maintenance of inventory Cost of keeping inventory data up to date, including inventory surveys and data entry Maintenance of equipment Cost of maintaining data collection equipment, including vehicles Record keeping and presentation Cost of collecting and presenting records Preparation of annual maintenance programme Personnel costs Production of scheme plaits Personnel costs

Source: UKPMS

Note: In some cases, these costs will treed to reflect the annualised cost of items not expended in every year

20

Box 4.4 Typical benefits of implementing a road management (planning and programming) system Better economic management: Priorities based on economic return; treatments selected on the basis of highest return on investment in different maintenance options within a constrained budget. Priorities based on projection of pavement condition; treatments selected for the time that they are applied, based on projection of pavement condition rather than existing condition.

Improved advice: Availability of information on historic and future trends to assist in establishing budget requirements and formulation of investment policies. Possibility of authority-wide assessment and management for different pavement types, footways, cycle tracks, and so forth.

Better design: Selection of treatment being based on the separation of defects into those relating surface, structure and edge (on asphalt pavements) or joints (on concrete pavements). Incorporation of historical data into the design process. Interfaces with other highway management systems in use. Ability to consider combinations of machine-derived and visual condition data.

Improved monitoring: Automatic recording of original defects, design decisions and information, and works records Improved general management: A more comprehensive and complete system than those commonly in use, applicable through all stages of the pavement management process from defect identification to the implementation of works, and suitable for use by all the different types of highway authority. Ability to define different inspection cycles to meet needs of different authorities; encouragement for the use of network and project level surveys which make better use of engineering resources. An assessment system which can operate at appropriate but different levels of detail on major and minor roads. Ability to collect data at different levels of detail depending on the importance of the highway feature being assessed. Output provided in a more readily understandable form, using text and graphics in conjunction with a flexible enquiry system.

Source: UKPMS

21

Box 4.5 Examples of system implementation priorities in different situations Well-established road administration An administration that is experienced and efficient may have operations and preparation systems already in place. In that event, the prime objective may be to improve the way budget needs are determined and to develop a strategic planning capability. The priorities would then be as follows: 1 2 Development of a programming system to improve the needs-based determination of budgets Implementation of a planning system to enable scenario analysis of strategic planning options

New or relatively inexperienced road administration In an administration where road management practice is not well advanced, and where the control of activities is relatively weak, the most important need may be to ensure that costs and work practices are managed more efficiently. In this situation the following order of system implementation may be more appropriate: 1 2 3 Ensuring adequate identification and control of costs through the use of an operations system Budgeting and setting priorities under budget constraint by the use of a programming system Introduction of a planning system to provide forecasting tools for assisting with policy formulation and identifying longer-term budget needs Implementation of a preparation system, which would receive the lowest priority since most administrations already deploy reasonably effective manual systems for works preparation.

Note that operations systems are appropriate mainly in administrations which undertake works in-house, though they can be used to assist with the letting and management of contracts.

22

System specification
Box 5.1 Typical users of road management systems

5.1 After deciding the types of system that the road administration will need, and defining their requirements in terms of objectives and priorities, the next stage in the design process involves the specification of systems - in other words, determining the outputs which each system will need to deliver and the categories of data and models that it will use to produce the intended outputs. 5.2 The recommended procedure for system specification is as follows: 1 2 3 Identify the prospective users of the system. Confirm the outputs that its users will require. Identify the categories of data and models needed to produce these outputs: this step will entail a process of continual iteration because data and model requirements are interdependent.

Network information system All staff in the road administration Planning system Top management of the road administration Planning or economics unit within the administration Funding agencies

Programming system Professional staff in the road management/maintenance department Budget holders

5.3 This chapter sets out practical guidance on each of these steps as they apply to computer-based road management systems in general. Detailed points of advice on the specification of systems for network information, planning, programming, preparation and operations respectively are offered in the five chapters that follow (Chapters 6-10). 5.4 The form of approach recommended here, in which outputs are defined before data and models, is the logical procedure for system specification. It is fundamentally at variance with the way in which in the past, where the emphasis has been on defining models and data requirements at the start.

Preparation system Road management/maintenance department Design department Contracts/procurement department

Operations system Works supervisors

Identifying the system users 5.5 This initial step draws on the findings of the institutional appraisal (3.6) to identify which staff in the administration will wish or need to use a computer-based system. Talking to staff, both in workgroups and individually, to determine their attitudes to computerisation and their degree of familiarity with computer processes, is an essential part of this task. External assistance - for example, from IT specialists also may be useful. Box 5.1 indicates typical categories of user for different road management systems. 5.6 Many users will regard the system as a collection of `black boxes', and they will have no wish to understand its detailed structure or contents, nor should they be required to do so. Their main concern will be to have available a system which produces results that are believable and in keeping with their engineering judgement. The design of the system needs to reflect this priority.

Confirming system outputs 5.7 Once the users of the system have been identified, they need to be consulted about the specification of its proposed scope and outputs. While consultation with users is the key to this task, it is important to maintain a practical and realistic view of the results that the system can be expected to achieve and to resist over-ambitious demands. Wherever possible, cost-benefit analysis should be used to assess the likely value of the proposed outputs (4.7-4.11). 5.8 The types of output required will depend on the policy of each administration, on its institutional arrangements, and on the detailed objectives of the proposed system. For this reason, it would be unrealistic to be prescriptive on this point. Chapters 6-10, which deal with the specification of particular categories of system, offer guidance on typical forms of output; but these are put forward only as examples which serve the purpose of demonstrating the types of issues that arise in designing outputs to meet user needs. The actual outputs obtained from particular systems will be determined by the objectives of the system and the specific requirements of its users, and may differ widely from those shown in the examples. 23

5.9 Outputs will often need to be provided at more than one level of detail: some managers will require only summary information, while other will also want to receive supporting data. Additional outputs should be available to show data inputs, often in the form of data sets with dates and times of processing. All outputs should be clearly and precisely referenced to avoid confusion. 5.10 Most outputs will be produced in a tabular format: all outputs should be available in this format, though some will also be amenable to graphic presentation. For example, the outputs from planning systems, which are intended for use by senior management, can often be communicated more effectively in the form of line graphs, histograms and pie charts. Similarly, many of the outputs from programming systems lend themselves to reproduction as strip diagrams, which provide schematic representations of a length of road. 5.11 Some systems produce outputs using map-based graphics depicting the whole or part of a network, accurate to scale in two dimensions. While these outputs may be visually attractive, the costs of collecting, maintaining and updating the spatial data needed to support them can be substantial. Before a system with this facility is procured, the additional benefits that are expected to be obtained from map-based graphical presentation need to be reviewed to make sure they will outweigh the recurring data costs. Identifying data and models 5.12 The outputs of a system are produced from a combination of data and models. Specifying these outputs will determine the data items to be collected and stored within the system and the types of model, in the form of algorithms and relationships, needed to process these data. Since the annual costs of data collection are typically five to ten times as high as the costs of purchasing the computer hardware for a system, accurate data design is essential if the system is to be costeffective. 5.13 Table 5.1 identifies information groups which might be used as a basis for classifying data items relevant to road management activities. 5.14 Box 5.2 comments on the key criteria to be used in selecting data items, which are: Relevance Appropriateness Reliability Affordability

Table 5.1 Information groups

Element Road inventory Aspects Network/location Geometry Furniture/appurtenances Environs Pavement structure Pavement condition Structures inventory Structure condition Volume Loadings Accidents Costs Budget Revenue Projects Interventions Commitments Personnel Materials Equipment

Pavement

Structures

Traffic

Finance

Activity

Resources

Source: Paterson arid Scullion 1990

need to be provided between old and new forms of data to preserve the value of historic information and allow it to be used in trend projection. 5.16 Systems have often fallen into disuse because of their onerous requirements for data collection and processing. Staff may collect huge quantities of data simply because the system has a vast potential for storage. The urge to store every piece of data must be balanced pragmatically against the costs involved in the process, the demands placed on the time of relevant staff, and the practicality of actually making use of the outputs generated from the data. Information quality levels 5.17 In 2.4 it was observed that as the road management process moves from planning, through programming and preparation to operations, the level of detail required in the system data increases (though the extent of the network to which the data refers decreases). Determining the appropriate level of detail therefore depends on the road management function for which the data will be used. 5.18 An example of the change in data requirements is that of cost estimation, which is a key task in each management cycle. Both the accuracy of cost information and the level of detail will need to become sharper as the process of road management moves through its successive stages. Overseas Road Note S (TRRL Overseas Unit 1988a) reviews different methods of cost estimation.

5.15 As road administrations develop and extend their resources, they will be in a position to adopt improved methods of data collection. Suitable links 24

Box 5.2 Criteria for selecting data Relevance Every data item collected and stored must have a direct influence on the output that is required from the system. Other data items which may be considered as desirable, interesting or possibly useful in the future, should be omitted in favour of those that are essential, relevant and of immediate use unless a very good cost-benefit case can be made for their collection. Appropriateness The volume of data and the frequency of updating them are major determinants of the cost of operating the management system. Some types of data are collected at different times in a staged process, and the intensity and detail of measurement may differ between these stages, usually adding progressively more detail to the basic information acquired originally. For example: for pavement structural assessment as part of a strategic planning process, data on road condition would need broad coverage across the network, but would have a low sampling rate; however, for engineering design of a project at the later preparation stage, intensive sampling over the limited extent of the project would be necessary to refine the design and contract quantities. The technology and resources involved in acquiring, processing and managing the data should be appropriate to the administration's capacity for maintaining the equipment, conducting the surveys, and sustaining the data processing. Reliability Data reliability is determined from the following: The accuracy of the data, defined by a combination of precision (the error associated with repeated measurements made at separate times or places, or by separate operators and/or instruments) and bias (the degree to which the mean measurement reflects the range and variability of all data points). Their spatial coverage; for network-level planning, low intensity sampling is adequate whereas, for engineering design of projects at the preparation stage, intensive sampling is needed with full coverage of the project area. Completeness of data is important because missing items degrade the reliability of the outcome. Currentness ensures that data which change rapidly from year-to-year, or which have a large impact on the ultimate decision, are kept up-to-date more than data which do not change so rapidly or are less sensitive.

A balance between the reliability of data and certainty of outcome should be sought. For example: high precision, intensive sampling of entire networks, such as can be obtained using mechanised methods, may represent overinvestment if the results are only to be used for broad planning. Affordability The volume and quality of the data items, and the associated data acquisition, must be affordable in terms of the financial and staff resources available to collect data and keep them current. The scope and quality of data are choices that must be weighed against the resources required to sustain them in the long-term, and against the value of the management decisions that rely upon them. Available resources and skills vary between road administrations, and may change over time. For small organisations, or where skills and resources are scarce in a larger organisation, simple and basic types of data, quality and collection methods must suffice. Where skills and resources are more abundant, a wider range of data, including the use of automatic collection methods, may be appropriate. Problems arise when administrations with very limited resources are responsible for managing large road networks. Source: Paterson and Scullion 1990

25

Table 5.2 relates the use of these methods to the component functions of road management. 5.19 The management function that the particular system is intended to support can be related to all the information groups shown in Table 5.1 to provide a rigorous basis for classifying data needs. This is achieved using the concept of information quality levels (IQLs), as shown in Table 5.3. Expressed in simple terms, the relationship between appropriate information quality levels and system types is shown in Table 5.4. Reference should be made to the World Bank document Draft guidelines on system design and data issues (Paterson and Scullion 1990) for detailed recommendations on the appropriate level of data detail to be used for each IQL for many of the information groups shown in Table 5.1. Appendix C provides summaries of data requirements at different IQLs for some information groups based on the Draft guidelines. 5.20 Most road administrations collect and store network inventory, traffic and finance data at one level only. The data may then be summarised for use in the different applications that compose the system, or selected values only may be used. The exact level of detail of the data collected for road inventory and traffic will be governed by the nature of the road hierarchy. It makes sense for data values to be recorded in only one place in the system since this

will help maintain data integrity when values are updated. The network information system should have the ability to reflect these strategies, and to store and report on data at different levels of detail. Box 5.3 outlines typical strategies for data collection. Types of model 5.21 Models, which contain relationships and algorithms, are often embedded in the design of the system. In road management systems, models are typically used for the following purposes: representing physical sections of road; summarising pavement condition as an index based on measured defects; projecting road conditions over time; selecting appropriate maintenance treatments based on condition; estimating and assigning costs to activities; predicting traffic, in terns of both flows and damaging effects; congestion analysis; analysing road user costs; deriving works priorities under budget constraints; allocating funding to geographic areas, budget heads, projects, and so forth.

Table 5.2 Cost estimation methods

Table 5.3 Information quality levels

26

Table 5.4 Application of information quality levels

System type Planning Programming Preparation Operations Information quality level IQL-IV IQL-III/IV IQL-II/III IQL-I/II

planning and programming. These models are comprehensive and many of them represent the current state of knowledge in their specific areas. But they are not universally applicable, and care should be taken to ensure that they are used only in appropriate situations, and with local calibration, which can be timeconsuming. One disadvantage of these models is that they impose substantial data requirements, typically at IQL-I and IQL-II levels. Before incorporating these models in a system specification, road administrations need to give careful thought to their ability to meet the data requirements on a continuing basis.

Criteria for model selection 5.22 Since each model has its own data requirements, models should be selected on the basis of the same criteria used to select data -namely, relevance, appropriateness, reliability and affordability (Box 5.2). An extensive range of models is available, which it would not be practical to attempt to review in this Note: where an administration does not possess inhouse the expertise to evaluate different models, external consultancy assistance may be helpful. 5.23 Models from the World Bank's HDM system form part of several road management systems for

Model calibration 5.24 Many models contain calibration factors which allow them to be adjusted to suit local conditions. If these calibration factors are not determined properly, the outputs from the model are likely to be inaccurate and inappropriate. For this reason the calibration of relationships within the model is an important step in the system design process. Box 5.4 outlines a recommended approach to calibration.

Box 5.3 Typical strategies for data collection Strategy A High level condition data (typically IQL-IV) are collected across the whole network each year. The data are used for planning and programming purposes. The programming exercise then collects more detailed data (typically IQL-III) on those sections where works are likely to be undertaken. Further detailed data (typically IQL-II) are then collected on some of the sections for which designs are produced, or for which works are undertaken. As more detailed data are collected on any section, they replace the data collected in the earlier phase, with the result that different sections in the database store data at different levels of detail. Strategy B Relatively detailed data (typically IQL-II/III) are collected across parts of the network on a rolling programme, perhaps with a cycle of three to five years. Each year, programming decisions are taken either by using current data for individual sections, if available, or by projecting condition data from previous years. All condition data tend to be stored at the same level of detail, though data collected as part of the works design or execution processes may also be stored. Other strategies Combinations of these strategies are also used, including the following examples: Annual data can be collected on the primary road network, whereas a cycle of data collection may be used on roads lower in the hierarchy. The cyclic approach can be used for the whole network, collecting data at low levels of detail (IQL-III/IV). Cyclic collection methods can be used without projection of condition. Detailed data can be collected annually across the whole network, though this approach is probably not cost-effective and is not recommended.

All these strategies have different implications for the detail of data stored in the system database.

27

Box 5.4 Understanding network behaviour Many models have been created to suit the conditions of a specific environment. Others have been developed and extended to cover a range of different environments and this is often achieved by including within the models calibration factors which permit their local adaptation. The important point is that the use of any model requires a correct understanding of the behaviour of the road network. This will ensure that money is not wasted on inappropriate maintenance and rehabilitation measures. To achieve a sound knowledge and understanding of network performance, an administration should consider monitoring a representative sample of the network on a long-terns basis. The monitoring will probably be carried out at the IQL-I level of detail. A representative sample should be selected to span the full range of traffic, materials, environments and construction methods as appropriate. This activity is often performed by a local research institution or similar organisation.

28

Specification of network information systems

Figure 6.1 The function of a network information system

6.1 A network information system forms a central and integral part of the decision-support systems used in road management, as shown in Figure 6.1, and can also be used as a system in its own right. Its purpose is to provide a single point of storage for data on every characteristic of the road network. By definition it is a database which can be interrogated to produce reports. The likely users of network information systems are noted in Box 5.1. 6.2 The standard outputs available from a network information system are likely to include: Gazetteer of road sections, in user-defined order, giving attributes of sections (such as label, description, and other data) Lists of sections based on user-defined selections of section attributes (for example, all sections in one geographic district, all gravel roads and their total length, all paved roads in a particular district carrying more than 1000 vehicles per day).

6.4 Boxes 6.1-6.3 illustrate three examples of outputs from network information systems. Box 6.1 shows a typical roads gazetteer for a district. This is a simple list of all sections in the district road network. Users should be able to select which types of sections in the network to include in the output, and to specify the details that are to be shown for each section. In this particular example the gazetteer includes the section label, section length, start and end node references, and a description of the section. The node references, given as six figure grid co-ordinates, enable the sections to be located spatially. 6.5 The information in the gazetteer is shown plotted to scale in Box 6.2. This graphical output indicates each road section with its label, the direction in which data are to be collected on the section, and the labelled nodes. For a plot like this to be produced, strings of grid co-ordinates need to be stored at regular intervals along each section. These are normally obtained by digitising data from maps. Data recording quality needs to be high if this type of graphical presentation is to be meaningful, since errors in the location of sections are obvious when plotted spatially. There are particular problems involved in digitising data from maps where sections extend over -matching is a necessary but tedious process. The plot also shows the position of certain footpaths. If road administrations are responsible for the maintenance of footways and footpaths, sections for these should be created in the same way as for carriageways.

6.3 It should be possible to report on most of the information stored in a database. User-friendly systems employ customised query languages which give users ready access to reports (sometimes known as ad hoc reports) on relevant data items. In other systems it will be necessary to use the query language of the database software itself. This latter approach may permit a large measure of flexibility in report production, but it is likely to be more difficult to apply than a system-specific report generator.

29

Box 6.1 Example of output of roads gazetteer

6.6 Box 6.3 shows details of a user-selected list of sections in the district. In this example, all details of all unclassified roads have been requested. Details of carriageway length and width, surface, road base and subbase materials are shown, as are shoulder width and construction type. Certain sections do not have shoulders or, if they do, the relevant information has not been recorded. Traffic has been grouped by ranges (hierarchy or class), indicated for each section on the output. The information system may also store data on other attributes of the section, for which reports would be available in other forms of output. 30

6.7 The categories of data stored in a network information system will be determined by the particular management strategy for data collection used by the road administration (5.12-5.19). The key data will belong to the inventory data group shown in Table 5.1. They will need to be stored at the most detailed IQL value required to support the applications included in the system. 6.8 For decision-support systems, all the data required by the system models can be stored with the inventory data. It should be possible to produce output reports on these together with the inventory data.

Box 6.2 Example of graphical output of roads gazetter

31

Box 6.3 Example of output showing section pavement and traffic details

6.9 Network information systems do not use models, although advanced systems may perform arithmetic calculations and store and apply this information to assist in decision making. For example, some systems can estimate cyclic maintenance requirements. These systems can accumulate the total quantities of certain entities or their attributes from the inventory data, and combine these with stored information about standards or intervention levels to determine maintenance quantities. This can be taken further by applying unit costs to the quantities to build up maintenance budgets. In this situation, the network information system is being used as a programming system.

32

Specification of planning systems

Figure 7.1 Example applications of planning systems

7.1 Planning systems assist in the strategic management of the road network. For example, a planning system might be used to help determine appropriate treatment standards for the various road hierarchies within the network, so as to minimise the life cycle costs of road construction and maintenance and reduce road user costs; or to help examine the likely effects of different budget levels on future road conditions. Figure 7.1 summarises the management cycle of decision-making for these typical applications. Planning systems are used for analysis of the entire road network, typically categorised by traffic level and road condition; individual sections are not identified. The likely users of planning systems and their ouputs are noted in Box 5.1. 7.2 The outputs of a planning system are likely to include: Projected annual capital and recurrent budget requirements to meet road administration standards over a user-defined period into the future Projected road conditions resulting from the application of pre-defined annual budgets for a user-defined period into the future Projected road administration costs and road user costs for pre-defined standards, or annual budgets, for a user-defined period Incremental net present value (NPV) of adopting one set of standards compared with another, or of adopting one particular stream of annual budgets compared with another.

7.3 Two types of forecast commonly produced by planning systems are those for road conditions and budgets. An example of a road condition forecast is shown as a tabular output in Box 7.1 and graphically in Box 7.2. The example indicates the effect over a ten-year period of a fixed budget level on the surface and structural condition indices of Class B roads and unclassified roads. The tabular output form is of more general use, enabling reports to be produced on indices of four types of road. Analysis can also demonstrate the effect of different budgets being specified for different years. With the particular maintenance and priority regimes adopted in this example, road conditions are predicted to decline over time. Structural conditions will deteriorate more rapidly than surface conditions, perhaps reflecting a maintenance policy which focuses on surface treatment rather than structural repair. Note also the predicted acceleration of structural deterioration in the second half of the projection period. 7.4 Boxes 7.3 and 7.4 show examples of typical outputs from an analysis which determined the annual budget necessary over a ten-year period to ensure that the surface and structural condition of the network remained unchanged. In this example, budget levels would need to rise steadily over the first part of the projection period, and slightly more rapidly over the second period. This reflects the deteriorating structural condition illustrated in Boxes 7.1 and 7.2. The budget figures normally used in such outputs are stated at constant prices, with no allowance for inflation.

33

Box 7.1 Example of output of projected road condition for given budget

7.5 Because a key purpose of planning systems is to illustrate trends, graphical outputs are particularly useful. The graphs in Boxes 7.2 and 7.4 could be shown as histograms. Other types of graphical outputs can be produced. For example, pie charts are a useful medium for communicating proportional data about the road network, perhaps indicating different types of construction, or sections in different states of repair. A good planning system should be sufficiently versatile to offer the user a range of possible graphical presentations. 7.6 The data used in planning systems normally cover the entire road network at a coarse level, probably IQL-IV (Table 5.3). Table 7.1 lists the information groups likely to provide data for a planning system. 7.7 The models required by a planning system will depend on the details of the applications involved. Those required by the two applications used as examples would probably include the following: Traffic growth: Models for traffic growth normally require data grouping by vehicle class. Ranges of existing traffic levels are required,

and future traffic is normally based on simple percentage growth rates for each class of vehicle. Loadings are modelled in terms of equivalent standard axles for the purposes of pavement design and strengthening, and in terms of mean gross vehicle mass for the purpose of bridge design. Road deterioration: Predictions of road deterioration are needed for the projection of conditions into the future, and for projecting historical condition data to present-day values when current information is not available. The various models are normally based on an analysis of past trends, using either deterministic or probabilistic methods. A range of mathematical formulations is available for this purpose. Treatment selection: Models for treatment selection are used to decide appropriate works to correct road defects, reflecting the judgement of a practising engineer. They are often termed `treatment selection rules'. The methods available are typically defined in terms of matrices of the various parameters that affect the treatment selected, or in the form of decision trees.

34

Box 7.2 Example of graphical output for projected road condition for given budget

Box 7.3 Example of output of budget required to maintain condition

35

Box 7.4 Example of graphical output of budget required to maintain condition

Table 7.1 Information groups likely to be used in planning systems

36

Prioritisation: A number of models are available for setting priorities when budgets are constrained. These models offer varying levels of sophistication, require different amounts of data and computation time, and produce recommendations that differ in terms of their impact on the longer-term condition and works requirements for the network. Methods which relate priorities principally to the severity of road condition are not recommended, since they tend to lead to conditions that worsen over time and to increasing works cost requirements. Other methods have the characteristics shown in Box 7.5. Where road user cost and pavement performance models are used, treatments can be prioritised only for the road pavement itself. Offroad defects cannot normally be prioritised with models involving concepts of treatment life. The choice of model will therefore depend on the feature being prioritised, as well as on the ability of the road administration to satisfy the data and computational requirements of the various methods.

7.8 Various planning models used in road management are described in Road maintenance management: concepts arid systems (Robinson and others 1998).

Box 7.5 Prioritisation methods

Methods based on treatment choice Priorities are related to treatment type and road hierarchy or importance. The methods recognize that some treatments are more cost-effective than others if applied at the right time. They have modest requirements in terms of data and computation time. Overseas Road Note I includes an example of this approach. Cost-effectiveness methods Priorities are related to the `cost-effectiveness' ratio of treatment life to treatment cost. These methods reflect the higher value of a treatment which lasts longer. They too can be relatively modest in terms of data and computational effort. Optimisation methods The optimum combination of a number of different works options is selected to achieve a given objective, which might be to minimise the life cycle costs on the network or to maximise the quality of road condition. Costs normally include both road administration and road user costs. These methods provide solutions based on a long-term view of the network, but are demanding in terms of data and computational effort. One example of this approach is the Expenditure Budgeting Model (EBM) used in conjunction with HDM-III. Reduced time period analysis methods These methods address the theoretical concern that life cycle optimisation methods base their decisions on long-term predictions which might turn out to be inaccurate. For example, budgets may change from those predicted, or traffic levels and road conditions may differ significantly from those projected. This type of method addresses these concerns by basing decisions on a limited time period of analysis. Their performance is similar to that of optimisation methods, but they are less demanding in terms of data and computational resources.

37

Specification of programming systems

Figure 8.1 Example application of a programming system

8.1 Computer-based programming systems are used in the short to medium-term planning of road management: they serve a tactical role as distinct from the strategic role played by planning systems. In their most common application, they help decide which road sections are likely to warrant treatment in the next budget period, and they assist in prioritising treatment by indicating the best combination of options that can be funded within budget constraints. Figure 8.1 shows the management cycle for this application. These systems are used also to produce rolling programmes of work, typically for three to fiveyear time horizons. Programming is performed on road sections across the entire network. The likely users of programming systems and their outputs are noted in Box 5.1. 8.2 The outputs of a programming system are likely to include: List of sections, in priority order and in section order, showing recommended treatments and costs that can be funded in the budget year under predefined capital and recurrent budget constraints. It should be possible for the user to work interactively with these lists so as to amend treatments, costs and the order of projects in the priority list. List of user-selected sections, in section order, showing conditions and recommended treatments. List of user-selected sections, in section order, showing traffic, axle loading and road user costs.

Projected rolling programme of work over a threeyear period, reflecting any user modifications to the list of sections to be treated.

8.3 Box 8.1 shows a specimen output for all sections in the network, summarising condition, and indicating recommended treatments and costs. Sections are listed in label order. Section lengths are also shown to help relate the scale of costs to the work to be done. Various methods of summarising condition can be used: in this particular example, condition indices are quoted for road surface, pavement structure, road edge and shoulders. Outputs from programming systems often show the priority index against each section where treatments have been recommended. 8.4 Treatments recommended by the system are generic. Since relatively coarse data are used for this network-level analysis, it is not normally possible to design detailed treatments; these are produced instead at the preparation stage. For example, the `Olay' treatment is a generic overlay, and no recommendation is made at this time about its thickness or composition. Similarly, the treatment costs too are generic. Because it is necessary to plan for uncertainty prior to detailed design, 20 per cent more treatments than can be funded are usually taken forward to the preparation stage. Treatments and costs are then refined, allowing the final programme to be confirmed.

38

Box 8.1 Specimen output for section treatments and costs

8.5 Box 8.2 lists sections where treatments have been identified, in priority order. In many instances the priority index (for instance, cost/benefit ratio) might also be shown. It is normal to indicate cumulative costs in this type of output so as to make it clear when the budget cut-off has been reached. In the example set out in Box 8.2, the road engineer has been working interactively with the system to select projects from those recommended for inclusion in the programme. The selected projects are shown in the final column. Since the available budget is no more than 20 000, it will not be possible to fund the

projects ranked second and third, and the engineer has selected the projects ranked in 4th and 5th place to complete the programme. 8.6 Data will typically be at level IQL-III/IV (Table 5. I ). The information groups from which data are likely to be needed are shown in Table 8.1. 8.7 The models required by programming systems are similar to those used in planning systems. The comments made in Chapter 7 are applicable also to programming systems. 39

Box 8.2 Specimen output for section priority list

Table 8.1 Information groups likely to be used in programming systems

40

Specification of preparation systems

Figure 9.1 Example applications of preparation systems

9.1 Preparation systems perform a variety of road management tasks at the stage when works are being packaged for implementation. Typical applications include the detailed design of works or treatments, and the letting of contracts or issue of work instructions. Figure 9.1 shows the management cycle for these applications. In this stage, detailed site investigations may be undertaken, detailed specifications, quantities and costings are likely to be determined, and any costbenefit analysis that formed part of the requirements phase (4.7-4.11) is reviewed to confirm the feasibility of the final works. In addition, preparation systems may be used in contract packaging to combine works from adjacent or near-by sections into projects of a size that is cost-effective for works execution. This category of system operates at the road section level. The likely users of preparation systems are noted in Box 5.1. 9.2 The outputs of a preparation system are likely to include: Detailed design information. Work packages. Works orders for projects, including bills of quantities.

a more detailed condition assessment of a project selected for potential funding. The output provides information on the locational, geometric and structural features of the section, and summarises at 100m intervals the condition measurements for a range of parameters. In this example, the areas of cracking and ravelling have been recorded, as have the number of pot-holes, the length of deteriorated edge on each side of the road, and the depth of ruts. Though the output includes a column for deflection measurements, none have been recorded on this section, perhaps because the parameter was considered inappropriate for a Class B road. The final column indicates roughness results. Data such as those recorded in the example provide the road engineer with valuable background information when designing or verifying specific treatments. 9.4 Box 9.2 shows an interactive treatment design application, related to surface dressing The output states that the design has been undertaken using the method described in TRL Overseas Road Notes 2 and 3 (TRL, 1982, 1985). The road length and width are obtained from the road inventory and are used to calculate the area to be treated. The engineer has worked interactively with the system and entered certain data: the system has then computed the design implications. The interactive process continues until the engineer is satisfied with all aspects of the design. During this process, the

9.3 Three examples illustrate these outputs. The first, shown in Box 9.1, includes the results of

41

Box 9.1 Example of output showing condition assessment

engineer would have access to condition data, as shown in Box 9.1, and to other information stored in the system - for example, on traffic volumes, accidents, or sources of materials. 9.5 The output shown in Box 9.2 has been obtained in the following way. Under the heading of `chippings', the engineer has entered the road surface type and the lane traffic category, and the system has determined that 10mm chippings are required with an application rate of 14kg/m2. Normally, for each input value, the engineer would be able to call up from the system the allowable range of possible inputs to assist with the selection. Under `bitumen', the engineer has entered the traffic constant, the existing surface, type of chippings, and the climate. The system has used a `look-up table' to determine the design constants relevant to each of these 42

characteristics; the engineer has then entered the road surface temperature and, using all of these values, the system has determined the grade and the application rate of the bitumen. The total quantities of both chippings and bitumen required for the section are then calculated by the system. 9.6 Box 9.3 offers a third example, which shows a bill of quantities for surface dressing produced by the system. The engineer has decided to combine the works on the three sections selected at programming stage into one contract package, which is reflected in the bill of quantities. The preparation system has to include a library of works items to enable these outputs to be produced. 9.7 The data required for preparation applications are likely to be quite detailed, but related only to a few individual sections. They will

Box 9.2 Example output for interactive treatment design

probably be at the IQL-11/111 level. The information groups from which data are likely to be needed are listed in Table 9.1. 9.8 Where preparation systems are used to assist with the preparation and letting of contracts, one item of data that may be needed is the type of contract to be applied. Standard forms of contract, such as those produced by FIDIC and the New Engineering Contract, are normally used for road works but local forms also may be in use. It is beyond the scope of this Note to enumerate the various types of local forms: in these instances, data may also include standard bill of quantity items required for different works, libraries of costs, suppliers of materials, and approved contractors.

9.9 Typical models required by preparation systems include: Traffic growth (as described in relation to planning systems). Road deterioration (as described in relation to planning systems). Works design: Models used for this purpose will be standard design methods for activities such as surface dressing, pavement and overlay design, and geometric design. Some of these models may be available as computerised systems, either integrated into the road management systems, or produced as stand-alone systems.

43

Box 9.3 Example of output from a bill of quantities

44

Table 9.1 Information groups likely to be used for preparation systems

45

10 Specification of operations systems

Figure 10.1 Example application of operations system

10.1 Operations systems assist with the management of on-going activities, supporting decisions that are typically made on a daily or weekly basis. Operations are focused on individual sections or sub-sections of road: typical tasks include work scheduling; monitoring the use of labour, equipment and materials; and recording completed work. Figure 10.1 shows the management cycle for a typical application. In addition, operations systems may contribute to cost accounting and financial management, equipment management and facilities management. 10.2 A road administration will need an operations system only if it has in-house works units. Where work is contracted out, it is the contractor who is likely to need to use operations systems. In this case, operations management by the road administration becomes a matter of project management: since computer-based systems for project management are widely available and their use is already well documented, they are not included in the scope of this Note. Typical users of operations systems are noted in Box 5.1. 10.3 Operations systems are likely to have the following outputs: Performance standards for works, defining the minimum requirements and specification for activities, and including schedules and costs for equipment, materials, and labour. Work instruction/accomplishment. Weekly labour time summary by person and budget head. Weekly cost summary by activity and budget head, with totals.

Annual cost summary by road section, activity and budget head, with totals.

10.4 Performance standards are used to ensure quality and consistency when managing operations. The example in Box 10.1 sets down the basic procedures for an individual activity, in this instance surface dressing. It includes a method statement, and defines resource requirements, costs and the expected productivity. Performance standards of this type should form part of the road administration's quality management system. They will need review and updating to reflect changes in costs or the introduction of modified work practices. 10.5 Performance standards can be used as the basis for issuing work instructions. Some systems combine the two documents by showing target output, and providing a box where the work achieved can be entered by the supervisor or foreman. This information is needed by the system to enable performance to be monitored over time. Inclusion of both pieces of information on the same form simplifies data entry into the system, and ensures that those involved are aware of any shortfall in performance. Recording information in this way allows weekly, monthly and annual summaries to be produced for monitoring purposes. 10.6 Box 10.2 shows a weekly staff time summary. The information is produced directly from time sheets which need to be completed daily. These can either be paperbased, or maintained in electronic form, with staff entering details directly into the system. The weekly staff time summary can feed into the staff payment system and into progress and performance monitoring systems.

46

Box 10.1 Sample output of a performance standard

47

Box 10.2 Sample output for weekly staff time summary

10.7 A key aspect of operations systems is that they allow monitoring of both achieved output and cost. Weekly monitoring (sometimes extended over fortnightly intervals) is necessary in most situations to avoid under-achievement and cost over-runs escalating out of control. Weekly reports can be produced for activities under all budget heads showing target and actual output and expenditure, under the headings of labour, equipment and materials. Spend-to-date and remaining budget should also be shown. Similar reports are normally produced on an annual basis, typically for costs by activity and section. All of these reports can be produced automatically from weekly work achievement returns. 10.8 Box 10.3 provides an example of a report for an annual cost summary. It relates to the capital budget allocation for a district, and is disaggregated by activity codes. This example shows annual targets and actual outputs, with the percentage achievement for each activity. Costs are disaggregated by labour, equipment and materials. For each activity code, the budget and actual expenditure are shown, with the percentage spend against budget.

10.9 - This particular example indicates that there was under-performance for Activities 04021 and 04024, with low productivity and an over-spend of budget. The reasons for this would need to be investigated by the engineer, though the problem would have been apparent from the monitoring reports produced through the year which give an opportunity for more speedy remedial action. For Activity 04025, the budget figure was met with a 32 per cent increase in productivity. For Activity 04029, there was 24 per cent over-production at a cost of only 43 per cent of the budget, suggesting that the targets set for the activity were low. Information fed back from outputs such as this provide a sound basis for investigating specific problems affecting individual activities or sections, and defining more realistic targets for the following year. 10.10 The data required for operations management are likely to be highly detailed -probably IQL-1/11 but they will apply only to a short length or sub-section of road. Table 10.1 lists the information groups from which data are likely to be needed for operations systems.

48

Box 10.3 Sample output for annual cost summary

Table 10.1 Information groups likely to be used for operations systems

10.11 While the recommendations derived from operations systems tend not to be based on the use of models, and it is unusual for these systems to include models, their outputs are used to support

decisions made by users. For this reason, operations systems can properly be regarded as decision-support systems.

49

11 Computer requirements
Sources of procurement 11.1 Earlier comments on the use of manual techniques are reiterated here (see 2.9). However, once the decision has been taken to adopt a computerised approach, hardware and software requirements need to be addressed. Road administrations are often keen to develop their own in-house software for proposed road management systems. Sometimes a `not-invented-here' attitude leads to a reluctance to use proprietary products, or to adopt systems that are widely used elsewhere. But the systems discussed in this Note may have to manage substantial volumes of data: unless highly efficient software is applied to this task, their running costs can be extremely high. Few road administrations have the competence in software development to produce for themselves systems that will be powerful and efficient enough to perform successfully. In most cases, procurement and customisation will offer the most economical solutions. 11.2 Any software product for use in road management has to be robust and dependable, particularly where large databases are involved. Some of the computer applications that are commercially available do not meet these criteria. For this reason, administrations need to assess rigorously the viability of the software and the competence of its supplier before committing themselves to its procurement.

integration through a common data bank. This modular structure has to reflect the administration's road management procedures as grouped into functions and tasks. Some proprietary systems fail to match adequately the needs of an administration because they lack the potential for modularity. 11.5 The road information system or data bank, which provides the core of the road management system, requires a network referencing system around which is built an inventory of the road network to which all other information can be related. Figure 11.1 depicts this modular framework, which includes data items from the information groups listed in Table 5.1 and the five types of information and decision-support systems identified in this Note (2.11). An integrated approach of this kind has to be a long-term aim. In the short to medium-term, most road management systems may contain only part of the framework, so the software that is procured needs to be flexible enough to accommodate future change and growth. Programs and databases 11.6 The modular framework shown in Figure 11.1 is implemented through a series of application programs operating in conjunction with a database. Application programs are needed for input, output and processing (models). All management information required for decision-making is held centrally, while data and technical processing may be decentralised. The structure allows the flow of information between modules to be controlled, so that data are checked to ensure quality and consistency before being used by other modules. 11.7 Ideally, the functions of the core database should be built around a relational database management system (RDBMS), preferably using fourth generation (4GL) programming languages. Applications programs can be written in the same 4GL as the database, or they can interface to the database through what is known as `middle-ware', which is purpose-written software for linking the applications programs to the database. This approach offers two key advantages: first, it allows modules to be procured from a number of different sources at different times; second, it avoids having the road management system tied down to one software vendor through its proprietary RDBMS. For example, a proprietary project management package could be used for work scheduling, with a standard database providing the data storage. The choice of RDBMS will depend on the availability of specialist expertise and the degree of sophistication needed to support the proposed applications. 11.8 The disadvantages of this approach are that considerations of the long term will be dictating

Customisation 11.3 Any proprietary product that is worth buying has to offer, above all else, the potential for customisation to meet the precise needs of the purchaser. This factor should in large measure determine the choice of software to be adopted. Some software products are designed in a flexible way and need to be customised before use by setting parameter values known as 'meta data'. Though the process of customisation can be undertaken by the road administration itself, it will normally be more economical to have the system supplier do this. Less flexible software can be customised only by the supplier changing the software code, which is likely to prove more expensive than customising parameter-based software. Some software is designed in such a way that anything other than superficial customisation is impossible.

Modular approach to software 11.4 The most efficient and flexible solution for software procurement is normally the adoption of a modular software structure, which achieves

50

Figure 11.1 Modular system framework

short-term actions, so that the initial solution may be more expensive and complicated than a dedicated application, while the development of middle-ware demands sophisticated programming resources which are expensive and may not be available. 11.9 Since the most costly part of any road management system is the data, as noted earlier, it is essential to make sure that any future upgrading of the system is able to use the existing network information. An administration introducing computer-based systems for first time will have most to gain by adopting a simple approach that is in scale with its institutional capability. Once the simple system becomes institutionalised, and as technology advances, the system can be replaced. Providing the system uses an RDBMS or a spreadsheet to store data, it is a relatively straightforward exercise for a computer specialist to transfer these data (possibly after transforming them) to a new, upgraded system. Hardware 11.10 The final decision to be made when planning the implementation of a road management system involves the choice of hardware. This approach - leaving hardware to the last - is in marked contrast to the approach taken in the past by almost all projects to develop and implement computer-based road management systems in non-industrialised countries. 11.11 The choice of hardware will be influenced by the operating system selected, which depends

on whether multi-user access to the management system is required. For mufti-user access, either a local area network (LAN) or a mufti-user operating system such as UNIX will be needed. For single user access, which will be the case in most situations, conventional systems using a Windows operating environment will be sufficient. Trained personnel will be needed to maintain the selected operating system, drawn either from the road administration or from local computing companies under contract. 11.12 Once the requirements for the system software and operating system are defined, the choice of hardware will usually be self-evident. In most instances a microcomputer-based system will be preferred because it allows easy access to hardware maintenance. But the use of workstations should not be overlooked, particularly where large volumes of data will need to be processed, since workstations can substantially improve the efficiency of data storage and operation at little additional cost. 11.13 In all cases, the system has to include adequate facilities to back-up hardware and software in the event of unforeseen failures. Software back-up can be achieved through magnetic or optical devices for the storage of system information and road management data. The minimum requirement for hardware back-up, in the form of un-interruptible power supplies (UPS), is the ability to allow the system to be shut down properly if a power failure occurs.

51

52

Part C: System operation

53

54

12 Training
12.1 Training improves job performance by extending knowledge, improving skills and modifying attitudes. It enhances the ability of personnel to work in the most economical, efficient and satisfying way to achieve the objectives of the organisation. Training in the operation and use of road management systems has to be viewed against this background. 12.2 The staff of the road administration will need to receive specific training in the skills that will equip them to work with the management system and make the best use of its outputs. Once the system has been implemented, they will require a continuing programme of further training. If specialist consultants are used to perform the training function, they should concentrate on training counterpart staff who will then be able to undertake future training unassisted. This approach calls for the assignment of consultants and counterparts who are well motivated and who have high levels of interpersonal skills. 12.3 Detailed guidelines for training are beyond the scope of this Note, but some recommendations on training needs analysis are provided. Training of staff by Thagesen (1996) addresses the subject in more depth.

Extent of responsibility and authority for - planning, programming, preparation, operations - finance, staff and equipment. Management skills required. Computer skills required. Technical/engineering skills required.

12.6 The analysis is likely to identify requirements for training in excess of the available funding. This means that priorities will have to be set, and this should be undertaken in accordance with cost-benefit principles so that the choice is made to provide the training that will yield the best return on investment.

Training levels 12.7 Road management systems include a broad range of applications, while their potential users will possess a variety of skills and backgrounds. For this reason, training has to be approached in a flexible way. The concept of training levels, outlined in Box 12.1, offers a useful means of building flexibility into training provision. As the training level rises from appreciation to ability, there is a corresponding increase required in the depth of knowledge of the systems, but a decrease in the number of potential users. 12.8 Training courses at one level can assume competence at the preceding levels, so as to keep the training programme focused and avoid repetition. At the first level of `appreciation', training is likely to be more of a dissemination exercise. Much of what is needed at this level would be undertaken at the commitment stage of the system implementation process (3.9). The second and third levels - `knowledge' and `experience' -will relate to the majority of the users of the system, with Level 2 addressing staff whose use of the system is infrequent and Level 3 aimed at regular users. Level 4 - `ability' - will be required by those who have the responsibility for maintaining the system and undertaking other specialist tasks, such as model calibration.

Training needs analysis 12.4 There is a high degree of correlation between training issues and institutional issues. The institutional appraisal undertaken at the start of the design stage (3.6) should identify clearly the training which the road administration will need to provide to help achieve its objectives. Since the institutional objectives will determine the required training interventions, it is important that no training is planned or implemented without first conducting an institutional appraisal. This will identify any human resource constraints and indicate the necessary remedial measures. 12.5 The training needs analysis and the institutional appraisal should, ideally, be undertaken simultaneously. The training needs analysis, normally based around a structured interview, compares the performance level required from an individual with the level recorded at the time of the analysis. The gap between the two levels indicates the training requirement. 1n addition to collecting factual information about the individual in terms of qualifications, experience and so forth, the training needs analysis should identify the following points:

Training topics 12.9 Though requirements for preliminary and ongoing training will be based on the results of the training needs analysis, it is likely that training needs will be identified in the general areas of Management techniques (Box 12.2). Computer awareness (Box 12.3). Systems operation (Box 12.4). Technical matters (Box 12.5).

55

Box 12.1 Training levels

Box 12.2 Course outline: management techniques

56

Box 12.3 Course outline: computer awareness

Box 12.4 Course outline: systems operation

Box 12.5 Course outline: technical matters

57

12.10 Detailed advice about the content of training courses is outside the scope of this Note. Guidance on training programmes for management, general coordination, field engineers and road inspectors is available in a report produced by the OECD (1995).

Monitoring training 12.11 To make sure that training objectives are being met, continual monitoring will be required. The results will be used to improve and strengthen the training programme. Monitoring can be undertaken through tests, questionnaires and feedback from participants, as well as performance assessments by personnel responsible for training. These assessments should be made at regular intervals throughout the training. 12.12 Meeting training objectives does not necessarily mean that the road administration's objectives will be achieved. The road administration may fail to draw on the outputs available or use the new systems; some trainees may not want to share their new knowledge with others. Issues such as these become evident only after training participants have been working for some time in their assigned positions. A post-evaluation is seldom made, but it can be a source of valuable information for the road administration generally, and its findings can be applied specifically in the design of subsequent training programmes. 12.13 It has to be admitted that training has a poor record of success in supporting the implementation and operations management of systems. Emphasis has often been placed on inputs (such as the number of people trained) rather than impacts (such as the effects of training). Training should always be designed to meet clear objectives, with achievement targets that are both manageable and measurable.

58

13

Systems management

13.1 Managing the road management system is a function that needs to be treated like any other operational activity undertaken by the road administration. The benefits which a computer-based system can offer, in terms of improved management of the road network, will be lost if this aspect of its operation is neglected. Systems management involves the following sectors of activity: Defining responsibilities. Controlling systems and data. Monitoring and feed-back.

13.6 When new systems are introduced there may be advantages in starting with a specialised unit, and then extending systems operation more widely within the administration. This approach has attractions in terms of producing both early and longer-term benefits. However, once specialist cells are formed, they often create their own vested interests and are difficult to disband. They may also make the subsequent institutionalisation of the system more difficult because inter-departmental rivalries may develop during the early stages of implementation. The initial institutional appraisal (3.6) needs to examine these options and decide on the best long-term alternative.

Control of systems and data User access 13.7 Where the system is intended to be used by many individuals within the administration, controls need to be set up so that different users have access only to the specific applications and data groups that are appropriate to their work. Even where system use is centralised in a specialised unit, access control is no less a requirement because several members of the unit may be working with the system at any one time. 13.8 Control can take several forms: Full access with the ability to run applications or to enter and modify data. Read only access restricting users to viewing or producing reports. No access.

Defining responsibilities 13.2 There are two possible approaches to the operation of a road management system within a road administration. One approach is to make the use of the system widely available throughout the organisation; the other is to concentrate its use inside a specialised unit which supplies system outputs as reports to decisionmakers. 13.3 The first of these approaches has the advantage of developing across the administration a sense of `ownership' of the results, and hence of the system. This is likely to yield long-term benefits since the system will become closely integrated into the operations of the administration. On the other hand, it will incur higher costs - since more staff will need to be trained to run the management system, and more computer hardware will be required - and it will generally take a long time for the benefits to be realised. 13.4 Operation through a specialised unit allows systems to be introduced more quickly. With activities focused within a small group, fewer people need to be trained in system operation, and hardware requirements will be reduced. The benefits of system use will probably be more immediate, but it will be more difficult to institutionalise the system in the long term. 13.5 Regardless of the mode of operation, there is usually a need to appoint a system administrator. This person should be a computer specialist who has overall responsibility for the functioning of the system. The responsibilities of the post will include ensuring that the system is running correctly and that operating procedures are adhered to; maintaining data security, access control, back-ups and archiving; issuing passwords; and managing modifications and upgrades to the system. In a central unit responsible for the management system, the system administrator will often be the head of the unit. In an organisation-wide operation, an individual in the Computer Department will usually be an appropriate person to fill the post.

13.9 This control is a basic security feature and will normally be implemented through the use of passwords. It also permits the results of sensitive reports to be viewed only by nominated persons in the administration.

Data updating 13.10 Because data become obsolete over time, the management of the system has to include procedures for their updating. Data that change rapidly, such as details of road condition, will need to have an updating regime that reflects the road administration's management strategy (5.20). For data that change less frequently, such as inventory information, there are two options: Continuous updating, either as an on-going exercise or as part of the process of condition surveys. Updating at fixed intervals of time, either for the whole road network or for defined portions of the network on a cyclical basis.

59

13.11 Continuous updating is often preferred on the grounds that the alternative would result in the inventory falling into disrepute. On the other hand it is difficult to capture all inventory changes in this way, and a major periodic updating exercise may still be required. In some instances, a blend of the two options may offer the most practical course of action. Data control 13.12 Data security is achieved by defining and observing strict procedures for regularly backing up the system database. Generally, at least part of the database will need to be backed up each day. Nowadays, hardware is also available to support recovery from the inevitable system crashes and power interruptions. 13.13 The regular archiving of data that are no longer current can help minimise the need for system storage, reduce hardware requirements and increase the speed of database access. Procedures should be introduced which specify archiving functions and define the actions to be taken on those occasions when the archive needs to be accessed. 13.14 Data integrity is often neglected, but is particularly important when databases are used. For example, when changes are made to the physical road network it should not be possible to change section details in the database without also changing all other data relating to that section. The system needs to validate all new data entries to ensure that they are consistent with the data already stored, and that values lie between defined tolerances. Data integrity becomes critical in a system offering multi-user access because more than one user may be working on the data for an individual section of road at the same time.

13.16 Monitoring should provide a check that the policies, objectives, budgetary processes and final works programmes are linked together coherently. There will inevitably be a need to modify system outputs before final implementation owing to the practical realities faced by an administration. Even so, the following questions should be kept under constant review: Are strategic objectives and desired levels of service being achieved? Do works programmes reflect the results of the road management system? Is value-for-money being obtained? 13.17 Failure to address these institutional issues will have technical repercussions (Figure 3.1). For example, the same works implemented at a later date on roads which should have been treated earlier may prove inadequate, and significantly more expensive treatments may be required. In this case it will be necessary to reexamine the administration's wider maintenance strategies.

Technical issues Data 13.18 Implementing a computer-based road management system generally entails large numbers of surveys, measurements and observations. It is essential to maintain up-to-date data (13.10-13.11), otherwise the administration's perception of the network grows outdated, and the system gradually but inevitably loses credibility, leading to its ultimate demise. 13.19 Updating the database includes the following activities: Annual or mufti-annual surveys collecting monitoring data. Collecting works records when activities on the road have been completed. Models 13.20 Planning and programming systems, in particular, are likely to include models based on assumptions that may need to be verified. This is critically important in the case of techno-economic models such as HDM-III, even where they are calibrated properly. The consequences of using unverified models include: Works failing to improve the condition of the road as predicted. Road sections considered as non-urgent deteriorating more quickly than expected.

Monitoring and feed-back Institutional issues 13.15 Introducing a road management system can have a significant impact on the institutional arrangements of the administration (see 3.9). For example, the definition of a mufti-year programme of works is likely to have implications for the administrative structures, finance, staffing and other resources needed for its preparation and implementation. The required institutional structures need to be in place and functioning as planned from the outset, and typical problems that may need to be addressed include: Difficulties in financing successful operation of the system. Difficulties in adapting administrative methods. Over-estimating the ability to undertake data collection surveys.

13.21 The effects of such disparities may not be apparent in the short term. For this reason a selected sample of the network needs to be monitored in

60

detail to establish differences over time between predicted and observed behaviour (see Box 5.4). In addition, those responsible for maintaining and operating the system need to ensure that the technical methods employed by the system continue to reflect the technical approaches being adopted by engineers and managers in the wider organisation.

System operation 13.22 Monitoring the operation of the system is essential if it is to sustain the ability to meet objectives which may alter over time. This final step in the management cycle is an on-going activity, identifying where the system is not meeting requirements and acting as a trigger for action and improvement. 13.23 In some cases the action required may involve minor modification to the system or to the procedures which the system is supporting; but on occasions a more significant system upgrade may be necessary. An upgraded system needs to be subject to the same steps of commitment, requirements, specification and procurement as a new system. The results of all monitoring activities should feed into the process for reviewing the administration's policy framework.

Audit 13.24 This provides a physical check, usually on a sample basis, that work has been achieved in conformity with standards and procedures, and that costs and other resources have been accounted for properly. The feedback from road management systems provides a key input to this process.

61

14

References

Commission of the European Communities (1993). Project cycle management. Brussels: Commission of the European Communities. OECD (1995). Road maintenance management systems in developing countries. Paris: Organisation for Economic Co-operation and Development. Paterson W D O and Scullion T (1990). Information systems for- road management: draft guidelines on system design and data issues. Infrastructure and Urban Development Department Report INU 77. Washington DC: The World Bank. Robinson R and others (1998). Road maintenance maintenance: concepts and systems. London: Macmillan. Thagesen B (ed) (1996). Highway and traffic engineering in developing countries. London: Sport. TRRL Overseas Unit (1987). Maintenance management for district engineers. Overseas Road Note 1. 2nd edition. Transport Research Laboratory, Crowthorne. TRRL Overseas Unit (1985). Maintenance techniques for district engineers (2nd Edition). Overseas Road Note 2. Transport Research Laboratory, Crowthorne. TRRL Overseas Unit (1982). A guide to surface dressing in tropical and sub-tropical countries. Overseas Road Note 3. Transport Research Laboratory, Crowthorne. TRRL Overseas Unit (1988a). A guide to road project appraisal. Overseas Road Note 5. Transport Research Laboratory, Crowthorne. TRRL Overseas Unit (1988b). A guide to bridge inspection and data systems for district engineers. Overseas Road Note 7- Volume 1. Transport Research Laboratory, Crowthorne.

62

Appendix A: Glossary of terms

Activity Audit Any work or intervention that is earned out on the road network. A physical check, usually on a sample basis, that work has been carried out, where specified, to pre-defined standards or procedures, and that costs and other resources have been accounted for properly. Normal pre-defined headings under which expenditure is allocated by a Ministry of Finance. The government budget normally used to fund major projects. A parameter that combines individual defect measurements to provide a summary indication of defectiveness. The ratio of `effectiveness' to `cost', where effectiveness is a measure of the future value or worth resulting from a decision that is taken, and cost is the present-day cost of implementing that decision. A formal comparison of costs and benefits to determine whether or not an investment is worthwhile. Scheduled works whose needs tend to be dependent on environmental effects rather than traffic. These works are programmed in advance and include such activities as, for example, culvert cleaning. That feature of data that relates to its completeness and internal consistency. A computer-based collection of data that normally uses formalised rules for the way that the data are stored. A computer-based system comprising applications modules to process data and provide enhanced information on which informed decisions on road management can be based and, ultimately, implemented. The class of decision making processes where outcome is predicted as a precise value on the basis of mathematical functions of observed or measured inputs. Projects planned at discrete points in time that result in improved road capacity or changes in alignment. Works undertaken to clear a road that has been blocked. An advanced computer programming language, usually used for interrogating databases. The `broadest brush' category of cost-estimating technique which relies on libraries of achieved costs of similar works; eg cost per kilometre of bituminous resurfacing. The physical components of a computer system, including processor, keyboard, monitor and printer. The `Highway Design and Maintenance Standards Model Version III', which is a computer-based decision-support system, developed by the World Bank, and used for economic appraisal of road projects.

Budget head

Capital budget Condition index

Cost-effective

Cost-benefit analysis

Cyclic works

Data integrity

Database

Decision-support system

Deterministic

Development works

Emergency works Fourth generation (4GL) programming language Global cost

Hardware

HDM-III

63

Information quality level (IQL)

Criteria developed by the World Bank for grouping data in terms of their level of detail and other attributes to assist in specifying data collection that is cost-effective when used in conjunction with road management systems. A computer-based system that collects, organises and stores data. All aspects related to the use of computers to assist with management or other activities. An investigation of an organisation that identifies its strengths and weaknesses, success in meeting defined aims, and the constraints under which it operates. A record of the physical attributes of the road network or other asset being managed. A system of linking computers in fairly close proximity, such that each can have access to common peripherals, data and software. A written description used as a starting point in developing computer software. This includes a detailed description of the functions, processes and data structures of the processes for which computer software is to be written, but described in such a manner that is independent of the programming language to be used by the software or the hardware on which it will run. A computer-based system for assisting with the management of maintenance (note that in UK English, this term will often be used synonymously with the term `pavement management system' whereas, in US English, the term will normally refer to an `operations management system'); to avoid confusion, the use of this term should be avoided. A series of well-defined steps which take the management process through the decision making tasks. Typical steps would be i) define aims; ii) assess needs; iii) determine actions; iv) determine costs and priorities; v) implement activities; vi) monitor and audit. The process typically completes the cycle once in each periodic cycle of the particular management function. A means of defining a management task based on its objective. Management functions are undertaken so that the requirements of the policy framework are met; examples are planning, programming, preparation and operations. Computer output that shows features and attributes of the road network spatially against a map background. Parameters input to a computer system that define the fundamental processes of operation of the system. Computer software that is used to interface between applications software and an RDBMS, such that the way that the applications software can be written is independent of the RDBMS. This outlines, in broad terms, the goal of the organisation responsible for the road network, and justifies its existence. A mathematical function or algorithm that is used to simulate real life effects. An entity that is broken down into discrete parts that can be developed, tested and operated independently of each other.

Information system Information technology (IT)

Institutional appraisal

Inventory

Local area network (LAN)

Logical design

Maintenance management system

Management cycle

Management function

Map-based graphics

Meta data

Middle-ware

Mission (statement)

Model

Module/modular

64

Multi-year programme

A schedule of road works planned to take place in discrete years into the future. A particular grouping of roads for management purposes; examples are the national road network; trunk road network; paved road network, etc. An information system that stores data about the road network and its inventory. The process of breaking the road network down into successively smaller links, segments and sections, each of which can be defined uniquely for road management purposes. Software mounted on a computer that provides an interface between applications software and the computer hardware. The operating system enables a user to make use of the hardware facilities of the computer, for example by organising where information is stored in memory, by storing, retrieving and copying files, and by looking after the input and output. The on-going activities of an organisation, for which management decisions are made on a near-term basis. Examples include the scheduling of work to be carried out, monitoring in teens of labour, equipment and materials, the recording of work completed, and the use of this information for monitoring and control. A fundamental cost-estimating technique that builds up the total cost of the work from its component activities described by the method statement and programme, in terms of labour equipment and materials. A computer-based road management system, typically used to assist with planning, programming or preparation; to avoid confusion, the use of this term should be avoided. This specifies the quality of finished work for an activity, and builds up a consistent description of the activity based on a preferred method of working, and requirements for equipment, labour and materials. Works planned on a regular basis to take place at intervals of several years. This involves an analysis of the road system as a whole, typically requiring the preparation of long term, or strategic, estimates of expenditure for road development and conservation under various budgetary and economic scenarios; predictions may be made of expenditure under selected budget heads, and forecasts of road conditions, in terms of key indicators, under a variety of funding levels. A set of statements, normally comprising a mission statement, objectives and standards, that define in detail the aims of an organisation and how it proposes to achieve these. The near-term planning stage where road schemes and projects are packaged for implementation. At this stage, designs are refined and prepared in more detail; bills of quantities and detailed costings are made; together with work instructions and contracts; detailed specifications and costings are likely to be drawn up. Addition of a thin film of surfacing to improve surface integrity and waterproofing that does not increase the strength of the pavement.

Networks

Network information system

Network referencing

Operating system

Operation(s)

Operational cost

Pavement management system

Performance standard

Periodic works

Planning

Policy framework

Preparation

Preventive works

65

Priority index

A parameter whose value gives an indication of the priority of the associated activity. The class of decision making processes where outcome is predicted as a probability function of a range of possible inputs. A method of problem solving that works backwards from a problem statement, breaking this down into more detailed components, and then develops these to find a solution. A documented series of steps for carrying out a particular activity or task. The preparation, under budget constraints, of mufti-year works and expenditure programmes in which those sections of the network likely to require treatment, and new construction possibilities, are identified and selected; a tactical planning exercise. Works responding to minor defects caused by a combination of traffic and environmental effects. The government budget which is normally used to fund those works that are needed every year, including such items as staff salaries, running costs of a road administration, and maintenance works for the road network which are undertaken on a regular basis. A method of analysis used in computer-based prioritisation of road works that uses an approximate approach to life cycle costing that is less demanding of computer processing requirements. Computer software designed to construct, modify and maintain a database, where the database stores data in a structured way in a number of database files, and where the detailed relationship between specified data items is defined. A grouping of road sections according to pre-defined rules, often based on issues of ownership, function, funding source, etc. The process of maintaining and improving the existing road network to enable its continued use by traffic efficiently and safely, normally in a manner that is effective and environmentally sensitive; a process that is attempting to optimise the overall performance of the road network over time. A computer-based system used to assist with road management. Minor works that need to be undertaken each year. A set of instructions that can be stored in a computer and used to carry out pre-defined tasks; often referred to as a `program'. Data that include geographic co-ordinates among their attributes so that they can be displayed using map-based graphics. Works the frequency of which cannot be estimated with certainty in advance. A detailed description of the attributes of the output from an activity, or of the steps by which that activity is earned out. Those with a vested interest in the performance of the road administration, including the road users, industry, agriculture and commerce, who are its `customers', plus the road administration itself and the road engineering industry.

Probability/probabilistic

Problem tree analysis

Procedure Programming

Reactive works

Recurrent budget

Reduced time period analysis

Relational database management (RDBMS) system

Road class/hierarchy

Road management

Road management system Routine works Software

Spatial data

Special works Specification

Stakeholder

66

Standard (maintenance)

A requirement, sometimes legally enforceable, that a road administration is obliged to meet as part of its road management activity. Pertaining to actions, often wide ranging, designed to achieve defined objectives in the long term. A report, often in the form of a computer output, that shows road features and attributes relative to the longitudinal, but not cross-sectional, position along the road. A structured group of discrete entities which interact for a particular purpose; examples are: A `computer system', which is a collection of software and hardware designed to carry out a particular function A `management system' which is a set of procedures designed to assist the management process (which may also include a computer system)

Strategic

Strip diagram

System

Tactical

Pertaining to actions designed to achieve defined objectives in the short to medium term. A formalised approach to classifying skills and needs of individuals that assists in identifying training needs. A cost-estimating technique based on the traditional bill of quantity approach to pricing engineering work, typically relating to aggregate quantities of work to be carried out, measured in accordance with an appropriate method of measurement. A device for storing electricity, that can be placed between computer hardware and a mains power supply, that enables continued operation of the hardware for a reasonable period in the event of a failure of the mains power supply; the device normally has the further ability to `smooth out' power surges and peaks in the mains power supply that could otherwise damage operation of the hardware or corrupt the software. A proprietary computer operating system designed for efficient and operation with large software systems and where many users are accessing the software or data at the same time; often used in connection with mini-computer or `work station' hardware, but also available on micro-computers. All construction and maintenance activities.

Training levels

Unit rate

Un-interruptible power supply (UPS)

UNIX (UNiplexed Information Computing Service - UNICS)

Works

67

Appendix B: Institutional appraisal check list

(Developed from: Brooks, D M and others, 1989. Priorities in improving road maintenance overseas: a checklist for project assessment. Proc Institution of Civil Engineers, Part 1. 86 (Dec), 1129-1141.) 1 1.1 1.1.1 1.1.2 1.1.3 1.1.4 1.1.5 1.2 1.2.1 1.2.2 1.2.3 1.3 1.3.1 External (socio-political, contextual or environmental) factors Legal and regulatory Is there a Roads Act which defines clearly the role and responsibility of the road administration? Which organisations are responsible for different classes of road? Are the legal powers adequate and, generally, are they understood by the road administration? What is the influence of government or civil service procedures on operations of the road administration? Is the road administration subject to external monitoring and audit? Socio-cultural Are there historical, social or cultural factors which affect the road administration's operations? Is there a recognisable management culture in the country? Are decisions in government independent of the influence of nepotism, favouritism, or corruption? Political What priority is given to road management compared with other sectors, and with other activities in the road sub-sector? To what extent are road administration staff political appointees? To what extent are decisions made influenced by political factors? Economics and resources To what extent is the economy governed by market forces, and to what extent is it planned or controlled centrally? What impact does the general economy have on the ability to manage the road effectively? Is a budget awarded, is it adequate, and can it be relied upon? Are operations independent of foreign exchange constraints?

1.3.2 1.3.3 1.4 1.4.1

1.4.2 1.4.3 1.4.4

68

1.5 1.5.1 1.5.2 1.6 1.6.1

Overall employment policies Are there sufficient personnel available at all levels? Are they adequately educated? Private sector What is the government policy towards private sector, and how is this implemented in practice? What is the availability of domestic project financing? What is the availability of international/donor project financing? What locally-based road consultants exist and where are they located: Domestic? International? What locally-based roadworks contractors exist and where are they located: State-owned domestic companies? Joint venture state/private? Private domestic? International? What is the workload of: Consultants? Contractors? Competence: How well qualified and experienced are the domestic staff of: Consultants? Contractors? What equipment holding do contractors have? What is the quality of work produced? Are bidding procedures: appropriate? transparent and free from corruption? How do contract documents compare with international norms, such as FIDIC? Is payment for works timely? Does the client have the ability to supervise, measure, test, etc? Is there access to arbitration? Related institutions Is the performance of the road administration dependent on the output of other organisations over which it has no control? To what extent is the road administration subject to external competition? What external incentives to performance exist?

1.6.2

1.6.3

1.6.4

1.6.5

1.6.6

1.7 1.7.1

1.7.2 1.7.3

69

2 2.1 2.1.1 2.1.2 2.1.3 2.1.4

Internal institutional (organisational, managerial and human resource) factors Policy framework Does the administration have a mission statement, objectives and standards, and are these appropriate? Is the policy framework appropriate and achievable? Is policy based on economics, or other factors? Are road plans produced for: Construction? Maintenance? and at what frequency? Organisational and administrative structure Is there an administrative structure that is appropriate for and capable of managing roads? Are responsibilities defined, and are staff aware of these? Is there an unambiguous chain of command? Procedures Do documented procedures exist for: Planning? Programming, including budgeting and setting priorities? Project preparation? Management of on-going operations? Are the procedures understood and applied? Resource management Is there a regular and formal budgeting process? Is the budget based on assessed or measured need? Is the budget related to actual costs and the ability to disburse? Are procurement procedures appropriate, efficient and independent of corruption? Does full financial control reside within the maintenance organisation? Are accounts independently audited? Human resources Is there a manpower inventory, and career development records for all staff? Is there a manpower plan which is used as the basis of staff recruitment, deployment, utilisation and advancement? Are compensation, benefits and incentives adequate to motivate staff? Is systematic staff appraisal carried out against performance standards as an input to the career development process? Do adequate sanctions exist to deal with indiscipline, and are these implemented? Are staff adequately trained, and is there an internal training scheme?

2.2 2.2.1 2.2.2 2.2.3 2.3 2.3.1

2.3.2 2.4 2.4.1 2.4.2 2.4.3 2.4.4 2.4.5 2.4.6 2.5 2.5.1 2.5.2

2.5.3 2.5.4

2.5.5 2.5.6

70

3 3.1 3.1.1 3.1.2 3.1.3 3.1.4 3.1.5 3.1.6 3.1.7 3.2 3.2.1 3.2.2 3.2.3 3.2.4 3.2.5 3.3 3.3.1 3.3.2 3.3.3 3.3.4 3.3.5 3.3.6 3.4 3.4.1 3.4.2

Technical (engineering and physical) factors Data Is the network referenced, and is this up-to-date? Does an inventory exist for the road network and its structures, and is this up-to date? Is there adequate traffic and axle load data, and is this up-to-date? Does sufficient up-to-date condition data exist to enable needs to be assessed? Is road accident data collected and analysed in a systematic manner? Is unit cost data available and up-to-date? To what extent are computer-based information and management systems available and used? Materials and supplies Are appropriate materials and supplies of the right quality available as required? Are the properties of materials used fully understood? Are there adequate testing facilities, and is the quality control of products and materials adequate? Are appropriate materials always used? Does an adequate system exist for ordering, storing and stockpiling materials arid supplies? Plant and equipment Is there a fleet of plant and equipment of the size and composition required? Is the availability adequate? Is the utilisation adequate? Are the workshops and stores adequate to support it? Is there an organisation capable of managing the fleet cost-effectively? Is adequate financial provision made for replacement and repair? Operations Do appropriate design methods exist, and are they used? Are there specifications for work, and are these met in practice to achieve adequate quality control on site: For works carried out by direct labour/force account? For works carried out by contract? Are contract procedures satisfactory and enforced?

3.4.3

71

3.5 3.5.1 3.5.2 3.5.3 3.5.4 3.5.5 3.6 3.6.1

Monitoring acid feed-back Is work done measured and costed? Are costs realistic in terms of overheads, equipment, materials and labour? Is cost and monitoring information collected centrally and used for planning and budgeting purposes? Is there a physical inspection and audit of work done? Is productivity measured? Is there adequate access to current research work and good practice from other organisations or research centres? Is relevant research currently carried out within the organisation? Are new techniques and practices introduced as a result of research results?

3.6.2 3.6.3

72

Appendix C: Example applications of the information quality level concept

(Source: Paterson, W D, 1991. Choosing au appropriate information system for road management. In: PIARC. 19th World Congress of PIARC, Marrakech, September 1991. Paris: Permanent International Association of Road Congresses). Table C.1 Road inventory

73

Table C.2 Methods of measurement for pavement structural evaluation

Table C.3 Characteristics of options for deflection measurement

74

Table C.4 Minimum spatial sampling rates and methods for pavement structural evaluation

Table C.5 Traffic volume and axle loading for different information quality levels

75

76

ISSN 0951-8797

OS-E

Guidelines for the design and operations of road management systems

ORN 15

Transport Research Laboratory Old Wokingham Road Crowthorne, Berkshire, RG45 6AU

Overseas Development Administration 94 Victoria Street London, SWIE 5JL

OVERSEAS ROAD NOTE 16

PRINCIPLES OF LOW COST ROAD ENGINEERING IN MOUNTAINOUS REGIONS, WITH SPECIAL REFERENCE TO THE NEPAL HIMALAYA

Subsector: Theme: Project title: Project reference:

Transport T2 Environmental impact of road construction in Nepal R 6026

Copyright Transport Research Laboratory 1997. All rights reserved. This document is an output from a project funded by the UK Overseas Development Administration (ODA) for the benefit of developing countries. The views expressed are not necessarily those of the ODA.

Transport Research Foundation Group of Companies Transport Research Foundation (a company limited by guarantee) trading as Transport Research Laboratory. Registered in England, Number 3011746 TRL Limited. Registered to England, Number 3142272 Registered Offices Old Wokingham Road, Crowthorne, Berkshire, RG45 6AU.

ACKNOWLEDGEMENTS This Note was written and prepared by G J Hearn of Scott Wilson Kirkpatrick & Co Ltd (SWK). C J Lawrance of the Overseas Centre, Transport Research Laboratory (TRL) contributed text sections and acted as editor. Additional text was provided by R M Weekes of SWK. Technical comments were received from S Handsley, T Hunt, D Jordan, D G Williams and L G Willliams, all of SWK, and R Finn, formerly of SWK. J H Howell of Fountain Renewable Resources Ltd and Indu S Dhakal of the GeoEnvironmental Unit, Department of Roads, Nepal, provided comments on vegetative methods of slope protection. Professor P G Fookes, who presents the Foreword, is an independent consultant with long experience of engineering geological problems in hazardous environments. This Overseas Road Note was originally the concept of Mr J W F Dowling, former Deputy Head of the Overseas Centre, TRL. Nepal, its culture and its geotechnical challenge always held a special affinity for this geologist. The cover photograph was taken by W G Heath, of TRL.

First Published 1997 ISSN 0951-8797

OVERSEAS ROAD NOTES Overseas Road Notes are prepared principally for road and transport authorities in countries receiving technical assistance from the British Government. A limited number of copies is available to other organisations and to individuals with an interest in roads overseas, and may be obtained from: Transport Research Laboratory Crowthorne, Berkshire, RG45 6AU United Kingdom

Limited extracts from the text may be reproduced provided the source is acknowledged. For more extensive reproduction, please write to: Programme Director, Overseas Centre, Transport Research Laboratory.

FOREWORD
Great things are done when men and mountains meet; This is riot done by jostling in the street. Gnomic Verses, William Blake (1757-1827) I believe that this Note is needed because there has been nothing so comprehensively available hitherto and the subject matter is important to those who build in high mountains. It deals with the mountains themselves, their behaviour, investigation and construction on their slopes, and represents a stateof-the-art survey in both conceptual and practical aspects. Although the case histories are drawn extensively from Nepal, the Note is applicable to mountains elsewhere, particularly those with a similar monsoonal climatic regime. Accordingly, it should be of considerable value to road planners, designers and constructers. The unifying theme is road works. However, a lot of the conceptual and practical detail will be of value to academics, engineering geologists and geotechnical, civil and hydraulic engineers, working in high mountain environments. The famous through mountain routes of history usually occupy convenient, strategic interconnecting valleys and passes. Mountain roads of today are necessarily built in inhospitable terrain because of the need to link communities and access remote valleys. When I first went to Nepal, in 1962, it was by the gruelling weekly flight in an ageing DC3 from Patna (Northern India) to Kathmandu, along with the mail and goods occupying much of the rear of the aircraft. Obtaining a visa was difficult and required a period in residence at Patna to be near the Nepalese Ambassador. The DC flight was the easy option compared with the land route! My first view of the spectacular mountains was cloud! In the 1960s and early 70s there were few roads capable of taking vehicles even into the foothills of the Himalayas. Movement of goods was, and still is in probably the majority of Nepalese countryside today, by backpacking porters. At that time route planning was exceptionally difficult and a poorly supported undertaking. Mapping coverage was minimal and air photographic coverage usually had problems because of the distortion produced by big changes in the severe relief. Remote sensing imagery was still half a lifetime away, and the hard won experience gained on the Indian designed and constructed Raj Path to Kathmandu was slow to trickle through. Living and working locally was not easy for foreigners, as it still is today away from the cities and bigger towns. The first major British-funded road into the Nepal Himalayas was a low cost, sealed road, from Dharan to Dhankuta. Although the distance between these towns is only 17km, the length of the road is 52km. We had little published information to draw upon in creating the design, but every metre required careful thought and I consider that the road when completed was an engineering triumph. The overall concept was to climb in a heavily engineered corridor, usually stacks of hairpins, and make distance on the higher, flatter slopes. Experience on that

road culminated in an innovative paper (Fookes and others, 1985 - see Bibliography) which distilled experience from that road, other locations in South-East Asia, the Andes, and the slowly growing world literature on the subject. Mountain roads in the late 1980s and 90s, particularly in developing countries, have become more common and more venturesome. This Note builds upon and takes forward the experience and conceptual approach to mountain road design published in 1985 and, I believe, will be of immense practical value to the new breed of mountain roads. The mountain environment stems directly from its geology and climate, current and past. The climate, particularly rainfall, dictates the way in which mountain slope erosion systems act and develop. Therefore, high mountains in different climatic zones tend to look and behave somewhat differently from each other and require variations in engineering approaches to investigation, design and construction. The bulk of the experience that has gone into the snaking of this Note has come from the Nepal Himalayas where the climate is warm and monsoonal in the Lower and Middle Mountain Zones. Extending the wealth of information contained in the following to mountains elsewhere in the world is particularly relevant for geologically young mountains with a similar climatic regime. Nevertheless, the majority of the fund of practical design and construction advice contained in the Note will be of direct help or give clearer guidance on ways of tackling severe relief, climate and materials in mountainous regions in different states of geological development and/or in different climatic regimes. The geological construction of mountain systems is inextricably linked to plate tectonics. Nepal typifies the geological zones and variety of rocks that make up an orogenic mountain chain. Orogenesis (from the Greek meaning `mountain creation') results when one of the earth's crustal plates rides over another, causing thickening and crumpling of the overriding plate. The metamorphic rocks that result form the rocks of the Himalayas: low-grade metamorphic rocks in the Lower Zones and igneous and high grade metamorphic rocks in the Middle Zones. The Siwalik Hills, little-folded, weak, weatherable young rocks, and extensive alluvial plains of the Terai are the results of rapid erosion of the rising mountain chain. The product of the monsoonal climate and the rapid, albeit jerky, uplift of the Himalayas with its massive rates of erosion dictates their special conditions and those of other high mountains of a similar nature. The Himalayas are among the youngest (about 10 million years) major mountain systems in the world. Slopes are very steep and largely covered in granular rather than clay-rich tropical residual soil more common on less steep slopes. These granular or taluvial slope soils are continually on the move downwards. The extensive slope instability common to young orogenic mountains, the frequency of gullies leading to bigger drainage systems to get rid of monsoon rain, and the overall deep incision of valleys keeping pace with the uplift, produce characteristic investigation and construction difficulties which, when flood and earth-

iii

quake events are added, bring about most severe engineering conditions. In these conditions investigatory techniques are principally based on geotechnical mapping supplemented by conventional pits and observational methods, all of which have to be further supplemented during construction. Access and major difficulties of conventional site investigation drilling are dominant, with continuing problems in a physical environment which, at best, is commonly only marginally stable. Typically, the higher the design standard, the greater the problems of construction. My experience is that high road design standards will lead to extremely high capital construction costs, for example in massive retaining works, tunnels and big bridges, and also to much higher repair and maintenance costs. I believe time is beginning to show that the most successful low cost roads are those where maintenance is frequent and well done and the road has been designed and constructed in sympathy with the physical environment. The latter does not necessarily imply high design standards but appropriate engineering for the natural circumstances. The Transport Research Laboratory has been actively involved in maintaining a professional interest in the engineering geological and geo-environmental aspects of road engineering in unstable mountainous areas. They have collaborated with engineering consultants and government agencies on a number of projects. Building on this and on experience by others, this Road Note is aimed at both private and public sector engineers and planners. The objective has been to produce a manual which can lead, in a straightforward manner, to planning, investigation, design and construction of road works that are in sympathy with the mountain environment to standards that are appropriate for that environment, and costs which make such roads affordable. I see this as having been achieved by the

concise style, with many to-the-point practical tables and diagrams, all of which should be straightforward to follow providing the reader has some understanding of the geological and climatic setting which govern the processes operating in the landscape. One of the strengths of the Note is that the chapters are both self-contained and dovetail into each other. Emphasis on aspects of design for mountain roads is different from that for conventional roads in lowlands: drainage is still the paramount design requirement but with extensive use of offsite drainage. Not only has a huge amount of water to be removed rapidly and efficiently, but the conveyer belt of debris being brought down the mountainside, particularly in times of high rainfall, has to be dealt with. The concepts developed for mountain gully and valley works are not usually required for other landforms. This is also true of hairpin stacks and flood protection arrangements. Earthworks, to a large extent, are not essentially different from those of flatter terrain, only perhaps there are more of them, and retaining structures, slope protection and stabilisation measures are ubiquitous. However, they must be well drained, flexible, extensive and put in quickly in short lengths to prevent a disaster during construction. The secret of successful high mountain road design and construction is to work with and not against the massive forces of nature. Frustrating most of the time, but I love it. Prof P G Fookes, F.Eng. Winchester September 1996

iv

CONTENTS
Page FOREWORD PREFACE 1 2 INTRODUCTION GEOGRAPHICAL SETTING OF THE NEPAL HIMALAYAS GEOLOGY AND TOPOGRAPHY CLIMATE GEOMORPHOLOGY HIMALAYAN ZONES OF NEPAL Siwalik Hills Mahabharat Lekh Middle Himalaya High Himalaya 3 ENGINEERING PROBLEMS GENERAL CONSIDERATIONS TERRAIN HAZARDS Landslides Slope erosion Stream erosion Sediment transport and flood plain scour River flooding Rain-fed floods Landslide dams Glacier lake outburst floods (GLOFs) Earthquakes Road construction impacts 4 PLANNING AND DESIGN OVERVIEW INTRODUCTION PROJECT IDENTIFICATION iii viii 1 3 3 3 3 8 9 9 9 9 10 10 10 10 14 14 15 15 15 15 15 16 16 17 17 20 5 FEASIBILITY STAGE Route corridor assessment Design standard and design life Erosion protection Feasibility study report DESIGN Detailed survey and alignment design COST ESTIMATION CONTRACTS Contract conditions and specifications Contract administration CONSTRUCTION ASPECTS Rock blasting Construction materials and water Gravel or sealed surface Rate of construction and construction sequence Pilot tracks Quality control and supervision Improvement and upgrading Maintenance practice DESK STUDY AND RECONNAISSANCE SURVEY INTRODUCTION DATA SOURCES MAPS Topographical maps Geological maps Soils maps Terrain classification maps Page 20 20 21 22 22 23 23 24 24 24 26 27 27 28 28 28 28 30 30 31 32 32 32 32 32 32 32 32

Page AERIAL PHOTOGRAPHY SATELLITE IMAGE INTERPRETATION RECONNAISSANCE SURVEY LANDSLIDE HAZARD MAPPING Introduction Common methods of landslide hazard mapping Applications of landslide hazard mapping Landslide hazard mapping for route alignment in Nepal Rock avalanche hazard mapping in Papua New Guinea 6 ENVIRONMENTAL IMPACT ASSESSMENT INTRODUCTION IMPACT CATEGORIES Geomorphology Ecology and natural resources Land use Socio-economic factors ENVIRONMENTAL IMPACT ASSESSMENT PROCEDURE Project description Base-line study Comparison of alternatives Mitigation and compensation measures Environmental management and monitoring 7 GEOTECHNICAL ASSESSMENT INTRODUCTION GEOTECHNICAL MAPPING GEOTECHNICAL INVESTIGATION 33 50 53 53 53 55 55 59 59 65 65 INTRODUCTION 65 DESIGN LIFE 65 RAINFALL ANALYSIS 65 RAINFALL - RUNOFF ANALYSIS 67 67 68 68 Flood envelopes 68 FLOOD FREQUENCY ANALYSIS 68 71 71 72 Bridges 72 Scour 72 Sediment transport 74 SLOPE-AREA METHOD GLACIER LAKE OUTBURST FLOODS (GLOFs) HYDRAULIC DESIGN Culverts REGIONAL FLOOD DATA AND FORMULAE Published area-specific discharge data Empirical discharge-area equations 8 Trial earthworks Sampling and laboratory testing Geophysics Logging of excavations as construction proceeds GEOTECHNICAL ANALYSIS AND DESIGN Introduction

Page 74 74 76 76 76 76

Assessment of soil strength parameters 76 Slope stability analysis Slope design Rock slope stability analysis and design HYDROLOGY AND HYDRAULIC DESIGN 77 79 81 82 82 82 83 84 90 90 90 90 92 92 93 94 94 94 94 96

vi

Page 9 EARTHWORKS CHOICE OF CROSS-SECTION Mostly full cut cross-section Balanced cut and fill cross-section Mostly full fill cross-section EARTHWORKS DESIGN AND CONSTRUCTION Cut slopes Unretained fill slopes Rock blasting Spoil disposal 10 DRAINAGE GENERAL PRINCIPLES ROAD SURFACE DRAINAGE AND SIDE DRAINS Crossfall Side drains Culvert location Culvert design Stream course protection Check dams Cascades Channel linings Drainage of hairpin stacks 11 ROAD RETAINING WALLS INTRODUCTION WALL TYPES Dry masonry Masonry with dry stone inclusions (composite walls) Masonry walls Gabion walls 97 97 97 97 97 99 99 100 101 101 103 103 103 103 103 104 106 108 108 108 113 114 116 116 116 116 121 121 121 14 Bio-engineering Impermeable membranes and surface coatings Revetments Surface drainage SLOPE STABILISATION Introduction Sub-surface drains Support for soil slope failures Support for rock slope failures 13 ROAD CONSTRUCTION ALONG VALLEY FLOORS INTRODUCTION FREEBOARD FLOOD PLAIN SCOUR AND EMBANKMENT PROTECTION RIVER TRAINING CROSS DRAINAGE AND TRIBUTARY FAN CROSSINGS BIBLIOGRAPHY 12 Foundations and sub-surface drainage Surface drainage SLOPE PROTECTION AND SLOPE STABILISATION INTRODUCTION SLOPE PROTECTION Embankments Cut slopes Reinforced earth Buttress-supported road slabs Mass concrete and reinforced concrete walls STABILITY, CROSS-SECTION AND ARRANGEMENT Backfill

Page 122 122 123

124 124 124 127 127 127 127 127 127 136 136 136 139 139 139 140 141 142 142 142 142 143 144 147

vii

PREFACE
In the world's poorer countries. there is an increasing requirement to build roads into the less accessible parts of the country as a means of facilitating development projects in these regions. Where these inaccessible areas constitute a mountainous interior, the problems and costs of road construction multiply. The increased costs of road construction are not simply related to the high capital cost of building an alignment in complex, possibly unstable terrain. Of much greater concerti to international funding agencies is the long term cost of road maintenance in areas where instability is a natural process and inevitably can be expected to affect the road at some time in the future. The Overseas Development Administration's objectives in terms of road design for development in mountainous regions are threefold. The first is to design an alignment that is affordable within the current development ethic. The second is to adopt a design that is appropriate to the terrain, with a good expectation that it will stand up to the forms and intensity of instability experienced in the region. The third is to design a road that can be maintained to a good standard by the host nation, within the limits of its financial resources and technical capabilities. This last may initially need to be accomplished with assistance, but under ODA's policy of helping nations to help themselves, responsibility for maintenance would, in time, be expected to return to the host nation. ODA has strived to construct roads that fulfil these requirements, but thirty years ago, when the Administration first began to plan its road construction programme in Nepal, relatively little was known about the pace and vigour of instability in the Himalayan region. Initially, ODA consulted TRL in matters of alignment and drainage design, and through this work was realised the importance of designing `with the terrain' in such an active environment. For management of the construction of later roads in Nepal, ODA has appointed consultancies with leading expertise in geomorphological and geotechnical investigations. As a result, ODA's roads in Nepal have a record of stability that equals the best. For a variety of reasons, organisations and individuals who have become involved with road design and construction in Nepal have remained on the scene, on and off, for many years. These organisations and individuals have been in the fortunate position (for posterity) of being involved throughout the phases of route planning and survey, road construction and beyond. As a result, they have been able to see the effects of design decisions taken perhaps fifteen or twenty years previously. They have also been able to monitor the effectiveness of a wide range of maintenance practices over periods of years.

The organisations who are behind Overseas Road Note 16 are among this group. By virtue of its association with ODA, TRL has had unrivalled opportunity to follow progress in road construction in Nepal. Through ODA's funding to TRL, TRL has been involved in providing advice to ODA on road engineering in Nepal and other mountainous regions, first in the late 1960's and periodically ever since. Through visits to Nepal over this long period, TRL staff have kept up observations more or less annually on ODA-funded roads and, incidentally, on other major roads in Nepal. TRL has been present on roads under construction and in the aftermath of occasional natural disasters. Latterly, TRL has carried out research into bioengineering methods of slope protection. Also, in collaboration with Scott Wilson Kirkpatrick it has used ODA research funds to measure rates of natural instability on slopes in east Nepal. In unstable mountainous environments there is no substitute for practical experience. The importance of rigorous application of standard design practice is not in question. But in an environment where unpredictable natural forces are always at work, reliance on standard design solutions should be tempered by a preparedness to use the eye and intuition to vary the 'book' design in places where the terrain demands it. The terrain and the forces of nature are paramount in a mountainous region. They dictate design, and any design that does not take account of the ever-changing environment is doomed to failure. The experience gained by the authors and their colleagues from observation of successes and failures on roads in Nepal and elsewhere has been brought together in this Overseas Road Note. The principles of good engineering practice described in the guide are derived from extensive experience and a deep affinity for landscape held by its contributors. Through its publication, ODA's three objectives of designing roads for development, mentioned above, can perhaps become more widely attainable.

C J Lawrance Overseas Centre

viii

PRINCIPLES OF LOW COST ROAD ENGINEERING IN MOUNTAINOUS REGIONS, WITH SPECIAL REFERENCE TO THE NEPAL HIMALAYA
1 INTRODUCTION
identification of ground rules for good engineering practice in steep and unstable terrains generally. 1.5 Therefore, although the Nepal Himalaya forms the main subject area of this guide, the discussions and recommendations should be equally applicable to the adjacent Hindu Kush-Karakorams, parts of the Andes and, perhaps with some modification to account for differing engineering technologies and land use patterns, most of the fold mountain chains of South East Asia. The guide will also be relevant to many aspects of road engineering in volcanic terrains, although volcanic hazards themselves are not dealt with. High altitude glacial mountains are excluded from this guide, as are problems of snow and ice on mountain passes. 1.6 Road design, construction and maintenance under Himalayan conditions requires a different approach to that conventionally adopted for less severe flat or rolling terrains. While the engineering principles remain the same, design parameters and design priorities are quite different. For instance, the shortest road alignment is not necessarily the easiest, quickest or cheapest option to construct or maintain. Frequently, topography, slope stability, flood hazard and erosion potential are likely to be the most significant controls on the choice of the most suitable alignment and design of cross-section. Variations in geology and slope greatly influence road design and hence the cost of construction, and these variations can occur over very short lengths of alignment. Geology, geomorphology and hydrology, therefore, are key factors in the design, construction and maintenance of roads in these regions. Nevertheless, an appreciation of these factors alone is not enough if roads are to be constructed in an environmentally sympathetic way, without premature damage. Road geometry, earthworks, retaining structures and drainage measures must be designed in such a manner as to cause the least impact on the stability of the surrounding slopes and natural drainage systems. Excessive blasting, cutting, side tipping of spoil and concentrated or uncontrolled road drainage often lead to accelerated instability and erosion. Although many of these effects are often unavoidable to a certain extent, the design and the construction method adopted should aim to minimise them. 1.7 Bearing in mind the destructive nature of slope and drainage processes that fashion the landscape in these areas, financial and technical commitment to road maintenance is as important as environmentally-compatible engineering design. The practicability and 1.1 Geologically young fold mountains, especially those experiencing humid climates with seasonally intense rainfall, are among the most steep and unstable landscapes in the world. They are consequently the most difficult in which to construct and maintain roads. Common problems encountered include steep and irregular topography, and difficult excavation and founding conditions due to the deeply weathered and transported nature of many of the soils. Intense and prolonged rainfall lead to locally high groundwater tables. saturated soils and large quantities of surface runoff during the wet season. These problems are usually compounded by landsliding, erosion, river flooding. river incision and periodic seismicity. 1.2 Unfortunately, these conditions occur in many developing or newly industrialising countries that contain a relatively poor and remote mountainous interior. In these situations, engineering approaches frequently have to be applied on a low cost and low-technology basis. Often, relatively high population densities co-exist with an extremely fragile terrain. Cultivatable land is at a premium and these communities usually practice intensive terrace and irrigation farming systems that create additional problems for drainage management along road alignments. 1.3 The Himalayas constitute one of the most tectonically active fold mountain belts and experience some of the highest rates of erosion in the world. The difficulties presented by conditions of adverse geology, topography and land use are made more acute iii the humid sub-tropical and humid warm temperature zones, where rapid rock weathering and heavy rainfall act to induce landsliding and erosion. 1.4 Nepal typifies these geographical conditions. Nepal is a country where the road network has expanded rapidly since the 1950s, in connection with the large number of development projects that have been implemented there since that time. In support of ODA's own development programme in Nepal, the authors' organisations have gained considerable practical experience of road construction methodology and techniques in this fragile environment. The experience has provided the technical expertise upon which much of this guide is based. For this reason, the report is illustrated with examples mostly from Nepal. However, observation of the condition of slopes along most of Nepal's roads over many years has led to the

cost of road maintenance in many developing countries must be borne in mind when deciding upon road standards and allocating project funds. Again, geology, geomorphology and hydrology will exert significant controls on what can be designed, constructed and maintained within a low cost framework. 1.8 The design of roads under Himalayan conditions is a relatively little-explored and documented subject. Road construction has been carried out by many agencies, often under development aid, but many using local resources. Some of these roads have remained in place due to costly investments in construction and maintenance, while others have succeeded because they are models of sympathetic engineering design in terms of alignment, road width and cross-section to suit the terrain encountered. At the other end of the spectrum, some roads have suffered such instability that they have ruined their environment and have become impassable for long periods. Alignments are sometimes abandoned due to inadequate planning, or trigger such instability and erosion that they cannot be maintained. Road drainage systems are then allowed to fall into disrepair, thus exacerbating the problems. One of the principal lessons to be learnt from these road failures is the fact that consideration of the design of a road cannot be limited to the right of way alone, but must take adequate consideration of ground conditions throughout the landscape in which it is to be constructed.

1.9 It is not enough to simply apply practices evolved in the mountains of Europe, North America and the steep terrain of Hong Kong. In these industrialised regions, the difficulties are surmounted by intensive geotechnical analysis and the most modern and capital intensive construction materials and techniques. These approaches are generally inappropriate under fold mountain conditions in other regions where neither the money nor the expertise are available to build and maintain roads in this manner. In addition, the frequency with which destructive geomorphological processes occur and cause road damage, no matter what level of engineering investment is employed, demands a different philosophy to design. 1.10 Although users of this guide should be aware of the full range of potential problems and design considerations described, some aspects will be more pertinent than others, depending upon the particular circumstances at hand. For instance, resources required to be spent on geotechnical hazard assessment and mitigation will vary from one project to the next. Furthermore, while the guide has been prepared for practising road engineers, both in the private and public sector, some of it is aimed specifically at specialist support staff, such as geotechnical engineers and hydrologists. Consequently, the layout of the guide has been designed to allow selective reference. 1.11 This is the first edition of the new guide, compiled from the extensive experience of many workers but not yet benefitting from the opinions of the readership. TRL would welcome comments from readers on the usefulness of the guide and any suggestions they may have for improvement.

GEOGRAPHICAL SETTING OF THE NEPAL HIMALAYAS

GEOLOGY AND TOPOGRAPHY

2.1 The Himalayan ranges (Figure 2.1) extend in a 2400 kilometre-long arc between the Indus River in Pakistan to the west and the Brahmaputra River in Assam to the east, and are between 250 and 400 kilometres in width, being bordered by the Ganges Plain to the south and the Tibetan Plateau to the north. The Himalayas evolved through collision of the Indian and Eurasian continental plates, compressing a deep basin of sediments that lay between them and thrusting these sediments up into a series of great folds or nappes (recumbent folds). Although the main period of land rise is over, uplift continues at a mean rate of approximately 1 nun per year, but local rates vary widely, with rates of about 2mm per year in eastern Nepal and up to 5mm per year reported for the foothill ranges of Pakistan, for example. 2.2 The lithology and structural geology of the Nepal Himalaya (Figure 2.2) reflect this compression and thrusting through the formation of continuous northwest-south-east lineations. These lineations have controlled the distribution of uplift and elevation, differential weathering, drainage patterns and river downcutting to the extent that the major elements of the topography comprise a series of roughly northwestsoutheasterly trending ridges and valleys that extend the entire length of the mountain chain. The ridges have been dissected by the major southward flowing antecedent rivers which have cut through the emerging northwest-southeast relief lineations as the land mass has risen. Earthquakes and ground uplift continue to occur predominantly in association with seismic events at shallow depth along active thrust faults, and particularly the Main Boundary Thrust (Figure 2.2). 2.3 The geological and relief lineations described above have combined to create five main physiographical zones (Figure 2.3), separated by the principal thrust faults, within which climate, geology, slope morphology, soils, drainage and land use patterns are broadly similar. These zones are described in paragraphs 2.11 to 2.18, and comprise, from south to north: the Terai (Gangetic Plain); the Siwalik Hills; the Mahabharat Lekh (these two mountain ranges collectively constitute the Low Himalaya); the Middle Himalaya or Middle Hills; and the High Himalaya. The glacial snow and ice landscapes of the High Himalaya are excluded from thus guide as road construction is rare in these areas.

The rainfall regime of the Low and Middle Himalayas is controlled by the southeast Asian monsoon with as much as 80% of the 1,000-4,000mm (and occasionally in excess of 5,000mm) annual precipitation falling during the summer months of June through to September. However, even during the summer, lengthy dry periods are common, and are often followed by rainstorms during which daily totals frequently exceed 100mm and occasionally 300mm. In the east of Nepal and neighbouring Indian Himalaya, catastrophic downpours amounting to 500-1,000mm in 3-4 days are reported to occur on average every 20-25 years and are accompanied by widespread flooding and slope failure. These problems are recurrent throughout the Low and Middle Himalayas of Nepal and India. 2.5 Annual rainfall tends to decrease from east to west, owing to the orientation of the Himalayan ranges with respect to the southeasterly direction of the monsoon. The Low Himalaya (especially the Mahabharat Lekh) is the first major topographic barrier to this air flow and tends to receive the most intense rainfall (10-20 minute duration intensities of 100150mm/hour are not uncommon). Annual totals vary between 1,500 and 2,000mm. The rain shadow cast by the Mahabharat Lekh is often responsible for a much drier climate in the Middle Himalaya immediately to the north, with annual rainfall frequently less than 1,000mm. Nevertheless, the rain shadow effect tends to be localised, so rainfall amounts and intensities can still vary considerably within distances of only a few kilometres. Further north again, on the higher ridges of the Middle Himalaya, rainfall tends to increase to between 2,000 and 4,000mm per annum, although intensities are considered not to be as high as they are over the foothill ranges.

GEOMORPHOLOGY
2.6 Cycles of landscape uplift and drainage incision, followed by periods of relative stability have given rise to sequences of steep slopes, that were once adjacent to incising and eroding drainage lines, separated by valley side benches that represent earlier floodplain terrace levels and erosion surfaces. However, these sequences are frequently disrupted and often replaced by landforms dominated by geological structure, by major valley side instability and by deep tributary catchments that have undergone extensive erosion and slope failure. Figure 2.4 shows a mountain model based on these geomorphological concepts of landscape development. The model is described in Box 2.1. Generally, lower valley sides are by far the most unstable. For example, between 80% and 95% of landslides mapped from aerial photographs in parts of the Mahabharat Lekh occur on the lower valley sides where river erosion regularly gives rise to slope undercutting, and where groundwater tables approach the ground surface most frequently. Although steeper slopes at higher elevations in the landscape are generally more stable, they are often in an equilibrium state and therefore remain susceptible to disturbance by external processes.

CLIMATE 2.4 The climate of the region varies from humid subtropical over the Low Himalayan ranges and on the valley floors and adjacent slopes of the main Middle Himalayan rivers, through humid warm temperate in the remainder of the Middle Himalaya to alpine over the High Himalaya.

Figure 2.1 The Himalayas and their regional setting

Figure 2.2 Summary geology of Nepal (based on data published by Department of Mines and Geology, Kathmandu)

Figure 2.3 Geography of Nepal showing the principal Himalayan zones

Figure 2.4 A Model for young fold mountains (after Fookes et al, 1985)

2.7 The production of large volumes of debris from eroding gullies and slopes, and its transportation through the drainage system, can occur in response to rainstorms and floods of only moderate magnitude. Within a given area, these floods may occur every year in small catchments (1-5km2) and every 5-10 years in medium-sized catchments (10-20km2). More widespread flooding and erosion in large and multiple catchments may recur every 20 years or so. Events that result in extensive slope failure, erosion, flooding and significant drainage and geomorphological readjustments, may recur over a period of 50-100 years. 2.8 Rainfall intensities in excess of 50-100mm per hour sustained over a 30 minute period are reported to be the threshold above which slope failures and erosion take place, although this threshold will vary widely according to slope and drainage conditions. Even in such an active landscape as the Himalayas the intensity of these processes varies markedly according to geographical and physiographical zones described in paragraph 2.3. In the Low Himalaya, for instance, denudation rates of 2-5mm per year are considered not uncommon, and in small catchments may reach double figures, whereas in most of the Middle Himalaya rates of ground lowering by erosion etc may be considerably less than l nun per year. 2.9 Quite clearly, an early appreciation of these various physiographical and geomorphological controls on slope and drainage processes and rates of erosion is crucial to the sensible location of road alignments and road design in general. 2.10 The proportion of land covered by forest in Nepal varies according to slope steepness, accessibility (for felling), population density and, more recently, conservation measures. Slopes in excess of 35 are generally too steep for cultivation and are usually forested although, where close to main trails and populated areas, they have usually suffered depletion for fuel, fodder and building construction. The least forest cover is found in the less steep and intensely cultivated parts of the Middle Himalaya. Further north, in the higher altitudes of the Middle Himalaya, it is in excess of 60% due to the lower population densities, steeper slopes and poor soil conditions that are generally unsuitable for cultivation.

Box 2.1 The five-unit mountain model Geologically young fold mountain areas typically contain the following features: a prevalence of older, fairly stable land form features on the upper valley sides, steep lower valley sides affected by instability, and sediment-filled valley floors. The instability is kept active by periodic episodes of vigorous downcutting by the rivers. The concept of fold mountain landscapes containing this range of land forms in ordered assemblage has led to the development of a terrain model for these regions (Fookes and others, 1985). The model is illustrated in Figure 2.4 and its units are described below. The Himalayas follow this pattern. and its regions are given below as examples. Unit a. Glacial and periglacial topography (typical of the High Himalaya). The rock and ice terrain of the high peaks. Unit b. Free rock faces and associated debris slopes (typical of the High Himalaya and highest elevations of the Middle Himalaya). Steep rock ridges and cliffs, with screes and boulder fields at the base. The kinds of instability frequently observed include rockfalls, rock avalanches, rockslides and wedge and toppling failures. Unit c. Degraded middle slopes and ancient valley floors (typical of parts of the Mahabharat Lekh and the lower elevations of the Middle Himalaya). This is a denudational topography, in which prolonged weathering and less active erosion have produced gentler slopes bearing thicker soils. Slope stability problems are much reduced, although rilling and gully erosion are rife. The unit is densely populated and cultivated. Note that the unit is at moderately high altitude, `sandwiched' between the rock ridges and steep slopes of unit 2 and the deep valleys of unit -1. Unit d. Active lower slopes (typical of many parts of the Mahabharat Lekh and some of the more confined slopes adjacent to the major rivers in the Middle Himalaya). These are a continuous feature, occupying almost all lower valley slopes. They have been formed as a result of a period of renewed uplift (attended by renewed downcuttmg of all the streams). Long, steep, straight valley sides are covered by a thin mantle of colluvial soil (transported slowly by gravity) over weathered rock. Instability is normal in this terrain; shallow translational landslides and gully erosion are the commonest forms. Unit e. Valley floors (typical of the Low Himalaya and, to a lesser extent, the Middle Himalaya). The valley floors are almost always occupied by large quantities of debris from the slopes above in the form of fans, terraces and the bed of the liver. The chief engineering problems present in this part of the landscape are high flood levels, debris-laden flows,

HIMALAYAN ZONES OF NEPAL

Siwalik Hills 2.11 On the southern margin of the Himalayas, the Siwalik Hills rise abruptly to between 500m and 1,000m above the Gangetic Plain along the Himalayan Frontal Thrust. Dun valleys occur at intervals within the Siwalik chain and form enclosed alluvial basins surrounded by hills. The Siwalik rocks themselves are formed from debris derived from the erosion of the rising Himalayas to the north, and comprise soft sandstones, mudstones and siltstones.

shifting river courses and shifting accumulations of debris (see Chapter 13).

2.12 High levels of slope instability and erosion in the Siwalik Hills are attributable to steep slopes, intense rainfall, the softness of the rocks and the high rates of chemical weathering under the humid sub-tropical climate. The finegrained rocks and weathered materials are subjected to cyclical wetting and drying which causes cracking and encourages further ingress of water into surface materials. Saturation is often enhanced by poor drainage and a seasonal rise in groundwater tables. The more granular slope materials are easily eroded upon weathering and the removal of the vegetation cover. Furthermore. underlying bedrocks are highly fractured with frequently steep and unfavourable bedding orientations within fold structures. Mahabharat Lekh 2.13 The Mahabharat Lekh forms a steeper and more rugged ridge system to that of the Siwalik Hills and rises 1,800-2,000m above sea level. The height and steepness of this zone is due to the uplift associated with continued tectonic movements along the Main Boundary Thrust and rapid incision of the drainage network. Relative relief in the deeper valleys may be up to 1,500m, with average slope angles in excess of 35. The Mahabharat Lekh is composed of a sequence of limestones, shales, quartzites and phyllites (fine-grained sedimentary and low-grade metamorphic rocks). Significant changes can occur over very short distances in rock type and structure, weathering grade, strength and depth of overlying soil mantle. Although rocks exposed at the surface are often in a weathered and fractured state, the constant stripping of this soil mantle due to surface runoff, landslides and erosion on steeply-sloping ground frequently prevents deep soils from developing. Soils are usually up to lm in depth, although residual soils on rounded and stable spurs can be in excess of l0m deep. Locally, landslide and colluvial deposits can exceed 30m in depth. 2.14 Vast quantities of water and sediment are transported on a seasonal basis. Erosion is clearly evident throughout much of this zone, although problems of instability, sediment accumulation and scour can be extreme in lower valley side and riverside locations. Most erosion can be attributed to natural forces; the influence of man in building well-managed agricultural terrace systems on virtually every slope less than 35 may have little more than a minor effect, if any, on erosion rates. However, mismanaged land uses have had significant and immediate effects locally. For instance, small catchments that have undergone widespread deforestation, overgrazing and poor land management can erode at rates of between five and ten times those that would be expected under normal slope conditions. Increased rates of erosion, siltation, widening of river beds and a tendency towards more flashy and ephemeral river flow are direct consequences of poor land management.

Middle Himalaya 2.15 The Middle Himalaya rises to an average elevation of 1,000-2,000m above sea level, although elevations increase to between 3,000 and 4,500m asl on the ridge tops in proximity to the High Himalaya. The rocks of the Middle Himalaya are predominantly schists and gneisses. They are mantled by weathered colluvial deposits and residual soils on the more gentle spurs and sloping ground at lower elevations, but they frequently form comparatively fresh and strong outcrop on the steeper slopes and in the zones of higher elevation to the north. 2.16 Generally, the Middle Himalaya is more stable than the zones to the south, because the landscape is less steep and the intensity of rainfall less extreme. With the majority of presentday seismic activity occurring in the vicinity of the Main Boundary Thrust, the Middle Himalaya generally falls within a seismic shadow and does not display the high rates of uplift and down cutting that are apparent in the Mahabharat Lekh. This zone is intensely cultivated and irrigated, and thus water management in agricultural areas is a key issue with respect to erosion and slope stability. 2.17 Along the main rivers, steep valley sides are often subject to rockfalls and rockslides, especially where under-cut by river erosion. This situation becomes more pronounced in the deeper and steeper-sided valleys towards the north, where the topography comprises cliffs, steep slopes, ravines and deepseated slope failures. Many areas of flatter ground are likely to represent the remnants of ancient failure deposits. Slope erosion is far less a hazard here due to the general lack of erodible soil mantle, stronger underlying bedrock and a less intense rainfall and runoff regime. However, almost all main valleys emerging from the High Himalaya carry some threat of glacial lake outburst flood (GLOF), as demonstrated by a number of events during the last few decades. High Himalaya 2.18 The High Himalaya comprises the great peaks that rise to 8,000m above sea level and associated inter-montane basins. The rocks of the High Himalaya are mostly high grade metamorphic gneisses and intruded granites, and the climate is periglacial and glacial. These regions are characterised by snow fields, drifting snow and snow avalanches, glaciers, scree slopes, rock falls, rock slides and rock avalanches, meltwater and flooding. Although road engineering in this zone does not form the subject of this guide, some references contained in the Bibliography provide background information.

ENGINEERING PROBLEMS
Box 3.1 Definitions of soil and rock terms Soil and colluvium Soil is completely weathered material that has lost all trace of rock structure and now consists only of mineral particles. Because of rapid weathering and constant movement downhill under gravity, most sub-tropical mountain soils are lacking in plastic clay minerals and are structureless. They also tend to be very stony. Colluvium is the name given to soils that lie on mountain slopes. Colluvium typically has the above features. It is 0.5-2m deep and often contains rock debris of different types, mixed up as the soil creeps down slope. The base of the colluvial soil profile tends to pass abruptly into weathered rock. Weathered rock. Weathered rock, in practical terns, is any rock that has undergone chemical degradation by the action of water, aided by warm temperatures. The effect of weathering is to weaken the rock and open up micro and macro fractures, allowing more water to get in and hasten the weathering process. Rock may be only slightly weathered, in which case its strength and integrity will be little affected. But when rock is highly weathered it becomes very weak and highly permeable. In this condition it may behave more like a soil than a rock. For instance, when failure takes place, shearing may take place along a plane through the rock mass (like a soil) rather than along pre-existing planes of weakness, as in a hard rock. These hybrid materials, which may represent the bulk of materials in an area, do not lend themselves to analysis by standard methods. Slopes often contain hard rocks, moderately soft rocks and very soft rocks, all within the same sequence. A weathered rock can be described loosely as a soft rock, referring to its low strength. In this case the two terms have the same meaning, but 'soft rock' also refers to a specific type of material (below). Soft rock Soft rock can be defined as a rock whose fabric (body) is inherently weak. This weakness is due not to weathering but to inherently weak bonding between the mineral particles or assemblages of minerals. The rock can have a fully developed system of joints (internal parallel fracture sets). Geologically very recent rocks tend to have these characteristics; geological processes have not acted long enough to fully harden the rock. Hard rock Hard rock can be defined as any rock in a little-weathered state which, in failure, would shear along one or more of its natural planes of weakness. Standard techniques of rock slope analysis would apply.

GENERAL CONSIDERATIONS
3.1 The various facets of the physical environment described in Chapter 2 present a range of difficulties and hazards that require careful consideration at all stages of a road project, but particularly during feasibility and design. Some of the more important considerations are listed below: choice of practicable and stable alignment choice of cross-section in difficult ground conditions with due regard of cost and alignment specifications drainage control and prevention of flood damage control of sediment and prevention of erosion foundation stability for retaining walls and other structures in colluvial or weathered soils, especially in riverside locations vulnerable to scour construction materials seismic design for major structures environmental impact, land acquisition and compensation, and drainage management through irrigated land

3.2 These considerations are discussed in the relevant chapters of this guide. A summary of engineering problems found in the Himalayan zones and mountain units defined in Figures 2.3 and 2.4 is given in Table 3.1. A brief description of terrain hazards that pose regular problems along Himalayan roads is given in the next section. Definitions of soil and rock terms used throughout this guide are given in Box 3.1.

TERRAIN HAZARDS
Landslides 3.3 Table 3.2 summarises the common forms of landslides and mass movements found in the Himalayas and outlines their typical engineering significance. Slope failures may involve two or more mechanisms, occurring either at different places on the slope, at different depths or at different times due to changes in ground conditions once initial failure has occurred. Furthermore, the engineering significance of a slope failure will vary according to whether it is a first-time or reactivated failure. First-time failures have an immediate effect on roads in their path but may not represent a continual maintenance problem because the failed mass may come to rest at an angle significantly lower than that from which it failed, and remain stable unless disturbed by toe erosion or seismic shaking. Reactivated slope failures on the other hand, and those that are expanding upslope due to progressive failure, are often the most problematic for road construction and maintenance

10

11

Table 3.1 Design considerations by physiographic zone in Nepal

12

Table 3.2 Common forms of failure and erosion found in the Himalayas

Table 3.2 Common forms of failure an erosion found in the Himalayas (continued)

13

because they have a Factor of Safety equal or close to unity, and are highly susceptible to comparatively minor changes in slope conditions brought about, for instance, by some road construction and maintenance practices. However, rates of slope movement may be relatively low and tolerable over short lengths of alignment. 3.4 Shallow soil falls, rock falls and rockslides, up to a depth of a few metres, tend to occur with intermediate or high frequency. Soil falls occur frequently from undercut river terrace banks and steep cuttings in soil, while rock falls are common on steep slopes formed in fractured rock, and especially from cut slopes during heavy rain. Shallow rockslides usually occur as planar failures along adverselydipping bedding, foliation or joint surfaces, or as wedge failures along intersecting joint planes. However, in highly jointed rock masses typical of the Himalayas, potential failure surfaces are frequently encountered whatever the bedding/ foliation orientation with respect to the slope. 3.5 Debris slides are shallow planar failures in granular soils. Despite being the most frequent hazards encountered, debris slides and shallow slope erosion usually have little more than a nuisance effect on mountain roads. Debris slides usually occur in the soil or weathered mantle in response to toe erosion, rapid saturation of granular soils or a release of negative pore pressures during heavy rain in more fine-grained materials. Slip surfaces often occur along the interface between the weathered mantle or colluvium and the underlying rock. Debris slides are frequently found in the heads of eroding gullies and are subsequently quickly incorporated into an expanding drainage system. For instance, in the eastern Mahabharat Lekh, between 80 and 95% of debris slides have been found to occur adjacent to stream channels. Their life span is typically up to five years, although failure can continue for longer if toe support continues to be removed by erosion. 3.6 Mudslides are shallow planar failures in fine-grained, cohesive materials such as weathered mudstones and shales. They are often triggered by undrained loading. caused by failed material being deposited from above, or by toe erosion. Due to the predominance of granular non-cohesive soils on mountain slopes, mudslides tend to be infrequent. 3.7 The term rotational slide describes the mechanism of circular failure in fine-grained, cohesive soils and argillaceous rocks. Shallow rotational failures are almost exclusively confined to the more clayey colluvial soils and weathered mudstones. Deep-seated rotational slides are comparatively rare due to the preponderance of structurally-controlled planar or wedge failure, although they can form important components of complex failures in highly fractured rock masses when a single discontinuity or wedge structure is unable to control the entire failure mechanism. 3.8 Deep-seated planar rock slides usually take place along major bedrock discontinuities, or weak zones that are adversely orientated relative to the slope direction. They

usually defy any low cost attempts at stabilisation or control, but fortunately they are comparatively infrequent. Landforms resembling those created by deep-seated rock relaxation (sackung), reported from the European Alps and North America, have also been identified on steep (70"), high valley sides in the mountainous Middle Himalaya of east Nepal. Sackung is caused by plastic deformation of rock masses due to gravitational stresses. It leads to gradual downslope creep and the creation of parallel head scarps above the failing mass. Loss of support by river erosion or seismic shaking, or even road excavation, can then trigger rock avalanche-type failure of the dilated rock mass. Slope erosion 3.9 Slope erosion occurs almost exclusively where silty and sandy soils on slopes greater than 20 or so have no protective vegetation cover, or where sub-surface erosion pipes collapse to form gully heads. Most soils of the Low and Middle Himalayas fall into this category. Residual soils developed on phyllites, schists and gneisses, that have had their- clay minerals leached out in groundwater, are often prone to erosion in this way and are sometimes associated with an intricate pattern of rills and gullies that can erode to depths of 15 metres or more. Erosion usually occurs wherever surface drainage is concentrated onto unprotected slopes and is common below paths, roads and leaky irrigation canals. Slopes are at their most vulnerable to erosion during pre-monsoon and early monsoon rains when vegetation is least developed and storms are often most intense. Stream erosion 3.10 The morphology of drainage systems adjusts to the moderate or high magnitude floods to which they are regularly subjected. The ephemeral and seasonal flows of the intervening periods have comparatively little effect on the drainage regime. Thus, the erosional effects of large storms and associated slope failures can remain visible and comparatively fresh for perhaps many years, until modified by another high magnitude event. The frequency of these events appears to be highest in the Low Himalaya, and may represent a recurrence interval as low as 510 years. Scour depths of up to 5 metres during these storms are not uncommon in this zone. Box 3.2 Stream erosion in Nepal Measurements taken before and after a particularly heavy storm of perhaps 50 years recurrence interval in the eastern Mahabharat Lekh indicated that stream channel cross-sections had increased by between 200 and 400%. By contrast, measured rates of channel erosion in small tributary catchments of the Arun Valley in the Middle Himalaya amount to an increase in cross-sectional area of between 7 and 15% per year. These lower rates of channel growth are thought to be representative of `normal' conditions.

14

Sediment transport and flood plain scour 3.11 Sediment tends to move in pulses through the drainage system, in response to intense storm runoff or the burst of a landslide dam or Glacier Lake Outburst Flood (GLOF). As the water subsides, the sediment is rapidly and unevenly deposited. The distribution of sediment on the valley floor is constantly changing, thus modifying the flow pattern of the river itself. The engineering consequences are threefold: the river scours different parts of the valley from season to season, removing terraces and undercutting the side slopes, causing new landslides deposited sediment can bury structures or block outlets the hydraulics of stream flow around bridge abutments and piers can alter significantly.

outburst. Although spring snow melt can significantly increase river levels, the effects on river hydrology are comparatively minor further downstream. Rain-fed floods 3.15 These are by far the most common and are generated by (i) short, high intensity storms that have a comparatively limited geographical spread but have devastating flooding and erosion effects locally, and (ii) less intense but prolonged rainfall that frequently occurs over large areas of the Low Himalaya. In extreme cases, flood discharges of between 10 and 15 cumec/km2 of catchment can be expected. Landslide dams 3.16 On occasions, slope failures can be large enough to temporarily block river courses. This phenomenon is common in tributary valleys up to 501an2 in area, but less so in larger trunk valleys. A landslide dam is especially likely to occur at places where a constriction in the valley floor coincides with an unstable valley side. The problems for road alignment on valley floors or lower valley sides in this case are: the river level on the upstream side of the landslide dam rises rapidly, resulting in inundation, sedimentation, and disturbance of the hill slopes around the margins of the temporary lake a flood occurs, due to the sudden breach of the dam by overtopping or bursting through. The flood can turn into a debris flow super-charged with sediment and capable of transporting large boulders.

3.12 Where sediment yields are high, the abrupt decrease in slope angle at the junction of steep tributary streams and the adjacent valley floor results in the formation of debris fans. Fan surfaces may rise by 2m or more during the course of a single storm when floodwaters are insufficient to remove the debris being supplied. Fan incision during the following storm may remove most of the deposited material before further aggradation occurs. This cycle of deposition and erosion can occur on perhaps four of five occasions per year and give rise to problems of foundation scour and bridge or culvert blockage. Boulders up to 5m or more in diameter have been observed to move across fan surfaces during extreme flooding. Rapid fan growth from unstable tributaries has been known to temporarily block trunk rivers, impounding water and giving rise to a flood wave when the debris is breached in a manner similar to that of a landslide dam (paragraph 3.16). River flooding 3.13 River flooding can have the following effects on roads and associated structures built in vulnerable locations: overtopping and associated scour of gravel road surfaces creation of temporary head and conditions of instability behind permeable road retaining structures if the rate of drainage dissipation during flood recession is slower than the fall in river stage temporary submergence of drainage structures with possible damage to bridge soffits by drift material blockage of drainage structures by sediment, and scour of embankments, retaining walls and flood control /protection structures undercutting of lower valley side slopes to create conditions of slope instability beneath road alignments.

Glacier lake outburst floods (GLOFs) 3.17 A GLOF is a large surge of water caused by the breach of a dam of glacial debris impounding the waters of a glacier lake, high in the headwaters of a river valley. Valley glaciers in the Himalayas are located at altitudes of 3,000-5,000 metres above sea level and have been retreating since the turn of the century. Glacial moraine deposited in front of the retreating glaciers has, in some cases, caused lakes to form. It is estimated that there are at least fifty of these lakes in the upper Arun valley alone. Breaching of a moraine dam can occur when: heavy rainfall causes the lake level to overtop the moraine dam and scour the outfall a sudden rise of lake level occurs, due to an ice fall or landslide into it the moraine dam ruptures due to the melting of an ice lens in the core, or an earthquake.

3.14 Severe flooding can be either rainfall-generated or triggered by the failure of a landslide dam or glacier lake

3.18 GLOFs can cause great destruction and modification to river courses for many kilometres downstream. They occur without warning and with no easy means of prediction. Other effects associated with the passage of a GLOF include the transport, deposition and redistribution of large

15

volumes of sediment and the temporary damming of tributary rivers. The passage of a GLOF can also have a long-term influence on suspended sediment concentrations, a four-fold increase being reported in the Tamur River in east Nepal following a GLOF in 1980. 3.19 The potential impact of a GLOF can be devastating to any engineering structures in its path. The impact depends upon the volume of water, the rate at which it is released, the attenuation of its flood wave downstream, the preexisting base flow in the trunk river and the volume of sediment carried and picked up during its flow. Case histories suggest that between 5 million and 20 million m' of water can be released within a period of 1-2 hours, with equivalent volumes of debris. Box 3.3 Magnitude of some recent GLOFs in Nepal A flood surge in the River Arun created by a suspected GLOF in 1969 had a maximum stage of up to 10 metres above normal river level in the more confined reaches. It is estimated that as much as 5-10 million m3 of water and a similar volume of debris flowed out in 1-2 hours, leading to a peak discharge of 4,000 cumec: an amount of water approximately equivalent to the runoff generated by a storm with a 100 year return period. A GLOF in the Bhote Kosi and Sun Kosi rivers in 1981 destroyed approximately 251rn of the Lamosangu-Kodari road, much of which was within 10m of the normal river level Earthquakes 3.20 Earthquakes and ground tremors are common in Nepal, the majority of epicentres (the point or zone on the earth's surface directly above the origin of the earthquake in the crust) being located close to active thrusts or faults. Although it may not be practicable to design mountain road alignments to avoid these active zones, it is advisable to give consideration to the potential effect of seismic shaking when designing high retaining structures, and essential when designing bridges. 3.21 It is often the secondary effects of ground settlement, liquefaction and slope failure that cause the greatest problems for engineering stability. Earthquake-induced landslides, especially rockslides, rock falls and rock avalanches, often pose a greater hazard to life and engineering structures than earthquakes themselves, and can often occur several months after the seismic event, when the perceived risk is over. Rock falls and rockslides often occur when water penetrates comparatively strong rocks that have dilated during an earthquake. The quartzites, limestones and other competent rocks that crop out along the Mahabharat Lekh and the hard granite/ gneisses along the Main Central Thrust appear to be particularly prone to such failures. 3.22 The three most important variables that control the behaviour of saturated cohesionless soils during seismic shaking are:

relative density (low density materials reduce in volume during shaking, causing deformation in engineering structures) confining pressure (low confining pressure allows the material to dilate) drainage conditions (pore pressures can build up rapidly during dynamic loading, causing the soil to loose strength or even to liquefy).

Box 3.4 Seismicity in Nepal Seismicity contours indicate that the highest concentrations of earthquakes in Nepal occur in the foothills between Kathmandu and Dharan, especially along the Main Boundary Thrust. Prehistoric rock fall and rock avalanche deposits are frequently found at higher altitudes in the Middle Himalaya and many of these may have occurred as a result of earthquake shaking and dilation of surface rocks associated with seismicity along the Main Central Thrust. The impact of the August 1988 earthquake in east Nepal illustrates points in the text. The earthquake epicentre was 65km west of Dharan with a shallow focal depth, a strength of 6.6 on the Richter Scale and a reported intensity of VIII-LX in Dharan on the Modified Mercalli Scale. The last comparable earthquake in Nepal occurred in 1934 and destroyed large areas of Kathmandu and neighbouring towns (which are built upon finegrained lake sediments). Although in the Dharan-Dhankuta area there is evidence for deep-seated rock relaxation, especially in the vicinity of the Main Boundary Thrust, the majority of failures occurred in the quartzites of the Mahabharat Lekh and on scree slopes where severe shaking caused failure of road retaining walls. However, the event probably resulted in extensive dilation of surface rocks along the Mahabharat Lekh because heavy monsoon rains a few weeks later caused widespread erosion and shallow slope failure. The DharanDhankuta road was blocked in over 30 places, and a total of more than 2,OOOm'/km of slip material had to be cleared from the road surface. Further north, where the rocks are less fractured, the surface effects of the earthquake were minimal, although isolated rock falls occurred on the steeper rocky flanks of the Arun valley in the Middle Himalaya. Road construction impacts 3.23 Road construction can have significant effects on slope stability, drainage, erosion and sediment supply to drainage networks. Studies from the Himalayas in Nepal and India indicate that cut slope failures after construction can generate an average of 500mYkm/yr of debris, and that up to 2,000m'/km can be generated during single storms with 10-20 year recurrence intervals. Erosion rates in small catchments significantly affected by road construction can be at least 10 times those expected under natural conditions. Environmental impacts of road construction are discussed further in Chapter 6.

16

PLANNING AND DESIGN OVERVIEW

INTRODUCTION

4.1 This chapter describes general matters that relate to planning road construction in mountain areas. The sequence of activities that constitutes the progression of a road project from inception through to operation is summarised below and in Figure 4.1:
project identification

rather than to retain fixed ideas about designs from early on. The engineer should be prepared to modify the design in the light of information as it becomes available. The informationgathering stages are thus the most important part of the design process, because they steer the engineer towards an optimum design in circumstances in which design decisions are often very difficult to make. 4.4 The evolution of a final design is achieved by investigating the site in several phases, at each phase selecting smaller areas to look at in greater detail, using techniques appropriate to each. The phases are: desk study and reconnaissance surveys during the feasibility phase to summarise terrain conditions and to identify and compare route corridors, design concepts and costs, and review environmental considerations geotechnical mapping, based on geomorphological or landform mapping along route corridors to identify constraints on detailed alignments and to define broad geotechnical parameters for design identification of preferred route corridor hydrological investigations and hydraulic design that are carried out in progressive detail from regionalspecific at the feasibility stage to site and structurespecific during detailed design geotechnical investigations at selected locations along the chosen route to further define geotechnical parameters design of alignment, cross-section and associated drainage and retaining structures geotechnical investigation during road construction, carried out as cuttings and foundations are exposed, from which the final designs are produced.

feasibility studies, relating to both the basic engineering and an assessment of the route corridor (includes the desk study) design of engineering works, usually in two phases (preliminary and detailed) commitment of funds, which often takes place in a series of stages, followed by invitations to tender and negotiation with contractors, potential financiers and suppliers implementation of construction, typically either by direct labour or by one or a series of construction contracts operation, maintenance and monitoring.

4.2 A systematic approach to project assessment and design should be adopted on all roads irrespective of their intended function or standard. However, for reasons of technical complexity, sparseness of geo-technical data, physically difficult communications and physically constrained working, planning procedures cannot be followed so logically as when building in lowland areas. Although the phases of the Project Procedure comprise a logical sequence, in practice, modification may be required to the timing of phases and the extent to which the various studies within each phase are carried out. For instance, feasibility and preliminary design often overlap and combine into a single progressive phase. Similarly, final design is often not concluded until ground conditions are fully revealed during construction. These considerations are illustrated in Table 4.1, which summarises the factors to be taken into consideration during the various phases of the Project Procedure. A relative order of importance or priority is assigned to the inputs within each project phase. 4.3 The most important effect of a flexible approach to design is that final design of the cross-section and retaining structures may be deferred until the time when the ground is opened up in the early stages of road construction; up to that point all designs should be regarded as preliminary. The aim of the feasibility study and design should be to work towards a final alignment and thence to a design,

4.5 In carrying out these investigations, it is important to be aware of the following factors concerning road construction in active fold mountains: problems are likely to be encountered whichever route is chosen and the choice of alignment is very often a tradeoff between one constraint and another, or one hazard area and another, and thus it is important to carry out comprehensive feasibility studies to establish hazard and risk assessments for decision-making not all hazards and ground problems will be identifiable prior to construction and therefore remedial action and contingency measures should be allowed for

17

Figure 4.1 The Project Procedure

18

Table 4.1 Procedural checklist

19

road construction across unstable slopes can result in construction costs that are over twice those incurred on stable ground of similar slope angles and materials, and maintenance costs will also be higher earthworks and retaining walls together can represent up to 50% of the total construction cost, and both items increase disproportionately with increasing slope steepness and formation width the average cost per year of routine maintenance in real terms amounts to approximately 3-5% of construction cost it would appear from the limited data available that, every 10 years or so, the occurrence of an earthquake or an intense rainstorm is likely to necessitate substantial road reinstatement works that can cost between 5 and 10% of the original construction cost allowing for inflation.

feasibility studies geotechnical studies detailed survey detailed design construction construction supervision construction management.

FEASIBILITY STAGE
4.9 The feasibility stage combines a desk study with preliminary walkover and reconnaissance mapping surveys to review the cost/benefit, practicality, stability and environmental consequences of the proposed works, and leads to the identification of a preferred corridor and engineering design approach. This phase is intended to quickly identify all factors that are relevant to design (Table 4.1) and should be multidisciplinary. It is usually the most crucial stage of a project, when decisions are made that are fundamental to achieving an alignment that can be constructed and maintained with least cost, minimum instability and tolerable environmental impact. 4.10 The feasibility study should also include a wider range of issues such as: geometric standard and design life. comparative cost estimates traffic forecasts (these are outside the scope of this guide) socio-economic and environmental assessments road safety assessment (this is outside the scope of this guide). Route corridor assessment 4.11 Broadly, route corridor options can be divided into the following categories: ridge top compound alignments that cross several types of topography, with vertical climbing sections (hairpin stacks) valley floor.

PROJECT IDENTIFICATION
4.6 The government organisation responsible for road works should take an active and informed part in project decisions, especially in defining the scope of works to be undertaken. It will be necessary, for instance, to define: the fundamental objectives of the project factors governing the selection of a route corridor detailed design standards a project implementation strategy.

4.7 It is important to list the specific project objectives in order of perceived priority. These might vary from the provision of rapid access to the site of a planned mine to the construction of all-weather access to villages and towns as part of a broader regional development programme. A clear order of priorities determined at the outset can avoid conflicts of interest that could arise later, and help establish a design standard most appropriate to the intended function of the road and the resources available for its construction and maintenance. 4.8 In developing countries, when a project requires large resources or the provision of specialist expertise, the host government or funding agent usually elects to employ external consulting and contracting organisations. Activities which normally require the appointment of such companies in association with local firms typically include: contract strategy studies environmental impact studies photogrammetric mapping

a) Ridge top alignments 4.12 These are often the most stable and least costly, and they are also favoured on socio-economic and environmen-

20

tal grounds, as they usually follow established lines of communication and habitation. Usually, however, steep slopes and abrupt changes in ridge-top elevation dictate that alignments are frequently required to traverse side-long ground beneath ridge tops. When there is a choice as to which side of a ridge an alignment is to be located, it is recommended that the choice be made on the basis of the following, in order of preference: select the side on which bedding, foliation or the principal planes of rock weakness dip into the hill-side, so that they are not undercut by excavation. However, in such situations, slopes frequently tend to be steeper and irregular, thus posing practicable problems for road construction. On the other hand, under more gentle slope conditions, deep soils may keep the outcrop of rocks in cuttings to a minimum, reducing the importance of rock structure in influencing earthworks stability the leeward slope, sheltered from the prevailing direction of rainfall a sunny aspect, so that soils are drier and rocks are less weathered.

lack of suitable spoil disposal sites and access difficulties for plant can pose difficulties during construction instability and erosion can easily extend from one loop of the road to another, both upslope and down slope storm runoff tends to become concentrated, for which large-capacity drainage structures and erosion protection works are required. The risk (cost) associated with failure of any part of the drainage system is usually high.

4.16 If the topography allows, the problems associated with stacked hairpins can be reduced by creating offset stacks, in which the hairpins are not immediately above one another but are staggered across the slope. This will minimise drainage problems and limit the danger of instability to fewer hairpin loops. c) Valley floor alignments (see also Chapter 13) 4.17 The advantages of a valley floor alignment are: that relatively little climbing and descent are involved, thus making route alignment easier and shorter, with correspondingly lower vehicle operating costs a ready supply of construction materials is available (although its quality may not be ideal) control of spoil disposal and construction of pilot tracks can be less stringent.

4.13 In the Himalayas the last two preferences are often contradictory as the prevailing direction of rainfall is usually from the south, in which case the leeward slope is likely to be more advantageous. b) Valley side alignments 4.14 Climbing sections of mountain roads can be designed as gradual traverses of side-long ground at a ruling gradient, or as vertical climbing sections (hairpin stacks), or as a combination of the two. The hairpin stack has the following advantages: a greater flexibility in route corridor location can normally be achieved the crossing of difficult ground, and in particular steep and unstable lower valley sides, can be minimised or avoided altogether the use of hairpin stacks to connect lengths of relatively easy ground, such as river terrace or ridge tops, can in some cases lead to a more direct alignment with a saving in overall route length.

4.13 Against these advantages, routes in the confines of valleys generally have little opportunity, other than by expensive bridging from one side of the river to the other. to avoid problems of awkward topography and instability, and they may be susceptible to river flooding and scour, as well as debris fall from instability above. Despite the attractiveness of a more direct route and low gradients, the construction cost of a valley floor alignment may be significantly higher than a hill route alternative, even though the valley route may be substantially shorter, because of the high costs of bridging, cross-drainage, scour protection and rock excavation in confined areas. If flooding and slope instability on the lower valley sides are recurring hazards, maintenance costs can also be significantly higher. On socio-economic grounds, valley routes may be less favoured if the majority of villages are located on ridge tops. Design standard and design life 4.19 In mountain areas, selection of geometric standard for low cost roads should take full account of the constraints imposed by the difficulty and stability of the terrain and, if necessary, reduce locally the design standard in order to cope with exceptionally difficult terrain conditions. For

4.15 The main problems associated with the hairpin stack approach are: on slopes steeper than 30 limited space to construct cut and fill slopes, necessitates either a relaxation in geometric standards or more expensive retaining structures

21

instance, to avoid or minimise the crossing of difficult ground, maximum average ruling gradients of 5% for trunk roads and 7%n for feeder roads could be increased to up to 7% and 9%n respectively, with a maximum gradient of up to 9% and 12% over distances of 300m. A formation width of 6.5 metres including the side drain could be reduced to 5.5 metres in particularly steep and unstable areas. On steep ground, a horizontal curve radius of as little as 10m may be unavoidable, even though a minimum of 25m is usually specified. As outlined in Box 4.1, the construction of a high design road will usually result in high repair and maintenance costs later. 4.20 Usually, some elements of design are considered to have an infinite design life, eg the embankment, while others have a finite design life, eg the pavement. Those elements that have a finite design life gradually deteriorate at a known rate under a predicted amount of wear and tear. These finite-life elements have different design lives according to their capital cost and the nature of the stresses imposed on them. Bituminous pavements are designed to last for 15-20 years, and bridges for 50-120 years. 4.21 In mountain areas, elements which are normally thought of as having an infinite life may run a finite risk of being damaged or destroyed by external forces of erosion and slope failure within, say, 25 years. Thus, while drainage structures are designed according to an acceptable design life, retaining structures and earthworks slopes are designed to an acceptable factor of safety based on a comparison of anticipated construction cost, and the cost and risk implications of failure or recurrent maintenance.

Box 4.1

New roads in Nepal damaged by natural disasters

In the Middle Himalaya of Nepal, two years after a mountain feeder road had been completed, a major storm caused the loss of 600 metres of road. The scale of the destruction could not have been foreseen during construction, and even if it had, the cost of the extensive protection works that would have been required would not have been justifiable within the prevailing budget. The total cost of reinstatement works amounted to the equivalent construction cost of ten kilometres of road (approximately 10% of the alignment). On a comparatively high standard road in the Mahabharat Lekh, earthquake and storm damage in one year resulted in a reinstatement cost that, allowing for inflation, again amounted to approximately 10% of the original road construction cost. Case histories from the Indian Himalaya tell of similar disasters well into the maintenance period. For instance, two storms to Sikkim caused the loss of 18 kilometres of road and ten bridges. Expenditure on disaster repairs may recur every 5-10 years, and is superimposed onto a `normal' maintenance cost that in any case may be 35%n per annum of the total construction cost after inflation. The limited evidence available suggests that costs of maintenance in unstable mountainous regions are, in the long term, proportionate to the costs of construction. In a marginally stable environment, high design standard does not necessarily insure against damage. Some form of damage to the road to future is almost inevitable. Thus, a higher design standard results in higher repair and maintenance costs later. Add to this the fact that a high design standard inevitably disturbs the physical environment to a greater extent than a lower standard. Such disturbance means that adoption of a high design standard can actually increase the risk of failure, and the cost of reinstatement will be the greater also. Feasibility study report 4.23 The feasibility study report will usually comprise the following: project identification, objectives and scope of study an account of broader development objectives identification of corridor and/or engineering options, with justification of preferred options and conceptual designs road safety issues related to design review of environmental issues including potential benefits and disbenefits

Erosion protection 4.22 Under conditions of limited funds, faulty construction practices or inadequate vigilance, any delay in implementing essential protection or precautionary measures can lead to serious problems later, requiring more expensive remedies. Furthermore, the effectiveness of maintenance organisations cannot always be guaranteed, thus emphasising the importance of adequate investment during construction. However, it is usually not cost-beneficial or even technically feasible to construct a road in an unstable mountain area that will be entirely free of instability problems. It is inevitable that damage or loss will occur to parts of the road, and perhaps within only a relatively short period after construction. The only sensible option is to decide upon a level of slope protection that reduces normal rates of damage to an acceptable level. The objective should be to design a road that can be built and maintained at affordable cost, within the technical and operational capabilities of the country, that functions at an acceptable level of safety to the public. If it is not economically possible to maintain long term road stability. then the decision to proceed at all with road construction should be seriously questioned.

22

review of existing traffic volumes (vehicular, population and livestock) import/export and origin-destination surveys, and traffic forecasts reductions in vehicle operating costs in the case of road improvement projects socio-economic impacts economic analysis tabulated comparisons of cost, stability and environmental aspects of corridor(s) cost and logistical summary of preferred corridor, review of funding requirements, project management requirements, construction planning and maintenance programme method of implementation.

Detailed survey and alignment design 4.26 With the route corridor confirmed, the alignment engineer, preferably accompanied by an engineering geologist or similar specialist, with a survey team, will flag the approximate centre-line. Considerable savings in time can be achieved if an approximate alignment is drawn onto photogrammetrically plotted contour maps (see below) and enlarged prints of aerial photographs in the office prior to embarking on detailed fieldwork. 4.27 If slope stability is critical to the alignment, then geotechmcal mapping surveys should be undertaken at scales of between 1:1,000 and 1:5,000. It will be easier for personnel to locate themselves with the required accuracy if an approximate centre-line has been set out, but the engineer should be prepared to modify the location of the centre-line in the light of the geotechnical survey. In very difficult ground, these surveys should ideally be carried out prior to the centre-line flagging exercise using aerial photograph enlargements or compass traverse as a means of location positioning. Box 4.2 Photogrammetry

DESIGN
4.24 The design stage is often divided into preliminary design and detailed design. Preliminary design is employed in order to avoid too much effort going into detailed design before a project is given the final go-ahead. A preliminary design would be appropriate if, for instance, the funding agency requests a more thorough review, than is usually achieved by the feasibility study alone, of the construction timetable, implementation programme and cost profile before committing financial resources to the project. The preliminary design can be carried out either as a separate exercise or as a logical extension to the feasibility phase. The additional requirements of a preliminary design are to: further investigate the engineering and cost implications of terrain hazards and difficult ground conditions carry out ground survey and/or photogrammetric mapping from aerial photographs to provide a more accurate ground model for estimation of quantities carry out detailed survey and, where possible, geotechnical or geophysical investigations at major bridge sites. 4.25 The above activities should only be undertaken within the preferred corridor and local variations to it, The various stages of the design phase are discussed below, although it must be reiterated that design rarely ends when construction commences, but rather evolves with experience as construction proceeds. The processes of detailed survey, alignment design and setting out are time consuming, especially if changes to the alignment are made later owing to unforeseen ground conditions or changing design criteria. The use of photogrammetry can speed up these procedures and provide the flexibility to allow additional off-site engineering works such as access to borrow pits, spoil disposal sites and slope drainage works, to be designed at a later date. As an example photogrammetry from aerial photographs of 1:25,000 scale can yield uncontrolled contour mapping at a maximum scale of 1:5,000, with contours at 5 metre intervals. It is advisable to correct the contour model by establishing two ground control points in each stereo pair, by tying points on the photographs either to the national or local grid, or by GPS. The main problems associated with the use of photogrammetry relate to the lack of ground definition in areas of shade, cloud or dense forest cover. 4.28 With the alignment confirmed, detailed design of all subsequent works can proceed. Design of the detailed vertical and horizontal alignments will require topographical mapping at a scale of 1:1,000 with contour intervals at a maximum of 2m, using ground survey, photogrammetry or a combination of the two. Ground survey may be preferable at this stage due to the greater survey accuracy required. The use of photogrammetry will require the establishment of a base line traverse and the commissioning of air photography at a scale of between 1:5,000 and 1:10,000. Plan and profile drawings, and schedules of earthwork and retaining wall designs and quantities can then be produced for contract documentation.

23

4.2 One of the most important aspects during detailed design is to ensure that the right-of-way is defined adequately, so that the road can be constructed along the optimum alignment. Further detailed topographic survey can then be carried out after site clearance but before construction in order to optimise road geometry and minimise quantities. A design capability should therefore be provided on site during construction, in order to attend to the needs of the detailed engineering and geotechnical assessment.

Box 4.3 Road construction cost comparisons front Nepal At 1996 prices, the construction cost per km for 5.5m formation width sealed roads in Nepal varies between US$0.1-0.3 million. If construction is to be carried out by an international contractor using machine-intensive methods, these construction costs can be more than doubled. In this case, the contractor's establishment cost can be 30% of the total billed items. There is a geographical pattern to the range of construction costs given above: construction costs in the Low Himalaya can be almost double those for the same design specification in the Middle Himalaya, due to generally steeper ground and significantly greater drainage and instability problems encountered depending upon topography, middle valley side alignments can be between 30% and almost 100% more costly to construct than ridge top alignments. Lower valley side and valley floor alignments can be up to 300% more expensive than those located on gently undulating ridge tops from the limited data available, construction to a 6.5m formation width can result in up to 70% increases in per-km costs over a 5.5m formation to all geographical cases.

COST ESTIMATION
4.30 Cost comparisons should be based on the expected costs of design, erosion protection, construction, maintenance and environmental mitigation. At the preliminary design stage, construction costs are usually estimated by sub-dividing the alignment corridor into segments according to slope angle, preferred cross-section and estimated quantities of soil and hard rock in excavations. The placement of fill on slopes steeper than about 30 requires the construction of an embankment retaining wall, the size and cost of which is estimated per centre-line metre for each combination of cross-section and slope angle. At the detailed design stage, the site-specific selection of cross-section, slope design and drainage and retaining structures is combined with the ground model and the designed alignment to yield quantities for cost estimation. 4.31 Table 4.2 gives approximate quantity estimates for the various combinations of design cross-section and slope angle. These data were derived from simple trigonometrical calculations based on the various assumptions given in the table, and are included here for approximate comparisons only. Clearly the quantities are also a function of formation width and for a road design in Nepal, reducing the formation width from 5.5 to 4.5m, led to a reduction of 20% in pavement and retaining wall quantities, and 15% in cut and fill volumes. Noncontinuous quantities such as culverts and slope protection works can be estimated by reconnaissance mapping or on a prorata basis from other roads constructed under similar slope conditions. In many projects, it is often found that cost estimates increase by 10% or more between successive phases of the estimating process and there have been occasions when the estimate at the feasibility stage has been less than 50% of the detailed design cost. These underestimates are particularly prevalent in mountainous areas and usually stem from oversimplification of topography, changes to alignment during design and inadequate geotechnical assessment. 4.32 A broad indication of road construction costs in Nepal, and their geographical variation is given in Box 4.3.

CONTRACTS
Contract conditions and specifications 4.33 Any provisions peculiar to the site will be allowed for in the special conditions of contract and the special specification. In Nepal, and in mountainous areas generally, the following topics are likely to require attention: the need for the contractor to regularly clear silt washed down from newly-completed construction operations, from drains and drainage structures, as part of his maintenance operations

the possible need for the contractor to work alongside

traffic

limitations on the use of pilot tracks and the requirement to protect adjacent slopes from erosion control on the dumping of spoil material erosion control and other environmental protection measures associated with land clearance, earthworks, drainage works, borrow areas, temporary stockpiles, the construction of work camps and the use of forest products

24

Table 4.2 Earthworks and retaining wall quantities according to cross-section type and slope angle

25

the establishment of nurseries to provide plants for slope revegetation requirements to provide accommodation and provisioning for labour requirements for programming of site surveys following the provision of design information.

the reliability of the contractor's own estimate given the extreme nature of the terrain the remoteness of the site, particularly if the supplying sea port is in another country, with the consequent length of time required to import plant, spare parts and materials difficult access to and within the site area extremes of weather uncertainties over the supply of locally produced commodities such as fuel and cement geotechnical uncertainties over the extent of unstable ground and foundation conditions at the sites of major structures uncertainties over the availability of construction materials and haulage requirements severe time and operational constraints imposed under the contract the effectiveness of escalation clauses the scale of liquidation damages.

Contract administration 4.34 At an early stage of the project it is necessary to decide how to proceed with constructing the works. The main issues are: direct labour or contracted work single large contract multiple small contracts

contract risk and claims labour intensive or mechanical methods of construction the extent to which allowance should be made for the development of design during the construction stage scheduling of construction contracts in relation to the wet season.

a) Direct labour versus contracted work 4.35 In the case of direct labour, the engineer, whether it be the government agency concerned or an appointed consultant, generally has greater control over methods of construction and more flexibility to make design changes. Modifications to design can be made during construction without incurring contractual claims. However, the larger the project is, the less likely it is that the client will be willing to undertake construction using his own resources, giving preference to engaging the services of contractors. b) Contract risk and claims 4.36 Having decided to employ contractors, the most appropriate contract strategy should be identified. This will determine, for instance, whether the works will be let to a single contractor or to several contractors in a series of smaller contracts. Different forms of contract apportion the risks involved in a construction project differently between the contractor, the employer and, to some extent, the insurers. The amount of risk that a bidder is required to carry can have a considerable effect on his bid price. Given that the risks attached to engineering works in unstable mountain areas tend to be high, any action that can be taken by the employer or engineer to reduce the risks is likely to be reflected in a reduction in price. Problems that are likely to present significant risk for a contractor include:

4.37 Some of these risks can be reduced or transferred from the contractor to the employer by the adoption of an appropriate form of contract. A representation of the allocation of total project risks apportioned to the employer and the contractor by using different forms of contract is given in Table 4.3. 4.38 In a price-based contract the contractor is required to carry the risks, and he will include a contingency in his bid price which the employer will pay whether or not the risk event transpires. The price may not bear much relation to the actual cost of construction. In a cost-based contract the

Table 4.3

Relative degrees of risk associated with different forms of contract

26

employer pays the actual cost of constructing the works. While the employer may initially be alarmed by the prospect of carrying the risks, the clear advantage to him is that he only pays for risk events that actually occur. Cost-based forms of contract can cause difficulties for funding agencies, particularly in respect of the procurement of goods. c) Labour-intensive methods versus mechanical methods of construction 4.39 In theory, it is possible to carry out most of the activities involved in road-building by manual methods, and in a labourabundant and foreign capital-scarce economy, it makes economic and social sense to do so. However, mass haul of spoil and construction materials, compaction and road surfacing are usually best undertaken by mechanical means. 4.40 It is important to anticipate the extent to which labourintensive construction will be necessary on a project. Such operations may require several thousands of labourers, both skilled and unskilled and, particularly in remote areas, it is likely that labour will have to be imported, possibly placing severe strains on local resources. Clear provision should be made under the contract for the supply of accommodation, fuel, water and sanitation facilities as well as the importation of goods, particularly foodstuffs. 4.41 By combining mechanical and labour-intensive methods, certain advantages can be gained. Operations such as the construction of retaining walls and cross-culverting can be undertaken using materials that can be carried in by porter, and can therefore be started in advance of the main earthworks. Operations such as earthworks and pavement works are better carried out by machine, working at an advancing road head. Designs suitable for porterage include the use of bolted structural steelwork for bridge substructures and superstructures, and corrugated metal piping for culverts. d) Development of design during construction 4.42 It is usual, particularly in the case of conventional admeasure contracts, for the design to be well developed before contracts are let. However, when working in difficult ground conditions there may be good reason to develop the design as far as an alignment and standard details, and then complete the detailing of the final design during construction. Advantages include: the designer is more free to modify designs due to unforeseen ground conditions, as often arises the design can be completed in a better knowledge of the relative costs and performance of alternative forms of construction

e) Scheduling to take account of the wet season 4.43 The significance of the timing of letting construction contracts in relation to the wet season is often underestimated. For instance, the effective working season in the Low and Middle Himalayas is between eight and nine months long and it is important that the time available to the contractor should contain as many complete dry seasons as possible. Ideally, the letting of a construction contract should allow sufficient time for full mobilisation before the commencement of the dry season. The mobilisation period will vary according to the size of the contract, the location of the site and the home base location of the contractor, but for a major international contractor it could be as long as six months. At higher altitudes, the occurrence of frost and snow will also reduce rates of construction.

CONSTRUCTION ASPECTS
4.44 There are a number of construction issues which should be addressed during the design phase, because they affect issues of construction management and environment. They are: rock blasting methods construction materials the choice between bituminous surface treatment or gravel running surface rate of construction and construction sequence temporary access for construction (pilot tracks) quality control and supervision disposal of spoil.

Rock blasting 4.45 Uncontrolled rock blasting is atypical feature of road construction in Nepal and other developing countries where labour-intensive methods of excavation rely on a totally fragmented rock mass for easy removal. However, bulk blasting, as it is termed, usually results in significant overbreak and the creation of a highly fractured rock mass. The angle of the dressed slope is invariably lower than the optimum, and the slope usually weathers and ravels back to an even shallower angle within a short period of time. Pre-split (pre-shear) and cushion blasting remove smaller quantities of rock along predetermined planes of weakness so that overbreak is minimised and the remaining rock mass is left intact, and hence is more stable. However, these methods require a greater design expertise and a large number of closely-spaced drill holes to be accurately orientated.

the designer can draw maximum benefit from the contractor's strengths.

27

Construction materials and water 4.46 The steepness of mountain slopes tends to limit opportunities for the accumulation of significant deposits of naturally occurring aggregates and gravels. River beds and river terraces offer a ready source of aggregate, although materials are often inconsistent and haulage uphill can represent a significant project cost. 4.47 For pavement design in areas where suitable materials are scarce, the following options should be considered: open a central quarry of good quality aggregate and budget for increased haulage costs blend materials to achieve a material that is within specification use materials that are outside the normal specification and plan to replace them more frequently if no source of surfacing aggregate is economically available, surface the road with gravel and maintain frequently. 4.48 Mica is present in a large proportion of the higher grade schists and gneisses of the Himalayas and to the natural aggregates derived from them. The presence of mica in fine aggregates can adversely affect the workability of concrete, and the addition of water to increase workability will only lead to a reduction in strength. The quantity of mica is controlled by careful selection of the source material. However, it is rarely possible to find naturally occurring fine aggregate completely free of mica and it is better, therefore, to avoid specifying the use of high-strength concretes. Where high strength concrete is a requirement, for example to bridge works and for culvert pipes, then fine aggregate can be manufactured by crushing rock. 4.49 The availability of water for construction purposes can become a major consideration if haulage uphill is required in the dry season to satisfy compaction requirements.

4.51 There are other disadvantages to the use of gravel surfacing in mountains: the use of gravel surfacing, with its poor resistance to skidding, in conjunction with tortuous alignments is potentially dangerous. Also, vehicles cannot climb steep inclines on a loose surface gravel is easily dislodged by tyre scuffing, and washed away by rain. Rapid gravel loss leads to high maintenance costs, and attrition of the poorer gravels will generate large quantities of dust, posing an environmental and safety hazard. 4.52 One of the main advantages with gravel road construction relates to the fact that embankment fill can settle significantly in the early years after construction. Such movements can easily be accommodated by a gravel surface which can be replaced later by a more durable pavement.

Rate of construction and construction sequence 4.53 Actual rates of construction vary widely according to the relative proportion of labour and machine-intensive activities, volumes of earthworks, frequency and size of culverts and retaining walls, and the ability to construct pilot tracks (see below). Adopting a labour-based approach will provide easier access to remote workfronts and therefore the rate of construction will depend upon the size of the labour force that can be managed and sustained, and on the maximum rate at which construction can be supervised by the engineer, although factors such as availability of materials, spoil disposal and the contractor's working method are also relevant. Dry season labour-intensive construction rates might be of the order of 1km/month. 4.54 In the case of predominantly machine-intensive construction, the rate of progress of advancing workfronts might be between 2 and 3km/month, depending on formation width, access and contractor's working method. 0.5km/ month is probably the maximum progress likely to be achieved in the wet season. Attention to drainage and erosion control is crucial to any activities carried out in the wet season and it is usually advisable to limit most construction activities to the dry seasons and ensure that the works are adequately detailed and protected from erosion during the intervening wet season. 4.55 The recommended sequence of cross-section construction is shown in Figure 4.2.

Gravel or sealed surface 4.50 The choice between a gravel and a sealed road surface is not simply one of traffic level. The general paucity of aggregates applies especially to good quality wearing-course gravels to such an extent that this may become an important factor in the choice between providing a road with a gravel or a bituminous surface. Good wearing course gravels may be difficult to obtain due to poor rock, deep soils, the difficulty of quarrying on steep slopes and long haulage distances from river beds. Wearing-course gravels require not only a good grading but also a limited plasticity, and it is often this last element that is difficult to fulfil. Without some plasticity, gravel wearingcourses erode more quickly, perhaps at a rate of up to 50mm per year, requiring replacement more frequently, thus aggravating the gravel supply problem still further.

Pilot tracks 4.56 The speed with which a road can be constructed or upgraded depends to a large extent on the number of work fronts that can be established. The provision of temporary construction access tracks can enable plant and materials to be supplied to remote work fronts and allow a faster rate of

28

Figure 4.2 Typical sequence of cross-section construction

29

progress. However, the steepness of the terrain and its vulnerability to erosion often make it inadvisable or impossible to force through a motorable construction access track. Exceptions where advance access can be put in safely are: along the approximate centre-line of the alignment in advance of the main work front, provided that drainage and erosion protection are incorporated as soon as the track is opened up, and completed prior to the onset of the wet season. If this is not achieved, then very severe erosion can occur, proving both difficult and expensive to rectify on rock slopes where the potential for instability and erosion is judged to be low along valley floors. The location and permanency of valley floor pilot tracks during the construction period will depend upon flood levels during the wet season. Access tracks along valley floors require either temporary or permanent bridging of tributaries. The location of access tracks in such situations should be designed to minimise disturbance to cultivated land and neighbouring communities.

ment to geometric standard are limited, as it is impracticable to make minor modifications to the geometry without having to undertake significant earthworks and retaining wall construction, involving high costs and the risk of reactivating slope instability and erosion. 4.60 Disregarding patching and re-sealing operations as being strictly maintenance works, road improvement and upgrading may consist of one or more of the following: drainage rehabilitation pavement overlay pavement reconstruction road widening and minor improvements to road geometry complete or partial reconstruction.

4.57 Where rapid construction progress is required, aerial support by helicopter or aircraft may be an alternative though expensive means of establishing remote work fronts and construction camps. The more remote these sites are, the longer will be the period of aerial support required to supply and maintain them. Quality control and supervision 4.58 Although the general and special specifications may contain sufficient stipulations and safeguards to ensure, on paper at least, that construction will be carried out to required standards with minimum and controlled environmental impact, it is the task of the engineer to supervise and oversee that these standards are applied in practice. In the case of multiple work front construction especially, failure to adequately supervise all aspects of work can have severe consequences when the road comes under stress from rainstorms or floods. This is particularly true of `hidden' work such as the preparation of foundations, the compaction of soil and backfill, and the filling of gabion boxes and construction of masonry walls behind the front face, which if not carried out to specification, cart lead to failure of the whole structure. Improvement and upgrading 4.59 In gentle terrain, there is often scope for upgrading the standard of a road by stage construction, if predicted growth in its use is uncertain, or if construction funding is limited. When the time for reconstruction is reached, its geometry and design can be reappraised. In mountains the possibilities for upgrading through widening or improve-

4.61 While access to the site will not be a problem on a road upgrading scheme, the fact that there is an existing road will mean that it will be necessary to accommodate the passage of public traffic throughout the construction period. In more gentle terrain this is normally achieved by providing diversion tracks, but in mountainous areas this is often impracticable, so that special provisions may be necessary to allow traffic to pass, which in turn places severe restrictions on the way some operations can be undertaken. For example, excavations for the installation of new culverts across the main carriageway cannot be undertaken as a single operation but will have to be performed in halves in order to keep the road open. Similarly, pavement layer construction cannot be undertaken as a fullwidth operation. 4.62 In developing countries, roads attract human settlement and in mountainous areas, where level ground is at a premium, homes are often built close to the roadside. This process is difficult to control and it is frequently difficult to gain access to sufficient width of land to carry out road widening. Also, land acquisition can be legally complicated and land prices may be inflated. 4.63 Geotechnical studies are made easier in existing road cuttings due to improved access for drilling rigs and the fact that soil horizons and geological strata are exposed. The assessment of foundation conditions for retaining walls, particularly those below the existing road, may be made significantly more difficult by the need to distinguish natural materials from construction spoil, a task that can be more difficult than it sounds. 4.64 It is often the case that earlier road construction will have exploited the most easily accessible sources of gravel and stone so that, unless the supplies are regularly replenished river deposits, good construction materials may be either difficult to locate or difficult to transport to site.

30

4.65 In steep ground there are often significant cost advantages in restricting road widening operations to one side only of the road cross-section. At any given point it is often better to encroach either wholly on the cut-side or wholly on the fill-side, particularly if retaining walls are involved. This will limit the extent of ground disturbance to one side of the road and it will make traffic control easier. However, this can make the achievement of a mass balance more difficult so that if borrow and spoil are problematic, then switching between cut and fill at regular intervals along the road may be necessary. If borrow is available, and the disposal of spoil is not a problem, the decision to achieve road widening predominantly by cut or by fill will depend upon slope conditions, the potential for cut slope instability and the stability of slopes below the road. It should be noted, however, that cut slopes often take a considerable time to weather back to a quasi-stable angle and further excavation is likely to trigger a new cycle of failure, erosion and readjustment. Although some cutting is usually unavoidable, a tendency towards fill rather than cut is preferable if foundation stability below the road permits. Maintenance practice 4.66 Although this note does not deal specifically with road maintenance, it is important in any road scheme to plan for sensible and timely maintenance in order to avoid creating unstable situations through either neglect or ill-advised maintenance practice. Some of the more common maintenance practices that create or aggravate instability problems are listed below: uncontrolled tipping of spoil materials arising from cut slope failures. This causes erosion and siltation below the road diversion of road drainage water around slip material on the carriageway causes uncontrolled runoff over the road edge

failure to repair eroded side drain inverts leads to deepening of the side drain and erosion of adjacent materials failure to clear drains of debris leads to overtopping of the drain and erosion of adjacent slopes quarrying into cut slopes for road maintenance materials, and inadequate quarry restoration, can create slope instability and erosion problems often, where the slope below is eroding and undermining the road, a hillside is progressively cut back in order to maintain passage. This de-stabilises the hillside. (It is usually always advisable to address the problem below the road as quickly as possible in order to prevent an unmanageable slope situation developing, and to avoid further cutting of the hillside above) if, through lack of funds or insufficient time in advance of a wet season, it is not possible to implement pavement protection or stabilisation works, then a drainage or slope problem can usually be kept under control by constructing temporary works proper design of walls and drainage works and good construction supervision are just as important during maintenance as during the initial construction period removal of construction materials (especially theft of stone from gabion boxes) and disturbance to road drainage systems in village areas and irrigated farmland by diversion of the water are problems that require rectification and monitoring.

31

DESK STUDY AND RECONNAISSANCE SURVEY

remotely sensed imagery - aerial photography and satellite imagery climatological records (Chapter 8).

INTRODUCTION
5.1 Desk study and reconnaissance survey are fundamental to the success of any feasibility study. The availability of desk study information, together with its scale and level of detail, will determine the extent to which desk studies can advance feasibility studies prior to embarking on detailed fieldwork. The importance of good preparation in a thorough desk study cannot be over-emphasised, although its scope will depend on the availability and quality of maps, aerial photographs and other published materials. Where sufficient data exist, the desk study can be a time and labour-saving process that allows the identification and analysis of options for route corridors and design concepts, and serves to focus attention on the most relevant factors to be considered during site reconnaissance. The more comprehensive the desk study is, the fewer problems may be expected later throughout the duration of the project.

MAPS
Topographical maps 5.5 Topographical maps are usually obtainable from the local survey department, or sometimes from specialist stockists in Europe. Their accuracy is sometimes questionable, and the scale may be too small to allow other than an orientation in the field. Contour heights may not be reliable, and maps may be out of date in respect of settlements and features that are prone to change, such as river courses. Their main advantage is in their consistent scale and clarity of representation, which enables areas, lengths and directions to be measured for purposes of feasibility study with a reasonable degree of accuracy. Geological maps

DATA SOURCES
5.2 The desk study comprises a review of published and unpublished information concerning the physical, economic and environmental characteristics of a study area. Although office-based, it is usual to combine the desk study with reconnaissance survey in order to reduce the uncertainties arising out of interpretations of often rather scanty published data. 5.3 Because of the difficulty of access and the frequent historical lack of economic interest in tropical mountains, background information about them tends to be sketchy and often inaccurate. At best, it may be limited to small scale topographical and geological mapping, aerial photography and satellite imagery. At worst, none of these data sources may be available. Some countries, in which tropical mountain ranges occur, do not keep stocks of maps and publications about these remote areas, and it may be easier to obtain information from sources in Europe and North America than in the countries themselves. 5.4 Desk studies are usually based on the following data sources: published literature covering a range of topics including road construction and maintenance case histories and geological, economic and environmental reviews. The scientific literature is often a good source of information, though the areas studied tend to be small topographical maps geological maps, agricultural soil maps and other natural resource maps 5.6 Geological maps at small scale are the most common type of thematic map found in both developed and developing countries. Most countries have a national geological map at 1:1,000,000 scale or smaller, although few have complete or even substantial cover at large scale. The main use of small scale geological maps is to provide both an overview of the broad distribution of rock types for construction purposes, and a background to interpretations of other forms of data, such as topography, terrain classification and drainage. Geological maps often do not show surface deposits overlying the solid bedrock, although major features such as river terraces and sand banks may be shown. Soils maps 5.7 Soils maps are usually produced only for areas of potential agricultural interest, so their availability in mountainous areas is patchy. Soil mapping is useful to verify or extend the information given in geological maps, with emphasis on materials occurring within the top one to two metres. Agricultural soil classifications can contain elements that are of interest to engineering and it is worth reading carefully the descriptions of the soil units to identify criteria of engineering significance. Terrain classification maps 5.8 Terrain classification maps have been compiled for various parts of the world as general-purpose planning tools. They are based upon the classification of landform combined with soils, drainage characteristics and parent materials. The landform component can be a detailed geomorphological interpretation, or more loosely defined in terms of general topographical properties (slope angle, slope shape, etc) and appearance in aerial photographs (air

32

Figure 5.1 Terrain classification for corridor selection purposes in east Nepal

33

photo pattern). The main advantages of this kind of mapping are its ability to summarise large areas of terrain and its applicability to a range of land use and development projects. 5.9 Published terrain classification has been used very effectively in geologically old landscapes with relatively dry climates, where the drainage and material assemblages have become ordered into clear patterns of terrain units. Its usefulness in active fold mountains, where ground conditions change rapidly and unpredictably over short distances, is much diminished. Pre-existing terrain classification maps are therefore unlikely to be available for use in route feasibility studies in these areas. 5.10 The objective of any route alignment feasibility study is to collect and analyse sufficient data to enable the broad corridor details, design concept and approximate cost associated with the preferred option to be identified, without recourse to costly and time-consuming field investigation over large areas. Therefore, if terrain classification can be carried out cheaply and reliably, it will prove useful in gaining an understanding of the terrain from which to plan a strategy and recording system for later reconnaissance surveys. The classification system adopted, and its extrapolation over the area of interest, should be based on the interpretation of aerial photographs combined with topographical maps and ground verification. The sophistication of the terrain classification system devised should be sufficient to allow route corridor alternatives to be identified and divided into manageable units within which terrain and engineering parameters are considered likely to remain approximately constant. These terrain units could be based on a simple combination of the terrain zones given in Figure 2.4 with a four-fold slope angle classification, and further differentiation between soil and rock slope categories. Alternatively, they could be based on a more sophisticated terrain classification system, such as that illustrated in Figure 5.1. Under the usual constraints of limited time and lack of sub-surface information, the simpler this system can be made the better. 5.11 The design of the classification will normally depend upon: the geographical extent and reliability of available data sources required to carry out the classification, and the time and budget available to derive primary data, often over large areas, if existing data sources are insufficient the extent to which recognisable landforms are repeated in the landscape, and the confidence with which assumptions regarding materials, processes, drainage and stability can be applied to them.

develop a classification that is not too rigorous - the simpler the better identify and record actual landforms, materials and processes by preliminary walkover and then devise a classification scheme that reflects these avoid using a classification scheme that has been produced previously for another terrain type. Mould the classification to apply strictly to the area under study, even to the extent of using more than one classification in a region where strongly-contrasting terrain groups exist abandon the hierarchical concept, upon which most terrain classifications are based, unless there is strong evidence that one does exist, and that the concept will be of value for design.

5.13 Table 5.1 describes some of the common landforms found in mountain terrain and illustrates the fact that contrasting processes and materials can give rise to the same landform type, thus reinforcing the need for ground verification.

AERIAL PHOTOGRAPHY
Aerial photograph interpretation (commonly shortened to air photo interpretation or abbreviated API) is not only a highly important technique for terrain classification and desk study evaluation of topography and ground conditions, but it also provides vital data for many other analytical and design activities, including: landslide hazard mapping route corridor identification and selection drainage assessment and river catchment mapping identification of potential bridge sites identification of potential materials photogrammetry for contour mapping land use classification mapping of buildings and property for land acquisition environmental impact assessment and environmental monitoring construction planning.

5.12 A classification could be drawn up with the following points in mind:

5.15 However, the view of the ground can be obscured completely by forest, cloud cover or shade, especially in steep and irregular terrain. Under these conditions, recourse to ground survey is usually the only viable option.

34

Table 5.1. Landforms in the Nepal Himalaya and their implications for road design 35

Table 5.1 Landforms in the Nepal Himalaya and their implications for road design (continued) 36

Table 5.1 Landforms in the Nepal Himalaya and their implications for road design (continued) 37

Table 5.1. Landforms in the Nepal Himalaya and their implications for road design (continued) 38

Table 5.1. Landforms in the Nepal Himalaya and their implications for road design (continued)
5.16 Air photo interpretation should always be undertaken by a person who is familiar with the landforms, processes and materials likely to be encountered. As this familiarity can only be gained properly by reconnaissance survey, air photo interpretation and ground survey should form an iterative process. Some of the possible misrepresentations arising from air photo interpretation without ground verification are illustrated in Table 5.1. An example of air photo interpretation undertaken for route alignment in Nepal is shown in Figure 5.3 (pages 41 and 42). 5.17 A factor to recognise in hilly and mountainous terrain is the exaggeration of relief when viewing aerial photographs stereoscopically. Relief exaggeration causes slopes within the stereoscopic model to appear much steeper than they really are. The greater the exaggeration of relief, the more difficult it is to make an accurate estimate of the true slope angle. The degree of relief exaggeration is governed by the distance, or air base, between successive photographs, in relation to the flying height. Doubling the air base doubles the relief in the stereoscopic model (Figure 5.4. The air base in the upper stereoscopic pair is half that in the lower pair). In extreme cases, the image of a mountainside can appear so steep that it is impossible to make any effective interpretation of ground conditions. By contrast, low relief features such as river terraces, fans and landslide deposits are more readily identifiable by the exaggerated relief.

39

Figure 5.2 Landsat satellite image of part of central Nepal

40

Figure 5.3 Aerial photograph interpretation for road alignment purposeSteroscopic pair of aerial photographs

41

Figure 5.3 Aerial photograph interpretation for road alignment purposes (continued) - map and legend

42

Figure 5 4 Relief exaggeration through increased air base between successive photographs

43

5.18 For those in a position to commission their own, it may be worthwhile taking photography with an 80% photograph overlap in order to be able to select an appropriate air base for the amount of exaggeration required. For interpretation in areas of high relief the viewer would use every photograph in the run (80% overlap, giving a short air base and minimum relief exaggeration). For areas of low relief the viewer could select every other photograph, to double the relief exaggeration. In extreme circumstances of very flat ground the viewer could select every third or even every fourth photograph, doubling the relief exaggeration each time.

5.19 The scale of photography is also an important factor to consider in the reliability and ground resolution of the interpretation. Table 5.2 indicates the optimum scales of photography required to perform various desk study and design tasks. 5.20 Table 5.3 summarises topographical and geomorphological indicators relevant to the identification of geological, stability and drainage features on aerial photographs. These guidelines are only indicative and each site will require its own evaluation of factors and processes.

Table 5.2 Air photo scales for various project tasks

Task Activity Feasibility Study: Route corridor identification Terrain classification Drainage/catchment mapping Landslide hazard mapping Contour mapping for preliminary estimation of quantities Preliminary Design: Detailed interpretation of chosen corridor(s) for geotechnical purposes Ground (contour) model for preliminary alignment design and quantities Detailed Design: Ground (contour) model for detailed alignment design and quantities 1:5,000 - 1:10,000 1:10,000 - 1:15,000 1:10,000 - 1:15,000 1:20,000 - 1:30,000 1:20,000 - 1:30,000 1:15,000 - 1:25,000 1:10,000 - 1:20,000 1:15,000 - 1:25,000 Optimum Air Photo Scale

44

Table 5.3 Appearance of terrain features in aerial photographs

Feature SOIL Depth Appearance in aerial photographs Shallow: High percentage of rock outcrop and marked structural control to slope morphology; slopes steeper than 40 are almost certainly to be formed in rock. Deep: Concavo-convex slope profile with lobate and gently rounded lower slopes. Dendritic drainage pattern. Presence of visible large boulders. Type Residual soil Red and reddish-brown colour on colour aerial photos, commonly occupying rounded ridge and spur summits and flat or gently sloping benches. Often intensely cultivated. Can become severely eroded on sloping ground where vegetation is removed. Rockfall/ Rockslide colluvium Undifferentiated colluvium (transported weathered slope materials GEOLOGY Rock outcrop Strong rock Deposits of boulders below rock cliffs due to rock fall or rock slide. Range in size from small isolated deposits to deposits extending over large areas. Identification: rock back scar or source area, chaotic jumble of boulders, uneven slope, immature drainage pattern with jungle and patchy cultivation. Gradual slope failures and shallow creep give rise to more uniform slopes at limiting stability, often with larger boulders towards the toe. Usually intensely cultivated, with shallow failure depressions, immature drainage systems and often wet ground at the toe. Lower slope may be regularly or permanently unstable. Steep slopes, structural control on topography, high light reflectance from bare rock surfaces, repeated pattern of structural surfaces, cliffs and waterfalls. Steep and rugged topography with steep-sided V-shaped gullies and valleys, knife-edge ridges, cliffs and fresh structural control to landscape. Rockfalls; rockslides. Weak rock Gentle slopes, rounded spurs and ridges, low relative relief, no visible rock outcrop, concave slopes, shallow slope failures (soil). Less marked or no structural control on topography. Rock Type Mudstone, shale Limestone Slopes as for weak rocks; immature and shallowly incised drainage system, surface ponding and wet, heavy soils, frequent mudslides. Small, relatively deep depressions on ridge tops and structural benches, internal drainage and lack of regular surface drainage, discontinuous ridges, castellated tor-like high points along ridges, outcrop in structurally-controlled cliffs, often dip slope/scarp slope landforms, frequently forested, poor soils reflected in sparse cultivation. Quartzite Phyllite/Schist Gneiss Narrow, jagged, irregular cliff-lines often with little structural control. Thin soils, sparse vegetation and high reflectance from rock surfaces. Intermediate between soft rock and hard rock landscape and landforms. Linear structures are often evident due to a pronounced foliation. Low elevations: subdued and weathered landforms with frequently deep residual soils. High elevations: steeper, rocky landscapes with high relative relief and structural control over topography.

45

Table 5.3 Appearance of terrain features in aerial photographs (continued) Table 5.3 Appearance of terrain features in aerial photographs (continued)
Feature Linear structures (Dip/joint orientations, faults, etc) Dip slopes Hard rock Appearance in aerial photographs Lineations of ridges, cliffs and streams, often crossing other topographical features of different orientations. Fault lines are often areas of concentrated erosion and instability. Deep gullies may form where faults run downslope. Lithological (rock type) boundaries may be marked by a fault. Planar and regular slope surfaces tending to occupy large portions of valley sides in steep terrain. Topographical lineations are parallel to strike. Streams are straight, often approximately equi-spaced and shallowly incised into bedrock. Soft rock Scarp slopes (Dip/foliation into hillside) SLOPE FAILURE Active Roughly planar surfaces showing translational slope failures, shallow failure deposits and deeper drainage lines. Slope profile is steep and irregular with short cliffs and benches. Gently-dipping strata give a stepped slope profile in steep terrain. Hard rock strata form ribs across a slope, parallel to strike, frequently with trellis drainage pattern. Fresh failure scar and slipped mass. High reflectance from unvegetated back scar. Landform uncharacteristic of surrounding topography. Rock spalls around margin of deep failure. Immature vegetation or bare ground in slipped mass. Springs, ponds and wet ground at toe and in slipped mass. Slope tension and compression (hummocky profile) within slide area. Toe of failure protrudes into river flow; river erosion removing toe. Lack of cultivation or disturbed cultivation/irrigation pattern. Old Subdued/weathered landform described above, mature vegetation, established cultivation. Mechanism Progressive soil creep Mudslides (translational failures in finegrained soils) Immature/uncharacteristic vegetation. Disturbed/hummocky ground surface; micro-relief ridges perpendicular to direction of movement. Discontinuous/uneven irrigation and cultivation pattern. Slow moving ground with distinct lateral boundaries and shallow, narrow, planar track. Vegetation/cultivation is often maintained though disturbed in pattern. Usually on slopes of less than 28. Weathered rock or fresh soil exposed in back scar. Toe may be lobate. Usually on lower valley slopes where groundwater is higher. Failing slope angle often similar to adjacent unfailed slopes. Debris slide (translational slides in coarsegrained soils) Debris slides usually occur as an initial sudden failure with or without irregular movements thereafter. Well-defined back scars and side scars and an irregular slope surface. Vegetation is often destroyed. Usually on slopes between 30 and 40 in angle. High reflectance from back scar and slip material if recent. Boulders may be identifiable in deposit. Failed slope or debris cone usually shallower in angle than adjacent unfailed slope. Springs often emerge from base of back scar and along toe of slipped debris.

46

Feature Debris flow (high velocity flows of debris and mud) Progressive rock creep ('sackung')

Appearance in aerial photographs Source area of slope failure or erosion on steep ground and depositional track or fan of debris on concave slope below. Common in unstable and eroding catchments. Mostly in fractured rock masses occupying high, steep slopes. Difficult to identify and confirm on aerial photographs. Ridges and trenches orientated normal to direction of maximum slope may form as a result of plastic flow within rock mass. Active creep causes rock spalls on margins of failure zone.

Rock fall

Rock fall scar (topple, wedge or plane) and a talus deposit below. Progressive rock fall gives slightly concave, uniform, talus slope of 33-38; larger particles accumulate towards toe.

Rock slide: Rapid (instantaneous)

Distinct margins of failure with cliffs. Planar failure surface may be exposed below the back scar with higher reflectance if recent. May be seepage, surface ponding and irregular slope drainage pattern within slipped mass. Failed mass of chaotic boulders and rafts of rock. May be a single block or a number of blocks, stability of each controlled by movements of block below. Slope angle of failed debris significantly less than that of adjacent unfailed slopes. Large valley-side failures may partially block river below or cover river terraces. Large blocks may persist in river.

Rock slide: Slow moving (progressive) Rotational slides: Soil

Hummocky and furrowed slopes in head of failure with or without compression ridges in toe. Slope rupture and small vertical displacement along margins. Arcuate back scar in plan. Concavo-convex slope profile in section from back scar to lobe at toe of failure. Well defined slip margins. Debris toe forms a spreading lobe on slope below. Rotated slipped mass may form a reverse slope below back scar, with possible ponds. Toe lobe may be eroded due to ground water released during failure or adjacent river scour. Lobate form of failed mass may create anomalous drainage paths diverging down slope along margins of slip. Springs may continue to emerge from base of back scar. Multiple back scars indicate either progressive failure upslope above original back scar or secondary failures within original slipped mass. Second case may give stepped slope profile exploited as farm terraces.

47

Table 5.3 Appearance of terrain features in aerial photographs (continued)

Feature Rotational slides: Rock Appearance in aerial photographs As for soil failures in principle, although: Slip margins may comprise a number of segments that approximate to arcuate form. Surface of failed mass usually of boulders. Irregular profile, possibly less lobate, probably no surface drainage; Secondary failure unlikely (unless undercut by river erosion below). The reduction in slope between the back scar and the failed debris is usually much more marked in the case of rock failures. Comments above for differentiation of rapid and slow moving rock slides apply to rotational slides. Rock avalanche (catastrophic rock fall) Original rock fall scar may not be apparent due to subsequent weathering. Failed mass is long, thin tongue of boulders on slope below, in valley floor or even high on opposite valley side. Flow lines and pressure ridges parallel to flow. Rock avalanche may alter drainage pattern. Failure volumes up to 100 million m3 can occur. DRAINAGE Slope Drainage Sheet wash Channel flow Pipes (shallow concentrated subsurface drainage and erosion) Springs At base of steep slope (strong pervious rock/weak impervious rock boundary). At junction between permeable and overlying impermeable layers (colluvium/ residual soil; rock strata). At base of slope failure back scar. Wet ground, often with dense vegetation. At heads of stream channels. Gully heads Eroding gullies Steep-sided erosion bowl at head of a gully. Irregular channel in plan and width shows variable erodibility of channel materials. High reflectance from bare surfaces. Obstructions such as fallen trees and boulders in channel bed. Sediment deposited downstream in channels and on fan surfaces. Flood plain and valley floor Active fans/high sediment yield Very low stream banks and evidence of sediment overspilling banks. Fan may have higher elevation than surrounding slopes. Lines of sediment orientated parallel to stream. Eroding source area for debris. Lack of vegetation on channel bed. Tributary fan encroaches into main river. Preserved recent depositional bed forms. Generally not identifiable on aerial photos. Shallow parallel erosion channels on bare slopes, becoming gullies below. Generally not identifiable on aerial photos until collapse occurs. Settlement along a pipe line may cause linearity in surface drainage. Subsur face flow may lead to wetter surface soils. Both conditions may cause lines of vegetation.

48

Table 5.3 Appearance of terrain features in aerial photographs (continued )

Feature Flood plain Flood plain terrace Appearance in aerial photographs Uniform or braided sediment-laden river bed. Unvegetated except for isolated islands of flood plain terrace. Immature or no tree cover. Usually uncultivated. Sediment deposited onto banks from recent flow. Recent erosion of terrace bank. Higher level terrace Active terrace bank erosion Usually either covered in mature trees or cultivated. Bordered by flood plain terrace or directly scoured by river. Terrace edge cuts across field boundaries. Boulders too large to be removed by erosion mark previous bank positions. Vertical banks with no vegetation: high reflectance. Successive aerial photos may show rates of bank retreat. Flood level Usually only identifiable on large scale air photos if taken within a year of flood. Strand-lines of vegetation, sediment and slope erosion. High flood in main valley may cause deposition of sediment at this level in tributaries. Lower valley-side scour by main river Fresh exposures of rock (high reflectance). Absence of vegetation. Instability on the lower valley-side caused by undercutting. Absence of terrace deposits. Large boulders deposited at the foot of the slope by river flow. High level fan Cone-shaped deposit at mouth of tributary valley above level of current main river action. No active or recent sediment on fan surface. Cultivated fan surface with signs of semi-permanent habitation. If fan surface is tree-covered rather than cultivated, surface flow may still occur intermittently. Fan may contain active channel incised to level of present-day drainage system.

49

Table 5.4 Comparison between the Landsat MSS, Landsat TM and SPOT satellites
Technical characteristic Landsat Multi-Spectral scanner USA 1972 to present 1 2 913km or 700km 9.30 am approx. 1 2 18 days or 16 days 4 185 x 185km Landsat Thematic mapper USA 1982 to present 700km 9.30 am approx. 16 days 7 180 x 180km SPOT

Country of origin Date of operation Satellite altitude Time of crossing Equator Revisit frequency Number of spectral bands Image size on ground

France 1986 to present 832km 10.30 am 3 26 days 3, plus 1 panchromatic 60 x 80 or 4 60 x 60km

NOTES 1. Landsats 1 and 2 2. Landsat 3 3. The satellite is capable of taking images of the same area at intervals of one to several days by directing its sensors right or left along preceding or future orbit paths. 4. According to whether satellite is looking vertically or obliquely (see Note 3).

Table 5.5 Advantages and disadvantages of the three main types of satellite imagery
Satellite Landsat Multi-Spectral Scanner Advantages World coverage on archive Coverage dates from 1972 Large image size (185 x 185km) Least expensive (archive) Landsat Thematic Mapper (TM) World coverage on archive 7 spectral bands available Resolution better than MSS (30m nominal) Large image size (185 x 185km) SPOT Best spatial resolution (multispectral 20m: panchromatic 10m. Nominal) Stereoscopic capability Image size is only 60x6Okm Only 4 spectral bands Not full world coverage (archive) Archive stereoscopic imagery Requires a lot of computer processing power to explore all 7 bands Disadvantages Rather low spatial resolution (80m nominal) Only 4 spectral bands

SATELLITE IMAGE INTERPRETATION

5.21 There are three principal types of satellite imagery, namely Landsat Multispectral Scanner (MSS), Landsat Thematic Mapper (TM) and SPOT. The technical characteristics of the satellites are listed in Table 5.4, but the choice of imagery often depends on what is available. The

advantages and disadvantages of each type of imagery are given in Table 5.5. SPOT panchromatic imagery can be acquired in stereo-pairs from which a digital terrain model with a horizontal accuracy of about 10m and a vertical accuracy of about 20m can be derived.

50

5.22 An example of a satellite image of mountainous terrain and its interpretation is reproduced in Figure 5.2 (page 40). Satellite images are compared with aerial photographs in Table 5.6, but the principal advantages of satellite imagery over aerial photographs include the ability to portray large areas of ground surface showing large scale geological structures, drainage patterns, river catchment areas and broad land use patterns. The preparation of an aerial photograph mosaic to cover an equivalent area would be time-consuming and very expensive, and would in any case suffer significant distortion and changes of scale in areas of high relief. 5.23 The main applications of satellite image interpretation to road feasibility studies in mountain areas relate to: measurement of distances and areas, especially catchment areas broad classification of terrain types assistance in the identification of route corridors

coarse differentiation between rock and soil slopes, broad soil and vegetation types, areas of high groundwater and large scale slope instability identification of erosion (even small areas, near the resolution limit of the image, are visible if they are of high contrast with their surroundings) identification of major geological structures and their influence on topography and regional slope stability identification and mapping of large water bodies that could give rise to Glacier Lake Outburst Floods (GLOFs) identification of broad changes in the course of rivers, land use and snow cover over time if sufficient archive data exist.

Furthermore, satellite imagery can be used as a means of monitoring some aspects of environmental impact, and especially land use change during and after project implementation.

Table 5.6 Comparison between aerial photography and satellite imagery

Image type Aerial photography Advantages Low unit cost, though total cost depends on size of area covered Widely available through survey departments Very good image detail on black and white photographs Colour enables drainage, land use and soils to be interpreted more readily Stereoscopic image Specialist training in use is minimal Familiarity as a conventional source of information Satellite imagery World-wide coverage (Landsat) Access to imagery not normally restricted Purchased by mail order from remote sensing centres Colour information Digital image analysis possible Constant scale Disadvantages National cover often incomplete

Access can be restricted in sensitive border areas No colour information Slightly poorer image detail on colour photographs Scale not constant. Cloud cover sometimes a problem Digital analysis not possible Large numbers of photographs needed at desk study stage High unit cost, although cost depends on size of area in image Relatively low spatial resolution Stereoscopic imagery rarely available (though likely to increase) Cloud cover sometimes a problem Specialist equipment necessary for digital analysis

51

5.24 Satellite imagery can be purchased in digital form (on tape or disc) or as photographic products. These two forms are compared in Table 5.7. Despite the potential applications offered by satellite imagery, if essentially cloud free, stereo aerial photography exists at a scale of 1:30,000 or larger, then the only practicable advantage in procuring satellite imagery for route feasibility and design purposes lies in the production of an essentially distortion-free photographic map of a given area that can be used for visual orientation and as a project planning base document. In the absence of topographical mapping it can also represent a useful means of measuring distances and areas. 5.25 Before a decision is made to purchase satellite imagery it is advisable to be aware of the following factors:

the percentage of cloud cover given in computer listings of archived imagery can be incorrectly stated, especially in areas where high ground is covered with snow colour prints of satellite images cost between 2 and 6 times more than black and white if photographic products are required it may be preferable to purchase film negative rather than film positive or paper print, so that extra copies can be made cheaply digital data are required for digital image processing and Geographical Information System (GIS) applications geometrically-corrected imagery will eliminate distortion, although the accuracy of the final image can only be as good as the Ground Control Points in the underlying co-ordinate system and the maps or other means by which they are identified as with aerial photographs, the reliability of the interpretation must always be verified by other forms of evidence.

if satellite imagery is not available from archive sources, it can be commissioned, although this procedure is more expensive and takes longer for delivery of images a higher spatial resolution is generally preferable all three types of imagery listed in Table 5.4 are impaired by atmospheric moisture and, in tropical areas especially, cloud cover can be a mayor constraint Imagery obtained in the dry season is more likely to be cloud-free and generally portrays a clearer definition of soil and rock conditions

5.26 Radar imagery is unaffected by cloud cover or atmospheric moisture, and therefore provides an alternative to conventional photographic satellite imagery in areas where

Table 5.7 Comparison between photographic and digital satellite image products
Advantages Photographic satellite images Prints are easy to annotate. Images are geometrically correct. Images can be enlarged to a scale of 1:100,000 without serious degradation. A full scene can be reproduced at maximum resolution of the photograph. If a negative is purchased, multiple copies are cheap to produce. Digital satellite images Colour can be optimised for differentiation of any type of feature. Resolution down to individual pixels can be achieved. Different combinations of spectral band can be displayed quickly. Processing and manipulation of the image is very quick. Mathematical and statistical processing can be carried out on the spectral data. Images can be fitted to any cartographic projection. Disadvantages Specialist photographic expertise is required to process or manipulate bands. Dynamic (colour) range is limited. A good colour balance for all parts of the image is difficult to achieve.

Image processing equipment is required for display and output. Hard copy has to be made from photographs of the display screen, colour printers or specialist film writing equipment. Only a fraction of the image can be displayed at full resolution on the screen.

52

these are limiting factors. The main disadvantages with radar imagery are a limited (but growing) availability and difficulty of interpretation.

preferred cross-sections according to terrain units rock types, geological structures, dip orientations, rock strength and rippability percentage of rock in excavations materials sources soil types and depths (a simple classification between residual soil and colluvium is useful at this stage) soil erosion and soil erodibility slope drainage and groundwater conditions drainage stability and the location of shifting channels and bank erosion land use and its likely effect on drainage, especially through irrigation likely foundation conditions for major structures approximate bridge spans and the sizing and frequency of culverts flood levels and river training/protection requirements environmental considerations, including forest resources, land use impacts and socio-economic considerations (Chapter 6).

RECONNAISSANCE SURVEY
5.27 When potential route corridors have been identified from the desk study analysis, then reconnaissance survey is usually employed to verify interpretations, to help determine the preferred corridor and identify factors that will influence the feasibility design concept and cost comparisons. Even at the reconnaissance stage, data collection should be detailed enough to identify the most significant sources of construction difficulty in the landscape. Allowing sufficient time and resources to identify and evaluate these problems during reconnaissance survey could avoid costly readjustments to the design concept or the route corridor later on. The team employed to carry out the reconnaissance survey should include adequate geotechnical expertise, familiar with the problems expected to be encountered. Hydrological and environmental assessments will also form important elements of reconnaissance survey. 5.28 In most cases, reconnaissance survey will significantly modify the desk study interpretations. In the absence of any worthwhile desk study data, or where dense forest cover prevents the successful interpretation of aerial photographs, reconnaissance survey will have to be relied upon as the sole means of route corridor identification and evaluation. 5.29 Reconnaissance survey data can either be recorded onto topographical maps or enlarged paper prints of aerial photographs. The latter can form a particularly useful means of recording data in the field, especially in areas where forest cover is relatively low. Minor ground features, such as houses, field boundaries, cliffs and water courses, can be readily identified. However, the scale of aerial photograph enlargements is highly inconsistent and alignment length should be approximated from topographical maps and then transferred onto the enlarged photographs. GPS can be an extremely useful means of determining coordinates for specific locations. 5.30 During reconnaissance survey, the following factors should be considered: terrain classification units the location of topographical constraints, such as cliffs, gorges, ravines and any other features not identified by the desk study slope steepness and limiting slope angles identified from natural and artificial slopes (cuttings for paths, agricultural terraces and existing roads in the region) slope stability and the location of pre-existing land-slides

5.31 Data collection should be designed according to the terrain classification units discussed in paragraph 5.8-5.13 and recorded directly onto maps and, where time permits, onto proformas according to the terrain classification system adopted. An example of a reconnaissance survey proforma is shown in Figure 5.5. Usually anything more detailed than this, or the use of multiple proformas, is time-consuming to compile and therefore not justifiable in terms of improved corridor assessment.

LANDSLIDE HAZARD MAPPING

Introduction 5.32 The aim of Landslide Hazard Mapping (LHM) is to portray the geographical variation in the susceptibility of slopes to failure. LHM is based on the premise that the relative potential for slope failure can be assessed by reference to the existing distribution of identifiable failures and by measuring and comparing the distribution of landslide-controlling factors. From the point of view of route location, LHM may be able to provide an early appreciation of relative slope stability over a comparatively large area at the feasibility stage, without going to the expense of detailed and extensive geotechnical ground mapping. However, if air photo interpretation, terrain classification and

53

Figure 5.5 Field data sheet for comparison of alignment feasibility

54

reconnaissance survey can be satisfactorily combined, or if a small number of feasible route corridor options can be easily recognised in the field, LHM may not prove to be a particularly cost-effective use of resources if carried out specifically for an individual road project. Furthermore, if large volumes of primary data have to be collected to complement insufficient desk study sources, then the speed with which LHM can be carried out will be significantly reduced. Common methods of landslide hazard mapping 5.33 LHM employs both direct and indirect methods of assessment. For direct mapping, a study area is zoned according to the location or density of recorded landslides, on the assumption that future landslides are more likely to occur in areas with similar slope conditions to those where landslides have occurred already. Indirect mapping relies on the evaluation of factors that are considered to be significant in the initiation of slope failure. Figure 5.6 lists some of the factors that contribute to the stability or failure of slopes in the Himalayas. Clearly, through lack of data, only a small proportion of these factors can be evaluated during route corridor feasibility study. The aim of LHM for corridor comparison should be to identify landslide locations and to compare this database with whatever terrain and factor classifications can be provided from topographical maps, geological maps, aerial photographs and reconnaissance survey, thus combining direct and indirect mapping methods. 5.34 Figure 5.7 presents a recommended procedure for carrying out LHM. Box 5.1 outlines the procedural stages that should be followed. A simple statistical comparison between the density of existing landslides and terrain classification units may be all that is required if a significant relationship is obtained. The availability of aerial photographs is the key to the speed and success with which this exercise can be performed. If aerial photographs are unavailable, then it may be more costeffective to revert to more detailed reconnaissance survey and geotechnical mapping of selected corridors, rather than to spend a great deal of time in the field generating primary data for LHM over large areas. 5.35 Wherever possible, differentiation should be made between landslide failure mechanisms, particularly the distinction between landslides and areas of erosion, and deepseated and shallow failures. This is an important consideration because these various distinctions pose different risk implications for road construction and maintenance, and their occurrence will usually be affected by different factors. 5.36 The main disadvantages with using direct LHM in isolation relates to the fact that it does not permit the potential for `first-time' failures to be evaluated, or for ground conditions to be compared in terms of broad susceptibility to disturbance during construction. For instance, a

weak rock type may show no visible signs of failure because slope angles are too shallow to initiate movement. During construction. however, slope excavation for road-works may give rise to widespread failure. Furthermore, aerial photograph interpretation, and even ground reconnaissance survey, may fail to recognise the more subtle failures, and therefore it is worthwhile evaluating contributory factors that can be readily identified from available sources. Box 5.1 Recommended outline procedure for Landslide Hazard Mapping determine the distribution of existing slope failures in a given study area, through a combination of air photo interpretation and reconnaissance survey identify the most significant factors that control or trigger slope failure in the study area divide the study area into slope units using terrain classification or geomorphological criteria systematically measure landslide controlling factors for each terrain unit and assign values to class intervals according to natural breaks or observed relationships in the landscape. For instance, slopes on rock type A may be known to fail when slope angles exceed 25, while slopes on rock type B may remain stable at slopes of up to 45 if foliation or bedding orientations are unfavourable to stability, and up to 60 where they are favourable assign a hazard rank to each factor class according to its correlation with the observed distribution of landslides (if the distribution of landslides cannot be reliably represented then the hazard rant: assigned to the various factor classes will have to be made on the basis of field observation and judgement) test the statistical validity of the final hazard classification against the distribution of landslides and by any other independent means, including ground verification the final hazard map should depict no more than five levels of hazard and should be accompanied by the landslide distribution map and an explanatory text.

Applications of landslide hazard mapping 5.37 LHM has been successfully applied to land use planning in mountain areas where economic and engineering development criteria have required geographical comparisons of landslide potential within the context of a regional development or land capability/land use management programme. Intensive data collection, rigorous modelling and a thorough knowledge of local landslide - controlling factors can then give LHM a reliable analytical basis, and the relatively high cost involved is justified.

55

Figure 5.6 Factors influencing slope instability and erosion potential in the Himalayas

56

Figure 5.7 Recommended procedure for landslide hazard mapping in remote mountain terrain

57

5.38 In the case of new road construction through mountain areas, the pre-existence of a LHM database would provide a useful point of reference for route corridor identification. However, it is usual to fund that such a database is unavailable, and that any given road project in remote mountain areas will have an extremely limited source of information at its disposal. Under these conditions, there are several reasons why, in practice. LHM may not provide the objective and reliable estimate of relative stability that is often claimed or implied (Box 5.2). 5.39 Despite these reservations, LHM may prove a useful means of imposing the need to evaluate a range of terrain characteristics in a methodical way, thus ensuring that all available information is taken into objective consideration. In general, the development and adoption of systematic guidelines for collecting, analysing and interpreting LHM data could improve the quality and reliability of the end-product. Furthermore, LHM and risk assessment can often be employed more effectively on a site-specific basis in evaluating hazard potential and runout distances on individual slopes, or in logging and classifying the stability of earthworks slopes during road construction and maintenance. Some illustrations of LHM applications are provided in the following sections and in Chapter 7.

Box 5.2 Landslide hazard mapping (LHM) for route alignment purposes The following points should be borne in mind when considering whether or not to embark on an LHM exercise for road alignment in remote mountain areas: there are no standardised guidelines for carrying out LHM LHM does not calculate an absolute value of slope stability, only a relative one. Unless rigorously applied. LHM may be incomplete in its assessment of hazard as defined by failure potential, depth of failure and failure mechanisms. Also, it rarely enables risk, or the absolute sensitivity of a slope to disturbance by road construction, to be determined landslide-controlling factors are difficult to measure or estimate accurately. Many are underground and some are time-dependent, eg groundwater fluctuations unstable mountain slopes are often characterised by multiple hazards of various mechanisms, and it may be difficult to represent them all on a single map with a common hazard ranking system. Furthermore, each will have a different risk implication for road engineering the problem of lack of data can, to a certain extent, be resolved by reconnaissance survey, but from a mountain road design point of view, it would be generally too costly and take too long to reconnoitre a large area in order to make a landslide hazard map merely for use as a means of identifying route corridors. A large proportion of the data would become redundant once the preferred corridor was identified leading on from the above, decisions regarding alignment options are usually required to be made long before LHM based on ground survey can be completed. Conventional air photo interpretation with reconnaissance survey is the most efficient way of identifying and selecting alignment options in young fold mountains, seismic shaking is an important trigger mechanism for slope failure and yet there are very few examples where thus factor has been incorporated into LHM. The main reasons for this are the indeterminate nature of earthquake intensity and the uncertainties regarding the manner in which slope materials will react to shaking.

Box 5.3 Distinction between `hazard' and `risk' The terms 'hazard' and 'risk' have often been used synonymously in the published literature, creating a source of confusion. The term 'hazard' defines the physical attributes of a potentially damaging event (landslide, flood or earthquake) in terms of type, mechanism, volume and frequency. 'Risk' is governed by hazard type, size and probability of occurrence, the value of the property, structure or population at risk and the vulnerability to the hazard, should it occur. In other words, 'hazard' describes the event and `risk' describes the probability and total consequences of the event occurring. Although risk can rarely be fully evaluated, some studies have attempted to assess the probability of landslides of given sizes and potential impacts, by reference to the return period of triggering events such as earthquakes and rainstorms. Even then. the relationship between seismicity and the timing and distribution of slope failure is largely unknown, while antecedent rainfall and seasonally fluctuating groundwater levels are often just as important in triggering deep-seated slope failures as 24 hour storm rainfall or maximum rainfall intensity.

58

Landslide hazard mapping for route alignment in Nepal

5.40 This study of relative slope stability for route corridor comparison in one of the most unstable parts of east Nepal was carried out as a separate research exercise after road construction to compare LHM with the original geotechnical reconnaissance survey. The latter did not have the benefit of good quality aerial photography or reliable topographical mapping, both of which were obtained later in the project. 5.41 The study area for the LHM comprised three main drainage basins covering a total area of 120km2. Black and white 1:25,000 scale aerial photographs, 1:50,000 scale topographical maps derived from the aerial photographs and 1:200,000 scale geological mapping, were used to derive a series of factor maps. Separate analyses were carried out for each of the three catchments in order to minimise the effects of extreme geographical variations in rainfall that could not be reliably quantified. The observed densities of landslides occurring within each factor category were compared with those expected had there been no control on the distribution of slope failure (a random distribution). Rock type, physiography and slope aspect were found to be significant factors, and the various factor classes were weighted according to their observed relative control on landslide distribution. Land use was found to be a significant controlling factor in two of the three catchments, while slope angles measured by parallax on the aerial photographs over grid cells of 230 x 230m proved too coarse to be significantly correlated with landslide distribution, even when analysed according to rock type. Figure 5.8 shows the sequential factor mapping undertaken for one of the catchments and the final hazard classification with respect to the constructed alignment. The hazard classification vindicated the chosen route and was found to be significantly correlated with the distribution of additional landslides mapped from aerial photographs taken seven years later, thus supporting its application as a predictive tool. 5.42 The LHM procedure described here took two manmonths to complete and yielded essentially the same conclusions with respect to route corridors as those reached by the reconnaissance survey, which was accomplished in approximately the same mantime duration. However, the reconnaissance survey yielded site-specific data that was incorporated into design later on and therefore represented a more cost-effective use of mantime, with or without the availability of desk study data sources. Although the landslide distribution map differentiated between hazard mechanism, to the extent that this is possible from air photo

interpretation, deep-seated slope failures and the factors that promote them were not adequately represented in the analysis. Shallow failures are much greater in number and usually far easier to identify.

Rock avalanche hazard mapping in Papua New Guinea

5.43 The objective was to provide an assessment of landslide and avalanche runout potential from a steep mountainside located above a mining township. Slopes are covered by dense rainforest, and consequently all data, covering an area of over 3km=, had to be dewed by geotechmcal ground survey using compass traverse at a scale of 1:2,500. 5.44 Fieldwork took approximately 5 man weeks to complete and was followed by a similar period in the office carrying out analysis, interpretation and reporting. Figure 5.9 illustrates the various stages of the mapping and analytical procedure. The geotechnical survey was used to: record evidence of active and previous slope failures divide the study area into a total of 494 terrain units derive a suitable database for landslide hazard mapping.

5.45 Tables 5.8 and 5.9 list the various rock and soil factors used in the LHM, the categories identified and the weightings employed. Separate analyses were carried out for rock and soil slopes. Each factor was assessed systematically for every slope unit and assigned a category value which was then multiplied by the weighting for that factor to yield the hazard score. The hazard scores were then summed for each slope unit and assigned to one of four hazard classes. The distribution of landslides identified by the geotechnical survey was used to test the validity of the hazard classifications. For both rock and soil slopes, it was found to be significant at the 95% level. 5.46 The source areas and failure deposits belonging to previous landslides were also recorded during the geotechnical survey. By categorising each landslide according to estimated volume and mechanism, and by measuring slope angle and slope length in the failure zone and slope angle and runout length in the depositional zone, a set of regression equations was established for predictive purposes. Estimation of the likely volumes of potential failures, and the relevant slope angles and slope lengths in the high hazard areas, allowed potential runout lengths to be predicted from these equations and precautionary measures to be recommended (Figure 5.9e).

59

Figure 5.8 Landslide hazard mapping for route alignment through an unstable river basin in east Nepal

60

Figure 5.8 Landslide hazard mapping for route alignment through an unstable river basin in east Nepal (Continued) 61

62

Figure 5.9 Landslide hazard and runout mapping in Papua New Guinea

Table 5.8 Rock slope hazard rating for a site in Papua New Guinea (on scale of H = 0 (least hazard) to H=3 (greatest hazard)

63

Table 5.9 Soil slope hazard rating for a site in Papua New Guinea (on scale of H = 0 (least hazard) to H=3 (greatest hazard)

64

ENVIRONMENTAL IMPACT ASSESSMENT

INTRODUCTION
6.1 Environmental impact assessment (E1A) is a formalised procedure for investigating, analysing and presenting the environmental implications of a proposed development and identifying mitigation measures required to reduce environmental impact to acceptable levels. The advantage of such a procedure is that environmental impacts and potential benefits are brought to light prior to implementation in an objective way, for discussion between concerned organisations such as the funding agency, the client, natural resource experts, community representatives and the engineering design management. Despite its apparent judicial role, EIA is intended to act in the best interests of all parties by integrating a project smoothly into the environment within the context of the wider conservation and regional development issues. 6.2 The issues addressed by an EIA may already be incorporated into a normal feasibility study, and many road schemes in the past have been designed and constructed without the benefit of EIA and with no apparent adverse environmental effects. Others, of course, have been less sympathetic. Whether or not a formal EIA is carried out it should be stressed that sound judgements in engineering, hydrology and engineering geology will always remain among the most important factors in successfully integrating design, construction and environmental compatibility in hilly or mountainous areas. It is the engineer's responsibility to ensure that, notwithstanding the need to accede to political and economic circumstances, this principle is adhered to, and that physical damage to the environment is minimised. 6.3 Most road projects funded under foreign aid are now required to have an EIA carried out as part of the aid agreement. Because EIA covers such a wide range of issues, it is normal to hire the services of a specialist consultant to carry out the study; this also helps to ensure impartiality. The environmental consultant is normally expected to liaise with the relevant governmental and non-governmental organisations in order to: establish or strengthen environmental legislation establish, confirm or improve upon recommended guidelines for EIA where the environmental consultant is from an overseas country, undertake training of national personnel in EIA and environmental management make maximum use of local man-power and technical resources, although this is usually a requirement of development programmes independently of environmental considerations.

6.4 Regional and national conservation and development issues, as well legislative and institutional aspects, are usually considered by government planning departments as part of project identification. They would not normally form part of an EIA for an individual road construction project, except where an environmental consultant is asked to advise on institutional strengthening and legislative policies. 6.5 The relationship between man and his environment is of course complex. However, the real issue behind many environmental concerns in remote (especially mountainous) regions is that separate social groups are brought into contact for the first time as a result of the development project and one can affect the other detrimentally, thus generating conflict between the groups involved. Frequently, one group exploits the natural environment in a way that degrades the habitat of the other, impairing the other's way of life.

IMPACT CATEGORIES
6.6 It is useful to classify a mountain road project EIA into the following environmental impact categories: geomorphology ecology land use socio-economic factors.

6.7 Impacts under these categories are expanded in Table 6.1, and their inter-relationships are shown in Figure 6.1. The discussions below illustrate how the impacts interact and show that no impact category can be examined independently of the others.

Geomorphology
6.8 The impacts of a road on landscape and geomorphology are among the primary concerns and responsibilities of the engineer. The geomorphological impacts of road construction relate to the effects of earthworks, spoil disposal and road drainage on slope and channel erosion and slope stability. Most geomorphological impacts of road construction are usually addressed as part of the design, as their effects are immediate and they have a direct bearing on engineering performance. The effects can extend for considerable distances outside the right of way, and can be devastating. Compensation is often either insufficient or non-existent, and thus these impacts can be among the most severe for rural farmers. Ecology and natural resources 6.9 Ecological impacts can be classified into four groups:

65

66

Figure 6.1 Cause-effect interrelationships between environmental impacts along mountain roads serving rural communities

loss, depletion or disturbance of faunal or floral species (animals and plants) that are rare or protected and are considered to be of regional, national or international importance. In some circumstances, ecological factors such as rare species and diverse forest or aquatic habitats may be sufficiently important to influence the detailed alignment or even the choice of route corridor loss, depletion or irreparable damage to rare or vulnerable habitats. Destruction of a habitat reduces a region's biodiversity (the number and variety of plants and animals living within the habitat). Biodiversity is important in maintaining the stability and resilience of the plant and animal populations in the habitat reduction in biodiversity (genetic, species and habitat) due to the above effects

An expropriation period equivalent to the design life, or any other time span, is difficult to justify because it is inconceivable that a right of way would be handed back to agricultural use once exceeded. 6.14 The compensation issue is complicated by the fact that the price of land often increases substantially within the route corridor (up to five-fold increases have been reported, and in the vicinity of rapidly expanding urban areas, hundred-fold increases may be common). In these circumstances, farmers who have been compensated for their land at the `normal' rate may not be able to afford the price of replacement land in the vicinity. Rehabilitation and reemployment programmes directed at these farmers and households can be proposed, along with maximum reclamation of the contractor's temporary working areas when construction is complete. 6.15 It is important to respect any existing agreements on the use of water supply for household use and irrigation. Road drainage works, especially hairpin stacks, can significantly affect water pathways. Socio-economic factors 6.16 These can include: traffic and infrastructural aspects, (this factor is directly the responsibility of the engineer) regional and national conservation and development issues economics, and changes in economic structure social, cultural, religious and ethnic factors demographic factors tourism mineral extraction (uncontrolled mineral extraction and waste dumping can be especially damaging) legislative and institutional aspects pollution (the pollution of water supplies is a particularly sensitive issue).

management of natural resources.

6.10 The pollution of lakes and water courses and the depletion of forest habitats are often the most important ecological concerns in sub-tropical mountainous areas. As Figure 6.1 shows, forest habitats can be adversely affected by a number of factors, therefore it is important to assess habitat stability under natural conditions before road-related impacts are evaluated. 6.11 Forest extension schemes attached to a road project may lead to an overall improvement in the management of forest resources, and road access may allow more effective policing of conservation areas. However, road construction usually results in an increased demand for forest products that can put the habitat under considerable pressure. 6.12 Natural resources are those components of the physical environment that are exploited by man. They are important to both ecology and socio-economics. Natural resources can be renewable (such as timber) or non-renewable (such as mineral deposits). Good management of renewal natural resources can enhance the health and productivity of the environment, including the non-resource environment (ie, those aspects of the environment that are not utilised by man). Bad management can deplete both the resource and the non-resource environment. Land use 6.13 Road alignments tend to occupy less steep ground that is usually of highest agricultural value. The livelihood of subsistence farmers can be severely affected by the loss of cultivable land. Financial compensation for acquired land is usually paid on the basis of cadastral survey and expected yields during the design life of the scheme. The current value

of agricultural land is comparatively easy to determine, though future yields may be difficult to assess.

6.17 Positive socio-economic benefits will accrue in remote areas only if a proposed road scheme is accompanied by a wellplanned and well-managed development strategy. The benefits of increased demand for foodstuffs, and hence higher prices, during construction may lead to a short-term improvement in farm incomes. Those who benefit most are farmers or landlords who own large areas of farmland, and those who are able to bring marginal land into agricultural production. Tenant farmers who survive at subsistence levels may experience no economic benefit.

67

6.18 In the long term the agricultural economy can be adversely affected if farmers become unable to compete with cheaper imported produce. Diversification is a possible solution to this, but it may cause the community to become less selfsufficient and therefore susceptible to external fluctuations in commodity price and availability. In addition, the full economic potential for diversification may not be realised when regions of similar agricultural or industrial output are connected. Under conditions where road access does not result in the import of cheaper food commodities, a rapid rise in population density alongside the road, combined with a general trend towards rural depopulation, may lead to food shortages and increased poverty. This may result eventually in significant migration of people out of the area. 6.19 Demographic issues include population growth, density, distribution and migration, and composition of ethnic, religious and age groups. Often, a development project such as a road scheme brings benefit to certain sections of the population and not others. according to caste or ethnic background, age, sex or education. Thus, disadvantaged groups may become increasingly marginalised and may be forced to move away from the road corridor, or out of the area entirely. A road scheme can actually increase inequality among socio-economic groups. 6.20 The young and the unemployed will tend to move into the road corridor to take advantage of potential employment opportunities, and this may lead to a breakdown in the sustainability of rural communities and a gradual erosion of traditions and cultural values. A further consequence is a rapid expansion of urban areas and shanty developments along the route corridor, within which basic living, sanitation and health facilities may be grossly inadequate. Similar outcomes may arise with the influx of migrant workers who form part of the temporary construction labour force, and this may lead to cultural and ethnic mixing and the potential for social unrest.

description of existing environmental base-line conditions, including the quantification of natural trends, such as land stability and rates of erosion, distribution of wealth and poverty, population growth, urban expansion, rural depopulation, deforestation and land use change identification and quantification of potential environmental impacts, both direct and indirect, including opportunities for environmental improvement recommendations for environmental management and monitoring of mitigation performance training of local personnel and institutional strengthening, where appropriate.

6.22 The various stages of EIA procedure are briefly described below. The EIA report should be accompanied by a non-technical summary, known as an Environmental Statement, that is set out according to the following format. The Environmental Statement should outline the manner in which the project will be managed within the wider context of the study area environment.

Project description
6.23 An overview of the proposed (or existing) scheme is required, in terms of its location, size, technical design and construction method, construction time-frame, estimated cost, labour force requirements and operational procedures.

Base-line study
6.24 The base-line study summarises the existing conditions against which the environmental effects of a proposed project are evaluated. Its objectives are to: describe the existing environment in terms of ecological zones, vulnerable and protected species and habitats, patterns of human occupation, land use types, and cultural and religious aspects quantify, wherever possible, the natural rates of environmental change in order to identify any pre-existing trends and provide the basis for assessing the potential impact of the proposed scheme. Projection of natural trends allows predictions to be made of environmental interactions in future, and the means of protecting and improving the environment to be identified.

ENVIRONMENTAL IMPACT ASSESSMENT PROCEDURE

6.21 The main donor agencies have published their own guidelines on EIA procedure and, nowadays, most projects cannot be implemented without environmental considerations having first been satisfied. The various guidelines correspond approximately to the following schedule of EIA activity.

Those in bold type relate directly to engineering:

description of proposed project systematic comparison of alternative schemes, sites, technologies and designs identification of appropriate mitigation and compensation measures, together with an assessment of their likely success and cost, where possible

Comparison of alternatives
6.25 Investment strategies, construction and operation technologies, design standards and route corridor options will be assessed as part of the engineering feasibility and preliminary design studies, but if EIA principles are incor-

68

Table 6.1 Positive and negative environmental impacts a) during construction

69

Table 6.1 (continued) Positive and negative environmental impacts b) after construction

70

porated into the design process, an environmentally compatible design will be produced. Comparison of alternatives is usually achieved using inventories of the various impact categories identified. Some categories, such as number of buildings demolished and land use classification within the right of way, can be quantified and compared relatively easily, while others, such as cultural and ecological effects, are less amenable to objective and tangible comparison. 6.26 Some impact categories will be considered more important than others, and the allocation of a weighting scheme helps to clarify the total impact associated with each alternative. However, a weighting scheme is subjective and may provoke emotive debate. The choice between route corridors is not always clear cut, and which of the environmental factors should take precedence is often a matter of opinion. Short term economic and political reasons for favouring a route that clearly contradicts engineering judgement should be strongly resisted.

Environmental management and monitoring

6.28 Environmental monitoring is the process of measuring changes in landscape, land use or cultural activity that arise as a result of the implementation of a development project. The Environmental Statement should outline the manner in which the project will be managed within the wider context of the study area environment, enabling maximum benefits to be derived from the scheme and allowing the performance of engineering design, construction, maintenance and mitigation measures to be monitored. 6.29 Environmental monitoring may form part of an environmental impact assessment, and indeed should do so in order to verify estimates of environmental impact made during the planning stages of a project. However, in practice this rarely happens. For projects where monitoring is seen as a practicable and desirable undertaking, rates of slope movement by failure, slope lowering by erosion and channel enlargement by scour and sediment transport can be measured by simple and low cost techniques. 6.30 Data collection in areas outside the sphere of influence of the scheme may allow firmer conclusions to be drawn regarding its environmental effects. However, with the possible exception of drainage and slope stability effects, the sphere of influence may be difficult to define, and comparisons may not be entirely valid. Monitoring will usually allow undesirable effects, not anticipated during project feasibility and design, to be identified, evaluated and mitigated as the project proceeds.

Mitigation and compensation measures

6.27 Once a route has been confirmed, mitigation and compensation measures should be identified to further enhance the integration of the scheme with its environment. The decision as to whether to proceed with construction will require that these mitigation measures are advanced at least to a preliminary design stage in order to establish their feasibility, overall benefit and likely costs. Mitigation measures that form part of the environmental protection programme but which are incorporated into the engineering design may need to be brought to the attention of the client in order to substantiate the apparent extra cost and head off possible accusations of `overdesign'.

71

GEOTECHNICAL ASSESSMENT

INTRODUCTION
7.1 Geotechnical assessment comprises mapping, investigation and analysis and is central to the design and evaluation of the following: slope stability assessment, slope stabilisation and road erosion control cross-section and cut and fill slope angles sub-grade conditions stability and bearing capacity for retaining walls and foundation conditions at sites of major engineering structures materials suitability as fill, masonry, aggregate and road surfacing the identification, characterisation and quality specifications for material source areas slope drainage detailed horizontal and vertical alignment (as influenced by the above considerations).

7.4

dip of lithology dip of foliation joint orientations joint spacing joint openness weathering grade permeability estimated rock strength. Soils: origin colour horizons depth grading plasticity density strength water, including inflow. Other aspects to be considered include:

Factual information
drainage features, such as springs and see pages, sinks, stream channels and areas of wet ground natural slope angles main topographical features, such as cliffs, terraces etc land use and details of irrigation the stability of slopes, with comments on mechanism, depth, extent and rate of movement of recorded failures, where known construction material sources including field descriptions and the locations of any samples taken for laboratory analysis the location and results of any field investigation and testing sites, such as trial pits, Mackintosh probe or Dynamic Cone Penetrometer (DCP)

GEOTECHNICAL MAPPING
7.2 The team `geotechnical mapping' is used here to cover engineering geological and geomorphological mapping procedures for recording and interpreting ground conditions for engineering purposes. Detailed geotechnical mapping is usually carried out only along a chosen route corridor identified from reconnaissance survey. Mapping scales of 1:2,500-1:10,000 are most common, although larger scales (1:500 - 1:1,000) are sometimes used at sites of major engineering structures or at higher risk landslide sites. The mapping should be accompanied by a report that summarises design and construction considerations. 7.3 The aim of geotechnical mapping is to convey information on surface and sub-surface ground conditions (so far as they can be ascertained) sufficient to satisfy the requirements for either preliminary or detailed design, while avoiding expensive ground investigation. It is usually most cost-effective to combine geological and geomorphological data on the same map, as slope stability assessments normally require both kinds of data. The range of factors usually considered is listed below and illustrated in Figure 7.1. Rocks: - rock type - outcrop pattern - location and trend of major geological structures

Interpretation
recommendations for stabilisation and erosion protection, where appropriate design and construction considerations, including foundation conditions in general, preferred crosssections, bridge sites and abutment/pier foundation and stability, drainage design, river protection, construction access, pilot tracks and spoil disposal.

7.5 Teats containing standard rock and soil engineering descriptions and classifications are listed in Chapter 14

72

73

Figure 7.1 Geotechnical map for part of a proposed road alignment in Nepal

(Bibliography). No single classification system is applicable to all circumstances, and it may be that none is entirely relevant to a particular case without some modification. The classification adopted should be based on the range of conditions found, with modifications to the standard if necessary.
7.6 During construction and maintenance, geotechnical mapping can be combined with inventory surveys to further classify ground conditions for purposes of hazard assessment and slope management. An example from Papua New Guinea is given in Box 7.1.

that the probe readings relate to actual ground conditions. For example, an apparent increase in soil strength may be due to the presence of a boulder, or an increase in the stone content with depth, rather than a genuine increase in soil density. 7.10 In clayey soils (uncommon in mountain areas, except under localised residual weathering conditions), peak and residual undrained shear strength can be determined directly from vane shear apparatus and hand penetrometer, although, as with the probes mentioned above, the presence of coarse particles in the soil will significantly affect the results.

GEOTECHNICAL INVESTIGATION
7.7 On low cost mountain roads, geotechnical investigations are usually limited to shallow sub-surface exploration using manual excavation and hand-portable equipment. Specialist drilling investigations are usually only cost-effective and practicable at major bridge sites, although they may have some application to high risk landslides along existing roads where access is comparatively straight-forward. The practical depth of hand-dug trial pits is limited by the increase in ground strength with depth, normally to around 2m. In some soils, 5m can be achieved, although shoring will be required to keep the pit open and maintain the safety of the pitting crew. Care must be taken in locating and digging deep pits in colluvium and unstable ground, particularly in the wet season or if the pit encounters ground water. 7.8 In addition to geological and geotechnical logging, the following data should be recorded during trial pitting: the location of the pit, sufficient to relate to the centreline if known, by grid co-ordinates and elevation, or by position from a temporary bench mark or permanent ground marker the date and weather conditions prior to and during excavation the depth and horizon from which soil samples are taken

Trial earthworks
7.11 There is often advantage in excavating small trial cuttings which can be left open for one or more wet season(s). This technique is particularly valuable in areas of complex geology and weathering profiles, and allows deeper investigation from a working platform than from the confined area of a trial pit. The depth of a trial cutting should be similar to the expected design depth of cutting, and it should be at least as long as it is deep. The cutting face is cut to an angle considered appropriate for the material exposed. The stability of the cut slope is then logged and monitored. It is important to ensure that the trial cutting and generated spoil do not themselves lead to instability or erosion. Trial cuttings are probably best undertaken when the contractor is on site and is available to contain any such occurrence. 7.12 Similarly, trial embankments and trial compaction tests can be undertaken with control on materials, layer thicknesses, water content and compaction levels in order to compare performance with laboratory test data, and to better define specification standards to suit the conditions found on site. However, these trials require specific contractual arrangements that may be unavailable within a low cost road construction programme.

Sampling and laboratory testing

7.13 The standard procedures for soil and rock sampling and laboratory testing are well known and do not require elaboration here. However, the observations listed below are especially relevant. 7.14 For routine geotechnical design it is usual to cant' out index and classification tests on disturbed samples. Where material is to be considered for use as construction material, then compaction and strength (CBR) testing may also be required. CBR samples should be tested in a soaked condition for areas of high rainfall and/or high groundwater. 7.15 Rock exposed at outcrop or in shallow trial pits is generally more weathered than rock at depth. Thus, an assessment of rock for masonry and crushed aggregate

7.9 Hand-operated probing equipment, such as the DCP and the Mackintosh probe, can be used to obtain estimates of soil strength near the surface based on experience with the soil being investigated. The DCP is a useful means of measuring in situ CBR, while the Mackintosh probe can be correlated to Standard Penetration Test (SPT) N values to assess relative densities in the same soil types. The DCP can only be used to a depth of 1 m, unless it is advanced by trial pitting, while in soft ground the Mackintosh probe can be used to a depth of 6 or 7m. However, both probes are of limited use in stony ground typical of most colluvial soils that mantle mountain slopes. If they are to be used in these areas, it is advisable to excavate some trial pits to confirm

74

BOX 7.1 Geotechnical inventory and slope hazard / risk classification during and maintenance: an example from Papua Guinea

75

made during walk-over survey may underestimate the strength and durability of material at formation level or when excavated from a quarry. 7.16 The requirement to carry out undisturbed sampling from trial pits is usually limited to the rare occasions when the result of the analysis is likely to cause a modification of the design. The engineer must have confidence in the sampling procedure, the test method and the capability of the laboratory to carry out the test properly. The practical problems in obtaining undisturbed samples from coarse-grained colluvial soils or partially saturated residual soils, in remote areas especially, are usually prohibitive. Undisturbed sampling from borehole investigations allows a much wider range of tests to be carried out, particularly with regard to shear strength and compressibility. However, core recovery in soil or weathered or fractured rock will be low or nil, unless sophisticated methods are employed. Undisturbed rock samples are generally only required for foundation design at major bridge sites or at the sites of high risk slope failure. 7.17 If good core recovery is obtained, it can provide a wealth of information on drainage, rock strength and weathering, fracture spacing and orientation, faults, crushed rock zones and shear zones. However, careful logging by geotechnical personnel is usually required in order to differentiate between different soil and rock types in areas of complex geology, including the identification of thrust materials, landslide

boulders are absent from the profile. The presence of large boulders complicates the interpretation of results.

7.20 Although, geophysical survey generally has little to offer low cost road design in mountain areas, there may be some potential for its uncorroborated use in situations such as: the detection of the boundary between comparatively fine-grained colluvium and an unweathered rock head surface the definition of distinct sequences of river deposits the detection of underground cavities and other karst features developed in calcareous rocks and deposits.

Logging of excavations as construction proceeds 7.21 The properties of materials can only be properly evaluated during construction. As excavation proceeds, observed ground conditions should be compared with those assumed in the design, and any necessary modifications implemented. For instance, where a soil profile is more weathered than originally thought, or where materials are found to be colluvial rather than residual, it may be necessary to reduce the cut slope angle, deepen a foundation excavation or widen a wall cross-section to accommodate the weaker materials. 7.22 Some geotechnical parameters that may not be fully evaluated until construction is underway are listed in Box 7.2.

colluvium, rock avalanche materials and crushed bedrock.

Geophysics
7.18 Geophysics is used to detect variations in seismic velocity and other physical conditions brought about by changes in strata or groundwater and the presence of underground cavities. The principal advantage of a geophysical survey is that it can record these conditions from the ground surface at relatively low cost compared with conventional ground investigation. Its main disadvantage is the difficulty associated with the interpretation of complex ground conditions, especially where subtle changes may go unrecorded. Geophysical data should be supported by borehole information to confirm the results and assist in their interpretation. 7.19 Seismic refraction is generally the most frequently used geophysical technique for engineering applications but it will be reliable only if: seismic velocity is constant, or increases with depth. This is not always the case in landslide materials, where there may be a low velocity along the slip surface there is a distinct boundary and a marked difference in velocity between the layers being investigated

GEOTECHNICAL ANALYSIS AND DESIGN Introduction

7.23 In the context of mountain road design, geotechnical analysis is used principally in evaluating slope stability for the design of earthworks and slope stabilisation schemes, and in the assessment of stability and bearing capacity for retaining wall foundations. This section deals mostly with slope stability aspects of geotechnical analysis; foundations for retaining walls are discussed in Chapter 11.

Assessment of soil strength parameters

7.24 In order to carry out any analysis, it is necessary both to assess the strength and depth of the materials forming the slope and the most likely worst-case groundwater and soil moisture conditions. Differentiation between colluvium and residual soil is of major importance owing to the differing strength

characteristics of these two soil types.

7.25 Colluvial soils are usually coarse-grained, and their approximate strength and other geotechnical characteristics can be assessed either from the results of SPT or other

76

Box 7.2 Geotechnical evaluation cannot normally be finalised until construction Geotechnical factors that are usually not fully evaluated until construction is under way include: soil depth and weathering grades at the soil/rock head interface the location of weaker and more weathered zones mineralogy (the presence of mica, chlorite and graphite, for instance, may increase the potential for failure) the distribution and orientation of rock discontinuities a classification of rock fabric microstructure and rock mass macrostructure the presence of pre-existing shear surfaces (although these may be difficult to identify even in excavations) soil moisture and drainage conditions bearing capacity and mobilised shear strength of insitu soils sub-surface drainage conditions and groundwater fluctuations suitability of excavated materials as aggregate or masonry susceptibility of soils to erosion following disturbance by earthworks and artificial slope drainage susceptibility of marginally disturbance by construction. stable slopes to

7.26 Residual soils, usually containing a higher proportion of clays, often occur on flat ridge tops and structural benches where in situ weathering has predominated over erosion and leaching of clay minerals. Unlike colluvium, these soils have not been remoulded and can exhibit an additional cohesion associated with the residual rock structure, thus increasing their strength. In these soils the PI/ relationship may not apply, and values so derived would be conservative in terms of cut slope stability. 7.27 Soil suction can play an important role in controlling the stability of slopes, enhancing the shear strength of the soil by an apparent cohesion. This is more pronounced in the generally finer grained residual soils than coarser colluvial materials. An increase in degree of saturation, by infiltration during rainfall, by seepage, or by a general rise in groundwater level, will lead to a reduction in soil suction and a corresponding decrease in strength and stability. Road construction can significantly alter the pattern of slope drainage and soil saturation, thereby modifying the susceptibility of adjacent slopes to failure. Soil cementation, caused by the secondary deposition of iron oxides and carbonates in groundwater can also significantly increase soil cohesion. 7.28 Any assessment of soil strength based on empirical correlations with grading, PI or probe results, must involve a degree of uncertainty. Frequently, a factor of safety of unity or less for apparently stable slopes can be obtained from stability analysis using strength parameters derived in this way. This may be due to a number of effects, including soil heterogeneity, or the suction, cohesion or cementation effects outlined above not being fully appreciated. These factors can be assessed by back analysis of existing slopes, preferably those that have suffered failure. Slope stability analysis Methods of analysis 7.29 Essentially, there are three methods that may be used to analyse a slope: assumption of an infinite slope slope stability charts computer programs

precise drainage and land management requirements where alignments pass through areas of irrigated terracing costs and performance of alternative forms of slope stabilisation and erosion protection works.

probe tests, or from their particle size. In fine-grained (usually weathered) colluvial soils, index tests (Atterburg Limits and natural moisture content) can provide a general indication of shear strength; a relationship between the effective internal angle of friction (') and Plasticity Index (PI) for remoulded clays can be used to estimate 0'. In mountainous areas, clayey colluvial soils are generally of very low plasticity with values of ' in excess of 30.

7.30 The infinite slope model allows planar failures to be analysed quite readily without recourse to computer programs, by assuming that in purely frictional soils (no cohesion) the shear strength can be represented by a single value of '. At the point of failure, when the factor of safety is equal to unity, the following condition applies:

77

tan.(1-ru.sec2.) = tan: where ru = pore pressure ratio (water pressure/total overburden pressure) internal angle of friction under effective stress slope angle at failure.

= =

computer allows the precise ground slope profile to be incorporated, together with any variation in groundwater profile. Thus, the individual slope characteristics, if known, can be accurately represented. Programs also allow a large number of potential slip surfaces to be considered rapidly; some allow random generation of non-circular surfaces, the most prevalent form of failure on steep slopes formed in granular and heterogeneous soils. 7.36 Computer programs allow the use of either a specified groundwater level or an ru value, or both. The assumption of a single groundwater level, above which all slope materials are dry is an oversimplification in most situations. Some computer programs enable perched water tables to be analysed, that represent seepage horizons superimposed on a deeper water table. A reasonable representation of groundwater conditions can be approximated by combining observed or assumed water tables with a knowledge of sub-surface variations in geology and soil permeabilities. 7.37 The concept of ru was first proposed when the majority of design was undertaken using charts. The use of a single ratio between water pressure and total vertical earth pressure at every point along a slope is a crude model of the actual variation in water levels, although it may be the most appropriate means of modelling water pressures where see pages are giving rise to more localised partial or total soil saturation, or where groundwater is not hydrostatic. Furthermore, computer modelling provides the means by which a slope cross-section is divided into discrete zones, each with its own set of soil parameters, including ru . Nevertheless, whilst there may be no choice but to use ru where no other information on groundwater is available, use of an observed or assumed water profile, perhaps parallel to rock surface, is usually preferable. Back analysis 7.38 Analysing a slope by varying soil strength parameters and groundwater levels within plausible limits to obtain a calculated factor of safety of unity, enables an appreciation of likely in situ strengths and water levels which can be used in design. Identification of unfailed bedrock levels assists in limiting the extent of trial failure surfaces and the range of failure mechanisms considered in the analysis. Parameters so derived can then be used for the design of new slopes, or for examining the stability of existing hillsides. 7.39 In steep terrain, it is possible to assume that an intact slope has a factor of safety close to unity and to derive soil parameters, but the calculated values would be conservative as the actual factor of safety must be higher than assumed (but to an unknown degree). Nevertheless, the derived parameters can often be greater than would otherwise be obtained from testing or empirical assessment, and can lead to more economic

7.31 The value of ru is zero when groundwater is below the slip surface and approximately 0.5 when at the slope surface. Intermediate groundwater levels, where known, can be assigned intermediate ru values accordingly. Thus, equates approximately to the natural slope angle at failure in the absence of water, and twice the natural slope angle at failure for water at the slope surface (the importance of identifying groundwater levels in the slope can be readily appreciated). However, it is rare that the profile of the slope can be approximated to a uniform gradient and it is common to find surface soils failing on a shallow bedrock surface parallel to the ground slope. This method therefore has limitations. 7.32 Stability charts, relating soil strength parameters, ru and factor of safety to slope height, were widely used before computers became more widespread, and can be useful for rapid assessment of stability. An assumption has to be made both on slope height and ru. value, both of which can be difficult in steep and complex topography. For instance, the ground slope above a proposed cutting is very rarely flat (as is assumed in most stability charts) and it may be necessary to assume a greater height of cutting at an intermediate cutting angle. Sensitivity analyses will be necessary to identify the worst case combination of geometry and ru. Charts are available for varying groundwater profiles rather than ru values, and these can be useful in some cases (TRL. 1997). 7.33 Design charts require the following parameters to be known or assumed: effective cohesion (c) effective internal angle of friction () slope height (H) groundwater conditions (based on ru ) required factor of safety.

7.34 Where a slope geometry or groundwater profile cannot be suitably represented by chart analysis, recourse will have to be made to computer programs. 7.35 A large number of computer programs are now available to determine slope factor of safety. The use of a

78

7.40 To back analyse an intact slope, it is necessary to assume the potential failure surface. This is greatly helped if the depth to rockhead or other competent material is known, either by mapping or by investigation. It is then necessary to undertake a sensitivity analysis to assess the most likely failure mechanism. The ranges in both groundwater levels and soil strengths also need to be assessed. Information on groundwater levels from standpipes installed in boreholes is useful, but it is rare for worst-case groundwater levels to be recorded, unless the monitoring periods spans a number of wet seasons. Worstcase groundwater levels, therefore, usually have to be assumed. 7.41 In the case of failed slopes, there is no doubt as to the slope having reached and passed the point of limiting equilibrium. However, difficulty remains in determining the failure surface and the groundwater conditions at the time of failure, particularly where stability is immediately improved by a release of groundwater or a significant reduction in slope angle when coming to rest. Ground investigation (boreholes, trial pits and trenches) may allow soil type, soil horizons and depths to rock head to be determined, but unless failure has occurred on an obvious surface, or variations in rock strata or foliation can be used to infer failure depth, it can be difficult to determine the exact depth of failure from ground investigation data alone. Furthermore, an estimate has to be made of the likely range in water level, with a considered view taken as to the probable water profile upon which to base the analysis. These considerations are illustrated in Box 7.3 with respect to recent geotechnical investigation, analysis and design in Nepal. 7.42 Where a slope failure continues to have a factor of safety close to unity, it will normally continue to fail on a regular basis, and consequently failure depth and groundwater conditions can be modelled more closely.

affect the final choice of parameters. 7.45 Although a computer program may generate failure scenarios and factors of safety that appear sensible, it is extremely important to ensure that ground conditions at failure suggested by the computer give a realistic picture of those that exist at the site. This may seem an obvious observation, but only too often a computer result is taken to be infallible, where in truth the conditions that it has modelled bear little relation to actual ground conditions. 7.46 Difficulty often lies in the choice of factor of safety to be achieved in the design. As the majority of adjacent slopes are likely to be near to a factor of safety of unity, it may be difficult to justify factor of safety values much greater than this, as might be expected in less severe terrain. Also, it is frequently very difficult to achieve high factors of safety without prohibitively expensive engineering works. 7.47 The choice of factor of safety will depend upon the reliability of the soil strength data and groundwater profile used, together with the consequences of failure should it occur. It is also dependent on whether the slope parameters chosen are average, conservative or worst credible. Each slope must be considered

as an individual case. A factor of safety of 1.1 is reasonable for most lightly-trafficked roads, where the consequences of slope failure are relatively low. Where the consequences of failure are high, such as in the situation where road loss could cause significant traffic disruption, then a factor of safety of 1.2-1.3 might be more appropriate. The philosophy of slope design and the choice of factor of safety is discussed further
with respect to cut slopes in Chapter 9.

Empirical methods
7.48 An examination of the relationship between the geometry, groundwater and materials characteristics of natural slopes can enable stable and unstable slope scenarios to be identified. This information can then be used to assess the stability of both new and existing slopes. 7.49 Table 7.1 is based on observational data from Nepal, and shows the range of limiting slope angles for the common soil types found there. The dry and wet cases for colluvial silts and sands correspond approximately to the ru = 0 and ru = 0.5 conditions respectively, mentioned earlier. The limiting slope angles for residual soils and sands are significantly higher, due to the presence of cohesion or suctions. 7.50 Near-surface throughflow, often induced by a shallow rockhead surface, can give rise to locally perched groundwater and ru values of 0.5. Mudslides in clayey silt residual soils have been found to fail across smooth bedrock surfaces on slopes as low as 12.

Slope design
7.43 Some form of analysis will be required for routine checks on the stability of earthworks, with more comprehensive analysis required where deep cuts or landslide stabilisation are envisaged. Vegetation can have a significant influence on the factor of safety, but its effects are not easily quantifiable and it is safer to exclude them from design calculations, unless planting or revegetation is specifically designed to achieve an acceptable factor of safety. 7.44 In carrying out this analysis, it is prudent to test the sensitivity in the factor of safety to small variations in the value of soil strength parameters and water levels. This will identify whether a particular slope design is more sensitive to some of the assumptions made than others, and it may

79

Box 7.3 Examples of geotechnical investigation, slope stability analysis and design from Nepal

80

Table 7.1

Range of limiting angles (in degrees) for natural slopes (based on observational data from east Nepal)

Rock slope stability analysis and design

7.51 The analytical and design approach adopted for rock slopes is dependent upon depth of cut, a qualitative assessment of its stability, and the consequences should failure occur. For cut slopes less than 10m in height it is usual to rely on engineering geological judgement and observational data. Table 7.2 has been compiled from cut slope inventory data gathered in east Nepal, by way of illustration. 7.52 For deeper rock cuts and unstable rock slopes, or those where the consequences of failure are severe, engineering geological mapping of rock type, rock discontinuity orientation, waviness and roughness, spacing, persistence, size, infill, rock weathering grade, groundwater and see pages will be required to build up a picture of the structural and mass strength of the rock slope. Identification and recording of existing rock failure mechanisms on surrounding slopes can give an indication of the most likely mode(s) of failure.

7.53 It is usual to analyse recorded discontinuity orientations by means of stability charts and sternest, and usually with the aid of a computer database and graphics. Stereonet analysis will indicate the most likely joint orientations upon which failure will take place. Although planar slides and rock topples are frequent mechanisms of rock slope failure in mountain areas, it is usual for a stereoplot of joint orientations to indicate that wedge failure, along intersecting joint surfaces, is one of the most likely mechanisms by which a slope will fail. The results of the stereonet analysis can then be used in conjunction with field data to judge the likelihood of failure in excavations cut to different slopes and orientations, and the possible extent of failures should they occur. Separate analysis can be carried out for different cutting angles. 7.54 Depending upon the consequences of failure, and the confidence placed on the analysis, further geotechnical investigation and slope stability analysis may be required. However, due to the costs involved to reducing geotechnical uncertainties, sophisticated methods of analysis and design are comparatively rare on low cost mountain roads.

Table 7.2

Range of maximum slope angles for cut slopes in rock (based on observational data from east Nepal)

81

HYDROLOGY AND HYDRAULIC DESIGN

INTRODUCTION
8.1 Hydrological analyses of rainfall and runoff data are required to establish relationships between flood flow and flood frequency for hydraulic design, for which the following steps are taken: classification of risk elements within the road drainage system (road side drainage, small culverts, large culverts, short span bridges and major bridges) allocation of a conceptual design life for each risk element based on the accepted probability of the element not being surcharged or overtopped during this period calculation of the discharge of the flood that corresponds to this return period.

the very localised nature of orographically-controlled rainfall can mean that a high magnitude storm may occur over one catchment, while an adjacent catchment receives little or no rain. The recurrence interval of the high magnitude storm may be long, but the following year a storm of similar size could occur over a neighbouring catchment. The recurrence interval of these storms, as measured by individual raingauges, could be correct (ie each raingauge might record a one in twenty year storm on average every twenty years), but a road traversing these catchments would experience both storms and an apparent decrease in recurrence interval (increase in storm frequency) over its length. This introduces the complex subject of combined probability. The combined probability of flood damage or road closure at any one of many crossdrainage structures is far higher than the probability at any individual site rainfall intensity will vary significantly throughout the duration of a storm, and between one catchment and the next, while floods of similar volumes in the same river can be accompanied by markedly different velocity distributions and hence scour potential due to variations in the flood hydrograph or changes in channel cross-section. Rainfall intensity and maximum flow velocities are among the most important parameters in hydraulic design, and yet are the most difficult to evaluate in terms of design life drainage structures in mountain regions will have to resist scour and transmit debris every year and failure to do so will damage or destroy them more effectively than a hydraulic surcharge. It is apparent from experience in mountainous terrain that culverts and bridges are rarely subjected to surcharge or overtopping before the foundations are undermined by scour or the waterway is blocked by debris, both of which can occur during the course of a single flood due to variable rainfall-runoff relationships, a storm of a given recurrence interval will rarely give rise to the same recurrence interval flood.

8.2 Unfortunately, in most remote areas the data needed to carry out these calculations are often extremely limited in terms of geographical spread and length of record. Rainfall and flood frequency analyses must be combined to yield an aggregate assessment and corroborated with field evidence, where possible, before the results can be confidently used for hydraulic design.

DESIGN LIFE
8.3 An evaluation of what is an acceptable design life for each category of road drainage structure will allow investment in construction design, maintenance and eventual replacement to be apportioned in the most cost-effective manner. In the case of side drains, culverts and short sections of road in flood-prone areas, this usually comprises 1-10 year events that are accommodated in design, 10-20 year events that are accommodated in design and maintenance, and 20-50 year events that are usually accommodated by reconstruction. 50-100 year events and greater are only accommodated in design by major structures and by the rest of the road where possible in adopting the safest available alignment. Defining the conceptual design life of any given hydraulic structure is comparatively straightforward; the main problem concerns the identification of the storm or flood magnitude with the equivalent recurrence interval. 8.4 Even if a reliable set of flood magnitude-recurrence interval relationships can be developed, there are other factors to consider: the very nature of probability theory with respect to flood-frequency (Table 8.1) means that the structure may not receive its design discharge throughout its life, or it may receive a large excessive discharge in the first year following construction

8.5 The concept of a design life for a hydraulic structure is therefore difficult to apply or to achieve in practice when there are so many other factors to consider. Nevertheless hydrological analysis should proceed through some and occasionally all of the following logical stages: choice of design life magnitude - frequency analysis of rainfall data rainfall - runoff conversion review of regional flood formulae flood frequency analysis slope area method and field verification.

82

Table 8.1

Percentage risk of encountering an event within a particular design life for different return periods

Design Life (L) (years) 5 20 36 49 67 79 89 96 99 10 10 19 27 41 52 65 79 88 96 99 -

Return period T (years) 20 5 10 14 23 30 40 54 64 78 92 98 99 30 3 7 10 16 21 29 40 49 64 82 92 97 99 50 2 4 6 10 13 18 26 33 45 64 78 87 95 98 100 1 2 3 5 7 10 14 18 26 39 53 63 78 87 95 99 200 1 1 2 3 5 7 10 14 22 31 39 53 63 78 92 99 500 1 1 2 3 4 6 9 14 18 26 33 45 63 86 1000 1 1 1 2 3 5 7 10 14 18 26 39 63

1 2 3 5 7 10 15 20 30 50 75 100 150 200 300 500 1000

Note: Percentage risk is calculated from R = 100 (1 - (1 1) ) T 8.6 Despite these logical steps in the hydrological analysis, flood frequency records, if they exist, are likely to form the most reliable basis for design. hydraulic structures further downstream in the drainage system will be minimal. The river flow regime at these larger structures will be determined instead by the rainfall pattern in the larger catchment or hydrological region as a whole. Therefore, as long as a regional network of topographically-representative rain gauges exists, it is length of record, data validity and recording interval, that become the chief concerns. 8.9 Most raingauges in remote mountain areas are operated manually and record rainfall on the basis of 24 hour totals only. 24 hour data provide little indication of rainfall intensity or the duration-frequency characteristics of intense rain that are most important for hydraulic design. For instance, the same 24 hour rainfall depth can indicate: prolonged rainfall of moderate intensity short duration rainfall of high intensity a portion of a total storm rainfall that was still in progress when the raingauge was read.

RAINFALL ANALYSIS
8.7 Although rainfall is usually recorded far more frequently than river flow, the distribution of raingauges in mountain areas is often grossly inadequate to record anything other than the regional pattern in rainfall variation. Even so, a raingauge spacing of as little as 1km may still be inadequate to record storm rainfall variability in some areas, brought about by orographic and rain shadow effects. Usually, the only practicable approach will be to identify broad areal groupings of rainfall data in order to classify hydrological regions. This regionalisation of rainfall for design purposes is not as great a problem as might at first appear, because the economic implications for underdesign (or overdesign) increase as the size (and usually catchment area) of a given hydraulic structure increases. The fact that a dense network of raingauges might not exist, therefore, may not necessarily adversely affect the validity of the hydraulic design. 8.8 Localised intense storms, giving rise to the surcharge of one or a small number of culverts, are bound to occur from time to time. These storms may have significant effects in individual catchments, but their impact on large

8.10 This problem can be partially overcome by using rainfall ratios or by comparing 24 hour rainfall data with the intensity and duration relationships obtained from continuously recording raingauges nearby or in the same hydrological region (see below). The usual procedures for analysis of both 24-hour and continuous rainfall data are summa-

83

rised in Figure 8.1. The method by which recurrence interval is calculated from 24-hour rainfall records is outlined in Box 8.1. Selection of the 24-hour annual maximum rainfall may exclude a number of other large storm rainfalls from a particularly wet year that may not be repeated during following drier years. This can be overcome by analysing partial series data that include rainfalls larger than a certain threshold. 8.11 Continuous rainfall data are usually analysed and presented in the form of intensity-duration-frequency curves, such as those shown in Figure 8.2. The rainfall intensity with a 24 hour duration can be calculated for each recurrence interval and compared with the corresponding intensities derived from the 24 hour annual maximum regression. An adjustment factor can then be used to derive intensity for durations of less than 24 hours from other daily rainfall in the same hydrological region. 8.12 In the absence of any continuous rainfall data, it may be necessary to assume an effective rainfall duration for hydraulic design purposes and refer to published rainfall ratio tables to identify suitable values for the constants contained in the following equation for rainfall ratio over a period (RRt):

Box 8.1 Calculation of storm recurrence internal from 24-hour rainfall data At least 10 years of rainfall record is generally required for annual series analysis. The highest 24-hour rainfall recorded in each year of record are ranked in order of magnitude and a recurrence interval is calculated for each rainfall based on the following formula: T = (N+1)/m where T is recurrence interval in years N is the number of years in the series, and m is the rank of each 24-hour record. The probability of a rainfall with a given recurrence interval occurring at least once in N years is expressed by the equation: P = 1-qN where q is equal to the probability of the rainfall not occurring in any particular year.

where b and n usually range between 0.2 and 0.5 and 0.5 and 1.0 respectively in the humid sub-tropics. 8.13 Design rainfalls derived in this manner must be treated with extreme caution and it is probably safer to combine results with worst case, short-duration rainfall intensities based on experience from other regions with similar climatic conditions. For example, an intensity of 120mm/hr for a 30 minute duration is often considered to be acceptable for the design of culverts for small steep catchments.

RAINFALL - RUNOFF ANALYSIS

8.14 Figure 8.3 illustrates common procedures for design discharge modelling. The simplest and most frequently used rainfall-runoff technique is the Rational Method: Q = 0.277CIA where Q is peak discharge in cumec C is runoff coefficient I is rainfall intensity in mm/hr A is catchment area in km2 0.277 is an arithmetic conversion factor that allows the above parameters to be expressed in the conventional units.

The recurrence interval (T) is plotted on a logarithmic scale against maximum 24 hour rainfall on a normal scale using Gumbel extreme value type 1 or Pearson type iii paper. Linear regression is then used to define the relationship between recurrence interval and peak 24 hour rainfall. It is worthwhile tabulating the 2-1 hour rainfalls calculated for each raingauge with 2 year, 2.33 year (approximate average annual rainfall), 5 year, 10 year, 20 year, 50 year and 100 year recurrence intervals. The mean and standard deviation between raingauges will assist in defining hydrological regions within the study area. 8.15 Because of the simplifying assumptions on which it is based, the Rational Method is most applicable to small catchments (usually of the order of 1km2). This size limitation is especially relevant to mountain areas where storm rainfall can be unevenly distributed within catchments much larger than this, and thus the runoff response is not catchment-wide, as the model implies. The value of rainfall intensity used in the calculation should not be the instantaneous peak, but the maximum average sustained during a period equivalent to the time to concentration of the catchment. The catchment time to concentration is defined as the time required for rainfall falling in the remotest part of the catchment to reach the catchment outlet, or culverting point, and can be calculated from one of a number of equations. It is typically well below 1 hour for very small, steep catchments. It is the variability in C, both in time and over short catchment distances, together with difficulties in determining design rainfall intensities, that can result in as much as a 50% error in design discharge computation.

84

Figure 8.1 Outline procedure for rainfall analysis

85

86

Figure 8.2 Intensity duration frequency curves from east Nepal

Figure 8.3 Commonly applied procedures for design discharge modelling

87

8.16 The runoff coefficient, C, is all-encompassing and, in theory, should embrace the following factors: catchment slope catchment shape soil type, depth and permeability bedrock permeability groundwater levels, antecedent moisture conditions and base flow vegetation cover and land use effects drainage density and hydraulic conductivity catchment storage rainfall intensity.

mountain catchments ought to be 0.7 and that, from the point of view of design, near 100% runoff should be expected in the most steep and rocky catchments during extreme rainfall. 8.19 Other models have been developed using the same basic concept as the Rational Method, but with the inclusion of several other indices to take greater account of rainfall variability and catchment characteristics. The Gupta Modified Rational Method and the Generalised Tropical Flood Model are summarised in Box 8.2. Generally, however, the Rational Method is the preferred approach for small catchments under the usual conditions of limited rainfall and catchment data. 8.20 Where the requisite data exist, storm flow modelling by hydrograph analysis is the preferred alternative to the Rationaltype approach. The Unit Hydrograph method relies on the derivation by continuous flow gauging of a flood hydrograph generated by a short duration, well defined, single-peaked storm uniformly distributed over the catchment concerned. This hydrograph then reflects the combined physical characteristics for the catchment, and the ordinates of future hydrographs with a common base time will be directly proportional to the total amount of direct runoff (excluding base flow). The principal advantage with the technique lies in the ability to model peak discharge more closely in catchments with similar runoff regimes to that in which the Unit Hydrograph was derived. Usually, however, due to lack of data in most remote mountain areas, the technique has limited potential application for hydraulic design.

8.17 It is obviously inconceivable for all, or even any, of these factors to be adequately represented by a single runoff coefficient. Thus, values of 0.7 or 0.8 are commonly used by default in mountainous areas. Nevertheless, a comparison of rainfall and runoff data can allow suitable values or ranges of C to be identified for different catchment types. 8.18 Table 8.2 summaries C values derived by `backanalysis' of storm runoff from small catchments in east Nepal. Although the values are only intended as a guide, they indicate that the minimum value of C for small

Table 8.2 Runoff coefficients based on data from east Nepal

Predominant land use Agricultural terrace (large storage capacity) Mixed terrace scrub and forest Forest (usually steep rocky slopes, thin soils)

Shallow (<25) 0.5-0.6

Moderate (25-35) 0.6-0.7

Steep (>35) Not represented

0.6

0.7-0.8

0.9-0.95

Not represented

0.8-0.9

0.9-1.0'

Note 1. Steep, saturated, rocky catchments

88

89

REGIONAL FLOOD DATA AND FORMULAE

Published area-specific discharge data
8.21 Maximum recorded area-specific discharge data. expressed in cumec/km2 of catchment are frequently published in the hydrological and geomorphological literature, and can provide useful additional information with which to formulate an aggregate overview of discharge potential. They should not, however, be used in isolation, especially as they usually offer no reliable indication of recurrence interval.

hydrological regions for which they were derived. This is especially important in sub-tropical mountains, where specific water yields are often higher than anywhere else in the world. As stand-alone techniques of flood assessment for hydraulic design, they should be used at the feasibility stage only with adequate safety margins. For detailed design they should only be used in a corroborative capacity. A selection of empirical equations developed in the Himalayan region is listed in Table 8.3. 8.24 The Dickens equation is probably the most commonly applied of empirical techniques, although it is often correlated with the Probable Maximum Discharge or other extreme floods that are associated with too high a recurrence interval for the design of most road drainage structures.

Empirical discharge-area equations

8.22 These equations are based on maximum recorded or maximum probable discharge. They take the general form of: Q = CAn where Q is peak discharge C is a runoff coefficient A is catchment area n generally ranges between 0.7 and 0.8, and is dependent on recurrence interval. 8.23 As these formulae equate discharge directly with catchment area, and do not take rainfall depth or intensity into account they should strictly only be applied in those

Flood envelopes
8.25 Flood envelopes are obtained by plotting maximum recorded discharge against catchment area on logarithmic graph paper and drawing a line that bounds the upper plotted points (Figure 8.4). Maximum recorded discharges within the same hydrological region are usually obtained from river gauging records, although it may be possible to supplement these data with field assessments of maximum discharge based on the slope-area method. These envelope equations are no different in their method to the empirical equations described above. Although some indication of recurrence interval may be possible from flood frequency data and historical accounts, it will generally not be possible to assign a recurrence interval to a flood envelope, as it represents a number of extreme flood events from different catchment areas. Flood envelopes based on historical floods

Table 8.3 Some empirical discharge-area equations from the Himalayan region

90

Figure 8.4 Envelope curves for maximum recorded floods in Nepal (Source: SMEC, 1993)

Figure 8.5 Flood frequency distribution for the Bagmati river, Nepal (Source: SMEC 1993) 91

tend to be conservative in their estimate, although large floods may be becoming more frequent due to land use and climatic change, or their frequency may have been underestimated from short term records.

terraces, trees, large boulders and houses, bearing in mind that night-time flood peaks are unlikely to be observed with any accuracy unless they affect occupied dwellings stranded sediments and vegetation on river banks, valley-side slopes and in neighbouring trees or bushes, although drift wood is usually quickly removed for fuel in populated areas erosional features, including scoured vegetation lines and indentations on river banks and valley slopes.

FLOOD FREQUENCY ANALYSIS

8.26 Where available, selected river gauging data provide the most accurate assessment of flood frequency for design purposes (Figure 8.5). Nevertheless, it is worth bearing in mind that the accuracy of flow depth measurements, and hence discharge calculations, during floods is often only described as `poor' or `fair' by gauging authorities due to the difficulty in reading graduated staffs during heavy rain, and frequently at night. Furthermore, flood flows will contain significant volumes of debris which will increase the apparent water discharge above that due to runoff alone. 8.27 Annual flood series analyses of flow records not less than 10 years duration can be undertaken to establish regression equations relating peak discharge to recurrence interval (eg Q2.33 (approximately mean annual flood), Q5, Q10, Q20, Q50 and Q100) for each gauging station. A comparison of discharge ratios, such as Q10:Q2.33, will provide a further test for homogeneity between catchments and flow records. Average longer period discharge ratios (eg Q50/Q2.33) can be calculated for catchments with long flow records and extrapolated to shorter period flow records from catchments within the same hydrological region. Where these catchments have no flow records at all, the technique of flood transposition can be applied if catchments are sufficiently similar with regard to rainfall, topography, soils and land use. The formula is written as: Q1 = Q2.(A1/A2)n for a given recurrence interval flood, where n varies between 0.8 and 0.5 for catchment areas of between 100km2 and 1,000km2 respectively. 8.28 The ratio A 1/A2 should not exceed 2 or be less than 0.5, as the case may be. When discharge data are available for more than one catchment area then the transformation calculation should be repeated to derive upper, lower and mean discharge estimates.

8.31 Once the peak flood cross-section is defined, peak velocity is usually calculated by reference to flow roughness equations, and traditionally the Manning formula has been the most popular: V = R 2 / 3.S 1/ 2.n-1 where V is peak velocity in metres per second R is hydraulic radius of flood channel in square metres S is water surface slope in percent n is an all-encompassing roughness coefficient. 8.32 This and similar roughness equations provide a rapid and useful means of flow computation in comparatively wide and uniform trunk rivers where channel cross-section and flow regime are controlled more by hydraulic considerations than by geology. However, a number of considerations should be borne in mind (Box 8.3) regarding their use in steep mountain streams with coarse bedload. 8.33 Published values of Manning's n for mountain streams and flood plains are reproduced in Table 8.4. The most effective way to determine a representative value for n is to back-analyse from measured velocity by flow monitoring. Even then, the value of n will vary according to the volume and size of sediment transported, as well as changes in channel and flow configuration during individual floods. Backanalysed n values for small mountain streams in Nepal varied between 0.05 and 0.2. 8.34 The Stickler method, which offers an alternative approach to the calculation of n from the median size of the bed material, is of little value in steep and irregular mountain streams because it ignores the effects of other forms of roughness. In steep, boulder-filled streams, the Strickler equation was found to estimate velocities that were, on average, 40% higher than those derived from a constant n value of 0.045. When applied to a 100m-wide, active flood plain with coarse bed material the equation yielded an n value of 0.020.03. A value closer to 0.1 is considered to be more appropriate. Obviously, the performance of any one

SLOPE-AREA METHOD
8.29 The identification of flood levels from site inspection allows peak discharge to be calculated from the cross-sectional area of stream channels combined with the maximum velocity inferred from a suitable open channel flow equation. This method of peak discharge assessment has fairly standard application at bridge sites, where it is important to augment design discharges based on hydrological modelling with actual field observation. 8.30 In the absence of flow gauging data the following evidence can be taken as indicative of previous flood levels: accounts of flood height given by local people with respect to readily identifiable features, such as river

92

Box 8.3 Use of the Slope-Area Method for discharge calculation in steep irregular streams unless flood levels can be correlated with a recent storm event, for which a recurrence interval can be assigned, there may be no reliable method of determining the frequency of the calculated flood discharge channel scour during subsequent flows can significantly increase the size of a cross-section while, by contrast, post-peak deposition of fine-grained material can significantly Deduce the accuracy with which the maximum flood channel is identified and surveyed some channels retain cross-sections that are related to earlier drainage patterns and higher base levels, and are consequently over-sized within the present flow regime a channel can be formed or greatly modified by a single debris flow event, due to the scouring effect of sediment-laden flow. This process of channel formation will generally bear no relation to uniform flow theory steep, rocky channels may be so irregular that measured relationships between velocity, roughness, slope and cross-section are meaningless. The floor of many mountain streams is often better represented by a series of cascades with flow interchanging rapidly between supercritical and subcritical. Relationships such as the Manning equation do not apply because of the rapidly varying flow, but they can be used as a surrogate by adopting a high roughness coefficient.

Table 8.4 Published values for Manning's roughness coefficient (n)

Channel Type

Roughness coefficient Good Average 0.040 0.055 Poor 0.050 0.070

Mountain Streams - Bottom mainly gravel - Bottom cobbles and boulders Flood Plains - Grass or low crops - Brush - Trees

0.030 0.040

0.030 0.050 0.075

0.040 0.075 0.100

0.050 0.100 0.150

flow roughness formula will vary from channel to channel and, wherever possible, it would be wise to test various formulae against flow data before any is used in preference to the others. 8.35 Computer programs can be used to model flow behaviour by computing water surface slope, flow velocity and Froude number from surveyed channel cross-sections and assumed channel roughness. As with flow roughness equations, the model will have its greatest potential application in streams that do not display rapidly varying flow conditions. 8.36 There are many instances where correlations have been derived between bank-full discharge, or channel capacity, and catchment area, or distance downstream. Although these equations may provide useful supplementary data to that derived from rainfall-runoff analysis and the Slope-Area Method at bridge sites, they offer little advantage for culvert design.

GLACIER LAKE OUTBURST FLOODS (GLOFs)

8.37 If an engineering structure such as a hydroelectric installation or major bridge is to be located in the path of a potential GLOF arising from one of a small number of glacier lakes, then a visit to the glacial region may be required, to assess the degree of hazard. The potential for lake burst can only be assessed by measuring moraine thicknesses and lake levels, melting rates of ice lenses, moraine erosion rates and the potential for slope failure and rock or snow avalanche into or near the lake. Where a GLOF could be generated from a number of widely dispersed sources, it would not be practicable to carry out investigations of this nature for an individual road project. Under these circumstances, volumes can only be estimated by reference to previous GLOFs with sufficient safety margins.

93

8.38 Once a GLOF hazard has been identified and its volume estimated, it is advisable to assume that the release of water into the drainage system will be instantaneous. The hydrograph generated by a GLOF can be modelled using the same routing procedures as applied to dam break floods. The following data are required for the analysis: GLOF volume and initial hydrograph shape base flow and tributary inflow at the time of GLOF valley floor long-section and representative cross-sections, nominally at 1km intervals downstream a value of Manning's roughness coefficient (0.1 may be appropriate to allow for flow turbulence and sediment load). 8.39 The GLOF hydrograph can be routed under unsteady flow conditions. Sensitivity analysis can be used to assess the performance of the model and the extent to which GLOF levels are affected by, for example, an increase in flood volume and a slightly lower or higher value of Manning's n. Bridge soffits should be designed with a freeboard well above the expected GLOF level. Road alignments should also be given sufficient freeboard, but even if freeboard is adequate for the passage of flood waves, a road may still be severely damaged by consequent slope failures and erosion that extend upslope.

narrow channel to minimise bridge length stable rock for abutment foundation straight and constant reach of river channel to reduce the possibility of flow direction changes and natural scour minimal backing-up of water from downstream due to tributaries and valley constrictions minimal intrusion of bridge structure and approach embankments into channel flow.

Scour 8.43 Scour is probably the most common cause of bridge failure. For major bridges it is necessary to calculate waterway width and potential scour corresponding to at least the 100 year flood with a margin for factor of safety. Where scour is predicted to be a major hazard, and where valley configuration, construction practicality and economics permit, it may be safer in the long run to design a bridge with a longer span and place the abutments at a higher elevation out of reach of the river, than to opt for a shorter span with protected abutments or piers positioned in the flow. If there is no option but to construct an abutment or pier within the design flood waterway width, such as on wide flood plains and outwash fans, scour protection becomes paramount. 8.44 Scour will occur when flow velocity and turbulence are sufficient to pick up bed and bank material, or when sediment swept along in the flow is large enough and travelling fast enough to abrade the channel boundary. The factors that control the pattern of scour are so difficult to quantify that there is no reliable physically-based method for assessing their combined effect. Methods that do exist generally relate to lowland rivers, or model studies, and in mountainous areas there may be no option but to resort to empirical models based on regime theory, together with experimentation carried out in less severe environments. Useful data can be derived from field assessment but it is important to bear in mind that scour holes may form and fill again on numerous occasions during the passage of a single flood, and observations made afterwards may be spurious. 8.45 Scour is conventionally subdivided into two categories, namely general scour of a channel cross section as a whole, and local scour due to specific flow perturbations such as meander bends, rock outcrops and bridge piers and abutments. Normal scour depth (Y) can be calculated from regime theory using the Kellerhals equation for gravel bed rivers: Y = 0.47Q 0.8. D 90 0.12 =, where Q = discharge (cumecs) and D 90 is the 90 percentile particle size diameter (metres).

HYDRAULIC DESIGN Culverts

8.40 Culverts are typically designed to run 75-90% full under inlet control to cater for storms of between 10 and 20 years in recurrence interval. Figure 8.6 shows the method of culvert sizing used for a proposed access road in Nepal. Culvert design is discussed further in Chapter 10.

Bridges
8.41 Bridges are designed with a soffit level normally between 1 m and 2m above the design flood to allow floating debris to pass unhindered. This soffit height can be varied according to the degree of confidence in the hydrological analysis. Preliminary design can be limited to a design concept and required aperture, but detailed design will require ground survey and site assessment of flow patterns, scour depths and scour distribution, bedload transport, flood levels and general cross-section stability. There are recent examples from Nepal where bridge decks have been lifted off their bearings by the combined effects of surcharge and entrapped floating debris, and deposited hundreds of metres further downstream. 8.42 The choice of a suitable bridge site is usually made on the basis of the following criteria:

94

95

Figure 8.6 Look up chart used for sizing culverts in east Nepal

8.46 Scour can be determined by subtracting the peak flow depth under non-scouring conditions from the value of Y obtained in the above equation. 8.47 Scour depths estimated by the regime method will be average for the cross-section as a whole, and it is necessary to multiply these values by a factor of between 1.25 and 2.00 to account for local variations across the sections. The graphical redistribution of scour is an accepted way of achieving this and will also enable reductions in bed width to be taken into consideration as scour proceeds. 8.48 An alternative approach to the regime method is to calculate scour potential using tractive force theory and the mean velocity expected through the bridge waterway during design discharge. Velocity is usually determined from the Manning formula and combined with a representative value of bed material size to assess whether or not entrainment will occur using accepted scour velocities. The representative bed material size is in the range of the median diameter (DSO) and the 80 percentile (D8o), depending on how graded the material is. Trial and error is then used to determine the average general scour level that will make the mean flow velocity equal the threshold velocity for erosion of the material exposed at that level. 8.49 In mountain streams and rivers with small to mediumsized bed materials this approach will yield conservative results because scour holes may become quickly filled with sediment from upstream and bed materials tend to become coarser and more dense with depth. For rivers with a high proportion of boulders, scour depth can be estimated on the assumption that the larger fraction of bed material will protect the smaller material beneath. Under these conditions, the following relationship will apply: D 50 /d = CF3 where D 50 is the median size of bed material (metres) d is the hydraulic flow depth in metres (area/surface width) C is 0.22 for a factor of safety of 1.0 F is the Froude Number, V/(gd) 0.5 V is mean velocity (metres/second) g is acceleration due to gravity. 8.50 Maximum scour- is usually assessed on a site by site basis, but for preliminary design is taken to be twice the normal scour depth. Normal scour is calculated from regime theory (Box 8.4).

they were derived, or are based on tractive forces and assumptions regarding velocity, depth and bed material size that are untenable outside flume laboratories, and particularly in mountain streams. Also, lack of data in most cases, will mean that parameters contained in these formulae cannot be properly evaluated. Predicted bedloads can vary by as much as ten-fold depending on the formula used. 8.52 Another problem with predicting sediment yields in mountain catchments is the fact that the highest concentrations are often the result of instantaneous injections of landslide debris into the channel flow, that do not form part of hydraulic considerations. Also, not all eroded material is transported through the catchment immediately; large quantities are stored for significant periods in terraces, debris fans and braided channel reaches. 8.53 The only practicable means of assessing sediment potential is to carry out a mapping exercise to determine the relative erodibility of channel material, the potential for slope failures, and the likely efficiency with which storm runoff is able to transport debris through the catchment. In addition, road earthworks and tipped spoil can lead to significant increases in sediment load, as can roadside erosion. Although these considerations can, to a certain extent, be quantified, there is no reliable method by which the data can be translated into sediment transport rates without monitoring records. 8.54 Realistically, the best approach is to identify those catchments that are likely to give rise to long-term problems of culvert or bridge blockage, and design the crossing structures accordingly. A box culvert with inlet guide walls and a slope equivalent to that of the natural stream bed is more selfcleaning than any other design, while a short-span bridge may be preferable to a large culvert where an alignment crosses an eroding gully with frequent debris flow activity. Chapter 13 describes the options that can be considered when designing a road across an active fan. Box 8.4 Regime theory and scour calculation From regime theory, the unobstructed waterway width (W) can be calculated from the equation: W = 3.26Q 1/2 = (Q = discharge in cumecs) If the waterway length of the bridge (L) is less than the regime width then the normal scour depth (Y') is increased according to the following equation: Y = Y(W/L) 0.61 (Y = regime scour depth)

Sediment transport
8.51 Despite the fact that culverts on mountain roads are frequently blocked by sediment or debris, there is still no reliable means of predicting sediment yields from mountain catchments for design purposes. Conventional bedload transport formulae are usually either empirically-based, and hence applicable only to the catchments from which Where the bridge and its approach embankments cause a constriction to the flood waterway, then the maximum scour depth obtained by application of the factors at paragraph 8.45 should be compared with Y.. given by the following equation. Y max = Y(W/L)1 56 and the greater of the two values adopted for design.

96

EARTHWORKS

CHOICE OF CROSS-SECTION
9.1 The choice of cross-section is usually second only to the choice of alignment and design standard in determining the stability, cost and environmental impact of the final construction. In gently sloping terrain, with ground slopes of less than 20 or so, there are generally few constraints on the choice of crosssection and a balanced cut and fill, which is generally the most preferable in terms of engineering practicability, cost and environmental considerations, is usually achievable. By contrast, in steep mountainous terrain a balanced cut and fill is virtually impossible to achieve over any significant length of alignment given geometric constraints and the abrupt and frequent changes in topography. Where there is no choice but to cross steep and irregular cliff and ravine topography, the need to select the most appropriate detailed alignment location and choice of cross section, on a slope by slope basis, becomes paramount. 9.2 Figure 9.1 illustrates the more important factors to be considered in cross-section design and identifies slope conditions under which various cross-sections are usually most applicable. These conditions are discussed briefly below.

dipping into the hillside (ie favourable to stability). Short-span bridges and masonry buttresses built across steep rock reentrants are cheaper and often more feasible than tunnelling through the adjacent spurs, although falling rock can cause significant damage. It is usual, therefore, to combine elements of full cut with road fill retaining wall in most steep ground situations.

Balanced cut and fill cross-section

9.6 Where there are no underlying stability problems, a balance between cut and fill in any one cross-section can usually be achieved on slopes up to 30 without the requirement for a road retaining wall. However, this usually requires a sinuous alignment to follow the slope contours, leading to an increase in alignment length and not necessarily the correct choice of crosssection in each case. Alternatively, designing for balanced cut and fill over a short alignment length will enable preferred crosssections to be combined with an undulating vertical alignment. Specifying balanced cut and fill assumes that plant is available to haul excavated material to adjacent fill sites. With labour-based construction methods, the maximum practicable haul distance may be little more than 50m, whereas plant intensive construction will allow almost unlimited haulage distances, albeit at a cost. Specifying balanced cut and fill assumes that plant is available to haul excavated material to adjacent fill sites. 9.7 In designing the cross-section, it may be necessary to allow for a cut-to-fill bulking factor of 20% and a wastage factor of 100, giving rise to a net surplus of 10% cut over fill volume.

Mostly full cut cross-section

9.3 Traditionally, low cost roads have been constructed across steep side-long ground in full cut with the excavated material side-tipped along the length of the alignment. This approach maximises the use of cheap unskilled labour in excavation and avoids the need to construct comparatively expensive retaining structures. Furthermore, in the absence of compaction plant, a road formation in cut will be more traffickable than one in till. 9.4 However, unless dictated on stability grounds (Figure 9.1), or where there is no practicable alternative, the widespread adoption of a full cut cross-section cannot be recommended for the following reasons: it inevitably reduces the factor of safety against failure of the natural slope above, and leads to a greater potential for failure and erosion in the cut slope itself it maximises the amount of spoil generated, resulting in problems of disposal and erosion below the road.

Mostly full fill cross-section

9.8 Where foundation stability permits, a full fill or retained fill cross-section is most appropriate under the following slope conditions: where persistent joint, bedding or foliation planes are dipping out of the slope and could give rise to failure in excavations where excavations for cut slopes in shallow loose scree or similar deposits at their limiting angle for stability could result in progressive slope failure where an alignment is located on eroding or highly erodible soils, the stability of which would be further reduced by cut slope excavation where a rock fill causeway is the only means of crossing the toe of an unstable slope that is periodically undercut by river scour or kept wet by irrigation (construction of the causeway may lead to an increase in stability through toe loading and scour protection). for purposes of pavement drainage, where the alignment is located on a terrace.

9.5 Alternatives to the full cut cross-section on steep rock slopes include tunnelling, half tunnelling, the use of short-span bridges, masonry buttress-supported road slabs (Chapter 11), and masonry road fill retaining walls. Unless tunnelling can be shown to be cost-effective in reducing alignment length, or in avoiding long-term maintenance costs, the high construction costs are usually prohibitive. Half tunnelling with lateral support is only feasible in strong and massive bedrocks with planes of weakness

97

Figure 9.1 Stability considerations in the choice of cross-section

98

EARTHWORKS DESIGN AND CONSTRUCTION

Cut slopes
9.9 Prior to excavation, and in the absence of detailed subsoil investigations, the design of cut slopes is usually only provisional. It is based on estimated soil and rock conditions from walkover surveys and trial pitting (Chapter 7), modified locally by topographical constraints and right of way limitations. In fact, the engineering behaviour of slope materials will be confirmed only when slopes have been exposed for at least one wet season. Even then, rainfall patterns, differential weathering of slope materials and, above all, the effect of changing land use and irrigation practices, can cause failures several years after construction. 9.10 The combination of geotechnical uncertainty, and the variability in ground conditions, frequently between one crosssection and the next, usually means that assumed or calculated strength parameters may be too low for one slope, leading to over-conservative excavation, or too high for the next, causing slope failure. If it is desired to keep the road open at all times with minimal landslide debris clearance, then the designed slope will need to be on the conservative side with a notional factor of safety of at least 1.2, and with adequate provision for erosion protection where vulnerable soils are exposed over large areas in the excavation. If temporary road closures and debris clearance can be tolerated and allowed for in maintenance, then a factor of safety of 1.1 and a steeper slope may be more economic, even in the long term. Table 9.1 shows a range of cutting angles with a notional factor of safety of 1.1, derived from stability charts for common soil types found in Nepal. 9.11 The factor of safety of a cut slope will changeover time as the slope adjusts to internal and external processes of weathering, drainage, landuse and vegetation. It is advisable, therefore, to combine theoretical considerations with observational data (Chapter 7) in order to make a rational choice of cut slope angles. 9.12 The aim of any low cost approach to earthworks design is to maximise cutting angles without having to resort to extensive use of cut slope support structures. Nevertheless, slope revetment and retaining structures will be unavoidable wherever: natural slope angles are close to their limiting angle for stability and there is no choice but to incorporate some cut in the cross-section alignments are required to cut through unstable slopes the constraints of alignment geometry dictate structural support to earthworks, for example where hair

pin bends and stacked loops are to be constructed on steep ground (35 or greater) in weak materials failures in the cut slope occur during construction and maintenance.

9.13 Where cut slopes are less than a metre or so in height, it may be preferable on environmental grounds to slacken them back to an angle that allows the earthworks to appear as part of the landscape. This may also provide additional materials for construction. The value of neighbouring land uses and right of way restrictions may be limiting factors in adopting this approach. 9.14 Cut slope profiles can be single-sloped, multi-sloped or benched. Single-sloped profiles are usually cut in uniform soils and rock slopes. Multi-sloped profiles are cut where an excavation encounters soil overlying rock, or a succession of river deposits of differing resistance to erosion or failure (Box 9.1). Where rock is encountered, persistent joint, bedding or foliation surfaces usually determine the final cut slope profile. Box 9.1 Terrace deposits Vertical variations in the particle size and density of alluvial deposits exposed in excavations may warrant a multi-sloped design. While medium dense gravels and cobbles can stand at steep angles (60-80) with little need for erosion protection, overlying unconsolidated sands should be cut back to an angle of about 40 and protected against erosion. However, sands that are partly cemented by iron oxides, through fluctuations in iron-rich groundwater, can usually be cut to angles of 50-60. Site conditions will dictate whether a single-sloped or multisloped design is adopted. 9.15 The principal advantage of benched cut slopes, from a stability point of view, lies in their ability to slow down the rate of surface runoff, and the fact that shallow failures are usually limited to one bench at a time. The steps on a benched cut slope should slope into the hillside and be provided with collector and cascade drainage systems. Properly designed benched cut slopes are expensive to construct, and erosion and slope failure can be accelerated if benches are not regularly maintained. Their use is comparatively rare in the construction of low cost mountain roads where a maintenance commitment can often not be guaranteed. Furthermore, vegetation is more difficult to establish on the steeper riser slopes than on a uniform slope profile. A benched cut slope is visually less appealing than a single-sloped excavation.

99

Table 9.1 Provisional design gradient/height relationship for soil cut slopes Cut slope gradient (V/H) Cut height (metres) 4-6 1.0 1.0 0.8 50.8 550.8 0.8 1.0 1.0 0.8 1.5 1.2 1.0

Soil type Clayey silts (transported)

Water table Low Moderate High Low Moderate High Low Moderate High Low Moderate High

0-3 1.5 1.2 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.5 1.2 1.0

7-10 0.8 0.5 NA >_0.8 5_0.8 NA 0.8 550.8 NA 1.0 1.0 NA

Silts

Coarse-grained colluvium

Silty clays (residual)

Note

The above slope angles have been derived from stability charts with assumed c' and ' values and an average factor of safety of 1.1. Slope gradients have not been given for 7-10 m high cuts with a high groundwater as this condition is unlikely to occur in granular soils, whose permeability is relatively high. This table is for illustration purposes only.

Box 9.2

Non-adherence to specification during construction can give rise to earthworks failures

Unretained fill slopes

9.16 The main considerations in fill slope and embankment design are the maximum permissible angle of side slope and the overall stability of the fill on the hillside. An examination of fill slope and embankment failures along a number of mountain roads shows that many are brought about by: inadequate under-drainage under conditions of pronounced seepage incomplete removal of vegetation and organic material prior to embankment construction construction of embankments on loose spoil material derived from earlier excavations erosion on slopes immediately below the embankment the presence of pre-existing shear surfaces beneath the embankment in fine-grained soils (infrequent in mountainous terrain) the presence of unfavourably orientated planes of rock weakness beneath the embankment.

Failures in cut slopes often occur as a result of unfinished earthworks rather than faulty design. Contractors sometimes oversteepen the base of an excavation in order to induce shallow failure so that earthworks can proceed at a faster rate. This can give rise to over-excavation and the exposure of a larger excavated slope to erosion. The same condition may occur where road widening is achieved by removing the toe of the adjacent cut slope. Slopes that are left with a convex profile (oversteepened at the base) are more prone to failure than those that are finished to a perfectly straight profile, as specified. On the other hand, concave cut slope profiles result in poor drainage at the toe, that may give rise to progressive failure by slumping and undermining. If slope drainage is an integral component of a cut slope design, then slope failure may occur if the works are not completed by the onset of the wet season.

100

9.17 Generally, the particle size distribution and the angularity of the particles are the only indicators of the shear strength parameters applicable to the design of embankment side slopes. Appropriate 0' values range between 25 and 40 for clayey and granular materials respectively. With low factors of safety (1.1-1.2) this gives side slopes of between 20 and 33. In mountain areas, where granular soils predominate, embankment side slopes are normally constructed to 1:1.5 (33) assuming that the specifications for material size, drainage and compaction can be met. In more gently sloping ground the side slope can be relaxed to accommodate weaker fill material, to provide a convenient means of spoil disposal and to allow the road to become visually less obtrusive in the landscape. 9.18 Adequate compaction of embankment fill can be problematic if it is undertaken by labour-intensive means. The use of hand rammers may achieve only 80% Proctor compaction in schist and phyllite granular fill. Settlement is therefore common, especially along the outer edge of embankments and around culverts. It is recommended that once fill slopes are constructed to formation level they are left for a wet season to settle before the pavement is added. This, of course, is not necessary if compaction plant is available. 9.19 The overall stability of a fill slope on a hillside is more difficult to assess. Before constructing a fill slope on side-long ground, it is necessary to terrace or step the formation in order to prevent a possible slip surface from developing at the interface between the fill and the natural ground. The potential for failure along a deeper surface in the ground beneath should be considered, although this rarely happens since the strength of soils tends to increase with depth. Unless fill is placed on colluvium, the weakest layer is likely to be just below the formation level. If thus material is similar to that used as fill, its stability on the slope will also be adequate as long as the stability of the embankment slope is satisfactory. Problems occur when strata or foliations in the rock masses beneath the fill are dipping parallel to the ground slope, or where the groundwater table is at or very close to the surface. Adverse rock planes can cause the fill to slide, triggered by increased load, or increased pore pressure along the failure plane. Groundwater can soften the founding material, or cause the fill material to be undermined through seepage erosion. In these situations fill slopes require under-drainage to keep groundwater moving freely down slope. 9.20 The design and construction of road fill retaining walls are discussed in Chapter 11.

array of shotholes to produce a clean break, nor generally does it take account of the fracture pattern of the rock to assist loosening. Bulk blasting usually results in significant overbreak and the creation of a highly fractured rock mass which usually ravels and fails to shallower slope angles for years afterwards, creating a permanent hazard and maintenance problem. 9.22 With pre-split (pre-shear) and cushion blasting, overbreak is minimised and the cut slope is left intact and less prone to rockfalls. Pre-split blasting is therefore preferable to bulk blasting but is far more expensive. It is more time consuming because many more holes are needed, and it calls for well-controlled drilling methods that produce a series of closely-spaced parallel shot holes. It also requires high quality explosives and detonators, and electronic detonation. When applied correctly, pre-split blasting produces a clean, straight rock face at the design angle. It also reduces the danger of flyrock to a minimum. However, depending upon the rock joint pattern, it may produce debris of a size that can only be removed by machine unless it is broken up further, and at additional cost, by hand. If properly designed, pre-split and bulk blasting can be used to good effect together to provide a clean rock face and small-sized debris. Pre-split blasting is especially recommended wherever rock mass failure is likely.

Spoil disposal
9.23 However much care is taken to minimise quantities of spoil, it cannot be eliminated altogether. Although all topsoil and overburden should be stockpiled for later reuse, surplus material may arise from a large number of situations including: surplus of cut over fill requirements cut materials unsuitable for inclusion in 'he works the establishment of an initial trace along the road alignment the result of operations taking place out of sequence with the main earthworks, such as excavation for structures overburden to quarries and borrow areas topsoil stripped and not re-used surpluses of imported construction materials, such as gravel in excess of requirements for pavement layer construction maintenance operations such as the clearance of slips or the removal of silt from side drains. 9.24 The control of spoil disposal is of utmost concern, because it can give rise to a variety of problems, including:

Rock blasting
9.21 Uncontrolled rock blasting is atypical feature of road construction where labour-intensive methods of excavation rely on a finely-fragmented rock mass for easy removal. Bulk blasting, as it is termed, does not use a regular

101

 erosion of the spoil tip itself the smothering or removal of natural vegetation (once stripped of plant and soil cover, slopes usually take between 3 and 5 years to revegetate, and as much as 10 years on steeper and more sterile slopes) instability within the spoil material itself, especially when infiltrated by water slope overloading and resultant failure disruption of existing runoff patterns and siltation of water courses and drainage channels disruption to agricultural practices.

provide a drainage blanket beneath a spoil bench where there is any indication of a spring seepage at or near the spoil site compact spoil benches during construction. While benches cannot be compacted in the formal sense, they can be constructed in definite lifts normally not more than 0.5m thick, with the top surface of each lift approximately horizontal. This will allow machines involved in spreading the spoil to track the surface and provide some degree of compaction where spoil benches are constructed on agricultural land, form the tip into a benched profile so that it can eventually be returned to agricultural production. In the meantime the risers between levels must be protected against erosion by constructing dry stone walls and applying vegetation where the top surface of the bench is large, runoff should be reduced by the provision of regular shallow interceptor drains. The slope of these drams should be constant as far as is practicable and should not be so steep as to induce erosion on completion, spoil benches should be left in their required shape and planted with grasses, shrubs and trees as appropriate, to encourage maximum stability and resistance to erosion.

9.25 There are two steps to be taken in minimising spoil problems within a construction project. The first is to identify those operations that will generate spoil, the places where it will be generated and the quantities involved, no matter how small. The second is to plan in advance for its disposal by the designation of safe tipping sites. The engineer is responsible for the designation of suitable sites, and his criteria for their selection should aim to avoid the problems listed above. When construction is being under-taken through a conventional construction contract, the engineer should ensure that both the contractor and the construction workforce are aware of the restrictions on the disposal of spoil, the location of approved spoil disposal sites and specific requirements for the management of these sites. Contract specifications regarding spoil disposal should be strictly enforced. 9.26 Spoil can be dealt with either by discarding it, or by turning it into landfill. The following guidelines should be observed: select areas for spoil tipping on steep slopes formed in relatively resistant bedrock, where tipping will result to no more than the removal of vegetation and shallow soil, with negligible slope incision thereafter. Bitumen drum disposal shutes can be used to convey the spoil down a short slope to a safe site below when creating a landfill site, make maximum use of terraces, level ground and spurs build many small spoil benches rather than a few large ones, to avoid slope overloading

Do not permit the following: tipping of spoil into stream channels other than major rivers, as the increased sediment load will lead to scour and siltation downstream tipping of spoil onto slopes where road alignments, housing areas or valuable farmland downs lope might be affected use of areas of past or active instability and erosion as tip sites the discharge of runoff over the loose front edge of a tip bench during or after construction tipping of spoil in front of road retaining walls where impeded drainage could soften the wall foundation.

102

10

DRAINAGE

GENERAL PRINCIPLES
10.1 Conservation of the natural drainage system around the road alignment should be one of the most important concerns during design and construction. By effectively creating a barrier to natural surface drainage that is only punctuated at intervals by culverted drainage crossings, road construction can lead to significant local increases in catchment area, perhaps by up to 100% in some ridge top localities. Furthermore, in the case of paved road construction especially, road drainage reduces the tune taken to reach maximum flow by shedding water from impermeable surfaces. Therefore, in addition to constructing side drains, culverts and bridges to convey their design runoff without surcharge, blockage by sediments, or scour (Chapter 8), great attention must be paid to strengthening those parts of the natural slope drainage system that experience increased runoff, and hence erosion potential, as a result of road construction. The main ways of doing this are to: control road surface drainage design culverts or drifts that convey water and the expected debris load efficiently maximise the frequency of drainage crossings to prevent excessive flow concentration protect drainage structures and stream channels for as far downstream as is necessary to ensure their safety

10.3 Occasionally, an outward-sloping road surface has been advocated for minor roads, on the grounds that, by allowing water to disperse gently onto the hillslope along the whole length of the road, the potential for erosion will be reduced. In practice, the opposite is true. The method undoubtedly offers very large financial savings in the reduction of drainage structures, but it is a highly hazardous form of design that cannot be recommended except in areas of very low erosion potential. The design carries the following weaknesses: in practice it is impossible to design a road geometry for a distributed flow of water (topography is the controlling factor) road settlement, which cannot be predicted, will in time change the design cross-fall road repairs will locally alter the cross-fall partial blockage of the road by debris results in a change of the flow pattern of drainage water, and instant local surcharge erosion can reach a disastrous level before maintenance crews can be mobilised slope protection from uncontrolled runoff would require a lengthy period of post construction monitoring and remedial works that is not practicable for normal contractor-based operations. road safety. Vehicles can slide sideways uncontrollably across a wet road surface and over the edge

plant vegetation on all new slopes and poorly-vegetated areas, around the edges of drainage structures and appropriately along stream courses, without impairing their hydraulic efficiency or capacity.

Side drains
10.4 Side drains serve two main functions: to collect and remove surface water from the immediate vicinity of the road and to prevent any sub-surface water from adversely affecting the road pavement structure. The latter function is achieved in the simplest case by leaving weep holes in the side drain (on both walls) set at 50-100mm above the invert. Groundwater in the subgrade can be released either by using a drainage blanket at the base of the pavement, or by incorporating gravel crossdrains (,grips) in the road base that exit via a weephole in the side drain backed with a piece of filter fabric. Deeper drains, comprising a filter-wrapped perforated pipe within a graded gravel backfill, can be constructed under very wet slope conditions to a depth of 1-1.5m below the level of the side drain invert, and led to the nearest culvert inlet. 10.5 Side drains (as well as the road itself) should have a minimum longitudinal gradient of 0.5%o, except on crest curves. Slackening of the side drain gradient in the lower reaches of significant lengths of drain should be avoided in order to prevent siltation. On potentially unstable slopes, side drains should be lined with heavy duty polythene, or

ROAD SURFACE DRAINAGE AND SIDE DRAINS Crossfall

10.2 Control of road surface drainage and side drain runoff should be a major component of the cross-fall geometry. On all minor roads, an inward-sloping road carriageway is the normal means of shedding water from the road surface. The inward slope incorporates an inherent factor of safety in retaining water that has accidentally escaped from the drainage system. A crossfall gradient of 3-4% is commonly adopted in order to prevent ponding on slack road gradients, and longitudinal scour of an unpaved road surface on long, steep sections of alignment. For reasons of traffic safety, outward cross-falls are required on some bends of trunk roads. In such situations it is usually necessary to prevent water from discharging overfill slopes in an uncontrolled manner by introducing some form of bund or upstand along the outside edge of the road.

103

some other impermeable material, before the masonry pitching is applied. This will prevent water entering the slope if the masonry becomes cracked by movement. The gap between the drain and the hill side must be filled with compacted material sloping towards the drain to minimise infiltration behind it. 10.6 The choice of side drain cross-section (Figure 10.1) will need to take consideration of hydraulic capacity, ease of maintenance, space restrictions and traffic safety. As far as traffic safety is concerned, a wide and shallow drain for a given flow capacity is preferable to a deeper one but in particularly steep ground the extra width required to achieve this may be impracticable, or too expensive. Side drain covers can be used to provide extra road width in places where space is limited: their widespread use, however, is not recommended as they can hinder the progress of routine maintenance. Under normal circumstances, the adoption of a trapezoidal cross-section will facilitate maintenance and will be acceptable from the point of view of traffic safety. 10.7 Design volumes of runoff are usually estimated using the Rational formula (Chapter 8). Flow velocities are calculated from the Manning equation using roughness values shown in Table 10.1. Most published roughness and velocity data are based on clean water flow, whereas sediment-laden water is more common along mountain roads. Side drains are left unlined or are lined with masonry according to the strength of the material in which they are excavated, and the velocity of runoff they are expected to carry. Usually, a lined drain is required when the underlying materials are soils or weathered rock. Concrete can be used for greater strength if vehicle trafficking is likely in narrow roadway sections. 10.8 When the cross-sectional area is less than about 0.1m2 and the gradient is gentle, drains can be lined with unbound masonry. Larger and steeper drains are lined with mortared masonry, although they can be up to ten times as expensive. Where masonry check dams are used to reduce flow velocities in side drains, there must be sufficient cross-sectional area above the check dam to take the maximum design flow. Cascades or steps in the drain long-section can also be a useful means of reducing flow velocity, although both check dams and cascades can impede the transport of debris, increasing the risk of blockage. 10.9 If there is no opportunity to safely discharge side drain water via culverts into existing stream channels, then it is usual to construct side drain turnouts, especially in bend locations (paragraph 10.40). However, it may be preferable to increase side drain capacity to convey runoff to the next available safe discharge point, rather than to construct side drain turnouts or relief culverts on erodible slopes.

Culvert location
10.10 The desire to maintain the existing slope drainage pattern as much as possible will require all watercourses to be culverted. This may result in an average spacing of one per 100m, and a range of spacing of between 30m, or less, to greater than 300m, depending upon topography. Where stream channels are poorly developed, for instance where permeable soils occupy gently sloping ridge top locations, the opportunities for discharging road side drainage into established stream channels, that will not undergo subsequent erosion, are much reduced. 10.11 Where there is no choice but to introduce relief culverts to the road drainage scheme, they should be discharged onto bedrock surfaces, or over convex slope profiles with appropriate protection works, in order to maximise energy dissipation with minimum erosion. The discharge of side drain relief culverts onto virgin soil slopes can result in the rapid erosion of the soil and weathered rock mantle to depths of 5m, or even 10m in extreme cases, which may ultimately undermine the road. 10.12 Road construction across irrigated farmland must not materially disrupt the pattern of water flow supplying the terracing. The principal irrigation channels should be taken beneath the road and this may require a larger number of culverts than normal. When the road is in cut, irrigation water can be conducted across the road via a cascade and culvert in the normal way. In box cut it is necessary to construct an appropriate diversion channel, employing an aqueduct if necessary. 10.13 There are occasions when a culverted stream crossing is impracticable. These are: culverts (narrower than 600mm) can become easily silted and prevent easy maintenance. In this case discharges must be small enough to be collected in the side drain and delivered elsewhere (without adversely affecting community water supply) when the local topography is too awkward to enable a culvert to be accommodated when a constant supply of debris transported through the channel would continually block a culvert. In this case a concrete drift with a shute below is usually the best solution.

10.14 Notwithstanding the above, the control of water and the prevention of erosion should remain the highest priority.

104

Figure 10.1 A selection of side drain types and their typical dimensions

105

Table 10.1 Flow roughness and velocity in open drains Material Roughness `n' value 0.022-0.025 Max vel m/sec

masonry arches are economic where vehicle loads are comparatively low, and where adequate stone is available (also, they look attractive in a rural environment) concrete pipes are comparatively cheap and preferable when sediment loads are not excessive. Vehicular access to the culvert site is necessary. The maximum pipe diameter is 1200mm corrugated metal pipes tend to be a more expensive form of construction, although they are portable and easy and quick to install. They are especially suitable in areas where stone for masonry work is in limited supply. Pipe diameter is effectively unlimited. Segments can be portered to site and assembled by hand, an advantage in places where culverts are built in advance of the vehicle access track causeways are preferable where sediment loads are high or where large-size debris is expected. The floor can be of masonry or of concrete if extra scour resistance is required concrete box culverts are necessary when the required opening exceeds more than about 4m= and when the sediment load is large and abrasive. Boxes are effective in channels sloping at up to 30, but it is important to prevent settlement and seepage erosion beneath the structure by constructing it on a bedrock foundation. In addition, an apron on the downstream side may be necessary to reduce erosion.

Sand, silt (unbound) Loam, fine Gravel Stiff clay Good grass Cover -easily eroded soil -other soil Coarse gravel Rock -smooth -jagged -soft -hard Masonry Concrete Culvert design

0.3

0.022-0.025

0.9

0.018-0.022 0.025-0.030 0.030-0.035

1.2

0.9 1.5 1.5

0.035-0.040 0.040-0.045 0.025-0.030 0.015-0.020

2.5 5.5

10.15 In mountainous areas culverts usually operate as hydraulically short drainage structures under conditions of inlet control (culvert flow Type 1, Figure 10.2). Typically, they are sized to flow 75-90% full, with measures to reduce velocities at the outlet. Culvert design procedures are fairly standard and use can often be made of standard graphs and charts. However, it is important to ensure that graphs and charts are appropriate to the culvert flow conditions, and that the correct flow type (Types 16, Figure 10.2) is assumed for design. 10.16 It is a false economy to reduce construction costs by minimising the aperture of culverts and allowing them to be surcharged or blocked too frequently. The decision as to whether to design a large culvert or a short-span bridge will depend on design discharge, anticipated sediment loads, configuration of flow, foundation conditions, ease of construction and cost. Short-span bridges may be the only solution when crossing ravines and waterfalls, debris flow channels and debris fans. 10.17 The choice of culvert type depends upon the local topography, sediment load, access to the site, and the availability of materials and local masonry skills. The main options and choice criteria are set out below:

10.18 Various options for culvert configuration are presented in Figure 10.3. Where sediment loads are low to moderate, the combination of a nominally 1 m deep catch pit inlet, a moderately sloping culvert long-section, and sufficient energy dissipation and erosion protection works at the outlet, is recommended (F). Culvert catch pits should be designed to be easily cleared of debris. Drop outlets (D) should be avoided wherever possible, unless the channel bed materials are erosion-resistant. Where there is no choice but to construct a drop outlet, or where scour protection is required beneath an existing culvert, then the apron protection shown in Figure 10.4 may be appropriate. The apron should have a downstand cut-off and raised side walls or wing walls to contain water splash. Where sediment loads are high, a chute inlet (Figure 10.3E), a wide culvert and greater erosion protection works at the outlet are usually required. 10.19 More elaborate methods for dealing with high sediment loads include: sediment retention schemes upstream. However, if their retention capacity is much less than the volume of material moving down the stream channel, and if they cannot be emptied regularly, they will simply be

106

Figure 10.2 Classification of culvert flow

Figure 10.3 Common types of culvert cross-section (not to scale)

107

overridden by new debris. Although they serve to reduce flow velocities through the creation of lower bed gradients, there is a risk that the raised bed level will cause erosion of the channel sides the construction of a reinforced concrete causeway at road level, accepting that the road will periodically become blocked by debris and thus require regular clearance the provision of a culvert or bridge with a waterway area substantially larger than that warranted by the design flow alone, in order to accommodate transported debris passing the road itself through a tunnel formed as a large culvert, to allow debris to flow over the top. This design is applicable only in extreme cases of debris fan aggradation (Chapter 13), and is inappropriate for low cost roads.

the need to find a stable cross-section with strong-points for founding and keying-in the structure, rather than by hydraulic considerations alone. 10.23 Because check dams need to be flexible and freedraining they are almost always constructed in gabion and to heights of up to 4-5m. It is important that their margins are adequately keyed into channel banks, although it may not be possible to achieve this in channels whose banks are composed of soil, river deposits or weak rock. Unfortunately, it is usually under these conditions where check dams are most needed. Where appropriate, bank revetments or side-walls should be constructed upstream and downstream of each check dam to provide additional support and to reduce the potential for scour around the sides of the check dam. 10.24 The strength of a check dam can be increased by constructing it with an arc shape in plan, its convex side facing upstream. However, this may tend to concentrate stream flow against vulnerable stream banks before the sediment wedge has been deposited, or afterwards, if it is later scoured out. Scour and seepage erosion on the down-stream side can be reduced by constructing a masonry or rip-rap toe apron. Where scour potential downstream of the structure is severe, larger scale protection works in the form of gabion or rip rap revetments, masonry sills and cascades can be considered. 10.25 If the primary function of a check dam is to reinforce a natural knickpoint that is being undermined by erosion, and where foundation conditions are good, check dams can be built of bound masonry. These are especially useful in places where space is limited, or where the geometry of the ground is too awkward for a gabion structure. For very small watercourses with low erosion potential, unbound masonry check dams, or check dams built of logs, may suffice. 10.26 Where erosion is already well-advanced, or where foundation conditions are poor, check dam construction may only be possible with large structures, placed on deep foundations or anchored into the bedrock at depth. Plans for such structures must be carefully considered in the light of cost, practicality of construction, probable life of the structure and overall effect.

Stream course protection

10.20 It is often impossible to make reliable predictions concerning the full extent of erosion protection likely to be required until the road drainage system is fully functioning and the slopes and drainage channels have responded to the new drainage regime. 10.21 The general design philosophy of stream course protection is to dissipate as much water energy as possible in the vicinity of the road itself, where erosion is likely to be worst, and protect the outfall channel down to a point where it is large enough or sufficiently resistant to withstand the increased flow. Outfall channel protection usually consists of check dams, cascades and channel linings (Figures 10.5 and 10.6). It is not uncommon to build protection works for 20-60m downstream of culverts, and there are instances where they have been constructed for distances of 500m, or more. If investment to this level of protection is considered necessary, it is clearly important to be sure that the measures will be effective. Protection of erodible channels upstream of culverts is usually accomplished by check dams and cascades constructed over much shorter lengths, and usually within 20m of the inlet.

Check dams Cascades

10.22 By trapping sediment on their upstream side, check dams create a stepped channel bed profile, thus reducing velocities and channel downcutting, and thereby ultimately halting the progress of erosion. The principle of check dam spacing is that each check dam is placed at the taper of the sediment wedge formed behind the check dam down-stream. However, mountain streams are generally too steep to allow channels to be protected in this way, as this would result in an excessive number of check dams. Furthermore, the final location of each check dam is determined more by 10.27 The principal function of a cascade system is to dissipate the kinetic energy gained by flowing water from a sudden drop in elevation. This is achieved by dissipating energy in increments at each step in the cascade. For discharge rates up to about lm3/sec per metre width of channel (depending on geometry), cascades are efficient and are an acceptable method of controlling water speed. When discharges are greater than this, the energy-dissipating effect of cascades is reduced, as the water tends to jump

108

Figure 10.4 Culvert spout outlet apron

109

Figure 10.5 Typical gabion check dam details

110

Figure 10.5 Typical gabion check dam details (continued)

111

Figure 10.6 Typical gabion cascade detail

112

over steps in the cascade, which then performs hydraulically in a similar manner to a straight chute. Very steep cascades, where the rise is much greater than the step, are ineffective for all but the smallest discharges. 10.28 Masonry cascades are recommended only where the channel is formed in relatively intact rock and good founding conditions are available, ie the channel conditions that usually least require protection. Masonry cascades can withstand the erosive effect of sediment-laden flow to a greater extent than gabion, although the latter can sometimes be protected by laying mortared slabs of flat rock on the cascade steps. 10.29 Gabion cascades are preferred where channel foundation conditions are poor, and where some flexibility in the structure is required. They also allow dissipation of hydrostatic pressure from behind the structure. The main disadvantage with gabion cascades is that they are permeable, and seepage erosion beneath the structure can lead to significant deformation. The provision of an impermeable membrane and filter fabric can control this effect. 10.30 The design capacity of a cascade must not be less than the capacity of the waterway in which it is installed. In confined stream courses this may mean that the sides of the waterway have to be cut back in order to preserve a sufficient cross-sectional area. Cascades should always be constructed with side walls that protect the channel banks against side splash erosion. Ideally, the channel banks should be trimmed back so that the side walls can be constructed flush with the original channel. It is important that side walls do not protrude above the height of the channel banks, as this will cause erosion between the bank and the side wall. The steps that form the cascade should be small enough to climb up and down for inspection purposes. Cascades should be wired or keyed into the adjacent road retaining walls on either side and extended down-stream to a strong point in the natural channel. If a strong point does not exist, the cascade should terminate in a stilling basin or plunge pool, possibly with channel protection below.

10.32 Channels can be lined with gabion, masonry, dry stone pitching, rip-rap or vegetation. Gabion mattresses can be constructed satisfactorily on channel beds with gradients of up to 30 as long as they are adequately secured at the top and pegged firmly into the channel bed. However, they cannot be made to fit the channel bed as closely as masonry. Their chief advantage is that they are sufficiently flexible to tolerate settlement caused by seepage erosion and scour from beneath. Gabions used for control of water must always be laid over a geotextile or gravel filter whenever the founding materials are potentially erodible. As with all gabion applications, abrasion by stream sediment can rapidly break through the gabion mesh, unless some form of screed is applied. However, a screed is not compatible with a gabions's flexibility, and the design must be based on site conditions. 10.33 Masonry linings can be constructed to fit the stream bed much more closely than gabion. They are also less easily abraded, but they cannot tolerate significant settlements, loss of support by seepage erosion or high groundwater pressure. Dry stone pitching is usually only suitable where discharges are lower than 1 m'/sec per metre width, and where sediment load is relatively fine-grained. 10.34 Channel rip-rap can be used to armour the bed and increase flow resistance, thus decreasing flow velocity. Rip-rap can take the form of boulders or cast-in-place tetrapods, whose size will depend upon the expected flow velocities and scour depths. These can be judged from field evidence of channel scour and transported sediment sizes in association with flow roughness and scour depth computations (Chapter 8). The main drawback with using rip-rap is the thickness of the layer to be constructed, which must be at least 1.5 times the size of the largest stone, and which may require excavation of the bed and banks to be accommodated. Large rip-rap is, therefore, only used to emergency cases, or as a bank armour where flow is concentrated, as on the concave bank of meander bends, or where a channel sharply changes direction. The use of rip-rap as flood plain protection is discussed in Chapter 13. 10.35 Grass, shrub and bamboo planting can provide some resistance to channel erosion and may be used where flow velocities are not expected to be too high. The introduction of vegetation to the channel bed and banks will also tend to reduce flow velocities, although channel vegetation should not be so widespread as to inhibit or divert flow, which could lead to bank scour. Shrubs and bamboo are likely to be the most effective, although the latter can usually only be cultivated in damp sites in warm climates. Where immediate effective protection is required, a structural solution is preferable to a vegetative one. 10.36 The winning of boulders and cobbles from gully beds for road construction materials can reduce the armouring effect provided by coarse material. If the bed material appears to be weathered and static for much of the

Channel linings
10.31 A channel bed lining is constructed in situations where the channel is at risk from scour, but where the bed is not steep enough to warrant a drop structure, such as a check dam or cascade. The lining can also be extended up the banks to prevent lateral erosion. When constructing a channel lining it is important to reproduce, as a minimum, the dimensions of the original channel. A dish-shaped cross-section to the bed lining is most preferable. The main disadvantage with channel linings is that a lower channel roughness leads to an increase in flow velocity and hence an increase in scour potential further downstream. In the case of masonry aprons, or gabion mattresses with masonry screeds, some reduction in velocity can be achieved by cementing protruding stones into the surface.

113

time, then its removal could expose more erodible sediments beneath. In such cases, extraction from the channel bed should be discouraged or prohibited. Conversely, where the entire bed deposit is fresh and evidently mobile. the removal of material may not have a significant effect on channel stability, especially if the quantities concerned are small compared to the volume of bed load.

Drainage of hairpin stacks

10.37 The disposal of water from hairpin stacks is unquestionably a major hazard, both for the road and for the surrounding hill slopes. It is imperative to ensure that water is discharged into channels that are fully protected. As much effort should go into the protection of these channels as into the protection of the road itself. 10.38 Side drain runoff at hairpin bends is often conveyed via contour drains to discharge into an adjacent catchment, frequently causing severe erosion problems on the slopes and to the channels below. It is preferable, therefore, to contain all water within the stack system itself, thus avoiding the construction of drainage structures remote from the road whose inspection and maintenance might otherwise be overlooked. 10.39 This approach presents two alternatives for design. One is to construct a large reinforced side drain around the

outside of each hairpin bend, the other is to install a relief culvert beneath the carriageway at each bend to take runoff into the inside side drain below. The main problem with the former option is the fact that drain failure will lead to erosion of the hill slope below and eventual undermining of the hairpin bend. Although the latter is the preferred option, its main drawback is one of awkward geometry. Side drain runoff is forced to make two ninety degree turns, and the reduced gradient between the inlet and outfall restricts the size of culvert that can be utilized. There is also the cost of these extra culverts and the required greater side drain capacity to be considered. 10.40 Where there is no choice but to discharge water onto a hillslope, the following sites should be sought in order of preference: gently sloping or terraced ground a slope formed in strong bedrock a concave soil slope to assist in energy dissipation.

A standard detail for side drain turnout flow dissipation and erosion protection is illustrated in Figure 10.7. Most turnouts curve in plan, which throws high flows to one side, thus concentrating discharge and increasing erosive power. The flow can be more evenly distributed by providing a flared and baffled outlet.

114

Figure 10.7 Suggested apron details for side drain turnout 115

11

ROAD RETAINING WALLS

extent to which the wall foundation excavations intrude into the roadway must be considered construction of retaining walls often takes place early in the sequence of works and porterage of constrution materials, and the availability of trained labour, may affect the choice of wall type permanent drainage is always an important consideration, but it may be necessary to consider the provision of temporary drainage measures if wall construction or backfilling will not be completed before the onset of a wet season particular care should be exercised in assessing foundation conditions in previously disturbed soils, such as is often the case on road improvement or reconstruction projects where material from earlier slides may have been tipped adjacent to the road. If it is contractually feasible to open up wall excavations before deciding on the form of wall construction, then this should be done road retaining walls are usually designed to a standard cross-section on the basis of assumed achievable bearing capacities, horizontal backfill slope, backfill friction, permeability and drained conditions. Toe walls with sloping backfills, and walls constructed to retain failed or failing slopes, do not lend themselves so readily to a standard approach, and usually will require separate design considerations.

INTRODUCTION
11.1 Retaining walls area common feature of road construction in mountainous regions and can account for 20% of total construction costs. For low cost roads, walls constructed to masonry or gabion are by far the most common, since the bulk of the materials can usually be obtained locally, leaving only cement or gabion wire to be brought from other sources. They are constructed for the following situations: to support a road either wholly or partly on fill when the ground profile is too steep (usually greater than 30 degrees) to allow an embankment slope to support the toe of a slope that has failed or is likely to fail to support cut slopes that would otherwise require a low, uneconomic angle of cut when there are constraints on the permissible plan extent of earthworks, as on hairpin bends, or hairpin stacks, and in densely populated areas as revetments to prevent erosion on steeply sloping cut faces (Chapter 12) as part of a slope stabilisation scheme.

11.2 Problems of access in steep terrain can make it impracticable to carry out sufficient foundation investigation during the pre-implementation stages of a project. This may prevent the compilation of definitive designs or even delay decisions on appropriate wall types. These circumstances lead to the following recommendations regarding the general approach to retaining wall design: it is often better not to design walls in great detail in advance of construction, but instead to provide standard designs and to make adequate provision in construction contracts for detailing designs at that stage; standard designs should cater for the range in the type of wall and cross-section that can be adopted under varying topographical conditions, foundation bearing capacity, and to suit locally available materials and expertise, thus encouraging competitive pricing.

WALL TYPES 11.4 Typical details of the most commonly used wall types are shown in Figures 1 1.l and 11.2.

Dry masonry
11.5 Dry masonry (unbound masonry or dry stone) walls are usually the cheapest form of walling and are suitable for heights of up to 3-4m. They should not be used as road retaining structures for heights greater than 4m, and preferably not greater than 3m, although there are cases where dry masonry walls have performed competently at heights greater than these. Wall failure is usually caused by lateral loads in the backfill, seepage pressures and vibration from heavy vehicles. In respect of the last factor, dry masonry walls should not be constructed within 1 m of the road edge. The width to height ratio varies between 1:1 and 0.6:1 for wall heights of Imto4m respectively. Skilled masons and suitable stone are required for the construction of quality dry stone walls, because a neat fit and good interlock of the stonework are critical to the wall's strength and stability. The selection of masonry stone should be subject to strict quality control, as should be the construction of the wall itself.

11.3 There are practical considerations in the design and construction of retaining walls: in the case of road widening and reconstruction, it may be necessary for construction to be corned out while maintaining access for traffic, in which case the

116

Figure 11.1 Typical details for masonry road retaining walls

117

Figure 11.1 Typical details for masonry road retaining walls (continued)

118

Figure 11.2 Typical details for gabion and reinforced earth retaining walls

119

Figure 11.2 Typical details for gabion and reinforced earth retaining walls (continued)

120

Masonry with dry stone inclusions (composite walls)

11.6 Composite masonry walls are similar to mortared masonry walls except that they have panels of dry masonry about 0.6-1 m square forming a grid on the face with 0.5m division strips. They are stronger than dry masonry walls and, at the same time, maintain the advantage of relatively free drainage. They are used frequently in many mountain regions as slope support and revetment structures in cuttings through weak rocks, but are difficult to specify and construct as road retaining walls in complex and undulating side-long ground. It is doubtful whether they result in any significant savings in construction cost over ordinary masonry walls as they are more time consuming to construct.

Gabion walls
11.11 Gabion walls are usually preferred under conditions of poor foundation, wet soils, high groundwater and slope movement due to creep, landsliding and seismicity. However, the density of gabion work is approximately 70% that of mortared masonry and hence a gabion wall has to be larger in volume for the same retained height. This increase in size could be a factor in the choice of wall type in situations where crosssectional space is limited. Gabion walls have the following characteristics: they can accommodate settlement without rupture they allow free drainage through the wall they can be constricted in short sections (2-3m at a time) which minimises the temporary loss of slope support during excavation the shape of a gabion structure can easily be varied to accommodate changes in ground conditions across the site a gabion structure is less easy than a masonry structure to fit into an irregular foundation because of the standard size and rectangular shape of the boxes.

Masonry walls
11.7 Mortared masonry walls are the most durable of the low cost wall options. They are especially suited to steep rocky ground where foundations are shallow and where the contractor's working area is restricted. 11.8 The base width of these walls is usually between 0.5 and 0.75 times the wall height, and their base should be stepped into a rock foundation or constructed on an unreinforced 300mm thick concrete base, if rock does not comprise the foundation. If the height of the wall varies along its length, (that is, if the base steps down and up to accommodate changes in founding level), the location of vertical joints should reflect the position of foundation steps, and the frequency should be adjusted to minimise differential wall movements. 11.9 Masonry walls are not tolerant of differential settlement. Their lack of structural flexibility and general impermeability, even with freely-draining back fill and weep holes, make these walls inappropriate on wet colluvial slopes and where ground movements are expected, although some cracking in a masonry wall can be tolerated provided it is not accompanied by major deformation. There are many instances where severe cracking and sometimes complete failure have resulted from wall construction across adjoining soil and rock foundation materials. Weep holes should always be connected to gravel interceptor drains in the back-fill, and the lowest weep hole should be no more than 0.2m above final ground level at the toe of the wall. 11.10 The skills required to construct masonry walls are widely available in many populated mountainous areas. They are easier to build than dry stone walls because any lack of fit between the stones is taken up by the mortar. Rounded masonry stone should represent a maximum of a third of the total stone content. The minimum dimension of tabular stone should be at least 50% of the maximum. It is important to ensure that the wall interior is not filled with dry stone rubble as a cost-cutting exercise.

11.12 Being an unbound structure, the strength of a gabion wall depends upon the mechanical interlock between the stones. The strength of a gabion wall where the boxes have been filled by hand can be far higher than that of one where they have been filled mechanically, because the stones can be packed to a higher density. For high densities to be achieved it is important that as much attention is paid to the packing of the infill stones as to that of the facing, although rigorous site control is necessary for this to be assured. The joints between gabion boxes should be spanned by frequent `stretcher' boxes, as for bonded brickwork, orientated both along the wall and from front to back. These will restrict pulling apart and bulging. Long, flat stones should be orientated from front to bark in order to resist diagonal shear forces that pass through the wall. Where river terrace deposits are likely to form the predominant source of gabion stone, rounded stone should be limited to a third of the total stone content, and evenly distributed throughout each box. All other stone should be angular, or dressed to make it so. Further details of gabion wall construction are given in Box 11.1. 11.13 Where a gabion wall is founded on an uneven bedrock surface, the lower boxes will deform into the hollows, and the sliding resistance will not be as high as for a wall with a masonry or a concrete slab base. If bedrock is encountered at a shallower depth than expected, which prevents the intended design level from being reached with boxes of standard height, the lowest course can be built of mortared masonry up to the level of a course of boxes,

121

Box 11.1 Construction of Gabions Gabion structures offer remarkable qualities in terms of strength, flexibility and free drainage. However, they must be built to a high standard if these attributes are to be attained. The following guidelines are recommended ensure drainage is provided from the lowest point of the foundations use high grade wire with a thick galvansing mesh should be either a proprietary welded mesh or a triple-twist hexagonal mesh of 100mm width and 120mm length panel frames should be made using 8 swg wire, and mesh should be made using 10 swg wire wire all gabion boxes together using additional wire of 10 swg during construction, add four or five cross-trusses (of 10 swg wire) per square metre in each horizontal dimension ensure that the minimum dimension of all stones is larger than the wire mesh size stones should be tabular, of even size, and angularity if the boxes are packed by hand, all of the stones should be carefully packed, not just the facings wire the lids down with additional wire of 10 swg backfill behind the gabion structure using even-sized filter material.

on the toe. Another effect is that the fill behind the wall will settle and could subject the wall to unpredictable shear stresses.

Reinforced earth
11.15 Reinforced earth walls have not been used to any great extent in developing countries, due primarily to a lack of design expertise and the lack of appropriate backfill compaction plant. The main advantages of reinforced earth walls are their ability to deform without significant loss of serviceability at heights of up to 16m and the fact that they can be constructed with a range of fills. They can be useful in places where stone for gabions is not available within economic haul distance of the site. 11.16 The principal disadvantages with these walls are:

the cost of, or difficulty of obtaining, the steel strip or geotextile reinforcement the need for a high level of compaction, including wall edges, which can only be achieved with mechanical plant the requirement for an adequate horizontal clearance between the facing of the wall and the natural ground in order for the reinforcements to develop the required tension resistance. This implies that a) the ground may have to be cut back further than for a conventional retaining wall, and b) there will be additional delay before the space between the wall and the hillside is backfilled, thus increasing the possibility of local failure developing in the hillslope a higher cost than masonry for walls less than 10m high.

above which construction can proceed in gabion as planned. The top of a gabion wall can be finished with a course of masonry if a seal is required between the road surface and the wall, in order to form an effective road drainage channel. 11.14 The preferred maximum height of a gabion road retaining wall is 10m, although walls of up to 14m have been constructed on occasions with only minor deformations. Under high lateral and vertical loads there is a potential for stone crushing, especially in the toe of the structure. Crushing and compaction of the lower courses of gabion boxes in this way can result in a volume loss of as much as 20/0. If the stones in the toe of the wall crush under the load of those above, the effect on the wall is to settle at the toe, causing the top to tilt outwards and increase the load

11.17 Reinforcing elements maybe difficult to obtain, and for most developing countries they would have to be imported. Steel reinforcing strips may be susceptible to corrosion and sufficient sacrificial allowance has to be made in determining strip thickness, depending on the aggressiveness of the back-fill material. Geosynthetics are not subject to corrosion, but they possess low stiffness relative to steel and the amount of deformation required to achieve maximum shear strength is higher. They also require protection from ultra-violet light. Geogrid does not require the use of expensive and unattractive concrete facing panels. Instead, the reinforcing can be wrapped around the outside face of the wall from one layer to the next. The face of the wall can then be covered with soil and vegetated.

Buttress-supported road slabs

11.18 Buttress - supported road slabs (Figure 11.3) have been used very successfully to extend the width of a road beyond the outer face of an existing masonry retaining wall,

122

in places where a cut into the hillside is unfavourable and where good founding material exists upon which to found the buttresses. The buttresses act as piers and carry a reinforced concrete deck. They may be used to heights of 12-15m, although this is dependent on good rock foundations. Designs that obviate the need for ground supported scaffolding as falsework are to be favoured. Buttress-supported road slabs could equally be used in new road construction in steep sidelong ground, where appropriate.

Box 11.2 Retaining wall design in Nepal Road retaining walls in Nepal are generally designed for static loads only, and consequently retaining walls and backfill must be in place soon after slopes are excavated. Stresses induced by seismic or shear loading are not normally catered for. The recently observed effects of seismic acceleration on retaining structures indicates that masonry walls, especially those with foundations less than 1.5m deep, are far more susceptible to damage and failure than gabion walls, because they tend to rotate outwards as rigid structures due to shaking of the wall and settlement of the backfill. By contrast, gabion walls up to 10m high have been observed to remain stable with foundations as shallow as 1-1.5m. Bearing pressures under seismic loading have been calculated to be in excess of 300kPa for masonry walls of 7m height. Foundation depths of 2-2.5m have been recommended for the design of retaining walls in earthquake-prone areas, but the favourable performance of gabion walls in this instance suggests that, for these structures, the recommendation can be relaxed. Back-sloping gabion walls have an inherently greater resistance to sliding, they have not been favoured in many situations, as the angle of batter makes construction work awkward, and the foundation of the wall needs to be drained in the same way as masonry, thus increasing the cost. 11.23 The type of wall to be built at any one location is selected on the basis of ground conditions at the site and construction cost, although the latter is almost invariably the deciding factor. While there is considerable latitude in matching the type of wall to ground conditions, there are circumstances under which the wrong choice of structure will lead to deformation or even failure of the wall. 11.24 The most suitable form of retaining wall and its cross-sectional area depend upon: whether the best design from a geotechnical standpoint provides an adequate roadway width available cross-sectional area of the site depth and volume of material to be excavated to foundation level lateral stress expected to be applied to the wall bearing pressure of the wall and bearing capacity of the foundation geology and groundwater conditions whether a flexible structure is necessary

Mass concrete and reinforced concrete walls

11.19 Neither of these wall types are generally suited to road construction in mountainous areas as they require large quantities of cement and crushed aggregate, and are relatively expensive. Reinforced concrete walls also require a greater diversity of technical skills to ensure that a good standard of construction is achieved. Reinforced concrete walls are most appropriate as toe retaining structures to unstable slopes in association with rock anchors, where foundation and anchoring conditions permit. Anchored sheet pile and caisson walls are rarely applied in a low cost situation, except where sheet pile is used as a temporary shoring measure.

STABILITY, CROSS-SECTION AND ARRANGEMENT

11.20 The design of retaining walls is covered in standard texts, some of which are referenced in Chapter 14 (Bibliography). Only those factors especially relevant to wall design on steep, wet and frequently unstable mountain slopes are considered here. Wall design and construction in riverside locations are discussed in Chapter 13. 11.21 Both standard designs and detailed site designs should have a sufficient factor of safety against overturning, sliding and bearing capacity failure. Under normal loads, where a wall is designed to support a horizontal backfill slope, a minimum factor of safety of 2 against overturning, 1.5 against sliding or shearing and 3 against bearing capacity failure is applicable. A minimum internal shear strength of 600kPa is desirable. 11.22 The choice between walls with a horizontal base (upright) or walls with a base slightly inclined into the wall (back-sloping) will require consideration of cost, the availability of fill, stability and ease of construction. Some of the more obvious advantages and disadvantages with various wall cross-sections under different ground conditions are shown in Figure 11.4. Although upright retaining walls are usually taller and therefore more expensive on steeply sloping ground than back-sloping walls, they are generally easier to construct and allow better compaction of backfill. Another advantage is that base pressures are more evenly distributed, and therefore compressive stresses at the toe of the wall will be lower. However, on slopes underlain by hard rock, this advantage may not be significant. Although

123

extent of subsurface drainage required availability of construction materials availability of construction skills method of construction.

11.28 For high walls (4-7m) the bearing pressures, particularly beneath the front edge of the structure can become considerable, often in excess of 200kPa under static loading conditions. For masonry walls of whatever profile, a toe slab or toe projection at the base should be specified to reduce compressive stress and scour due to runoff over the wall surface. 11.29 A 0.5m gravel and pipe drain should be constructed along the back of a masonry wall foundation to facilitate drainage. In addition, gravel drains (grips) should connect the rear drain to the front of the wall at not more than 20m intervals. This drainage feature is essential for walls with a back-sloping base. Where a wall has a stepped foundation to fit into a topographical depression, water will move along the wall, and accumulate at the lowest point in the footings. A subsurface drainage system is required here to collect water and deliver it safely into the outfall. Ideally, weep holes should be provided in the concrete base of masonry walls founded on soils to allow drainage. The foundation platform for gabion walls built without a basal slab must be either naturally freelydraining, or have drainage grips installed.

Backfill
11.25 A well compacted backfill is vital to the serviceability of the road above a wall, and possibly to the wall itself. The consequence of poorly-compacted backfill is that, in time, settlement will cause the road to subside and crack, causing unpredictable lateral shear stresses within the wall. Most walls are designed on the basis of a specified backfill value under drained conditions. It is important to ensure that these design parameters are achievable under site conditions. Ideally, backfilling and compaction should keep pace with the wall as it rises, to give as much space as possible to deliver a good compactive effort. Compaction in layers of 200-300mm is usually specified. Special care is required at the base of the excavation behind the wall, where it is difficult to achieve good compaction because the working area is very restricted.

Surface drainage
11.30 The top of the wall is often finished with an upstand that rises to about 0.75m above the road surface, or serves as the base for a series of roadside edging blocks or parapets. This edge prevents water from draining randomly over the edge of the wall where the road has an outward camber. A bituminous fillet can be formed to prevent water entering a gabion structure, and to convey it to the end of the wall. However, even a slight easing of a gabion wall, or cracking of the fillet, will allow water to ingress into the backfill, and it may be appropriate to install a filter drain behind the gabion walls to a maximum depth of lm below the road in order to prevent seepage erosion of the finer materials from the sub-base. 11.31 If the wall is on a sag curve, then water collected at the low point can be conveyed through a gap constructed in the upstand, down to a stream course with erosion protection, as required. 11.32 The slope at the end of the wall is vulnerable to scour from water running off the road behind the wall. The outfall should be protected by a slope revetment formed into a channel, and delivered to a safe discharge point. 11.33 The slope in front of a gabion wall, especially, should be free-draining (all upstanding or poorly-drained material will need to be removed).

Foundations and sub-surface drainage

11.26 If the toe of the excavation comes within about 0.5m of the slope surface it is convenient to dig away the upstanding portion of soil to form a continuous step along the front of the excavation. This platform provides a wider working area, and ensures that there is no obstacle to drainage in front of the wall. It is often left to the engineer on site to decide whether a particular wall should be founded at its design foundation level or whether excavation to a deeper, stronger level should take place. Simple hand probing tests, such as the Mackintosh probe and the dynamic cone penetrometer, when suitably calibrated, can provide a quick assessment of bearing capacity, and can augment the more qualitative assessments. 11.27 Where shallow colluvium (1-2m) is encountered, foundations should be excavated down to stable rock head or weathered rock/residual soil materials below the colluvium. Where the depth of colluvium is too great, a gabion wall will probably be the most appropriate choice. The allowable bearing pressure for any area of wall construction is dependent on: form of construction (a higher pressure can be used for settlement-tolerant structures) ground conditions at foundation level topography of slopes below the base of the wall.

124

Figure 11.3 Typical details for buttress-supported road slab

125

Figure 11.4 Comparison of common retaining wall configurations

126

12

SLOPE PROTECTION AND SLOPE STABILISATION

INTRODUCTION
12.1 It is important to differentiate between methods of slope protection and methods of slope stabilisation, as the former are designed specifically to combat slope erosion and shallow slope failure up to a depth of about 0.5m, while the latter are designed to rectify problems of deeper slope movement. Methods of slope protection comprise drainage control and surface treatments for soils and weathered rock that include masonry revetments, the use of vegetation and less conventional slope coverings including geotextiles, bituminous fabrics and gunite/chunam. 12.2 Methods of slope stabilisation normally involve more substantial engineering works, including slope regrading to reduce slope angles, retaining walls and drainage works. Usually, however, it is appropriate to combine elements of protection and stabilisation in order both to give their greatest chance of success and to prevent long term slope deterioration. Toe walls, localised slope grading, shallow drainage and the use of vegetation are the most common measures employed in a combined system.

shoulder drains or berms can prevent runoff from discharging over embankment slopes to those areas where erosion has already started, but this can have the effect of concentrating runoff elsewhere. 12.5 One of the most effective ways to control erosion on embankments and on natural slopes below a road is to take reasonable precautions to prevent its initiation. The following guidelines are recommended: avoid concentration of runoff wherever possible avoid disturbing the natural ground outside the areas to be used for construction do not allow construction plant to track natural ground in an uncontrolled manner (a track mark can be sufficient to concentrate flow and start an erosion channel) pilot tracks should be constructed only when there is a guarantee that erosion prevention measures will be in place prior to the onset of the next wet season every effort should be made to prevent spoil from being dumped outside the limits of designated spoil areas. Spoil is highly erodible; it can smother vegetation and serve to concentrate flow sufficiently to initiate erosion do not allow runoff to discharge, either temporarily or permanently, onto unprotected natural ground, other than in pre-existing drainage channels.

SLOPE PROTECTION Embankments

12.3 Embankment erosion is usually initiated as a result of one or more of the following factors: the side slope is too steep or too long for the embankment materials to withstand erosion embankment materials have not been compacted to specification concentrated road runoff is permitted to drain over the shoulder.

Cut slopes
12.6 Table 12.1 illustrates the range of measures regularly used to treat cut slope erosion and failure along mountain roads. Usually, factors of cost, availability of materials and practicality will limit the selection of measures to those that can be applied on a low-technology, labour-intensive basis. It is apparent from Table 12.1 that a number of measures may be required to solve each particular problem. These often involve a combination of slope trimming, slope support, revetments, slope drainage and vegetation applications. Many of these measures will be applicable to the prevention and control of slope erosion as well as to the stabilisation of shallow slope failure.

12.4 Embankment erosion starts very often at the road shoulder edge, where the level of compaction tends to be relatively low, rather than on the slope surface. Revegetation of embankment slopes is most rapidly achieved by planting with grass slips, by the spreading of collected topsoil containing roots and seeds or by sodding with turves. Turves should be cut from level, fallow fields, or from areas specifically cultivated for the purpose. Grass slip planting is frequently the most effective. A tough grass with a low form and creeping habit should be used, such as Cynodon dactylon which is a common pan-tropical species. Erosion of embankment slopes can also be prevented to some extent by careful selection of the material in which the embankment is constructed, if a choice is available. Well-graded soils with some cohesive fraction offer better erosion resistance than single-sized non-cohesive soils. The use of

Bio-engineering
12.7 Where a slope is subject to erosion or very shallow slope failure, bio-engineering methods of slope protection are appropriate. Bio-engineering is the use of living plants, either alone or in conjunction with engineering structures and nonliving plant material, to reduce erosion and shallow-seated instability on slopes. In bio-engineering applications there is an element of slope stabilisation as well as slope protection: for convenience in discussing bio-engi -

127

Table 12.1 Range of methods commonly applied to the treatment of cut slope erosion and failure

128

neering as a subject, both functions are included here. There is a wide range of techniques associated with bioengineering, which cannot be described here. Gray and Leiser (1982), Gray and Sotir (1996), Coppin and Richards (1990) and Schiechtl and Stem (1996) give good accounts of bio-engineering systems and management. 12.8 Vegetation is a very desirable means of providing slope protection for reasons of availability, relatively low cost, appropriateness of installation techniques and compatibility with a rural environment. It is particularly appropriate in situations where large areas of slope are affected, a common situation on road cuttings and over unstable mountain slopes. The enhancement of road-side vegetation also has a positive effect both visually and in terms of plant diversity. 12.9 Planting schemes should if possible be undertaken in cooperation with local farmers, to keep grazing of newly-planted sites under control and to minimise the effects of soil saturation and runoff from farmland above road cuttings. Concentrated seepage or runoff from irrigated land is a common cause of slope failure, even on slopes that have a well established cover of vegetation. 12.10 The principal advantages of vegetation from a slope protection point of view are: the presence of a vegetation cover protects the soil against rainsplash and erosion, and prevents the movement of soil particles down slope under the action of gravity vegetation increases the soil infiltration capacity, helping to reduce the volume of runoff plant roots bind the soil and can increase resistance to failure, especially in the case of loose, disturbed soils and fills plants transpire considerable quantities of water, reducing soil moisture and increasing soil suction

soil flowage. The uppermost soil layer can become saturated during periods of heavy rain, causing it to liquefy and flow. This process initiates within the top 50mm or so of the soil, but probably rarely progresses deeper than about 150mm over a whole slope surface before the heavy rain in a storm event ceases, or before a rill is developed and flow becomes channelled. The effect of soil flowage can be very destructive to minor engineering structures and young or even mature plants. localised translational shear failure of a soil slope. A depth of about 0.3m is perhaps the maximum practicable limit for effective physical restraint of a slope by plants. The total influence of plants, especially trees, on a slope goes much deeper. It is certain that the hydrological effects, both positive and negative, of plants go to at least one metre and probably to several metres as a matter of course.

12.13 Although vegetation cannot be designed or built to engineering specification in the conventional sense, it can be selected and arranged on the slope to perform a specific engineering function. This function should be identified as part of the process of bio-engineering design. The engineering functions of vegetative treatments are listed in Table 12.2, together with those of civil engineering structures for comparison. 12.14 The general characteristics required of bio-engineenng plants are: they should be adapted to the growing conditions of the general environment, as well as those of the individual site rapid growth long-living substantial root system easy to propagate and yielding abundant propagation material

12.11 In view of the above, it is evident that vegetation is important in the control of erosion and shallow forms of instability ( 1-3m depth at most), but that it plays no significant role in the stabilisation of deeper failures of soils or rock. Vegetation complements engineering structures, protecting them from scour and disruption, and preventing damage to the slope generally. 12.12 The mechanisms of failure on soil slopes for which bioengineering protection is appropriate are: erosion. Washing of soil particles over the slope surface. This process may be regarded as the removal of a skin of material only, to distinguish it from the deeper forms of slope failure that are given below, though where runoff becomes concentrated it can rapidly lead to deep gullying in erodible soils

appropriate to secondary considerations of socio-economic factors, eg a selected species should not be invasive or poisonous to livestock. 12.15 Ideally, a mixture of plant types should be introduced so as to give a range of rooting depths. This will tend to prevent continuous shear planes from developing in the upper soil layers, discouraging shearing from taking place. It is appropriate to use local species rather than imported material because native plants are more likely to be adapted to grow in the hostile conditions found on bare sites and be resistant to local diseases. Also, if a nursery industry is to be developed to serve the road project, this can be implemented more easily if the species are familiar to the nurserymen.

129

Table 12.2 Engineering functions of treatments

12.16 Revegetation should be carried out as early as possible in the rainy season, to maximise the time available for the plants to become established. Grasses planted late in the rainy season may not become well enough established to survive the following dry season. Grasses benefit from an increased growing season by being planted shortly before the rains and watered until the start of the wet season. Sites should be protected from grazing until the plants are large enough to tolerate some browsing. 12.17 Despite the apparently rapid growth of plants in the tropics and sub-tropics, large grasses and woody plants take at least three years to become well established. In subtropical countries with monsoonal rainfall regimes the dry season is undoubtedly a major factor causing this long period to maturity. However, grasses in particular can begin to contribute to slope protection within the first season.

Grasses 12.18 Grasses should preferably be a mixture of creeping and clumping types. Creeping grasses form a continuous root system (a root mat), whereas clumping grasses leave gaps between the plants that can be subject to erosion. The advantage of clumping grasses is that they can grow very large, with deep roots. 12.19 Grass can be established by hydro-seeding, turfing or by manual planting. Hydro-seeding is by far the quickest method of grass establishment on highway cut slopes and embankments. It is often sprayed onto a bio-degradable geotextile pegged to the slope surface, which helps to hold the seed mix in place until established. However, note that: hydro-seeding requires specialist equipment and application technology it is not possible to apply hydro-seeding at sites located far above or below a road, beyond the reach of spraying equipment

130

hydro-seeding is not appropriate for forms of slope failure involving soil flowage or deep mass movement.

ing stability conditions. Bamboos require careful propagation and raising in a nursery, and take about five years to reach maturity. Trees and shrubs 12.25 Trees and shrubs are placed as individual plants positioned at random on the slope. Trees should be of low height, or capable of good recovery after being lopped. A maximum height of about 5m is appropriate for trees on slopes, sufficient for them to achieve good stature without becoming too susceptible to leverage by wind. Trees and shrubs originate as woody cuttings taken from mature plants, as seedlings grown in tubular polythene bags (`poly pots'), or as seedlings lifted and brought with bare roots from a nursery. Cuttings are planted on site, and require no preparation other than to be planted without delay after cutting. 12.26 Cuttings can be rather slow to take. Although they may put out a good show of leafy branches, root development tends to lag well behind. Seedlings, although apparently less sturdy when first planted, produce a more even balance of roots and shoots. Also, although small they are fairly tolerant of disturbance. Species selection 12.27 The choice of vegetation species is dependent upon both the function which the plants have to perform and the local site characteristics. It is most important to choose a species that will thrive on the site. For this reason, and because of the wide variety of micro-sites usually found in mountainous areas, environmental factors may have to take precedence over engineering attributes of plants. However, in most tropical, subtropical and temperate areas there is normally a sufficiently wide range of plants available so that all criteria can be met without serious compromise. The selection criteria are: Engineering considerations engineering function of the system (Table 12.2) rooting type for the function required (ie deep or spreading) aerial structure. The main purpose should be to assist the engineering function, but the aerial part of the plant is also important for the ecological and social compatibility requirements listed below.

12.20 Turfing is rarely adopted in mountainous areas because usually there is so little grassland available that can be used as a source of turf without causing erosion or loss of grazing land. In remote, low cost road applications, planting or seeding of vegetation by hand is often the only practicable method. 12.21 It is not really known whether it is better to plant grass randomly or in lines on any particular site. Research on the subject is continuing: Figure 12.1 and Table 12.3 give some suggestions. Horizontal grass lines are efficient at catching moving debris, but the lines tend to retain water on the slope and increase the risk of soil saturation and flowage. Vertical (down-slope) grass lines release the water but suffer from erosion of the roots and death from drought. Thus, vertical lines cannot normally be recommended for slope protection. A compromise exists in diagonal grass lines, which retard the rate of water flow down the slope while encouraging enough infiltration for growth. 12.22 Another area of uncertainty in grass planting is whether to seed the slope or plant it with grass slips. (Slips are small complete grass plants divided off from a parent clump). The choice depends to some extent on the tendency of the species to spread by seed or division. Seeded slopes give a random planting configuration, and seeding has the advantage of speed of sowing. Grass slips have the advantage of establishing quickly and of forming definite lines from the start, which enables them to perform a function as physical barriers on the slope as soon as they are planted. Grass established from seed has little or no barrier effect until the second growing season. 12.23 Where grass is to be established by hand broadcasting of seeds the slope should be covered with a layer of mulch (cut plant material). Although this increases costs, mulch creates a protective environment in which the seeds can become established, and offers some protection to erosion-susceptible soils during the early stages of the wet season when revegetation is still in progress. If mulch is not provided, grass seeds can simply be washed from the slope. Mulching also protects very young, newly germinated grass plants against sun scorch. This can be a problem on exposed slopes in higher altitudes and tropical locations; it can also be damaging to early growth from some slurry-seed mixes. 12.24 Bamboo species can be very advantageous in places where ground conditions are suitable, and they are welcomed by farmers as having many uses. They grow best on damp, shady sites on fine soils. Bamboos are relatively light in weight and develop extensive and dense root systems. They can also tolerate erosion and adapt to chang-

Environmental considerations climatic conditions (temperature and moisture at the micro-site level) physical site conditions (stoniness, permeability and fertility of the slope material; see also Table 12.3)

131

Figure 12.1 Grass planting systems

132

Table 12.3 Selection of vegetative slope protection techniques

Slope angle START Slope length Material drainage' Good >15 metres Poor Site moisture 2 Damp Dry Damp

Optimal technique

>50 Good Poor <15 metres

Dry Any Damp Dry

Good

Any

>15 metres Poor 35-50 Good Any Any

<15 metres Poor Any

Good

Any

Diagonal crass lines Contour grass lines 1 Downslope grass lines and strengthened rills or 2 Chevron grass lines and strengthened rills Diagonal grass lines Jute netting and planted grass 1 Downslope grass lines or 2 Diagonal grass lines 1 Jute netting and planted grass or 2 Contour grass lines or 3 Diagonal grass lines 1 Horizontal bolster cylinders and tree planting or 2 Downslope grass lines and strengthened rills or 3 Grass seeding, mulch and wide mesh jute netting Herringbone bolster cylinders and tree planting 1 Brush mattresses of woody cuttings or 2 Contour grass lines or 3 Grass seeding, mulch and wide mesh jute netting 1 Diagonal grass lines or 2 Herringbone fascines and tree planting or 3 Herringbone bolster cylinders and tree planting 1 Contour strips of grass and trees or 2 Tree planting 1 Diagonal lines of grass and trees or 2 Tree planting Direct seeding of shrubs or small trees

<35

Any Poor Any

Any

Any

Any rocky 3 material

Notes 1. Material drainage is related to the permeability of soils and the likelihood of their reaching saturation and losing cohesion under intense rainfall, thereby starting to flow. 2. Assessment of site moisture should include a consideration of overall slope moisture. This is an environmental dryness factor related to the sum of local topographical and climatic variables. Rain shadow effects can be significant in this, although aspect is usually the most prominent factor, especially outside the tropics. 3. "Any rocky material" is defined as material into which rooted plants cannot be planted but seeds can be inserted in holes made with a steel bar This table was supplied by the Geo-Environmental Unit, Department of Roads, Kathmandu

133

adequate availability of seed or plant stock without the risk of depleting natural reserves appropriate propagation and planting method ecological compatibility between plants on the site compatibility of the chosen species with the community of local people (eg positively useful plants or undesirable plants such as weeds).

tally across a slope. Fences made of organic material have a life of only two or three years at most in a wet sub-tropical environment before rotting. Wire fences, made with steel posts and wire rails and mesh panels, last for 8-10 years. 12.33 Fences are intended to support placed soil or moving debris, providing a stable environment in which plants can grow. However, during heavy rain, water can build up in the soil behind these structures, causing the soil to flow out beneath. Otherwise, if the fence is weakened by decay, the pressure of wet soil may push it over. Because of this, and the relatively small amount of soil they retain, fences can be regarded as having limited application for slope stabilisation in hot, wet climatic environments. Jute net 12.34 Woven jute netting can provide useful temporary protection and a stable environment for grass establishment in landslide back scars and on eroding cut slopes. It is sometimes possible to have the net made locally from jute grown in the region. Jute netting can reduce the velocity of surface flow, and is capable of retaining small volumes of soil debris and a degree of soil moisture necessary for the growth of grass slips in drier areas. It has been applied to cut slopes as steep as 80. Jute netting is unsuitable where soils are wet or are undergoing active shallow failure. 12.35 Netting should be secured to a smooth slope with staples, live pegs or split bamboo. Live pegs offer advantages of low cost and the capability to grow. The life of the fabric can be extended from one to at least three years if it is soaked in a bitumen solution prior to fixing. Palisades, brush mattresses and fascines 12.36 Palisades, brush mattresses and fascines are made from live woody cuttings. Palisades are fences consisting of closelyspaced upright cuttings, the line of cuttings being placed horizontally across a slope. The continuity of the fence tends to break down over time through death of the cuttings by overcrowding and their subsequent decay. Brush mattresses and fascines are more durable, though less able to support debris immediately from the time of installation onwards. 12.37 Brush mattresses are formed from hardwood cuttings. The cuttings are laid with their lower end in a shallow trench and the aerial part sticking out above ground. For extra security they are often wired to wooden stakes driven into the ground at 1 m centres. An advantage of brush mattresses is that they allow excess debris to roll over them with minimal damage. This helps them to survive long enough to take root and grow into strong shrubs. 12.38 Fascines are bundles of live cuttings laid buried in soil. The cuttings can be placed in shallow cross-slope trenches or in earth supported by wattle fences. Fascines are capable of rapid shoot and root development and eventually

12.28 The selection of a vegetative technique for the protection of a particular slope depends on many factors relating to the site. It is necessary to consider these in as much detail as time and skill allow in order to arrive at the most appropriate technique. It is not practicable to quantify the many variables involved, so assessments rely principally upon experience. Table 12.3 gives a simplified guide to the selection of techniques on earthworks slopes in seasonally wet mountainous areas, and Figure 12.2 shows examples of bio-engineering systems. 12.29 It is normally possible to draw up a list of local plants that fit most of the above requirements for bio-engineering. The department of botany in a university may be able to give advice on the characteristics and distribution of species within the region. Departments of agriculture and forestry have been found less able to help because their expertise lies with economic species, which plants useful for bio-engineering purposes frequently are not. However, these institutions are excellent sources of advice on setting up and running nurseries, and on plant propagation and care. 12.30 The relation between the length of life of engineering structures and the time that a plant takes to grow to maturity is relevant to the question of combining plants and civil engineering structures effectively. If a treatment such as jute netting is to protect seedlings until they are established, the treatment must last long enough before decaying to be able to perform this function. If not, the slope will be at risk of being damaged before the plants are capable of protecting it. 12.31 Engineering structures made of dead organic products such as jute, wood or bamboo can be regarded as being `temporary' - defined here as having a serviceable life not exceeding five years. Materials such as jute netting and bamboo last only one or two seasons in a tropical environment, but a coat of bitumen will increase their life to three years or even more. 'Permanent' materials such as stone and cement-bound products have the potential to last much longer than five years, so the distinction between temporary and permanent structures is usually obvious. Slope fences 12.32 Slope fences are traditionally made of woven wattles (flexible stems) or split bamboo and placed horizon-

134

Figure 12.2 Examples of bio-engineering slope protection systems

135

form a dense shrubby barrier. Their purpose is to catch moving debris and provide support for the slope through the interlocked root systems. Impermeable membranes and surface coatings 12.39 Gunite has been used widely and successfully in Hong Kong and Malaysia for the prevention of erosion and rockfall from weathered or fractured rock slopes under conditions where immediate protection is required. It comprises cement-stabilised aggregate (shotcrete) sprayed onto a wire mesh slope covering. Although it is standard practice to incorporate weep holes in the sprayed surface, the technique is generally inappropriate under conditions of groundwater seepage. Its application to low cost mountain roads is limited by its expense and the predominance of colluvial soils and complex drainage conditions in many cut slope excavations. 12.40 Chunam is a lime-based screed which is usually applied to a soil slope by trowel. Again, it has been used successfully in Hong Kong to prevent erosion of slopes formed in residual soils. As with gunite, the use of chunam under conditions of groundwater seepage or high soil moisture content can have an adverse effect on stability, despite the incorporation of weepholes into the covering. In addition, where a chunam cover becomes cracked it can lead to concentrated runoff and rapid erosion of the slope beneath the cover. Although chunam

tion work are used on steeper sections of cut slope to support overhanging portions of rock or boulders, and to fill in any cavities in the slope surface formed by localised failure and seepage erosion. Gabion toe or breast walls are usually 1 or 2m in height and are designed to protect the base of a cut slope from fretting, or to provide a small retaining capacity to loose and usually failed materials. Surface drainage 12.43 Surface drains (Figure 12.4) are frequently constructed on hillsides, for purposes of slope stabilisation, or across the top of a cutting to prevent surface runoff from eroding freshly exposed materials. In the latter case, they are usually constructed between 3 and 5m behind the top of the excavation, and on slopes up to 35: above this angle construction becomes impracticable. They are normally built as a trench, lined with either mortared masonry or dry stone. Polythene-lined drains are cheaper and more flexible than masonry but they are often vandalised and punctured, especially when adjacent to footpaths and housing areas. 12.44 At sites where some seepage from the drain is allowable, earth bunds can be considered. These can be unlined and grassed, or stone pitched, and can be further reinforced with vegetation on the downslope side. Generally, bunds require more careful construction than trenched surface drains because they need to be accurately positioned in order to maintain a continuous longitudinal fall. 12.45 Obviously, surface drains are only effective when surface runoff rates are significant. Surface runoff can be expected only during high intensity rainfall on moderate to steeply-inclined slopes, on slopes of low permeability where vegetation is patchy, or where runoff from agricultural land becomes concentrated onto unvegetated soil slopes. If surface runoff is substantial, and there is a clear threat of erosion or slope failure further downslope, the use of surface drains is justifiable. However, they do represent a hazard in the sense that they are easily damaged or blocked by debris or leaf litter, and are often not seen and therefore not cleaned on a regular basis. In addition, differential settlement or ground movement will dislocate masonry drains, leading to concentrated

probably has a greater potential use on low cost mountain

roads than gunite because it can be applied by hand, it requires a maintenance commitment which usually cannot be relied upon. Also, on environmental grounds, it is a less preferable option than vegetation. Revetments 12.41 Revetments are designed as slope protection rather than slope support structures (Figure 12.3). They are most frequently constructed in dry stone, mortared masonry, composite dry/mortared masonry and gabion. Their construction at the toe of a cut slope can be of particular benefit to the reduction in seepage erosion and softening of materials that would otherwise lead to progressive erosion and failure of the entire cut slope. 12.42 Where see pages are encountered, it is preferable to employ freely-draining revetment structures, such as dry stone pitching and composite dry/mortared masonry grids. Mortared masonry revetments are frequently employed where slopes formed in weathered rock require protection against erosion induced by surface runoff rather than failure of the surface materials. Masonry buttresses and denti-

seepage, if they are constructed without polythene lining. If there is any doubt about their effectiveness, or whether they can be maintained in the long term, it i5 better not to build them than have them become forgotten and allowed to fall into disrepair. making drainage and instability problems worse. Some further considerations of surface drain construction are listed in Box 12.1.

136

Figure 12.3 Revetment structures

137

Figure 12.4 Common types of contour and cut off drains and their application

138

Box 12.1

Factors to be considered in the design of surface drains

water collected by the drain must be discharged safely, in a manner that will not initiate erosion elsewhere construction of masonry-lined drains should be limited to undisturbed slope materials. Differential settlement, which frequently occurs in made ground and particularly at the interface between natural ground and fill, will lead to rupture drain gradients should not exceed 10

analysing and remedying these landslides are not compatible with the low cost design philosophy of mountain roads, especially in remote areas where access for drilling and excavation machinery is often difficult (Chapter 7). 12.47 The following precautions and actions should be taken, in approximate order of preference, to avoid or minimise the engineering damage and traffic disruption caused by large landslide movements: avoid the landslide by prudent choice of alignment wherever possible minimise length of alignment affected and cross at most stable location design earthworks cross-section, ie proportions of cut and fill, so as to improve stability or minimise slope disturbance (Chapter 9) carry out measures to improve temporary stability and accept periodic loss of access and a long-term maintenance commitment stabilise locally, using a combination of comparatively low cost drainage, slope support and erosion control measures that, overall, lead to a satisfactory increase in stability in the vicinity of the alignment. (However, it may be necessary to accept that the road suffers further deformation or is sometimes closed) stabilise entire slope using capital and machineryintensive methods.

stepped drain outlets should be provided with a cascade down to the collection point drains should discharge into a stream channel wherever possible, and preferably into channels that already convey a sizeable flow in comparison to the drain discharge low points in the drain system should be designed against overtopping by widening or raising the side walls lengths of drain should be kept short by the construction of frequent outlets in order to reduce erosion potential should drain failure occur where it is not practicable to discharge cut-off drainage into an adjacent stream channel, cascades can be constructed down the cut slope to convey water into the side drain. However, these structures are often vulnerable to the effects of side splash, undermining by seepage erosion and concentrated runoff along their margins. They must be designed to contain the water, and their margins must be protected with vegetation or stone pitching.

SLOPE STABILISATION
Introduction 12.46 Large (greater than 50,000m3), deep-seated and active landslides on natural slopes are relatively rare due to the frequency of small failures and erosion processes that tend to remove weathered and unstable materials as soon as they develop. However, where large landslides are encountered, they can lead to repeated road loss or deformation. These consequences must be weighed against the likely costs, practicalities and long-term success of capital-intensive stabilisation works. In most cases of this nature, a flexible gravel pavement across the zone of movement with appropriate seasonal maintenance, is by far the best solution until the landslide becomes dormant. Generally, the costs and technical back-up required for investigating,

12.48 As with the prevention or control of erosion, slope stabilisation usually employs a number of methods to remove or reduce the causes of failure that are external to the slope, such as toe erosion protection (Chapter 13), together with measures to improve the stability of the slope itself, including drainage, slope support and reprofiling. Where capital-intensive measures are to be employed, it will usually be necessary to carry out some form of ground investigation (Chapter 7) in order to be able to define an acceptable factor of safety for the design of stabilisation works that are most likely to succeed. Sub-surface drains 12.49 Drainage is often the most important factor in slope stabilisation, and is usually undertaken by the diversion of water away from the slipped mass by surface drains (discussed above) and by drainage of the slipped mass itself. Drainage design should be based on an appreciation of the natural drainage patterns of the slope, especially the locations of springs and see pages. Filter or interceptor drains can be used independently or in association with surface drains. However, as with surface drains, drain dislocation for whatever reason can make matters significantly worse.

139

12.50 Herringbone drains tend to be the most common forms of shallow (usually lm deep) sub-surface drainage adopted to tackle instability problems in earthworks and natural slopes along low cost mountain roads. The use of herringbone systems, with diagonal feeder drains wrapped in wire gabion panels and an impermeable membrane on their invert and downslope side, will reduce the tendency for dislocation to develop. Nevertheless, significant movement of the drain will affect its hydraulic conductivity to the point that it may become locally surcharged. Furthermore, the excavation of sub-surface drains should not itself be allowed to disturb the stability of the slope. 12.51 To facilitate rapid extraction of water, a perforated plastic pipe is frequently placed on the floor of the drain and wrapped in filter fabric to prevent clogging by fines. Diagonal feeder drains should be lined on the uphill side with a geotextile filter fabric to prevent clogging. The choice of filter fabric depends upon expected rates of subsurface flow and the particle size distribution of the surrounding soil. Too fine a fabric may prevent free drainage of the slope immediately above the drain; too coarse a fabric will allow fines to enter the gravel core. 12.52 Deeper trench drains (up to 3m in depth) are occasionally used to obtain a greater drawdown of water. Generally, 3m is the maximum depth advisable for drain excavation by manual means. In reality, groundwater inflow, the frequent presence of large boulders and the problems of safety in both shored and unshored excavations mean that most trench drains are excavated to depths much shallower than this. The spacing of trench drains will depend on permeability and the drawdown required, but generally varies between 3m and 10m. 12.53 In large debris masses where periodic mass movement occurs during saturated conditions, deeper trench drains can be used very effectively. Trench drains can be used to stabilise the mass by removing excess water at critical periods, and thereby avoiding the loss of cohesion which comes with saturation. The simplest situation is where erosion of the debris mass has no adverse consequences, and open drains can be used to initiate a semi-natural system of gullies. In this case, unlimited drains can be used, with the option of vegetative protection of drain sides in the second wet season. In some cases, short sections of drain bed lining or drop structures may also be required. Where erosion of the debris is a potential hazard. then the drains must be lined or backfilled with freely-draining material. 12.54 Trench backfilling materials include gravel, rubble or no-fines concrete, with the last providing an additional buttress against slope movement. Perforated pipes can be used to facilitate drainage in the same manner as for herringbone systems. Ideally, trench drains should be excavated at right angles to the slope contours to minimise slope disturbance and facilitate rapid drainage. Backfilling should

commence as soon as deep drains are excavated in order to minimise the possibility of trench collapse and induced slope failure. 12.55 Horizontal drains are designed to intercept groundwater and concentrated seepage at depth Their use on low cost mountain roads is comparatively rare. Examination of drill core and logging of water strikes during drilling operations can greatly assist in the identification of see pages, although it is common to find that only a relatively small percentage of the installed drains are successful in lowering pressures within water-bearing strata. Because of the usual uncertainty over their likely performance, their effect in reducing water levels should not assume critical significance in the design factor of safety. 12.56 Installation of horizontal drains into failing soil or rock masses is cautioned as drain dislocation and leakage could lead to accelerated movement. Drains usually comprise 50-100mm diameter slotted PVC pipes inserted into drill holes on a rising gradient of 5-10%. Pipes should be wrapped in filter fabric to prevent clogging although, even with fabric wrapping, pipes may require cleaning by water flush every 5-10 years. Where horizontal drains are successful, drain discharge rates of 1,000 litres per day can occur.

Support for soil slope failures 12.57 Along with drainage, toe retaining structures, constructed in either masonry or gabion, are employed most frequently in soil slope stabilisation. It is necessary to design the wall against overturning, sliding and bearing capacity failure (Chapter 11), as well as to design it as an integral part of the overall slope stabilisation scheme. The success of any retaining wall depends upon the adequacy of its design, construction, foundation stability, key into adjacent slope materials, and scour protection, if positioned in vulnerable river-side locations. 12.58 Other techniques of slope support are occasionally used for which specialist services are required. These are briefly outlined below. 12.59 Anchored reinforced concrete retaining walls. These require a bedrock foundation and may be contemplated when the required restraining forces are such that a gravity retaining wall of sufficient dimension would be impracticable due to foundation or space limitations, and where a stable rock mass behind the failing slope can be used for anchoring purposes. The principal advantage with this design of wall is that it can be constructed in confined spaces, while its main disadvantage is high cost. 12.60 Cast-in-situ contiguous bored-pile walls. These offer a lateral resistance to slope movement while allowing drainage to pass through the space between each pile. They are almost always designed with a key below the slip

140

surface in rock and an anchored reinforced concrete bulkhead across the front face of the wall, anchoring it back into the hillside. Piles constructed at wider centres can induce an arching effect in moving soil masses that tends to increase stability, but the uncertainty over their design and performance, and the need to employ piling equipment in their construction, usually makes them impracticable. 12.61 Soil nailing. Soil nails are designed to transfer the load of an unstable soil mass in shear to the stable material below the failure surface. Soil nails may be up to 5m in length and they are installed either by high pressure injection or by drilling and grouting. Their main advantage is that they can be used to stabilise soil masses under space limitations that restrict the use of more conventional methods. Also, unlike anchors, once installed they require no maintenance. If the necessary equipment can be made available, the technique may have some limited application along low cost mountain roads, although factors of cost, access difficulties for machinery and the general success of more conventional methods of stabilisation will usually prevail. 12.62 Reinforced earth. Reinforced earth retaining walls are discussed in Chapter 11. Support for rock slope failures 12.63 Retaining structures. Large rock slides are generally not amenable to stabilisation within a low cost project

philosophy. Masonry retaining structures can be constructed to some effect if space permits, but the removal of failed rock in order to construct a toe wall may have little overall influence on stability, and might disturb the slope further during excavation. Anchors and anchored bulkheads can provide support against rock creep, dilation and gradual planar failure over relatively small areas (up to 400m2), although progressive failure will eventually affect all anchorage systems if they are not maintained. 12.64 Slope re-profiling. The trimming back of steep slopes to more gentle slope angles may have a locally beneficial effect on stability, but it is generally not feasible to do this over large areas. Slope unloading at the head of a landslide, to reduce the driving forces of failure, presents practical difficulties and is very costly in the case of large landslides. Toe weighting, using rock fill berms and road embankments, can significantly increase stability. 12.65 Rock fall containment. Rock fall debris can be contained by means of cut slope benches and slope netting, together with a rock trap ditch, wall or fence at the base of the slope. Rock fall containment measures are only applicable in critical areas. High impact wire nets and flexible post systems, developed in alpine Europe and North America, are relatively costly and require specialist installation and maintenance. Rock trap ditches and walls at the base of a slope take up space but are easy to build and maintain.

141

13

ROAD CONSTRUCTION ALONG VALLEY FLOORS

INTRODUCTION
13.1 As discussed in Chapter 4, where there is a choice it is usually preferable, on hydrological and stability grounds, to adopt a hillside rather than a valley floor alignment. However, the choice of corridor will depend upon the length and practicality of hillside and valley floor options, and the degree of hazard posed by slope failure and flooding along each. Furthermore, some valley floors are significantly more hazardous than others, and it will be necessary to carefully evaluate the risk implications of these hazards before an alignment is chosen. 13.2 Valley floor and lower valley side alignments can encounter some or all of the following landforms and hazards: broad rivers that may rise and fall rapidly by several metres on a regular basis, and by as much as 10m or more in response to high magnitude rainstorms and GLOFs (Chapter 8) rivers which are actively meandering and changing their plan-form, which could subsequently encroach on the alignment active river flood plains that are likely to flow full at least once a year. The erosive power against the banks of a river in flood is very great vigorous tributary streams that are usually highly erosive and capable of transporting large volumes of sediment fans from tributary valleys that are either i) high level, with or without active incision, ii) in equilibrium, or iii) rapidly aggrading (building up by accumulating debris) flood plain terraces that may be susceptible to river scour on numerous occasions during the wet season, and inundation once every 2-3 years higher level terraces that may be subject to scour on a regular basis where they protrude onto the active flood plain rock spurs or promontories that project into the flood plain, forming obstacles to river flow and road alignment steep, and often eroded, rock slopes on the outside of valley meander bends slope instability on the lower valley sides in general.

13.3 These conditions are most common on youthful valley floors, and especially those with gradients steeper than 1 in 20. The rivers that occupy these valley floors dram steep and frequently unstable catchments. Their flood plains will be either so confined and erosive that the development of terrace sequences has not been possible, or will be subject to cyclic erosion and side slope instability over engineering time-scales to an extent that any preserved terrace surfaces cannot be regarded as safe for road alignment. In such situations, valley floor road alignments should be avoided altogether, otherwise frequent loss of significant sections of road will be inevitable. 13.4 Where a valley floor is comparatively mature, and ancient high level terraces are well preserved, then a road alignment located at the back of these terraces, combined with intervening rock cut, may prove satisfactory. If valley side rock mass conditions are not especially adverse to stability, it is usually preferable to construct a road in full cut, or a combination of cut and retained fill through these rocky areas, with a freeboard above the highest anticipated flood level. Where valley side stability conditions are unfavourable, or where river flooding could cause erosion and slope failure to extend far enough upslope to undermine road foundations, it is advisable to examine the practicalities and costs of an alternative alignment altogether. 13.5 The various design considerations associated with road construction in valley floor locations are discussed below.

FREEBOARD
13.6 It is usual to provide the road surface and associated structures with a freeboard of 2m above design flood level to accommodate surface waves and to provide some leeway in the estimation of flood level. The freeboard can be reduced to 11.5m in cases where the hydraulic analysis is more reliable. However, the calculation of the design flood (Chapter 8) is a particularly difficult task when rainfall and flow gauging data are limited or non-existent, and where a catchment runoff regime is subject to short term fluctuations brought about by road construction, land use change, extreme rainstorms and cycles of slope instability, channel incision and aggradation. Although widely appreciated, it is important to remember that flood levels can be substantially higher on the outside of meander bends than anywhere else along a given reach.

FLOOD PLAIN SCOUR AND EMBANKMENT PROTECTION

13.7 Flood plain scour, flood plain deposition and valley side instability usually occur at predictable locations. However, external influences, such as tributary fan incursions onto the flood plain, temporary landslide dams and engineering structures, can cause significant short-term modifications to flood plain processes and flow patterns. These

142

should be identified and monitored during the course of construction and maintenance, with appropriate steps taken to protect or locally realign affected sections of road. 13.8 Notes on the calculation of channel scour are given in Chapter 8. It can be assumed that maximum velocities around the concave (outside) banks of river bends and in valley constrictions are between 1.5-2 times greater than average or calculated velocities. On highly active flood plains with mobile bed material, predicted and actual scour depths can frequently exceed 5m, and occasionally 10m. Foundation excavations for road retaining walls and other structures are often impracticable at these depths, given the bouldery nature of the bed material and the requirements for dewatering the excavation. 13.9 Mortared masonry walls are more durable than gabion walls in abrasive riverside locations. and they have the potential to arch over small areas of scour, where gabion walls are more likely to deform. Even when heavy duty selvedge wire is used, gabion boxes are easily broken open by debris-laden water flowing at velocities greater than 4m/ sec, which is not unusual. 13.10 Where there is no choice but to construct a retaining wall within the zone of highly erosive floodwaters it is worthwhile extending foundation excavations deeper than the depth required for bearing capacity considerations alone in the expectation that bedrock will be encountered, to obtain a stable foundation for a masonry wall. Alternatively, where the foundation is composed of a significant proportion (usually 50% or greater) of large boulders, the softer materials can be excavated and replaced by concrete to provide a stable foundation for a masonry wall. 13.11 However, it is frequently the case that neither of these foundation conditions are achievable within practicable excavations depths, and especially on the outside of river bends where scoured bedrock and boulders have been replaced by finer-grained materials. The potential for foundation scour in these situations will usually dictate that a flexible gabion structure is adopted, in preference to a more rigid masonry one (Chapter 11), and combined with whatever scour protection works are feasible under the circumstances. Foundation stability can be improved by constructing the retaining wall on a concrete raft, thus reducing differential settlements. Sacrificial walls, double thicknesses of gabion mesh, gabion mattresses (Box 13.1) and stone rip-rap are likely to prove effective during small and medium-sized flood events only, and will require regular repair or replacement. Reinforced concrete rip rap can be fabricated in situ if sufficiently large local stone rip rap is unavailable or cannot be transported to the site, as is often the case. However, the cost of fabricating rip rap to the required dimension (3m in some cases) is usually prohibitive, and it is usual to adopt a compromise solution under conditions of extreme scour potential.

Box 13.1 Gabion mattresses and aprons When constructing gabion scour protection works it is important to create a smooth and regular surface without protrusions and to extend the measures beyond the zone of immediate vulnerability. Gabion mattresses (up to 0.3m thick) are designed to deform into scour holes, thus providing some protection against scour progressing to the wall foundation. The inner edge should be anchored or lapped beneath the wall or slope, and the outer edge can be buried and anchored well below the flood level. A mattress can also be laid on fill slopes up to the design flood level. Gabion aprons are generally 0.5-1m thick and, although snore robust, are not capable of deforming effectively into scour holes. Where possible, they should be constructed with their top surface buried beneath the level of the flood plain, as close as possible to the expected scour depth, and if not to this depth, then at least no higher than the flood plain surface. All gabion structures built within reach of floodwaters should be laid upon or backed by filter fabric, or a graded gravel filter, to prevent fines from being washed out through the stones, and the structure undermined. 13.12 At some sites, reinforced concrete may be the only means of providing effective scour protection. This expensive option should only be contemplated at high risk sites, and where dewatered excavation for foundation can be advanced to maximum anticipated scour depths. If this is not achieved, basal scour and undermining may take place without detection from the surface, until large scale failure occurs.

RIVER TRAINING
13.13 Groynes, guide walls and other such structures will have a comparatively short life on active flood plains. However, training works can have some beneficial effects during low-moderate flow conditions and, even on hostile flood plains, they may survive for 4-5 years. Their serviceability at vulnerable sites for periods much greater than this is doubtful, although the life of a river training structure can often be extended by the use of rip rap protection, grouting and wide gabion aprons on the upstream side. 13.14 The siting and spacing of groynes should be determined by the flood flow regime identified from field evidence, although the practicalities of construction and excavation will often be deciding factors. If groynes are regarded as sacrificial, rather than permanent measures, it will probably be more worthwhile in the longer term to use capital resources to protect the road itself better. 13.15 River training can also be attempted by large-scale earthmoving on a flood plain to construct linear flood channels and boulder barrages, and to steepen the flood

143

plain cross-gradient away from the road alignment. Such practices require heavy earth-moving plant and may lead to unpredictable redistributions of flood plain scour and side-slope erosion in the short term. In the longer term, the river will recreate its original flood flow pattern.

order to allow the design of a road crossing that consists of: a relatively short span bridge approach embankments vented causeways and drifts (multi-culverted embankments, Figure 13.1) to cater for other distributary channels.

CROSS DRAINAGE AND TRIBUTARY FAN CROSSINGS

13.16 Where alignments are located on the lower slopes of steep valley sides, cut slopes can truncate drainage channels with the result that, during heavy rain, sediment and water may overshoot culvert inlets and discharge directly onto the road surface. This is usually remedied by constructing a large catchwall between the culvert inlet and the road edge, or by providing a dished concrete causeway. On occasions, concentrated runoff and slope failures will erode new gullies that will require some form of culverting or other drainage provision. A causeway is likely to be the only practicable remedy. 13.17 Tributary fans present a range of problems for road alignments. It is useful to differentiate between: mature and stable high level fans equilibrium fans immature and unstable fans.

13.21 The above combination will usually be cheaper to construct and represents a more convenient solution than a multi-span bridge or a lengthy detour into the tributary to find a suitable shorter span bridging site with stable abutment and pier foundations. However, river training and erosion protection of bridge abutments, piers and approach embankments will prove costly and difficult to maintain in the long term. In addition, vented causeways should only be considered where sediment loads are likely to be low. 13.22 Immature and unstable fan surfaces are usually characterised by cyclical regimes of erosion and deposition across the whole fan surface during the course of individual storms, and a general process of fan aggradation from one year to the next. The crossing of these fans presents severe problems to valley floor alignments. Sediments will accumulate wherever flow velocity decreases as a result of an abrupt concavity in the channel profile or a sudden increase in channel width. Ideally, valley floor alignments should cross fans immediately upstream of these concavities, and if possible at the fan apex. 13.23 Bridge clearance of at least 5-7m should be provided above the existing fan surface level wherever a fan is actively aggrading and its tributary catchment is unstable. However, it is not always possible to conform with this recommendation due to alignment constraints, unstable valley sides adjacent to the fan apex, or the fact that the apex is poorly defined and the fan itself extends upstream into the tributary valley. Under these circumstances, a combination of the following may be the only viable option: gabion checkdams in the stream channel above the fan and erosion control in the catchment above, to control the stream bed level river training and scour protection works upstream and downstream of the bridge, to control the stream course a commitment to regular maintenance and waterway clearing operations, to keep flow within the channel and to provide room for the accumulation of debris during fan-building episodes.

13.18 Mature and stable fan surfaces are usually preserved as old, high level landforms that have become incised by rejuvenated stream channels. The principal problem for road construction on high level fans, other than alignment constraints, is the choice of a suitable site to cross the incised channel, bearing in mind that its banks will be composed of unconsolidated and erodible materials. 13.19 In the case of equilibrium fan surfaces, all sediment supplied to the fan is transported out of the catchment by one or a number of well defined channels. Equilibrium fans, and their wide flood plains immediately upstream, will usually comprise a number of distributary flow channels with only one or two of them occupied during normal flow conditions. Other channels may become occupied every 2-3 years or so, in response to floods or landslide - generated debris flows from further upstream. These fans are usually associated with terrace sequences on the adjacent valley floor and, therefore, represent a stage of drainage development between mature high level fans and immature, active fans on flood plains (described below). 13.20 A thorough understanding of the flow patterns across equilibrium fan surfaces is required before a road alignment and bridging structures are designed. Artificially increased channelisation and bank protection of the normal flow channel may increase its definition and capacity, in

13.24 Design strategies for crossing unstable fan surfaces are summarised in Box 13.2.

144

Figure 13.1 Typical details for drift crossings

145

146

14

BIBLIOGRAPHY

ANON, 1993. Bio-engineering for road protection in Nepal. Eastern region road maintenance project, Nepal. Available from Geo-environmental Unit, Department of Roads, Kathmandu, Nepal. (Unpublished). BAKER, V R, R C KOCHEL and P C PATTON (Eds), 1988. Flood geomorphology. New York: J Wiley and Sons. BARKER, D H (Ed), 1994. Vegetation and slopes. Stabilisation, protection and ecology. Proceedings of the International Conference, University Museum, Oxford. London: Institution of Civil Engineers; Thomas Telford. BRUNSDEN, D, J C DOORNKAMP, P G FOOKES, D K C JONES and J M H KELLY, 1975. Large scale geomorphological mapping and highway engineering design. Quarterly Journal of Engineering Geology, 8(2), 227-253. BRUNSDEN, D and D B PRIOR (Eds), 1984. Slope instability. Chichester: J Wiley and Sons. CARSON, B, 1985. Erosion and sedimentation processes in the Nepalese Himalaya. ICIMOD Occasional Paper No 1. Kathmandu, Nepal: International Centre for Integrated Mountain Development. CHARMAN, J H and J S GRIFFITHS, 1993. Terrain evaluation methods for predicting relative hazards from mass movement, fluvial erosion and soil erosion in the developing world. In: SOCIETY FOR EARTHQUAKE AND CIVIL ENGINEERING DYNAMICS. Natural disasters. London: Thomas Telford. CHOW, V T (Ed), 1964. Handbook of applied hydrology. New York: McGraw-Hill Book Company. COPPIN, N J and I G RICHARDS, 1990. Use of vegetation in civil engineering. London: Butterworths. DEARMAN, W R, 1991. Engineering geological mapping. Oxford: Butterworth-Heinemann. DEOJA, B, M DHITAL, B THAPA and A WAGNER, 1991. Mountain risk engineering handbook, Part I: Subject background, Part II: Applications. Kathmandu, Nepal: International Centre for Integrated Mountain Development. DICKENS, C H,1865. Note on the flood discharge of rivers and streams with reference to determining the size of bridges from drainage areas. Medley Indian Engineer-, II, 133-136. FARRADAY, R V and R G CHARLTON, 1983. Hydraulic factors in bridge design. Wallingford, UK: Hydraulics Research Limited.

ESSERY, I T S and M W HORNER, 1978. The hydraulic design of stepped spillways. CIRIA Report 33. London: Construction Industry Research and Information Association. FOOKES, P G, M SWEENEY, C N D MANBY and R P MARTIN, 1985. Geological and geotechnical engineering aspects of low-cost roads in mountainous terrain. Engineering Geology, 21, 1/2, 1-152. FOOKES, P G and A MARSH, 1981. Some characteristics of construction materials in the low to moderate metamorphic grade rocks of the Lower Himalayas of east Nepal. Part 1: Occurrence and geological features. Proceedings of tire Institution of Civil Engineers, Part 1, 70, 123-138, (paper 8380). FOOKES, P G and A MARSH, 1981. Some characteristics of construction materials in the low to moderate metamorphic grade rocks of the Lower Himalayas of east Nepal. Part 2: Engineering characteristics, Proceedings of the Institution of Civil Engineers, Part 1, 70, 139-162, (paper 8380). FOOKES, P G and M SWEENEY, 1976. Stabilisation and control of local rockfalls and degrading rock slopes. Quarterly Journal of Engineering Geology, 9(2), 37-55. GALAY, V, 1987. Erosion and sedimentation in the Nepal Himalaya. An assessment of river processes. Canadian International Development Agency. Water Energy Com-mission Secretariat, His Majesty's Government of Nepal. GANSSER, A, 1964. Geology of the Himalayas. London: Interscience Publishers. GEOLOGICAL SOCIETY OF LONDON, 1972. The preparation of maps and plans in terms of engineering geology. Quarterly Journal of Engineering Geology, 5(4), 293-382. GERRARD, J, 1993. Mountain environments. An examination of the physical geography of mountains. Chichester: Belhaven Press. GRAY, D H and A T LEISER, 1982. Biotechnical slope protection and erosion control. New York: Van Nostrand Reinhold Company. GRAY, D H and R B SOTIR, 1996. Biotechnical and soil bioengineering slope stabilisation. A practical guide for erosion control. New York: Wiley & Sons. GUPTA, R L,1973. A general formula for determination of peak flood. International Symposium on River Mechanics, Bangkok, 9-12 January 1973. International Association for Hydraulic Research. Paper B 13, 147-159. HEARN, G J, 1987. An evaluation of geomorphological contributions to mountain highway design with particular

reference to the Lower Himalaya. (Unpublished PhD thesis, University of London).

147

HEARN, G J, 1995. Engineering geomorphological mapping and open-cast mining in unstable mountains. A case study. Transactions of the Institution of Mining and Metal-urgy, Section A: Minerals Industry, 104, A1-A17. HOEK, E and J W BRAY, 1982. Rock slope engineering. London: Institution of Mining and Metallurgy. HOWELL, J, J E CLARK, C J LAWRANCE and I SUNWAR,1991. Vegetation structures for stabilising highway slopes. A manual for Nepal. Department of Roads, His Majesty's Government of Nepal. (Unpublished, but available from Transport Research Laboratory) HUTCHINSON, J N, 1977. Assessment of the effectiveness of corrective measures in relation to geological conditions and types of slope movement. International Association of Engineering Geology Bulletin, 16, 131-155. ILRI, 1979. Drainage principles and applications. IRLI Publication No 16. Wageningen, The Netherlands: International Institute for Land Reclamation and Improvement. INGLIS, C C, 1949. Behaviour and control of rivers and canals (with the aid of models). Research Publication No 13, Part 1. Poona, India: Central Waterpower, Irrigation and Navigation Research Station. IVES, J D, 1986. Glacial lake outburst floods and risk engineering in the Himalaya. ICIMOD Occasional Paper No 5. Kathmandu, Nepal: International Centre for Integrated Mountain Development. IVES, J D and B MESSERLI, 1989. The Himalayan dilema. Reconciling development and conservation. The United Nations University, Routledge. JONES, T E and J D PARRY, 1993. Design of Irish bridges, fords and causeways in developing countries. Highways and Transportation, January, 28-33. KELLERHALS, R, 1967. Stable channels with gravel paved beds. Proceedings of the American Society of Civil Engineers, Journal of Waterways and Harbours Division, 93, 63-84. KRAEBEL, C J, 1936. Erosion control on mountain roads. USDA Circular No 380. Washington DC: US Department of Agriculture. LAWRANCE, C J, R J BYARD and P J BEAVEN, 1993. Terrain evaluation manual. Transport Research Laboratory, State of the Art Review No 7. London: HMSO. LINSLEY, R K, M A KOHLER and J L H PAULHUS, 1975. Applied hydrology. New Delhi: McGraw Hill Publishing Company Ltd.

MANNING, R, 1891. On the flow of water in open channels and pipes. Transactions of the Institution of Civil Engineers of Ireland, 20, 161-207. MAYNORD, S T, J F RUFF and S R ABT, 1989. Rip-rap design. Journal of Hydraulic Engineering, 115, 7, 937-949. MAZZOCCOLA, D F and J A HUDSON, 1996. A comprehensive method of rock mass characterisation for indicating natural slope instability. Quarterly Journal of Engineering Geology, 29(1), 37-56. NEILL, C R (Ed), 1973. Guide to bridge hydraulics. Roads and Transportation Association of Canada. SAVAGE, W Z, H S SWOLFS and P S POWERS, 1985. Gravitational stresses in long symmetrical ridges and valleys. International Journal of Rock Mechanics, Mining Sciences and Geornechanics Abstracts, 22, 5, 291-302. SCHIECHTL, H M and R STERN, 1996. Ground bioengineering techniques for slope protection and erosion control. Oxford: Blackwell Science. SCHUSTER, R L and R J KRIZEC (Eds), 1978. Land-slides. Analysis and control. TRB Special Report 176. Washington, DC: Transportation Research Board. SHAFFNER, U, 1987. Road construction in the Nepal Himalaya. The experience from the Lamosangu-Jiri road project. ICIMOD Occasional Paper No 8. Kathmandu, Nepal: International Centre for Integrated Mountain Development. SMEC, 1993. 1993 flood damage assessment. Final Report. Snowy Mountains Engineering Corporation, in association with CEMAT Consultants (P) Ltd. His Majesty's Government of Nepal. STARKEL, L, 1972. The role of catastrophic rainfall in the shaping of the relief of the Lower Himalaya (Darjeeling Hills). Geographia Polonica, 21, 103-159. STOCKLIN, J, 1980. Geology of Nepal and its regional frame. Thirty-third William Smith Lecture. Journal of the Geological Society of London, 137, 1-34. STRICKLER, A, 1923. Beitrage zur frage der geschwindigheits - formel und der rauhigkeitszahlen fur strome, kanale and geschlossene leitungen. (Investigation into the question of the swiftness formula and roughness number for streams, channels and pipes). Mitteilungen des Eidgenossischer Arntes fur Wasserwistshcaft. (Communication of the Administrative Bureau for Water Resources). Bern, Switzerland.

148

TRL, 1992. A design manual for small bridges. TRL Over-seas Road Note No 9. Crowthorne, UK: Transport Research Laboratory. TRL, 1997. Hydrological design manual for slope stability in the tropics. TRL Overseas Road Note No 14. Crowthorne, UK: Transport Research Laboratory. TURNER, A K and R L SCHUSTER (Eds), 1996. Landslides. Investigation and mitigation. TRB Special Report No 247. Washington, DC: Transportation Research Board. VARNES, D J, 1984. Landslide hazard zonation. A review of principles and practice. International Association of Engineering Geology Commission on Landslides and Other Mass Movements on Slopes. Paris: UNESCO. WATKINS, L H and D FIDDES, 1984. Highway and urban hydrology in the tropics. London: Pentech Press Ltd. WILLIAMS, L G, 1991. Drainage of a mountain road in Nepal. Journal of the Institution of Water and Environmental Management, 5, 513-518. ZARUBA, Q and V MENCL, 1982. Landslides and their control. Amsterdam: Elsevier, 2nd revised edition.

149

ISSN 0951 - 8797

OS-G

Principles of low cost road engineering in mountainous regions

ORN 16

Transport Research Laboratory Old Wokingham Road Crowthorne, Berkshire, RG45 6AU

Department for International Development 94 Victoria Street London, SWIE 5JL

OVERSEAS ROAD NOTE 17

ROAD SAFETY EDUCATION IN DEVELOPING COUNTRIES GUIDELINES FOR GOOD PRACTICE IN PRIMARY SCHOOLS WITH PARTICULAR FOCUS ON THE SITUATION IN GHANA

Subsector:

Transport

Theme: Project title: Project reference:

TI: Reduce accidents and increase road safety in rural and urban areas Road Safety Development and Evaluation of Educational Programmes R6029

Copyright Transport Research Laboratory 1997. All rights reserved. This document is an output from a project funded by the UK Department for International Development (DFID) for the benefit of developing countries. The views expressed are not necessarily those of the DFID.

Transport Research Foundation Group of Companies Transport Research Foundation (a company limited by guarantee) trading as Transport Research Laboratory Registered in England, Number 3011746 TRL Limited Registered to England, Number 3142272 Registered Offices: Old Wokingham Road, Crowthorne, Berkshire, RG45 6AU.

ACKNOWLEDGEMENTS This document was prepared by Gill Murray and Jill Guy, both consultant teachers, whilst working for the Transport Research Laboratory, Overseas Centre. The Project Manager was Ivan Sayer. The authors wish to acknowledge the numerous schools, teachers and children whose help was invaluable. The authors are particularly grateful to Mr L Clocuh, Director, Regional Education Office, Mrs P Anson-Yevu, Assistant Director, Accra Metropolis District Education Office and Mr J Amegashie, Director, National Road Safety Secretariat.

First Published 1997 ISSN 0951-8797

OVERSEAS ROAD NOTES Overseas Road Notes are prepared principally for road and transport authorities in countries receiving technical assistance from the British Government. A limited number of copies is available to other organisations and to individuals with an interest in roads overseas, and may be obtained from: Overseas Centre Transport Research Laboratory Crowthorne, Berkshire, RG45 6AU United Kingdom

Limited extracts from the text may be reproduced provided the source is acknowledged. For more extensive reproduction, please write to: Programme Director, Overseas Centre, Transport Research Laboratory.

CONTENTS
Page 1. INTRODUCTION Why Have These Guidelines Been Written? Who Are The Guidelines for? What is Road Safety Education? What Are the Main Aims of Road Safety Education? Why is Road Safety Education Important? Who is Responsible for Accident Prevention? Why Do Children Have Accidents? How Can Accidents Be Prevented? 2. ROAD SAFETY EDUCATION - A NATIONAL POLICY Making Road Safety Education work at a national level STEP 1 STEP 2 Collecting Information Creating a Road Safety Education Committee Forming Policy and Delivery Structure Implementing the Policy Monitoring and Evaluating 1 1 Ghana 1 Nepal 1 Papua New Guinea 1 2 2 3 4 6. REFERENCES 19 20 17 17 15 5. ROAD SAFETY EDUCATION - EXAMPLES OF GOOD PRACTICE Page

15

APPENDIX A: Summary of `Safe Ways' APPENDIX B: Aide Memoire APPENDIX C: Proposed Council Members for National Road Safety Committee in Ghana APPENDIX D: Life Skills Syllabus

22 23

6 6

STEP 3 STEP 4 STEP 5

6 7 7

3. ROAD SAFETY EDUCATION (RSE) - THE CURRICULUM What are the key points in teaching Road Safety Education? Curriculum links 4. ROAD SAFETY EDUCATION - CHILDREN'S LEARNING OBJECTIVES Are there Universal Objectives for Children Learning to be Safer Pedestrians?

10 10

12

12

ROAD SAFETY EDUCATION IN DEVELOPING COUNTRIES GUIDELINES FOR GOOD PRACTICE IN PRIMARY SCHOOLS
WITH PARTICULAR FOCUS ON THE SITUATION IN GHANA

1. INTRODUCTION
FACTS Road accidents are a primary and growing cause of death and injury to children in developing countries. Road Safety Education is limited in developing countries, but the Transport Research Laboratory's research has demonstrated that this can be a cost effective remedial measure if the principles of good practice are followed.

WHY HAVE THESE GUIDELINES BEEN WRITTEN?

1.1 In response to the above facts, TRL, funded by DFID (Department for International Development), recently developed a specifically designed Road Safety Education resource, called `SAFE WAYS', (Summary of `Safe Ways'. Appendix A), for upper primary children, which was researched in Ghana. Preliminary results are encouraging. 1.2 Arising from this and other overseas work, and from their considerable experience of the British Road Safety Education scene, the authors have drawn together these international `Guidelines' in the hope that people will be inspired by the good practice of others. Whatever cultural differences exist between countries the same fundamental principles of Road Safety Education underlie good practice. 1.3 The Guidelines illustrate the importance of Road Safety Education, and show how a developing country might put in place a system that ensures primary children receive adequate Road Safety Education.

after in-service Road Safety Education training, and would be suitable for the following people: lecturers in colleges of education headteachers classroom teachers.

WHAT IS ROAD SAFETY EDUCATION?

1.6 In broad terms Road Safety Education is teaching children to be safer road users. It is providing a foundation and a guide on which the child can develop its own strategy for coping with the dangers and hazards of the road environment. Road Safety Education is an area of work which fits naturally into most subject areas and topics. How this can be achieved will be illustrated within these Guidelines.

WHAT ARE THE MAIN AIMS OF ROAD SAFETY EDUCATION?

1.7 To help pupils become safer road users by developing: their knowledge and understanding of the road environment their decision making and behavioural skills necessary to survive in the road environment their knowledge and understanding of the rules that influence and control the actions of road users their knowledge and understanding of the causes and consequences of accidents responsible attitudes to their own safety and the safety of others.

WHO ARE THE GUIDELINES FOR?

1.4 The Guidelines will be helpful at a national planning and policy level to the following people: policy makers in Ministries e.g. Education, Transport, Health, Public Works, Police administrators in government services e.g. curriculum advisors in the education service.

1.5 The Guidelines give examples of what and how to teach and of Road Safety Education's place in a school's curriculum. They could be distributed at a district level

WHY IS ROAD SAFETY EDUCATION IMPORTANT?

FACTS 500,000 people are killed in road accidents each year, and 350,000 of these are in developing countries. Road accident fatalities in a number of African and Asian countries are increasing annually. Pedestrian accidents are a particularly serious problem in developing countries. For example, in Africa 42 per cent of the road accident fatalities are pedestrians compared with only 20 per cent on average in Europe and the United States of America.

1.8 The road and traffic environment affects our lives in a profound way on a daily basis, whichever country we live in, whether we are adults or children. 1.9 Children in developing countries spend much of their time as pedestrians and are extremely vulnerable road users. Road Safety Education, improved driver behaviour, engineering measures and good practice measures need to be taken to reverse this fact. Ghana has begun this task. 1.10 Ghana has a particularly serious pedestrian accident problem, pedestrians representing 43 per cent of road accident fatalities of which 47 per cent were children. Ghana's Ministry of Transport and Communications is aware of the characteristics of its accident pattern, and is

using education as one means of addressing the problem. Any country should, like Ghana, be aware of its accident pattern before it can take remedial measures. In general, casualties increase as children get older and become more exposed to traffic. In Ghana child casualties are mostly in the 6-10 year old age group, as are most deaths. Other studies have shown that males have a higher pedestrian casualty rate per head of population than females, and this is also true of Ghana. In most European countries accidents to pedestrians tend to happen close together on local roads in areas where visibility is obscured. In Ghana most pedestrian accidents happen on urban roads away from junctions. The resource `SAFE WAYS' is an example of a good practice measure written in the knowledge of the country's accident pattern.

WHO IS RESPONSIBLE FOR ACCIDENT PREVENTION?

1.11 Many different groups of people are responsible for accident prevention: Police They need to encourage safe behaviour, and prevent dangerous behaviour by applying the law.

Politicians They need to know about the problems and how to resolve them: this may mean passing laws to do so. Community members They need to set an example of how to behave safely. Children copy what adults do. They need to be involved in initiating and supporting local action to improve road safety.

Policy makers They need to be aware of the problems and be able to implement solutions through policy making at ministerial level. Teachers They need to receive initial teacher training, and in-service training in Road Safety Education in order to give planned educational opportunities to their pupils. Parents They need to set a safe example, protect and educate their children. Infrastructure authorities They need to consider children, particularly as pedestrians, when planning and developing roads. Drivers They need to be made aware of children's limitations in traffic, and measures need to be taken to improve driver behaviour. Official road safety body For Road Safety Education to be effective there needs to be a national commitment at all levels, indicated in the diagram above, and a means of ensuring that all efforts are coordinated into a cost effective programme. This national commitment can be facilitated by a national body responsible for communication, co-ordination and publicity within the road safety field. It is important to ensure, however, that any such organisation does not just exist in name only, but is effective and dynamic. 1.12 Ghana is fortunate to have the benefit of a powerful national committee (National Road Safety Committee) as its official body. It is active in a variety of road safety fields and has an overview of policy and practice.

WHY DO CHILDREN HAVE ACCIDENTS?

1.13 The nature of children is to be impulsive and unreliable. They find it hard to give their attention to two activities at once, and often they are absorbed in play perhaps making them less able to understand real life situations. 1.14 They have limited experience. It should be remembered that children develop skills only gradually - they have not been alive for very long! 1.15 They are smaller than adults. Their vision is restricted because of their lack of height. Their perception is immature. 1.16 They are often unsupervised or poorly supervised by their parents. From studies of pedestrian accidents we know that children play outside their homes, and because these areas are sometimes quiet, perhaps some parents think that children can cope alone. 1.17 Various studies confirm these points: they found that children display: poor powers of perception, concentration, attention, memory, and physical and emotional control poor knowledge and understanding of traffic poor behavioural patterns in the traffic environment.

1.18 These characteristics apply to all children both from developed and developing countries.

HOW CAN ACCIDENTS BE PREVENTED?

1.19 Road accidents can be prevented in four main ways, namely the four E's.

has introduced a scheme on a trial basis to improve accident prevention and reduce accidents. 1.23 To encourage the transfer of suitable technology in this field TRL has published `Towards Safer Roads in Developing Countries - a road safety grade for planners and engineers' (TRRL, 1991). It gives examples of good and bad practices around the world, and highlights the fact, that many underlying general principles for planning and engineering affecting safety are, to some extent, universal.

Engineering - by altering the environment. Enforcement - by imposing laws. Encouragement - by the use of publicity campaigns. Education - by gaining and using knowledge.
1.20 It should be recognised that education is one measure in a package of measures that can improve road safety.

Enforcement
1.24 Accidents can be reduced through the application of the law. To be effective police need to be well equipped, knowledgeable and able to respond quickly and safely to any traffic incident. Traffic police also need to earn the respect of other road users by setting good examples of driving and general road user behaviour. 1.25 With generally lower standards of road user discipline in many developing countries than in developed countries, the potential for improving driver behaviour through the use of policing is high. This is especially true with regard to moving violations. Introducing a highway patrol on a major intercity road in Egypt led to a reduction in accidents of almost 50%. In Pakistan, regular highway patrols on selected sections of roads resulted in a 6%

Engineering
1.21 Human error is the major causal factor in road accidents. However, engineering measures and planning improvements, such as, traffic signals, pedestrians crossings, junction layouts and speed control humps can affect road user behaviour in such a way that errors are less likely to happen or when they do happen the environment can make them less serious. 1.22 There has been very little research in developing countries into the relationship between highway design standards and accident rates; thus many developing countries have just modified schemes from developed countries. Ghana, however, is one of a few developing countries that

reduction in accidents. On similar roads with no patrols accidents rose by 37%. However, police presence at signalised junctions in Kenya and Pakistan showed drivers having little regard for police presence in terms of the proportion of drivers passing red signals and stopping for pedestrians waiting to use zebra crossings. 1.26 In Accra, Ghana, traffic police hold regular training and instruction sessions for traffic violators. Apprehended drivers attend police headquarters for corrective training lessons by a police instructor. Public transport drivers allowing passenger overloading have their vehicles impounded for a limited time and are fined. Encouragement 1.27 Encouragement is the role of publicity which attempts to inform and educate road users with a view to persuading them to change their behaviour to make accidents less likely. There are many channels through which encouragement can take place e.g. newspapers, radio, television, magazines, books, tapes, films and videos, leaflets, posters and stickers. Often many of these media are used to promote a road safety campaign based on an agreed chosen problem area. 1.28 In Ghana the National Road Safety Committee (NRSC) is the springboard for most road safety publicity, choosing one or two topics a year to focus upon. For example a campaign `Save the Child' was launched in 1995

to inform the driving public of the vulnerability of children. The NRSC is responsible for many other media areas encouraging road safety awareness. Recently it has produced a video and a flip chart for use in schools and designed bill board posters to promote the use of seat belts and motorcycle helmets. The Committee also features regularly on television and in the newspapers to convey safety messages and produces information leaflets for distribution at work places and churches. Education 1.29 Road accidents can be prevented by acquiring and using knowledge. It is recognised that Road Safety Education programmes should be graded and structured developmentally and that teachers need guidelines on what and how to teach. Such programmes need to be relevant to the particular developing country, although fundamental principles of Road Safety Education underlie all good practice. 1.30 Engineering, enforcement, encouragement and education give a framework in which a road safety team is capable of co-ordinating and integrating a wide range of road safety improvements. At a strategic level it is important that Road Safety Education as a remedial measure is planned and developed within this multi-disciplinary context. The remainder of this document focuses on the development and implementation of Road Safety Education as an effective accident prevention measure.

2.

ROAD SAFETY EDUCATION - A NATIONAL POLICY

MAKING ROAD SAFETY EDUCATION WORK AT A NATIONAL LEVEL

2.1 For progress to be made nationwide in accident and injury prevention road safety has to be designed as a national goal with efficient management structures set up at national as well as at local levels. 2.2 The diagram below shows in a very simplified way how Road Safety Education can reach primary school children by following 5 steps. It requires commitment, co-ordination and communication at all levels to reach implementation. STEP 1 - Collecting information Information needs to be collected and collated when assessing a country prior to implementing a suitable Road Safety Education programme. A co-ordinator needs to be appointed for this purpose. Ideally this person should be appointed by one of the ministries responsible for children's safety e.g. Education, Health, Transport. They should have an education background since it will be their responsibility to assemble the Road Safety Education Committee mentioned in STEP 2. Figure 2.1 'Organisational Inquiries' illustrates key areas to be addressed, key people to be visited and key questions to be asked by the co-ordinator. Figure 2.2 `Organisational Model' displays the information gathered from Ghana. With thus background information a country can begin the task of developing a structure to make Road Safety Education work. (See Figures 2.1 and 2.2). STEP 2 - Creating a Road Safety Education Committee A Road Safety Committee for education needs to be created by people with the power to generate change.

Representation should come from disciplines such as education, police, health, highways and transport, and the national road safety body, if such an organisation exists. The size of the committee should be restricted to 12 - 20 people (see Appendix B). In Ghana the National Road Safety Committee (NRSC) has assumed the leading role in Road Safety Education. Education is one part of its function. STEP 3 - Forming Policy and Delivery Structure "In order that Road Safety Education be really effective it requires a clear structure within a recognised curriculum with a planned, sustained and coherent programme of learning" (DOT, 1995). Therefore the policy and aims of Road Safety Education need to be decided by the multidisciplinary committee referred to in STEP 2. Details of content should be written by curriculum specialists based upon the core Children's Learning Objectives suggested in Chapter 4. A feasible delivery structure needs to be carefully thought out and agreed upon, in the light of the information gathered. Cultural and religious considerations also need to be made when thinking about delivery. For example, a country might have a strong network of women's groups which is an established way of passing on health and safety messages. Successful transference of Road Safety Education for primary children in the UK is through teachers in schools. Ghana has begun to follow good practice by: producing a policy report for Road Safety (Crown Agents, 1995). The Report suggests that the best approach to Road Safety Education in Ghana is to develop institutional provision in formal education, i.e. through curriculum development, and also to utilise non-formal education methods such as publicity.

Step 5 Monitoring and evaluating Step 4 Implementing policy Step 3 Forming policy and delivery structure Step 2 Creating road safety education committee Step 1 Collecting information

planning to establish Road Safety Committees iii the larger cities and in each region, and by appointing 10 regional Road Safety Officers to assist with the delivery of Road Safety Education through teachers in primary schools.

Ghana it is proposed to use NRSC staff as resource persons in teacher training colleges. STEP 5 - Monitoring and Evaluating The monitoring and evaluation of Road Safety Education should take place at all levels, for example at school as well as at ministerial level. Road Safety Education is difficult to monitor and evaluate, because it is impossible to decide whether an accident would have happened with or without an educational input. However it is important that the Road Safety Education Committee be assured that the expenditure is bringing an adequate return in terms of greater knowledge and safer behaviour, which should lead to reduced accidents in the long term. It is advisable that one person with a research background should be appointed to have an overview of the monitoring and evaluation. According to the Policy Report for Road Safety (Crown Agents, 1995) the NRSC will link with the University of Ghana to carry out specified in-depth research and suggest ways of monitoring and evaluating publicity and education initiatives.

STEP 4 - Implementing the Policy Government ministries need to accept the recommendations of the Road Safety Education Committee and introduce policies to enable implementation at national and local level. Implementation at local level to primary aged children is usually through teachers in schools, but can be through the police, health workers, or community groups such as women's groups or churches. Both the UK and Ghana have taken the education route. If an adequate school system is in place the education route is recommended with support from other disciplines. It is Ghana's intention through the National Road Safety Committee to give some in-service training to teachers. In order to facilitate the inclusion of Road Safety Education in the school curriculum teacher training colleges will need to be encouraged to include it in their programmes. In

Fig. 2.1 Organisational inquiries for road safety education (RSE)

Fig. 2.2 Organisational MODEL for Road Safety Education (RSE) ref Ghana

3.

ROAD SAFETY EDUCATION (RSE) - THE CURRICULUM

RSE should be PRESENTED frequently

3.5 Children behave more safely immediately after some training and less safe some months later (Limbourg and Gerber 1981). This shows the importance of presenting Road Safety Education frequently.

WHAT ARE THE KEY POINTS IN TEACHING ROAD SAFETY EDUCATION?

3.1 There are 5 important points (`the 5 Ps') to remember when teaching children road safety. They can be summarised as follows. It should begin be follow be have a

RSE should have a PLACE in the school timetable

3.6 Road Safety Education must have a proper place in the school timetable. It must be decided where it belongs: is it a subject in its own right? (`Road Safety') is it part of another subject? e.g. life skills, health education, environment is it to be integrated into other school subjects? e.g. mathematics, social studies, English, science (see Figure 3.1 below) is it outside the school curriculum? e.g. road safety clubs in Ghana, cycle training in the UK, both take place after school hours.

PRE-SCHOOL PRACTICAL PRINCIPLES of child development PRESENTED frequently to school children PLACE in the school timetable.

RSE should begin PRE-SCHOOL

3.2 To be effective Road Safety Education should be given systematically and continuously throughout a child's school life. Its effectiveness largely depends upon an accumulation of learning and experience, which begins before the child can walk and never really ends.

RSE should be PRACTICAL

3.3 Research has shown that practical training is essential if pedestrian behaviour of young children is to be improved. Theoretical teaching in a classroom situation has been shown to improve knowledge without affecting the child's behaviour near roads.

CURRICULUM LINKS
3.7 In Ghana Road Safety Education is at present in the Life Skills syllabus, and the `SAFE WAYS' programme as a whole could be used there. Parts of the `SAFE WAYS' programme also link with other subject areas in the Ghanaian curriculum as shown below. 3.8 The most important thing is that Road Safety Education exists somewhere in the school timetable and it is identified. 3.9 Ghana has begun to address these issues. At present some Road Safety Education is in the Life Skills syllabus of their national curriculum (see Appendix D). More Road Safety Education is being planned for the future and is likely to appear in the proposed new Environment curriculum. 3.10 Two new resources are to be introduced: 1) SAFE WAYS written by TRL for DFID, is for teachers of upper primary school children. 2) `Road Safety Education in Primary Schools'. A Guide for teachers by the National Road Safety Committee.

RSE should follow the PRINCIPLES of child development

3.4 To begin with a very young child should be prevented by an adult from going near roads, and eventually the young person is competent alone in a busy traffic environment. To move the child from one extreme to the other teachers have to use their knowledge of the principles of child development. For example a child under 6 years old cannot put himself /herself in someone else's position; so it is difficult to imagine whether a driver can see him/her or not. The child under 11 years old finds it difficult to focus on the relevant and ignore the irrelevant - so perhaps would be distracted by a friend calling out when he/she is trying to cross the road.

10

Fig. 3.1 `Safe Ways' Road Safety Education Resource

11

4.

ROAD SAFETY EDUCATION - CHILDREN'S LEARNING OBJECTIVES

child demonstrating good practice to adult total independence with young person being the decision maker.

ARE THERE UNIVERSAL OBJECTIVES FOR CHILDREN LEARNING TO BE SAFER PEDESTRIANS?

4.1 In considering children's needs there should be a gradual progression from total protection to total independence. This involves interaction with parents/carers, teachers and peer groups with their importance varying at different times. Children should move through the progression as follows:

4.2 Passive methods of Road Safety Education are known to be poor. Recent research in the UK has shown that 5 and 7 year old children had virtually no ability to separate safe from dangerous places to cross (K Ampofo-Boateng and J Thompson, 1993). This shows that practical learning in the real road environment is vital. This approach has been used effectively in both the UK (RoSPA. 1994) and Ghana (Sayer, 1997), proving popular with teachers and pupils alike. 4.3 Each stage builds upon and reinforces previous learning. Therefore only new objectives are set out below. The stages are only a guide and one must always be aware of an individual's starting point.

total protection with adult leading child adult demonstrating good practice to child

12

13

14

5.

ROAD SAFETY EDUCATION - EXAMPLES OF GOOD PRACTICE

GHANA
5.1 The SAFE WAYS programme presented school pupils in Accra with novel opportunities to learn, both in the classroom and school compound, and in practical situations outside near real roads. The programme aimed to increase their road safety awareness

culminating in the children making safer journeys to and from school.

Plate 5.1) Upper primary school pupils in Accra participate in Road Safety Education using `SAFE WAYS'.

Plate 5.2) Primary school teachers in Accra participate in workshops before using `SAFE WAYS'.

15

Plate 5.3) A teacher in Ghana uses a poster discuss crossing the road safely with her class.

Plate 5.4) A poster based upon the National Road Safety Committee's flip chart used to discuss safe places to cross the road.

16

NEPAL
Nepal uses puppets to teach Road Safety Education 5.2 Nepal has changed drastically in recent years. Kathmandu was virtually a village 30 years ago. The roads now carry a mixture of bicycle, buses, lorries as well as ox carts and rickshaws, with the additional hazards of many farm animals moving along. 5.3 A road safety programme was implemented in 1992 because of the large number accidents. With illiteracy rates of 80% for women and 50% for men, puppets were chosen as a means of portraying road safety messages to children. Teams of men were recruited through the Scouts and trained to work in

puppets are cheap, transportable and easily made. After the show a discussion takes place to test the children's understanding. Then a demonstration is given with the help of teachers preferably near the roadside, or in the school grounds. 5.5 This approach has apparently been very successful; not only have the children benefited but teachers and mothers have also become involved through the demonstrations.

PAPUA NEW GUINEA

5.6 Papua New Guinea has produced a road safety book called `Road Smart' aimed at school children. 5.7 A variety of road safety messages, rules and information is contained within its 40 pages. They are presented in appealing and exciting ways. (See Figure 5.1 below).

designated areas.
5.4 The puppet show lasts about 20 minutes and deals with a single problem e.g. running into the road. The

Fig. 5.1 Example of a road safety message

17

6.

REFERENCES

AMPOFO-BOATENG K, THOMSON J A, and others, 1993. A developmental and training study of children's ability to find safe routes to cross the road. British Journal of Developmental Psychology, Volume 11, Number 1, pp 31-45. DoT, 1995. Road Safety Education in Schools: Primary Schools Good practice Guidelines. London: Department of Transport. CROWN AGENTS Ref: TEA/A369, 1995. (Unpublished contractors report for Ministry of Transport and Communications, Ghana). Road safety study and the institutional strengthening of the vehicle examination and licensing division. DOWNING, C Sand others, 1991. Trials of a road safety booklet for a pre-school traffic club. TRRL Laboratory Report 992. Crowthorne: Transport and Road Research Laboratory. LIMBOURG M and D GERBER, 1981. A parent training program for the road safety education of preschool children. Accident analysis and prevention, Vol 13 No 3, pp 255-267. RoSPA, 1994 . Let's decide - walk wise. Birmingham: Royal Society for the Prevention of Accidents, ISBN 85088 071 9 RoSPA. SAYER I A and others, 1997. Improving road safety education in developing countries; Ghana. TRL Report 265. Crowthorne: Transport Research Laboratory. TRRL OVERSEAS UNIT, 1991. Towards safer roads in developing countries: a guide for planners and engineers. Crowthorne: Transport and Road Research Laboratory.

18

APPENDIX A: SUMMARY OF `SAFE WAYS'

LESSON 1 STEPPING OUT
(Where and how to walk) Classroom: Practical Work: Homework: Work: Accidents and pedestrian behaviour are discussed. Practice in walking and stopping takes place in the compound. Learning is reinforced and the children are encouraged to pass on their knowledge to younger children.

LESSON 2 LOOKING AND TELLING

(Observing and environment) Classroom Work: Practical Work: Rules for safe walking are discussed. This looks at what can be found on the footpath and road, and takes note of observations made from the compound and/or near the road. Observation work continues and includes road signs.

Homework:

LESSON 3 ZEBRAS AND OTHER CROSSINGS

(Using protected crossing places) Classroom Work: Discussion takes place on what protected crossing places are, their names, and how to make use of them. The Green Cross Code is featured. This takes place in the compound or outside the school, and includes a demonstration of how to use at least one of these crossings. The personal use of crossings is considered especially on the journey to and from school. Passing on knowledge to younger children is encouraged.

Practical Work:

Homework:

LESSON 4 SEEING AND BEING SEEN

(Crossing where there are no protected crossing places) Classroom Work: There is a classroom discussion on how to choose the safest place to cross when no protected crossing places exist. Practice in the selection of a "safe" place, and crossing in a safe manner is carried out in the compound and/or outside the school. Using the Green Cross Code is included, and the importance of explaining it simply to younger children.

Practical Work:

Homework:

LESSON 5 GETTING THERE SAFELY

Choosing the safest route between home and school) Classroom Work: Discussion takes place on the problems encountered on school journeys, and suggestions are made on how to make routes safer. A large poster is studied for this purpose. On the way home pedestrians are observed, and personal routes to and from school to home are considered for their safety value. Pupils evaluate and try to improve their school route.

Practical Work:

Homework:

19

APPENDIX B: AIDE MEMOIRS

20

21

APPENDIX C: PROPOSED COUNCIL MEMBERS FOR NATIONAL ROAD SAFETY COMMITTEE IN GHANA
1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. Chairman of the Council Ministry of Transport and Communications Ministry of the Interior Ministry of Health Ministry of the Environment Ministry of Local Government and Rural Development Ghana Police Service Motor Transport and Traffic Units (MTTU) Ghana Education Service Ghana Highway Authority Department of Urban Roads Vehicle Examination and Licensing Department (VELD) Building and Road Research Institute (BRRI) Police National Ambulance Service Environmental Protection Agency Accra Metropolitan Authority Automobile Association Ghana National Chamber of Commerce

22

APPENDIX D: EXAMPLE FROM THE LIFE SKILLS SYLLABUS IN GHANA

23

ISSN 0951 - 8797

OS-E

Road Safety Education in developing countries Guidelines for good practice in primary schools

ORN 17

Transport Research Laboratory Old Wokingham Road Crowthorne, Berkshire, RG45 6AU

Department for International Development 94 Victoria Street London, SWIE 5JL

Overseas Road Note 18

A guide to the pavement evaluation and maintenance of bitumen-surfaced roads in tropical and sub-tropical countries

ORN 18

First Published 1999 ISSN 0951-8797 Copyright Transport Research Laboratory 1999.

Subsector: Transport Theme: T2 Project title: Rehabilitation of roads with bituminous surfacings Project reference: R 6023

This document is an output from a project funded by the UK Department for International Development (DFID) for the benefit of developing countries. The views expressed are not necessarily those of the DFID.

TRL is committed to optimising energy efficiency, reducing waste and promoting recycling and re-use. In support of these environmental goals, thus report has been printed on recycled paper, comprising 100% post-consumer waste, manufactured using a TCF (totally chlorine free) process.

Transport Research Foundation Group of Companies Transport Research Foundation (a company limited by guarantee) trading as Transport Research Laboratory. Registered in England, Number 3011746. TRL Limited. Registered in England, Number 3142272. Registered Offices. Old Wokingham Road, Crowthorne, Berkshire, RG45 6AU.

ACKNOWLEDGEMENTS This Overseas Road Note was produced in the Civil Engineering Resource Centre of the Transport Research Laboratory (Programme Director Mr T Toole) on behalf of the Department for International Development. The research has been carried out with the active collaboration of highway authorities in many countries and their help and cooperation has been essential to the success of the project. The TRL project team responsible for this Road Note were Mr C R Jones (Project Officer), Dr J Rolt, Mr H R Smith and Mr C Parkman. The helpful comments of Mr P May of May Associates are gratefully acknowledged. Where necessary, use has been made of work published by other research and road authorities.

OVERSEAS ROAD NOTES Overseas Road Notes are prepared principally for road and transport authorities in countries receiving technical assistance from the British Government. A limited number of copies is available to other organisations and to individuals with an interest in roads overseas, and may be obtained from:

International Development Advisory and Information Unit Transport Research Laboratory Crowthorne, Berkshire, RG45 6AU United Kingdom

Limited extracts from the text may be reproduced provided the source is acknowledged. For more extensive reproduction, please write to the address given above.

CONTENTS
Page
1 Introduction Scope of this note Project appraisal 2 Pavement evaluation and maintenance procedure 3 Interpretation of existing data 4 Surface condition and roughness surveys Surfacing defects Bleeding and fatting-up Fretting and stripping Loss of stone Surface texture Aggregate polishing Cracking Type Intensity Position Width Extent Deformation Rutting Depressions Corrugations Other types of deterioration Potholes and patching Edge failures and shoulder erosion Deterioration caused by pool-drainage Roughness measurements 5 Localised surfacing defects 6 Performance charts 7 Additional tests Deflection tests Dynamic cone penetrometer tests Destructive sampling and material testing Skid resistance tests 1 1 1 1 3 4 6 6 6 6 6 6 7 8 8 9 9 9 9 9 11 11 11 11 12 12 12 13 14 17 17 19 19 20

iii

Page 8 Identifying the causes of pavement deterioration Rutting without shoving Rutting with shoving Wheelpath cracking - asphalt surfacing Wheelpath cracking - thin bituminous seal Non-wheelpath cracking - asphalt surfacing Longitudinal cracking Transverse cracking Block cracking Crocodile cracking Non-wheelpath cracking - thin bituminous seal 9 Maintenance and rehabilitation Analytical approach Structural approach Deflection approach Maintenance options Reflection cracking 10 References 11 Applicable standards Appendix A: Detailed surface condition survey Appendix B: Road roughness measurements Appendix C: Deflection beam measurements Appendix D: Deflection beam survey procedure Appendix E: Falling Weight Deflectometer (FWD) test procedure Appendix F: TRL Dynamic Cone Penetrometer (DCP) test procedure Appendix G: Test pit procedure Appendix H: Sand patch test Appendix I: The portable skid-resistance tester 21 21 21 21 30 30 30 30 31 31 31 33 33 34 34 35 36 41 43 44 45 50 53

54

56 61 65 66

iv

1 Introduction
Scope of this note 1.1 This Road Note gives guidance on road pavement evaluation procedures suitable for bituminous-surfaced roads in tropical and sub-tropical climates and reviews alternative methods of maintenance and repair. It is intended primarily for highway engineers who are responsible for maintaining roads in tropical and subtropical environments but the techniques and principles on which it is based are equally applicable in other environments. 1.2 Paved roads in tropical and sub-tropical climates often deteriorate in different ways to those in the more temperate regions of the world, because of the harsh climatic conditions and often a lack of good road pavement materials. In addition, roads in many countries often suffer from accelerated failures caused by variable quality control during construction, high axle loads and inadequate funding for maintenance. 1.3 This Road Note describes methods of pavement evaluation designed to establish the nature, severity and extent of the road deterioration. It gives guidance on the use of non-destructive and destructive pavement tests and describes how the results of these tests can be interpreted, both to identify the causes of the deterioration and to assess the strength of the existing road. It also reviews alternative rehabilitation design procedures and comments on their limitations and advantages. Project appraisal 1.4 The process of road project appraisal is described in detail in Overseas Road Note 5 (TRRL, 1988). It can be summarised in the following stages: Road project identification Feasibility and preliminary design Detailed design Implementation Evaluation

released in 1999). During the detailed design stage, the pavement evaluation is based on similar information but the frequency of measurement is increased, to validate the findings of the feasibility study and to optimise the design of each segment of the project road. This Note gives guidance on pavement evaluation procedures which can be used during both the preliminary and detailed design stages of a project to maintain or upgrade an existing road.

2 Pavement evaluation and maintenance procedure

2.1 The process of selecting appropriate methods of maintenance or rehabilitation is shown in Figure 2.1 and can be summarised as follows: Collect and interpret existing design, construction and maintenance data. Carry out surface condition, roughness and traffic surveys. Carry out structural and materials testing. Establish the cause of the pavement deterioration. Select appropriate method of maintenance or rehabilitation.

2.2 Each road authority will have a different approach to the management of the road network. Some authorities adopt a comprehensive approach with the support of formal road management systems and collect data on a regular basis for planning and programming purposes. The data collected as part of such a system are often sufficient for feasibility studies at project level but are rarely sufficient for detailed design. The procedures described in this Road Note are based on the assumption that very little data are available, however, in situations where this is not so, the recommendations can be easily adapted. For example, the stages prior to the detailed condition survey (Figure 2.1) may be carried out on a regular basis and therefore be completed already. Nevertheless, it is always advisable to verify the accuracy of data supplied from other sources before use.

1.5 During the preliminary design stage, the pavement evaluation study establishes the nature, severity and extent of the road deterioration, the cause of the deterioration and the strength of the existing road pavement. This information, together with the material test results, is used to identify alternative maintenance or rehabilitation strategies which can be considered in the subsequent project appraisal. This appraisal will consider the social impact, environmental impact and economic viability of each alternative. The economic viability is normally assessed using existing road transport investment models such as RTIM3 (TRRL, 1993a) and HDM III (Watanatada et al, 1987) and HDM-4 (to be

Figure 2.1 Road pavement evaluation and rehabilitation procedure

3 Interpretation of existing data

3.1 Design, construction and maintenance data, if available, can be used to establish the type and approximate thickness of the pavement construction. Using the data, those lengths of road having the same nominal thickness and type of construction are identified. Each length of road is then treated as a separate evaluation exercise. 3.2 The traffic loading (in terms of equivalent 80 kN standard axles (esa)) that the road pavement has carried since its construction should be calculated (TRL, 1993b). Often, historical traffic counts are available but reliable axle load data will not have been collected. If neither classified traffic counts nor axle load data are available then surveys should be carried out as part of the evaluation exercise in order to establish current values. Techniques for carrying out such surveys are described in Road Note 40 (TRRL, 1978). If historical traffic data are available, the total commercial traffic loading that the road has carried since construction can be estimated. If this information is not available then the total traffic loading to date can be estimated using traffic growth rates based on other information. These techniques are described in Overseas Road Note 5. 3.3 It is important that, wherever possible, axle load data should be separated by direction of traffic as any differences in axle loads can be useful in identifying the causes of pavement deterioration. Significant differences can occur on roads that lead to quarries or major ports where, for example, raw materials are being exported or imported.

4 Surface condition and roughness surveys

4.1 After dividing the road into lengths of nominally similar construction, it may be necessary to subdivide it again based upon the current condition of the road. This can be done by carrying out a windscreen survey. The best way to do such a survey is for the survey vehicle to stop at 500 metre or one kilometre intervals to enable the condition of the road pavement to be recorded accurately using a selection of the road pavement deterioration criteria described in paragraphs 4.7-4.27. Note that important aspects of road deterioration may be missed if the vehicle is not stopped and survey staff given the opportunity to inspect the road closely. The roughness of the road should also be measured at this stage in the evaluation (see paras 4.28-4.30). These measurements are necessary for the economic appraisal and are useful in defining sections of road in similar condition. The road can then be subdivided into shorter uniform sections based upon the following: time since construction; traffic loading; type of road deterioration; and topography.

compared with other non-destructive tests carried out at the same location (see para 8.5). 4.5 The resources and the equipment required for the detailed condition survey and the operational details are described in Appendix A. 4.6 The recommended form for recording the surface condition data is shown in Figure 4.1. It is designed to be as flexible as possible since the nature of paved road deterioration varies depending on factors such as the type of construction, climate and traffic levels. There are, however, a number of defects that tend to be common to all road pavements and these are described in Table 1. There are three blank areas on the surface condition form which should be used if the other defects, described in Table 2, occur. Table 1 Terms on the surface condition form

4.2 Detailed condition surveys of the sections are then carried out. When the uniform sections are relatively short, the detailed condition survey is best carried out over the entire length of the section. However, where resources are limited then a number of representative one kilometre lengths of road can be used to identify the cause of pavement distress (see para 4.30). The length of road investigated by this method should represent no less than 10 per cent of each section. 4.3 Before the detailed surface condition is carried out, the section or representative one kilometre length is permanently marked into `blocks' of equal length. For inter-urban roads the maximum block length should be either 50 or 100 metres, however, the length may be reduced to as short as 10 metres if the road is severely distressed. 4.4 During the detailed surface condition survey the nature, extent, severity and position of the following defects is recorded: surfacing defects; eg bleeding, fretting, stripping etc. cracking deformation (excluding rutting) patching and potholes edge failures

Rutting is recorded once at the beginning of each of the blocks. It is important that rutting is measured at a discrete point as its severity may need to be

5 Figure 4.1 Detailed surface condition form

Table 2 Other defects

beneath. These areas then disintegrate under traffic and develop into shallow potholes. The introduction of denser asphalt mixes and the use of cement and hydrated lime as filler has largely reduced the occurrence of stripping in asphalt surfacings. Although the mechanisms of failure differ, the result of both of these types of deterioration will be a shallow pothole or a series of potholes. Hence the extent of the defect can be recorded as shown in Figure 4.2. The following definition is recommended: Fretting/Stripping: Shallow potholes having a diameter greater than 100mm.

Surfacing defects Bleeding and fatting-up 4.7 Bleeding is usually observed first in the wheelpaths and is the result of bitumen being forced to the road surface by the action of traffic. Fatting-up of the surface is a less extreme form of bleeding where the surface becomes very smooth but there is insufficient binder to form a continuous film on the surface. The following definitions are recommended: Bleeding: Fatting-up: Continuous film of binder covering the aggregate. Smooth and shiny appearance but aggregate visible.

Loss of stone 4.10 The loss of chippings from a surface dressing resulting from poor adhesion between the binder and the aggregate appears early in the life of the surfacing. It starts in the wheelpaths but, with time, the problem may spread across the carriageway making it difficult to differentiate between this type of failure and bleeding. However, it can often be identified by an accumulation of drippings at the edge of the road pavement. The extent of the defect is recorded according to Table 3. The following definition is recommended: Loss of stone: Continuous film of bitumen visible due to the loss of aggregate. Table 3 Extent of the defect

4.8 Bleeding and fatting-up can often be discontinuous. In asphalt surfacings this can be the result of variations in the mixing process, local over application of tack coat or secondary compaction by traffic. In surface dressings it can be caused by variability in the prepared surface or poor quality control during the spray and chip operation. Fretting and stripping 4.9 Fretting is the progressive loss of fine aggregate from the road surface and occurs when the small movements of individual particles, under the action of traffic, exceeds the breaking strain of the bitumen. It tends to occur later in the life of the surfacing after the bitumen itself deteriorates with age and usually begins in areas of high traffic stress such as sharp bends. The loss of fine aggregate at the surface results in lack of mechanical interlock which can eventually lead to the loss of coarse aggregate and the formation of potholes. Stripping in asphalt surfacings is the result of the displacement of binder from the surface of the aggregate caused by the combined action of water and traffic. In most cases there is a migration of the binder towards the surface of the road resulting in localised bleeding at the surface and unstable poorly coated aggregate Surface texture 4.11 The ability of a bituminous surfacing to provide the required skid resistance is governed by its macrotexture and microtexture. The macrotexture of the surfacing, as measured by its texture depth, contributes particularly to wet skidding resistance at high speeds by providing drainage routes for water between the tyre and the road surface. The surface condition survey should include a qualitative assessment of texture in the wheelpaths so that it can be used to trigger quantitative testing if required. As a guide, the categories shown in Table 4 (CSRA, 1992) are suggested. Aggregate polishing 4.12 The microtexture of the surfacing, as measured by the resistance to polishing of the aggregate, is the dominant factor in wet skidding resistance at lower speeds. The assessment of polishing is more difficult

Figure 4.2 Extent of potholing and patching in a `block' (after CSRA, 1992)

Table 4 Visual assessment of surface texture

assessment will depend on the judgement of the technician, and Table 5 (NIT'RR, 1985) is suggested as a preliminary guide. Table 5 Visual assessment of aggregate polishing

than that of the surface texture, but will be unnecessary if surfacing aggregates having a satisfactory minimum Polished Stone Value were used during construction (Department of Transport, 1994a). When marginal quality aggregates have been used or if increased traffic flows have resulted in an increased state of polish, skid resistance will be reduced. The qualitative

Cracking 4.13 The assessment of cracking should fulfil two objectives. Firstly, it should identify whether the road pavement is suffering from load or non-load associated distress. Secondly, it should establish whether the severity of cracking will affect the 7

performance of any subsequent new pavement layer by causing reflection cracking (Rolt et al, 1996). These objectives are best achieved by identifying five characteristics of the cracking: type intensity position width extent

L - longitudinal cracks T - transverse cracks B - block cracks C - crocodile cracks P - parabolic cracks Intensity 4.15 The intensity of cracking is defined by six levels described below. If the intensity of crack in varies within any block, it should be the intensity-that predominates that is recorded. 0 - no cracks 1 - single crack 2 - more than one crack - not connected 3 - more than one crack - interconnected 4 - crocodile cracking 5 - severe crocodile cracking with blocks rocking under traffic.

Type 4.14 Although there is often no single cause for any type of crack, its appearance can provide a guide to its likely cause. The causes of cracking are discussed in more detail in paragraphs 8.9-8.29. It is recommended that five types of crack are defined. These are listed as follows and illustrated in Figure 4.3.

Figure 4.3 Types of cracking

Position 4.16 The position of the cracking is recorded. The cracking can be confined to either or both of the vergeside (V) and offside (O) wheelpaths, or can be spread over the entire carriageway (C/W). Width 4.17 The measurement of crack width is difficult, but it is important because the width partly determines whether a crack can be sealed effectively. Four categories are recommended as shown below (Paterson, 1987). The first three are for cracks which are not spalled; cracks with substantial spalling are classified as width 4. Initially, until technicians are familiar with the system, the width of the cracks can be measured with a simple `Go/No Go' gauge shown in Figure 4.4. The width of the cracks usually vary within any block, and so it is the width of crack that predominates that is recorded. 1 - crack width < 1mm 2 - 1mm < crack width < 3mm 3 - crack width > 3mm 4 - cracks with spalling Extent 4.18 The extent of the cracking is defined as the length of block affected as shown in Table 3. The extent of cracking should be recorded irrespective of intensity. Deformation 4.19 In terms of its assessment, pavement deformation divides into two groups. Firstly, those

defects with short wavelengths, where severity can be measured by the use of a simple 2 metre straightedge and calibrated wedge (Figure 4.5). Secondly, those defects with longer wavelengths that are best quantified by the use of more sophisticated road profiling instruments. This is discussed in paragraphs 4.28-4.30. Rutting 4.20 Rutting is load associated deformation and will appear as longitudinal depressions in the wheelpaths. It is the result of an accumulation of non-recoverable vertical strains in the pavement layers and in the subgrade. This type of rutting is not associated with any shoving in the upper layers of the pavement unless it becomes very severe. 4.21 The width of the running surface and the traffic flow govern the number of observable wheelpaths on paved roads. For example, a 3-metre carriageway will have two wheelpaths but at road widths greater than 6.5 metres there are generally four. At intermediate widths and low traffic flows there is the possibility of three wheelpaths, with the central one being shared by traffic in both directions. Rut depths should be recorded in the wheelpath showing most rutting. On most roads this is usually the vergeside wheelpath because here the road pavement is generally weaker as a result of higher moisture contents and less lateral support. The straight-edge is placed across the wheelpath, at right angles to the direction of traffic, and the maximum rut depth recorded as shown in Figure 4.5. If the ruts are greater than 40mm deep, the wedge can be held vertically and the depth recorded to the nearest l0mm.

Figure 4.4 Crack width gauge 9

Figure 4.5 Straight edge and calibrated wedge

In some countries there are many roads where distinct wheelpaths do not exist, for example, because of a large volume of non-motorised traffic. In these circumstances the pattern of road deterioration will be different and some of the important clues relating to the position of the deterioration on the carriageway will be absent. This document does not specifically address this situation but many of the techniques for evaluation and assessment described will be appropriate to such conditions. 4.22 Rutting can also be the result of shear failure in either the unbound or the bituminous pavement layers resulting in shoving at the edge of the road 10

pavement. Where the shear failure is occurring in the unbound roadbase or sub-base the displaced material will appear at the edge of the surfacing. Where the failure is occurring in the bituminous material, the displaced material will be evident in the surfacing itself. This is illustrated in Figure 4.6. The severity of the shoving is difficult to measure without taking levels. However its occurrence, together with the depth of rutting, should be recorded, thereby clearly identifying the cause of the failure. This can be simply done by putting a circle around the value of rutting recorded on the surface condition form.

Figure 4.6 Transverse core profile to investigate rutting

Depressions 4.23 Localised depressions, caused by settlement of the pavement layers, construction faults and differential movement at structures, particularly culverts, should be recorded. These are easy to see after periods of rain as they take longer to dry than the rest of the road. When the road is dry, they can also be identified by the oil stains that occur where vehicles cross the depression. The depth should be measured using the 2 metre straight-edge and calibrated wedge. Corrugations 4.24 Corrugations consist typically of a series of ridges perpendicular to the centre line of the road and usually extend across the whole width of the carriageway. Their spacing, or wavelength, is usually in the range of 0.5-1.0 metre but can, in some circumstances, be as much as 10 metres. In paved roads they are caused by instability in either the asphalt surfacing or in an unbound roadbase under a

thin seal. There is generally no need to measure the severity of the corrugations as it will not affect the selection of the remedial treatment. The extent of the defect is recorded as shown in Table 3.

Other types of deterioration Potholes acid patching 4.25 Potholes are structural failures which include both the surfacing and roadbase layer. They are usually caused by water penetrating a cracked surfacing and weakening the roadbase. Further trafficking causes the surfacing to break up and a pothole develops. Because of the obvious hazard to the road user, potholes are usually patched as a matter of priority. Although patches are not necessarily defects, they do indicate the previous condition of the road and are included in the assessment. The extent of potholes and patching is recorded as shown in Figure 4.2.

11

Edge failures and shoulder erosion 4.26 Edge failures are caused by poor shoulder maintenance that leaves the surface of the road pavement higher than the adjacent shoulder. This unsupported edge can then be broken away by traffic, narrowing the running surface of the road. Edge failures (F) are recorded when they exceed 150mm in width at their maximum point or when the vertical step from the surfacing to the shoulder is greater than 50mm (S). It is convenient to measure the defects with the scale on the side of the calibrated wedge, shown in Figure 4.5. The length of the road affected is recorded according to Table 3.

roughness of the road is usually measured using a Response Type Road Roughness Measuring System (RTRRMS) which must be periodically calibrated to allow the values of roughness to be reported in terms of IRI. Suitable methods of calibration include a rod and level survey (ASTM, E 1364-95) or a standard instrument, such as the TRL Profile Beam (Morosiuk et al, 1992) or the MERLIN (Machine for Evaluating Roughness using Lowcost INstrumentation) (Cundill, 1996). Both the roughness survey and calibration procedures are described in Appendix B. 4.30 The roughness of roads with similar pavement construction is a good measure of their relative pavement condition, but it does not identify the nature of the failures or their causes. However, if resources for the surface condition survey are limited, or if the sections of the road under investigation are very long, roughness and windscreen survey data can be used to establish those lengths of road having failures of differing severity. This allows representative lengths of road to be selected which can then be used to identify the cause or causes of deterioration.

Deterioration caused by poor drainage 4.27 Localised pavement failures are often caused by the poor design or maintenance of side and cut-off drains and cross drainage structures. When side drains and culverts silt up, water ponds against the road embankment eventually weakening the lower pavement layers. Conversely, if the water velocity in the side drain is too high it erodes the road embankment and shoulders. More general failures occur when there is no drainage within the pavement layers themselves. Paved roads do not remain waterproof throughout their lives and if water is not able to drain quickly, it weakens the lower pavement layers and results in rapid road failure. Where pavement deterioration is the result of poor drainage design or maintenance this should be recorded on the surface condition form.

Roughness measurements 4.28 It is well established that vehicle operating costs increase as the roughness of the road pavement increases (Hide et al, 1974) (Chesher and Harrison, 1987). Most of the road defects described above contribute in some way to increasing the roughness of the road pavement, although in its early stages cracking may cause little or no change. However, without proper maintenance, the cracked surfacing deteriorates and the resulting potholes and subsequent patching cause a rapid increase in roughness. Surface texture and variability in rut depth also have a significant effect on the roughness of a road pavement. 4.29 The standard measure of road roughness is the International Roughness Index (IRI) which was developed during `The International Road Roughness Experiment' in Brazil (Sayers et al, 1986a). It is a mathematical quarter car simulation of the motion of a vehicle at a speed of 80 kph over the measured profile and can be calculated directly from road levels measured at frequent intervals. Devices for measuring levels are usually either slow and labour intensive or fast, automatic and expensive. Hence, the

12

5 Localised surfacing defects

5.1 After the surface condition survey has been completed, the engineer interprets the results, decides where repairs are needed and what form of maintenance is required. To do this effectively the engineer must first identify the causes of the deterioration. This is important as it is likely that treating the symptoms of pavement deterioration

rather than their causes will prove unsatisfactory. When the road pavement is either rutted or cracked, a programme of additional testing is usually required to establish the causes. However, there are some surfacing defects, if localised, which can be treated at this stage without the need for further testing. Suggested treatments for these types of pavement distress are summarised in Tables 6 and 7.

Table 6 Surfacing defects - roads with thin bituminous seals

Table 7 Surfacing defects - roads with asphalt surfacings

13

6 Performance charts
6.1 Apart from the surface defects described in Tables 6 and 7, bituminous surfaced roads will generally deteriorate either by rutting or by cracking. It is important that the initial form of deterioration and its cause is identified, because this determines the type of maintenance that is most appropriate. After further trafficking, the initial cause of deterioration can be masked by subsequent deterioration. An illustration of this is shown in Figure 6.1, where the final appearance of the road deterioration is similar despite having different initial causes. It is also important to establish if the failures are localised, perhaps because of poor drainage, or whether they are affecting the road in a more general manner. 6.2 When an evaluation takes place there will often be considerable lengths of road that have reached a terminal level of deterioration similar to that shown in Figure 6.1. However, even within nominally uniform sections, road pavements are inherently variable, having a range of pavement thicknesses and material properties. This results in differential performance, with some areas deteriorating less rapidly than others, and it is in these areas that the initial form of deterioration can be most easily identified. 6.3 The cracking or rutting recorded during the windscreen or detailed condition survey may be displayed graphically in the form of performance charts. These enable the length of road affected by each form of deterioration to be quantified. The cracking and rutting can also be compared to other predominant forms of deterioration and this may help to promote a better understanding of the causes. 6.4 An example of the use of performance charts is illustrated in Figure 6.2 for a 20km section of paved road having a mechanically stabilised gravel roadbase with a thin bituminous surfacing. The initial form of deterioration was rutting which was associated with shoving whenever the failure became severe. Although there is some cracking which is coincident with high values of rutting, there is no cracking in areas of less severe rutting, suggesting that the rutting preceded the cracking. In addition to the rutting, substantial lengths of the surfacing are suffering from bleeding. However, the charts show that there is no correlation between the bleeding and the rutting, indicating that the shoving is in a lower granular layer, not the bituminous surfacing. 6.5 Using performance charts similar to those described above, the section is divided into subsections having failures of differing severity. A programme of additional tests (see Chapter 7) is then prepared to identify the causes of the differential performance between the sub-sections. There may be some cases where the complete section of road will have reached a failed condition, for example when

the road pavement has been under designed or where there are serious material problems. In such cases the cause of the deterioration can only be established by comparing the thickness of the road pavement or the material properties of the pavement layers with relevant design standards and material specifications.

14

Figure 6.1 The development of road failure

15

Figure 6.2 Illustration of performance charts

16

7 Additional tests
7.1 Deflection-based measurements and Dynamic Cone Penetrometer (DCP) tests are used to help identify the cause of differential performance between sub-sections and to provide information for the maintenance or rehabilitation of the section. In some cases the moisture content of the road pavement, especially the subgrade, changes seasonally. In these circumstances the tests should be carried out after the rainy season, when the road is at its weakest. The results from these non-destructive tests are usually confirmed by destructive sampling and material testing.

be too high and the pavement will deteriorate through the development of ruts. Under such circumstances the deflection will be correlated with rut depth, as shown in Figure 7.1, and such a correlation provides an indication of the reasons for failure. 7.5 Apart from the maximum deflection, there are other parameters and indicators from the deflection bowl that may be used to identify some of the structural differences between sub-sections and hence assist in identifying the cause, or causes, of failure. The radius of curvature (ROC) of the deflection bowl, shown in Figure 7.2, can be used to estimate the relative properties of the upper layers of the pavement. This may be used to identify relatively weaker surfacing layers where fatigue cracking is more likely. However, once cracking is apparent the ROC will decrease considerably hence care is required in interpreting the ROC data. Similarly the deflection values at the extremes of the deflection bowl are indicators of the relative strength of the subgrade. 7.6 Therefore there are advantages in using deflection equipment capable of measuring other deflection bowl parameters as well as maximum deflection and the Falling Weight Deflectometer (FWD) and the Deflectograph are the most widely used. The FWD, in particular, is growing in popularity as it has the advantage of being able to apply impact loads which more accurately simulate the effect on pavements of heavy vehicles moving at normal traffic speeds than the slowly moving load applications associated with the Deflectograph or the deflection beam. Procedures for using FWD equipment for road surveys are given in Appendix E. If, however, funds are not available to measure deflection bowl characteristics using one of the more sophisticated measuring devices, then consideration should be given to using a curvature meter (NIRR, 1970) in association with a deflection beam to measure both the ROC under the rear wheels of the deflection lorry and the maximum deflection. 7.7 Analysis of deflection bowl data is dependent on a suitable model to calculate the response of the pavement to the applied load. Most analysis programs are based on the assumption that the pavement behaves, in the first instance, like a multi-layer structure made up of linearly elastic layers. Using such a model, it is possible to calculate the elastic modulus of each pavement layer from a knowledge of the shape of the deflection bowl. This `back-analysis' procedure requires accurate deflection data extending from the central maximum deflection out to deflection values at radial offsets of as much as 2.1 metres. However, the linear elastic model is a very simple model of road pavements. Road materials display a variety of properties that do not comply with the assumptions of the model. For example, the elastic modulus of unbound materials is not a constant but depends on the stresses to which

Deflection tests 7.2 The strength of a road pavement is inversely related to its maximum vertical deflection under a known dynamic load. Table 8 lists the more common deflection devices, their loading regimes and output. Table 8 Deflection devices

7.3 The least expensive of these instruments is the deflection beam. This is a mechanical device that measures the maximum deflection of a road pavement under the dual rear wheels of a slowly moving loaded lorry. The recommended test and survey procedures for the deflection beam are given in Appendices C and D. TRL recommends the use of a 63.2 kN rear axle load, other authorities recommend different loads, most commonly 80 or 100 kN. Over this range of loads the maximum deflection is usually linearly relatedto the applied load. Therefore, for structurally adequate pavements where over-stressing is not a danger, deflection values can be measured with these higher loads and then normalised to any standard load for comparison purposes. 7.4 Maximum deflection under a slowly moving wheel load is a good indicator of the overall strength of a pavement and has been shown to correlate well with long term performance of pavements under traffic. For example, if a road is underdesigned for the traffic it is carrying for any reason (eg. incorrect assessment of subgrade strength or traffic loading) the stresses in the lower layers of the pavement will

17

Figure 7.1 Example of good relation between rut depth and deflection

Figure 7.2 Computation of radius of curvature the material is subjected at each point in the structure, i.e. the materials are not linear. This is a particular problem with the subgrade because the modulus of the subgrade has a strong influence on the shape of the entire deflection bowl. Errors or inaccuracies in the assumptions here, give rise to errors in the calculations of the moduli of all other layers. A further consideration is the capability of the programs to handle complex structures. The more layers that are present, the more difficult it is for the programs to identify a suitable unique solution. Overall, the acceptability of the results often depends much more on the skill of the analyst than the sophistication of the analysis program. Recent research (Strategic Highway Research Program, 18 1993a) has resulted in a set of rules and guidelines that can be used when estimating pavement layer moduli by backcalculation from deflection bowl data and it is considered that these provide a reasonable basis for the back-analysis of road pavements. 7.8 Alternatively FWD deflection data may be tabulated and plotted to show the variation of pavement response along the road. Certain parts of the deflection bowl are influenced by the different pavement layers. With reference to Table E1 (Appendix E), the chosen deflection criteria are usually dl, d6 and (dl-d4). The maximum deflection dl gives an indication of overall pavement performance whilst the deflection difference (dl-d4)

relates to the condition of the bound pavement layers. Deflection d6 is an indicator of subgrade condition. A typical deflection profile is shown in Figure 7.3. Although actual values of deflection will depend on the type and condition of the pavement layers, such plots show relative differences in their condition and give an indication of any structural weaknesses.

numeric which represents the combined strength of the pavement layers. This is done by calculating the Structural Number (SN) as shown in Equation 1.

Dynamic cone penetrometer tests 7.9 The DCP is an instrument which can be used for the rapid measurement of the in situ strength of existing pavements constructed with unbound materials. Measurements can be made down to a depth of approximately 800mm or, when an extension rod is fitted, to a depth of 1200mm. Where pavement layers have different strengths, the boundaries between them can be identified and the thickness of each layer estimated. The operation of the DCP and the analysis of the results are described in Appendix F. 7.10 DCP tests are particularly useful for identifying the cause of road deterioration when it is associated with one of the unbound pavement layers, eg. shear failure of the roadbase or sub-base. A comparison between DCP test results from subsections that are failing and those that are sound will quickly identify the pavement layer which is the cause of the problem. 7.11 In some circumstances it is convenient to convert the individual pavement layer thicknesses and strengths measured in the DCP test into a simple

The layer coefficients are related to standard tests for the pavement materials and are fully described in the AASHTO Guide for Design of Pavement Structures (1993). To take into account variations in subgrade strength, the modified structural number (SNC) can also be calculated (Hodges et al, 1975), as shown in Equation 2.

If it is suspected that the road failures are related to the overall structural strength of the pavement, the Modified Structural Number of different sub-sections can be readily compared to identify the weakness. Destructive sampling and material testing 7.12 When the results of the condition survey indicate that the properties of the asphalt surfacing could be the cause of differential performance between sub-sections (see paras 8.1-8.29) then this should be

Figure 7.3 FWD deflection profile

19

confirmed by further testing. Sufficient 150mm diameter core samples need to be taken from each subsection to ensure that representative values for the composition and properties of the asphalt surfacing arc obtained (BS 598, 1987). Prior to testing, the cores must be examined to establish the following: thickness of each bound layer; degree of bonding; occurrence of any stripping; and depth of cracking (if required).

during the detailed condition survey (see Table 4). A mean of ten tests, usually in the vergeside wheelpath, should be used to characterise each 50 metre section. Sections should also be chosen in hazardous areas such as the approaches and crowns of bends. These values can then be compared to national standards, where they have been established, to identify the lengths of the road that need resurfacing. If national standards do not exist then the intervention values proposed in the UK may be used as a guide (Department of Transport, 1994b). 7.16 The microtexture, in terms of the `skid resistance' value (SRV), can be measured using the portable skidresistance tester (RRL, 1969) (ASTM, E 303-93). The test procedure is described in Appendix 1. There are other instruments available which measure skid resistance more rapidly (and more continuously), for example SCRIM (Sideway-force Coefficient Routine Investigation Machine)(Department of Transport, 1994a) and the Griptester (County Surveyors Society, 1988), but these are more costly. A representative value of SRV can be obtained in a similar way to that described for texture depth, with the mean value of ten results being used to characterise a 50 metre section of road. These values can then be compared to national standards, where they have been established, to identify the lengths of the road in need of resurfacing. 7.17 If national standards are not available then those recommended in the UK may be used as a guide (Department of Transport, 1994a). The present UK intervention levels are now specified in terms of the Sideway-Force Coefficient (SFC) as measured by SCRIM. If only the portable skid resistance tester is available, then previous UK standards, summarised in Table 9, are suggested as a preliminary guide.

Where only the thickness of the asphalt surfacing is to be measured, then 50-100mm diameter cores are satisfactory. Similar cores can be used for transverse core profiles, such as those shown in Figure 4.6, which are used to establish whether shoving is the result of shear failure in the surfacing or in one of the lower unbound pavement layers. 7.13 When deflection measurements and DCP results indicate that either the thickness or properties of the lower pavement layers are the cause of the differential performance, then test pits are needed to obtain additional material information to confirm these results. The recommended procedure for carrying out test pit investigations is given in Appendix G. These investigations are used both to provide an explanation for the present behaviour of the pavement and to provide information for its rehabilitation. Each test pit will provide information on the thickness of each pavement layer and properties of the material. These can then be compared to specified values.

Skid resistance tests 7.14 When the detailed surface condition survey indicates that the surfacing has poor texture or polished aggregate (see Tables 6 and 7) then a quantitative survey will usually be required. This survey can only be dispensed with if the road is suffering from other failures that require the road to be resurfaced. 7.15 The texture depth of bituminous surfacings is measured by the sand patch test (BS 598, 1990). The test procedure is described in Appendix H. There are also other relatively low cost instruments, such as the Mini-Texture Meter (Department of Transport, 1994a), which give continuous measurements of surface texture and are quicker and more convenient to use. However, the results from texture meters need to be calibrated against the sand patch test if they are to be compared with specifications. The sand patch test gives a single value of texture at one point and therefore a number of tests are needed to give a representative value for the road. This is done by selecting sections of road, 50 metres long, which cover the range of severity of the defect recorded

Table 9 Suggested minimum `slid resistance' values

20

8 Identifying the causes of pavement deterioration

8.1 The next stage in the evaluation procedure is to establish the cause or causes of the pavement deterioration by interpreting the data collected during the surface condition survey and the additional testing. The causes of deterioration combined with the extent of the failures must be considered together when selecting the most appropriate method of maintenance or rehabilitation. 8.2 Besides the surface defects described in Tables 6 and 7, bituminous surfaced roads will generally deteriorate either by rutting or by cracking. To help identify the cause of the deterioration, rutting and cracking have been subdivided into six categories based on the nature of the failure, its position and the type of road construction. These are: rutting without shoving; rutting with shoving; wheelpath cracking - asphalt surfacing; wheelpath cracking - thin bituminous seal; non-wheelpath cracking - asphalt surfacing; and non-wheelpath cracking - thin bituminous seal.

it will then be necessary to show a relationship between the severity of rutting and the deflection of the road pavement at the time of the evaluation, as illustrated in Figure 7.1. If deflection equipment is unavailable, a similar analysis can be completed by relating the severity of rutting to the strength of the road, as measured by the DCP (see para 7.11). 8.6 If the severity of rutting does not relate to the strength of the road pavement, the most likely cause of the rutting is secondary compaction of one or more of the pavement layers by traffic during the early life of the road. In this case the rate of increase in rutting will decrease after the initial compaction phase.

Rutting with shoving (Figure 8.2) 8.7 Shoving parallel to the edge of the rut (see para 4.22) is indicative of a shear failure in one of the pavement layers and is caused by the pavement layer having inadequate shear strength to withstand the applied traffic stresses at that particular depth in the pavement. Unlike the rutting described in paragraph 8.5, the severity of the rutting will not usually be related to the overall strength of the pavement as indicated by either its deflection or modified structural number. 8.8 The failures are usually confined to the upper pavement layers where the applied traffic stresses are at their highest. A process of elimination is used to identify which layer has failed. If the pavement has an asphalt surfacing then a transverse core profile (Figure 4.6) can be used to establish in which bituminous layer, if any, the failure is occurring. If the failure is not in the asphalt surfacing then the DCP can be used to identify which of the underlying pavement layers is the cause of the failure. This is done by comparing the strength of the layers in failed areas with those that are sound. For roads with thin bituminous seals, a comparison of the ROC values from these different areas can also be used to identify substandard roadbase materials.

8.3 A method of establishing the probable cause or causes of pavement deterioration is given in the flow charts shown in Figures 8.1-8.8. These charts will not cater for all the types and stages of pavement deterioration. In particular, when a road has received a series of maintenance treatments or when the initial deterioration is masked by further progressive failures, the problem of identifying the initial cause of failure becomes more complex. However, the charts do provide a framework that enables highway engineers to develop their own pavement evaluation skills. The charts identify general causes of deterioration but do not attempt to establish specific material problems, as this can only be done by further destructive sampling and subsequent laboratory testing.

Rutting without shoving (Figure 8.1) 8.4 These ruts are usually wide as they are caused primarily by movement deep in the pavement structure, and there will be little or no evidence of shoving at the edge of the pavement. This type of rutting is the result of two possible causes, either insufficient load spreading or secondary compaction. 8.5 Insufficient load spreading is the result of the pavement layers being too thin to protect the subgrade. It is characterised by an increase in rutting with traffic loading. Where there is historical data on the progression of rutting and traffic, or where there is a significant difference in traffic loading between the two lanes, then this relationship can be established. More usually this information will not be available and

Wheelpath cracking - asphalt surfacing (Figure 8.3) 8.9 If cracking is caused primarily by traffic it must, by definition, originate in or near the wheelpaths. In severe cases it is sometimes difficult to be sure whether the failures start in the wheelpath or whether they are a progression of another form of cracking, as shown in Figure 8.9 (Dickinson, 1984). The initial type of cracking should be identified as described in paragraphs 6.1-6.2. 8.10 Short irregular longitudinal cracks in the wheelpaths are often the first stage of traffic induced fatigue of the surfacing which, after further trafficking, interconnect to form crocodile cracks (see Figure 8.9). Although caused by the flexure of the surfacing, they are not necessarily `traditional' fatigue cracks which start at the bottom of the asphalt surfacing and

21

Figure 8.1 Initial deterioration - Rutting without shoving

22

Figure 8.2 Initial deterioration - Rutting with shoving

23

24 Figure 8.3 Initial deterioration - Wheelpath cracking in asphalt surfacing

25

Figure 8.4 Initial deterioration - Wheelpath cracking in thin bituminous seal

26

Figure 8.5 Initial deterioration - Non-wheelpath cracking in asphalt surfacing

Figure 8.6 Initial deterioration - Longitudinal cracking in asphalt surfacing

27

28

Figure 8.7 Initial deterioration - Transverse cracking in asphalt surfacing

Figure 8.8 Initial deterioration - Block cracking in asphalt surfacing

propagate upwards. In tropical climates the bitumen at the top of asphalt wearing courses oxidises rapidly (Smith et al, 1990). This causes the material to become brittle and results in cracking being initiated at the top of the surfacing rather than at the bottom, despite the strains being lower (Rolt et al, 1986). 8.11 Where crocodile cracks are shown, by coring, to have started at the bottom of the asphalt layer, then they are likely to be `traditional' fatigue cracks caused by excessive strains at the bottom of the surfacing. Excessive strains can be caused by a weak subgrade, giving rise to large maximum deflections, or a weak roadbase leading to small radii of curvature. However, in both cases the cracking is frequently associated with rutting; in the former case, because of insufficient load spreading; in the latter case, because of shear failure in the roadbase (see paras 8.5 and 8.7). In practice this type of crocodile cracking very rarely occurs without any rutting. 8.12 In some circumstances traditional fatigue cracking can occur simply because the road has

reached the end of its design life; in other words no other form of failure has occurred beforehand. This is a relatively rare phenomenon and for this reason is sometimes difficult to identify because of the need to calibrate standard asphalt fatigue relationships for local conditions. However, the age of the surfacing and the traffic carried should provide the most important clues. 8.13 Poor surfacing materials can also result in crocodile cracking. Inadequate quality control exercised during the manufacture and construction of dense surfacings can lead to poor particle size distribution, low bitumen contents, segregation and poor compaction, all of which will make the material more susceptible to cracking. Failures of this type can occur in areas where deflections are satisfactory and where little or no rutting is occurring 8.14 If the bond between the asphalt surfacing and the underlying layer is poor then the surfacing can effectively 'bounce' under traffic. This quickly results in crocodile cracking in the wheelpaths and is

29

characterised by blocks of less than 200mm square. The cause of the poor bond can be ineffective priming of the roadbase, or deficient tack coat prior to placing an overlay. Often the cracking will progress to laminations, which are shallow potholes that are clearly the result of the surfacing `peeling' off. 8.15 Parabolic shaped cracks in the surfacing which occur in areas of severe braking such as the approaches to junctions and sharp bends are caused by slippage and are also the result of a poor bond. Small areas of parabolic cracking are not indicative of serious failure. However, if it is more extensive, the probable cause is an inadequate tack coat or the use of soft aggregate in the surfacing which, in breaking down, results in a poor bond and subsequent slippage. 8.16 Cracking in bituminous overlays, particularly in the wheelpaths, can be caused by cracks in the underlying layer 'reflecting' through the overlay. Reflection cracking will generally occur early in the life of the overlay and is often associated with pumping of fine material from a lower granular layer. Cores cut through cracks in the new overlay will establish whether they are being caused by existing cracks in a lower pavement layer. Wheelpath cracking - thin bituminous seal (Figure 8.4) 8.17 The bitumen film in surface dressings is very thick compared to that in asphalt surfacings and it is more tolerant to flexure under traffic. Errors in the design or construction of these seals are more likely to result in failures such as bleeding or loss of stone rather than cracking. However, as the seal gets older, age hardening of the bitumen can result in wheelpath cracking or fretting. If cracking is being caused by excessive flexure under traffic then it will be associated with areas of high deflections. 8.18 Where the surfacing has been used to seal an existing cracked asphalt layer, any subsequent cracking may be caused by the reflection of cracks from the previous surfacing. Slurry seals are particularly susceptible to reflection cracking. 8.19 Bituminous seals having a poor bond with the underlying roadbase will behave in a similar way to that of an asphalt surfacing. In this case any water going through the resultant cracking will aggravate the poor bond, resulting in the rapid formation of potholes. This can be a problem with stabilised roadbases if they are not primed effectively prior to surfacing. Non-wheelpath cracking - asphalt surfacing (Figure 8.5) 8.20 The cause of non-traffic associated cracking in an asphalt surfacing is largely established by identifying its type (see para 4.14). As traffic has

played little or no part in these road failures the cracks will not be confined to the wheelpaths and there will not be any substantial rutting. Non-wheelpath cracking can take the form of longitudinal, transverse, block or crocodile cracking. Longitudinal cracking (Figure 8.6) 8.21 Thermal stresses can cause cracks to appear along poor longitudinal construction joints and in areas of severe temperature gradients, such as the edge of road markings. In their early stages neither of these types of crack is particularly serious; however, if left unsealed, the cracks will eventually spread into the wheelpaths where they will result in more serious deterioration. 8.22 Where longitudinal and transverse cracks occur in combination, they are likely to be either reflection cracks propagating from a lower stabilised layer or cracks caused by thermal or shrinkage stresses in the asphalt. These are described in more detail paragraphs 8.24 and 8.26. 8.23 Longitudinal cracks caused by subgrade movement will generally be quite long and can meander across the carriageway. They can occur because of poor construction, swelling in plastic subgrade or embankment materials, and the settlement or collapse of embankments. Cracks caused by the slippage of an embankment will often occur in semicircular patterns and both these and cracks caused by subgrade movement will often be associated with a vertical displacement across the crack. Transverse cracking (Figure 8.7) 8.24 Transverse cracks in the surfacing of a road pavement which includes either a chemically stabilised roadbase or sub-base are likely to be reflection cracks from the stabilised layer, particularly if the stabiliser is cement. This form of transverse cracking is often associated with longitudinal cracks and, in severe cases, block cracking. 8.25 If the transverse cracks are irregularly or widely spaced they are likely to have been caused by some form of construction fault. Differential vertical movement caused by consolidation or secondary compaction adjacent to road structures and culverts can cause transverse cracks in the surfacing. These cracks will be associated with a poor longitudinal road profile caused by the differential movement. 8.26 Transverse cracks confined to the surfacing and occurring at more regular and shorter spacings are probably caused by thermal or shrinkage stresses. This type of cracking will most likely occur in areas subject to high diurnal temperature changes, such as desert regions, and will be exacerbated by poor quality surfacing materials. When cracks occur after

30

many years of good performance it is likely that progressive hardening of the binder has made the surfacing more `brittle' and therefore more susceptible to cracking. As transverse thermal cracks progress, they will link up with longitudinal ones to form block cracking as shown in Figure 8.9. Thermal stresses can also cause cracks to open up at transverse construction joints. Block cracking (Figure 8.8) 8.27 Block cracking, when confined to the bituminous surfacing, is usually the final stage of cracking due to thermal stresses. These cracks almost always start at the top of the surfacing and propagate downwards. Block cracking can also occur through reflection of the shrinkage crack pattern in lower chemically stabilised layers. Crocodile cracking 8.28 Crocodile cracking that is neither confined to the wheelpaths nor associated with rutting is indicative of a fault in the construction of the surfacing. The more common production faults are poor particle size distribution, low binder contents, overheated bitumen and the use of absorptive aggregate. Construction faults include poor compaction, segregation of the mix and poor bonding, either between layers of bituminous material or the granular layer beneath. In these cases the precise cause of failure can only be determined by destructive sampling and laboratory testing. Non-wheelpath cracking - thin bituminous seal 8.29 Roads having thin bituminous seals are less susceptible to the non-traffic associated failures described in paragraphs 8.21-8.28 because their thicker bitumen film results in a higher strain tolerance. Surface dressings, in particular, are less likely to crack either at construction joints or alongside road markings. They are also less susceptible to thermal or shrinkage cracking. Where strains are large, however, as in the case of reflection cracking from a stabilised roadbase or from subgrade movement, the surfacing failure will be similar to that described for asphalt surfacings.

31

Figure 8.9 Crack development patterns in bituminous surfacings (after Dickinson, 1984)

32

Maintenance and rehabilitation

9.1 The selection of an appropriate maintenance treatment or rehabilitation strategy is based on a number of considerations. Firstly, the cause of deterioration in the existing pavement must be correctly identified and its importance assessed. For example, the deterioration may result from some deep seated structural insufficiency or construction defect. In such cases consideration must be given to full or partial reconstruction of the pavement to correct the situation. Secondly, attention should be given to the nature, extent and severity of the deterioration to check what effect it will have on the treatments that are being considered. For example, thin asphalt surfacings on their own will not provide a satisfactory repair where reflection cracking is likely, nor will any form of thin surfacing provide a significant improvement to riding quality where this is poor. Finally, the strategy must be economically viable taking into consideration both the costs of maintenance and the vehicle operating costs over a number of years. 9.2 It should not be assumed that when a road is in poor condition it inevitably needs strengthening. When traffic is low, for instance, the existing road structure is often thick enough to prevent long term rutting. In this case the maintenance treatment selected should address the cause, or causes, of the deterioration without necessarily adding strength to the pavement. It is important, therefore, to check the ability of the existing road pavement to carry the predicted traffic loading using at least two of the methods described below. Where either of the methods are shown to accurately predict the present performance of the road under study then the method is equally applicable for the design of strengthening works in the event that the road is shown to be too weak to carry the future traffic. Analytical approach 9.3 The traffic carrying capacity of an asphalt pavement is governed by how effective the pavement layers are in preventing; fatigue cracking of the asphalt surfacing; shear failure of the granular materials; fatigue cracking or crushing of lightly cemented materials; and wheelpath rutting resulting from subgrade failure.

failure modes. The performance of road pavements has traditionally been dependent on the stress/strain values at two locations in the structure. The horizontal tensile strain at the bottom of the asphalt layer controls one type of fatigue cracking and the vertical compressive strain at the top of the subgrade controls rutting. 9.5 For roads having a thin bituminous seal the traffic carrying capacity is determined only by resistance to rutting. The performance of the surface seal will generally depend on environmental effects ratherthan traffic loads. The traffic carrying capacity of anasphalt surfaced road will be determined by both its resistance to fatigue cracking and wheelpath rutting. However, research has shown that the predominant form of surface distress of asphalt surfacings in tropical climates is not fatigue cracking starting at the bottom of the asphalt layer but `topdown' cracking which is initiated at the surface of the layer (Rolt et al, 1986) (Smith et al, 1990). The type and severity of this form of cracking is a complex function of material properties and both environmental and traffic stresses and its development has yet to be successfully described by means of a practical analytical model. `Top-down' cracks often develop long before other types of cracks and thus the performance of asphalt surfaced roads rarely agrees with the analytical models. Nevertheless rehabilitation design should take account of all possible modes of future failure and therefore it is important to ensure that traditional fatigue failure of the surfacing and failure through inadequate protection of the subgrade do not occur within the design life required. In order to do this, analytical procedures properly calibrated to local conditions provide a suitable method. 9.6 The analytical approach requires a suitable mathematical model to describe the pavement. Almost all methods make use of the multilayer linear elastic model, although more sophisticated models can also be used. This model requires, as input, the thickness, elastic modulus and Poisson's ratio of each layer of the pavement. Very thin layers such as an existing seal are normally incorporated with the underlying roadbase or ignored. Asphalt surfacings are usually assigned moduli based on mix constituents and binder properties at the design temperature although direct laboratory measurements of modulus can also be made on samples of material extracted from the road. Other moduli values can be either calculated from the back-analysis of FWD deflection bowls or assigned values following DCP testing and/or the laboratory testing of samples taken from trial pits. Stresses or strains at the critical points in the pavement are then calculated under the application of a standard load designed to replicate a 40kN wheel load (80kN axle load). These strains are then used to calculate the `life' of the structure using

9.4 Theoretical models to predict the behaviour of granular and lightly cemented materials under the action of traffic are not well defined and therefore specifications for such layers have always been set in such a way that failures are unlikely. This has mitigated against the use of lower quality materials and has theoretically restricted the range of likely

33

relationships (Powell et al, 1984)(Shell, 1985) between stress/strain and pavement life of the form: Asphalt fatigue criteria Log Nf = a + b Log r Where Nf = Fatigue life in esa r = Horizontal tensile strain at the bottom of the asphalt layer a and b = constants associated with material properties Subgrade deformation criteria Log Nd = a+ b Log z Where Nd = Deformation life in esa z = Vertical compressive strain at the top of the subgrade a and b = constants associated with material properties 9.7 Where the forms of these relationships are shown to predict the present performance of the road pavement, they can be used with more confidence to estimate the future traffic carrying capacity. After adjustment of the pavement model they can then also be used to determine overlay thickness, where necessary. Structural approach 9.8 In this method the traffic carrying capacity of the road is estimated by comparing the existing pavement structure and its condition with established design charts. The thickness of the various pavement layers should first be established using the DCP and trial pits, and the in situ strength of the pavement layers and the subgrade determined by a combination of deflection and DCP data. These tests should be carried out shortly after the wettest period of the year, when the pavement can be expected to be in its weakest condition. If this is not possible, adjustments will need to be made to the deflection data and material properties to reflect the season during which the data were collected. The in situ strengths of the pavement layers obtained in this way, in particular the upper granular layers, should always be verified by laboratory tests to ensure they conform to normally accepted specifications. The effective structural number of the pavement can then be obtained by using techniques described in the AASHTO Guide for Design of Pavement Structures (AASHTO, 1993). 9.9 The required strengthening measures are then established by comparing the effective structural number of the pavement with the required structural number of a pavement for the future traffic, obtained from an appropriate design method, at a representative value of in situ subgrade strength. If 34

the AASHTO guide is used then a mean value of the resilient modulus of the subgrade, suitably corrected (AASHTO, 1993), is used. If Road Note 31 is preferred then the lower 10 percentile of the in situ subgrade CBR should be used, measured when the pavement is in its weakest condition. Where the comparison of the effective structural number, past traffic and design recommendations is shown to be consistent with the present condition of the road pavement, then the engineer can be more confident in designing the thickness of any necessary strengthening overlay by this method. 9.10 There are presently a number of methods of determining the structural number of a road pavement directly from FWD deflection bowl characteristics (AASHTO, 1993) (Jameson, 1992) (Rohde, 1994) (Roberts and Martin, 1996). With the development and refinement of these procedures it is likely that the rehabilitation of road pavements using the structural number approach will become increasingly popular. Deflection approach 9.11 The representative maximum deflection is used by a number of road authorities to estimate the carrying capacity of a road (Kennedy and Lister, 1978) (Asphalt Institute, 1983). The deflection criteria curves recommended in these design procedures (i.e. the relationship between deflection and traffic carrying capacity) are not necessarily applicable to road pavements found in tropical and sub-tropical regions. However, it is clear that an overlay reduces the stresses in the lower layers of the pavement and therefore, to prevent deformation in these layers and the subgrade, appropriate deflection criteria can be developed (NITRR, 1983). Such an approach is particularly appropriate when investigations show that either the project road, or other roads of similar construction in the region, are rutting because of a deficiency in the overall `strength' of the pavement (see para 7.4). 9.12 The deflection and condition surveys must be carried out after the wettest period of the year when the road pavement can be expected to be at its weakest. The severity of rutting is then plotted against the maximum deflection at each test point and a best fit line and confidence limits calculated as shown in Figure 9.1. The value of critical deflection corresponding to a defined level of critical rutting is then determined for any particular level of statistical reliability. The 90th percentile is recommended with a critical rut depth of l0mm for roads with asphalt surfacings and 15mm for those with thin bituminous seals. Provided the past traffic loading is known, one point can be plotted on the deflection trafficloading graph. This point is unlikely to lie on an existing criteria curve, however, assuming a similar form of relationship, a `calibrated criteria curve' can be obtained by drawing a new line through the point and parallel to the existing curve as illustrated.

Figure 9.1 Diagrammatic calibration of deflection life criterion line (after NITRR, 1983)

9.13 The traffic carrying capacity of the road, in terms of rutting, can be estimated by comparing the representative deflection of homogeneous sections (see Appendix D) with the calibrated deflection criteria curve as shown in Figure 9.1. The traffic carrying capacity represents the total traffic loading that the road will carry from construction. Therefore the future traffic carrying capacity is the total traffic loading minus the traffic loading that the pavement has earned prior to evaluation. 9.14 The thickness of any necessary strengthening overlay can be determined based on reducing the representative deflection of the pavement to the design deflection obtained from the calibrated deflection curve. The relation between the thickness of a dense bituminous overlay and the reduction in deflection, under a 62.3kN axle load, has been shown to be: T = 0.036+0.818 Dr - Dd 0.0027Dr where Dd = Design deflection (mm) Dr = Representative deflection (mm) T = Overlay thickness (mm) (3)

This relation is valid between representative deflection values of 0.25 - 1.2mm and overlay thicknesses of 40 150mm. Maintenance options 9.15 If it is established that the road does trot require strengthening, the method of maintenance should be based upon the type of the existing surfacing and the cause of failure. Pavement maintenance will generally result in two operations. Firstly, those areas where failure has already occurred should be repaired by some form of remedial treatment and, secondly, the road should generally be resurfaced to prevent other lengths failing in a similar manner. Suggested methods of maintenance for the different types of pavement deterioration for roads having thin bituminous seals and asphalt surfacings are given in Tables 10 and 11 respectively.

35

Reflection cracking 9.16 Reflection cracking can have a considerable and often controlling influence on the life of thin bituminous overlays. The rate of propagation of these cracks has been shown to be dependent on the strength of the road, the severity of the cracking before overlay and the future traffic (Rolt et al, 1996). The complete prevention of reflection cracking through thin overlays is not possible. However, when the existing cracked asphalt surface is relatively thin, the most effective method of reducing reflection cracking in any subsequent overlay is to remove the areas showing cracking of intensity 3 or greater and to patch prior to construction. Where the existing surfacing consists of several previous bituminous overlays, it may be more cost effective to introduce a crack relief interlayer rather than to remove all the cracked material. Reviews of practice in North America (Sherman, 1982)(Barksdale, 1991) suggest that the most successful techniques are: asphalt-rubber interlayers; interlayers of open-graded bituminous material; or heater-scarification and recompaction of the cracked layer.

36

Table 10 Existing road surface - Thin bituminous seal

37

Table 10 (Continued)

38

Table 11 Existing road surface - Asphalt surfacing

39

Table 11 (Continued)

40

10 References
AASHTO (1993). Design of pavement structures. American Association of State Highway and Transportation Officials, Washington DC. Asphalt Institute (1983). Asphalt overlays for highway and street rehabilitation. Manual Series No. 17 (MS-17). Asphalt Institute, Lexington. ARRB (1996). Walking Profilometer. ARRB Transport Research Ltd, Victoria. Austroads (1992). Pavement design. A guide to the structural design of road pavements. Austroads, Sydney. Barkdale R D (1991). Fabrics in asphalt overlays and pavement maintenance. Synthesis of Highway Practice 171. Transportation Research Board, Washington DC. Beaven P J and Tubey L W (1978). The polishing of roadstone in Peninsular Malaysia. Supplementary Report SR 421. Transport Research Laboratory, Crowthorne. Bertrand C, Harrison R and Hudson W R (1991). Evaluation of a high-resolution profiling instrument for rise in road roughness calibration. Transportation Research Record 1291, Transportation Research Board, Washington DC. Chesher A D and Harrison R (1987). Vehicle operating costs: evidence from developing countries. Highway Design and Maintenance Standards Series. The John Hopkins University Press, Maryland. Committee of State Road Authorities (1992). Pavement management systems: Standard visual assessment manual for flexible pavements. Technical Methods for Highways TMH No 9. Committee for State Road Authorities, Pretoria. County Surveyors Society (1988). A directory of pavement assessment equipment. County Surveyors Report No 5/10. Devon County Council. Cundill M A (1996). The MERLIN road roughness machine: User guide. TRL Report TRL229. Transport Research Laboratory, Crowthorne. Department of Transport (1994a). Skidding resistance. Design Manual for Roads and Bridges, Volume 7: Sect 3: Part l: HD 28/94. Department of Transport, UK. Department of Transport (1994b). Structural assessment methods. Design Manual for Roads and Bridges, Volume 7: Sect 3: Part 2: HD 29/94. Department of Transport, UK. Dickinson E J (1984). Bituminous roads in Australia. Australian Road Research Board.

Hide H S, Abaynayaka S W, Sayer I and Wyatt R J (1974). The Kenya road transport cost study: research on vehicle operating costs. Laboratory Report LR 672. Transport Research Laboratory, Crowthorne. Hodges J W, Rolt J and Jones T E (1975). The Kenya road transport cost study: research on road deterioration. Laboratory Report LR 673. Transport Research Laboratory, Crowthorne. Jameson G W (1992). Development of the Philippines Asphalt Overlay Procedure. Report for ADB Technical Assistance Project TA No. 1426 PHI. VicRoads, SMEC and Queensland Transport. (Asian Development Bank, Manila, Philippines). Jones C R and Smith H R (1980). Deflection temperature relationships for bituminous surfacings in Kenya. Laboratory Report LR 936. Transport Research Laboratory, Crowthorne. Kennedy C K and Lister N W (1978). Prediction of pavement performance and the design of overlays. Laboratory Report LR 833. Transport Research Laboratory, Crowthorne. Kleyn E G and Van Heerden (1983). Using DCP soundings to optimise pavement rehabilitation. Annual Transportation Convention. Report LS/83 Materials Branch, Transvaal Roads Department, Pretoria. Livneh M, Ishai I and Livneh N A (1995). The effect of vertical confinement on the DCP strength values in pavement and subgrade evaluation. Transportation Research Board, 74th Annual Meeting. Washington DC. Morosiuk G, Jones C R and Osman M(1992). Calibration and standardisation of road roughness measurements using the TRRL profile beam. 7th Conf. of the Road Engineering Association of Asia and Australasia, Singapore. National Institute for Road Research (1970). The measurement of deflection and curvature of road surfaces. CSIR Technical Guide K16. Council for Scientific and Industrial Research, Pretoria. National Institute for Transport and Road Research (1983). Bituminous pavement rehabilitation design. Draft Technical recommendations for Highways TRH 12. Council for Scientific and Industrial Research, Pretoria. National Institute for Transport and Road Research (1985). Nomenclature and methods for describing the condition of asphalt pavements. Technical Recommendations for Highways TRH No 6. Council for Scientific and Industrial Research, Pretoria.

41

Paterson W D O (1987). Road deterioration and maintenance effects: models for planning and management. Highway Design and Maintenance Series. The John Hopkins University Press, Baltimore, Maryland. Powell W D, Potter J F, Mayhew H C and Nunn M E (1984). The structural design of bituminous roads. Laboratory Report LR 1132. Transport Research Laboratory, Crowthorne. Queensland Transport (1992). Pavement rehabilitation manual. Pavements and Asset Strategy Branch, Queensland Transport, Queensland. Road Research Laboratory (1969). Instructions for using the portable skid resistance tester. Road Note 27. Stationery Office, London. Roberts J D and Martin T C (1996). Recommendations for monitoring pavement performance. Research Report ARR 293. ARRB Transport Research Ltd, Victoria. Rohde G T (1994). Determining pavement structural number from FWD testing. Transportation Research Record 1448, Transportation Research Board, Washington DC. Rolt J, Smith H R and Jones C R (1986). The design and performance of bituminous overlays in tropical environments. 2nd Conf. The Bearing Capacity of Roads and Airfields, Plymouth. Rolt J, Hasim M S, Hameed M and Suffian Z (1996). The prediction and treatment of reflection cracking ill thin bituminous overlays. Second Malaysian Road Conference 1996, Innovations in Road Building. Kuala Lumpur. Sayers M W, Gillespie T D and Queroz C A V (1986a). The international road roughness experiment: establishing correlation and a calibration standard for measurements. Technical Paper No. 45. World Bank, Washington DC. Sayers M W, Gillespie T D and Paterson W D O (1986b). Guidelines for conducting and calibrating road roughness measurements. Technical Paper No. 46. World Bank, Washington DC. Shell (1985). Addendum to the Shell pavement design manual. Shell International Petroleum Co. Ltd., London. Sherman G (1982). Minimising reflection cracking of pavement overlays. Synthesis of Highway Practice 92. Transportation Research Board, Washington DC. Smith H R and Jones C R (1980). Measurement of pavement defections ill tropical and sub-tropical climates. Laboratory Report LR 935. Transport Research Laboratory, Crowthorne.

Smith H R, Rolt J and Wambura J (1990). The durability of bituminous overlays and wearing courses in tropical environments. 3rd Conf. The Bearing Capacity of Roads and Airfields. Trondheim. Smith R B and Pratt D N (1983). Afield study of in situ California bearing ratio and dynamic cone penetrometer testing for subgrade investigations. Australian Road Research 13(4) pp 285-294. Australian Road Research Board. Strategic Highway Research Programme (1993a). SHRP'S layer moduli backcalculation procedure. SHRP-P655. Strategic Highway Research Programme. National Research Council, Washington DC. Strategic Highway Research Programme (1993b). Falling weight defectometer relative calibration analysis. SHRP-P-652. Strategic Highway Research Programme. National Research Council, Washington DC. Transport and Road Research Laboratory (1978). A guide to the measurement of axle loads in developing countries using a portable weighbridge. Road Note 40. Transport Research Laboratory, Crowthorne. Transport and Road Research Laboratory (1982). A guide to surface dressing in tropical and subtropical climates. Overseas Road Note 3. Transport Research Laboratory, Crowthorne. Transport and Road Research Laboratory (1985). Maintenance techniques for district engineers. Overseas Road Note 2, Transport Research Laboratory, Crowthorne. Transport and Road Research Laboratory (1988). A guide to road project appraisal. Overseas Road Note 5, Transport Research Laboratory, Crowthorne. Transport and Road Research Laboratory (1990). A users manual for a program to analyse dynamic cone penetrometer data. Overseas Road Note 8. Transport Research Laboratory, Crowthorne. Transport and Road Research Laboratory (1993a). The road transport investment model - User's Manual. Overseas Centre, Transport Research Laboratory, Crowthorne. Transport and Road Research Laboratory (1993b). A guide to tile structural design of bitumen-surfaced roads in tropical and sub-tropical climates. Overseas Road Note 31. Transport Research Laboratory, Crowthorne. Van Vuuren (1969). Rapid determination of CBR with the portable dynamic cone penetrometer. The Rhodesian Engineer. Watanatada T, Harral C G, Paterson W D O, Dhareshwar A M, Bhandari A and Tsunokawa K (1987). The highway design and maintenance standards model. Highway Design and Standards Series. The John Hopkins University Press, Maryland.

42

11

Applicable standards

C 535-96

The British Standards Institution is the independent national body for the preparation of British Standards. Enquiries should be addressed to BSI, Linford Wood, Milton Keynes, MK14 6LE. BS 598 Sampling and examination of bituminous mixtures for roads and other paved areas Part 100:1987 Methods for sampling for analysis Methods of test for the determination of texture depth

D 3319-90

Test method for resistance to degradation of large-sized coarse aggregates by abrasion and impact in the Los Angeles machine Test method for accelerated polishing of aggregates using the British wheel Test method for measuring surface frictional properties using the British pendulum tester Test method for measuring road roughness by static level method

E 303-93

E 1364-95

Part 105:1990

BS 812

Sampling and testing of mineral aggregates, sands and fillers Part 2:1975 Methods for determination of physical properties Part 105:1990 Methods for determination of particle shape Methods for determination of aggregate crushing value Methods for determination of ten percent fines value Methods for determination of aggregate impact value Methods for determination of aggregate abrasion value Method for determination of soundness

Part 110:1990

Part 111:1990

Part 112:1990

Part 113:1990

Part 121:1989

BS 1377 Soils for civil engineering purposes Part 2:1990 Classification tests Part 4:1990 Part 8:1990 Part 9:1990 Compaction-related tests Shear strength tests (effective stress) In-situ tests

The American Society for Testing and Materials is a notfor-profit organisation which provides a forum for producers, users, ultimate consumers and those having a general interest to meet on common ground and write standards for materials, products, systems and services. Enquiries should be addressed to ASTM, 100 Barr Harbor Drive, West Conshohocken, PA 19428-2959. C 131-96 Test method for resistance to degradation of small-sized coarse aggregates by abrasion and impact in the Los Angeles machine

43

Appendix A: Detailed surface condition survey

The detailed surface condition survey is a walking survey carried out by a team of four technicians/ labourers and one support vehicle with driver. This team size should be able to complete 10 lane kilometres per day. Increasing output by surveying both lanes of a two lane highway simultaneously is not recommended. A safe working environment should be maintained at all times. Many organisations will have on-site procedures which should be followed. Where there are no local safety procedures those described in Overseas Road Note 2 are recommended (TRRL, 1985). The equipment needed by the team, after the road has been permanently marked, is: traffic control signs or flags; 2 metre straight-edge and wedge (Figure 4.5); crack width gauge (Figure 4.4); distance measurer; and surface condition forms (Figure 4.1) and a clipboard.

The results of the survey should be recorded on pre-printed forms as these provide a check list for the technician, telling him what items are to be examined during the inspection and so reducing the possibility that significant information is omitted.

44

Appendix B: Road roughness measurements

The many methods for measuring road roughness in use throughout the world can be grouped into four generic classes (Sayers et al, 1986b) on the basis of how accurately they measure the profile of the road and hence International Roughness Index (IRI). Class 1 - Precision profiles Class 2 - Other profilometric methods Class 3 - IRI by correlation Class 4 - Subjective ratings Class 1 - Precision profiles This class has the highest standard of accuracy. Class 1 methods are those which sample the vertical profile of the road at distances no greater than 250mm to an accuracy of 0.51rim for smooth roads. This profile is then used to directly compute the IRI. Class 1 methods are mainly used for the calibration and validation of other methods of roughness measurement. They can be used for relatively short sections where a high degree of accuracy is required but are not suitable for general roughness surveys. Examples of Class 1 methods include the rod and level survey (ASTM, E 1364-95), the TRL Profile Beam, the Face Dipstick (Bertrand et al, 1991) and the ARRB Walking Profiler (ARRB, 1996).

estimate IRI and also calibrate other RTRRMS. The MERLIN does not record the absolute profile but measures the mid-chord deviations over a predetermined base length for a section of road (see Figure B 1) and then relates a statistic from the frequency of those deviations to the IRI using a predetermined correlation. The roughness values recorded by RTRRMS depend on the dynamics of the vehicle and the speed at which it is driven. The dynamic properties of each vehicle are unique and will also change with time, for example as springs and shock absorbers wear. It is therefore essential that the roughness values obtained from a RTRRMS are converted to units of IRI by regularly calibrating it with a Class 1 or 2 device or the MERLIN.

Class 4 - Subjective rating This class has the lowest standard of accuracy. It includes methods such as subjective evaluation involving rideability and visual assessment. This is illustrated in Figure B2. The estimate of IRI has been found to be subject to errors of up to 40 per cent for new observers (Sayers et al, 1986b) and therefore this method should only be used when other methods are unavailable. Uncalibrated RTRRMS also fall into this category.

Class 2 - Other profilometric methods This class includes all other methods in which the road profile is measured as the basis for direct computation of the IRI, but which are not capable of the accuracy and/or measurement interval specified for a Class 1 precision profile. This class includes most high-speed profilometers.

Class 3 - IRI from correlation Devices in this class measure roughness but need calibration to convert the data into units of MI. The majority of road roughness data currently collected throughout the world are obtained with Response-Type Road Roughness Measuring Systems (RTRRMS). While these systems can take the form of towed trailers, such as the towed 5th wheel bump integrator, they more frequently involve instruments mounted in a survey vehicle. Examples of vehicle-mounted RTRRMS include the TRL bump integrator unit, the NAASRA meter and the Mays meter. These instruments usually measure roughness in terms of the cumulative movement between the vehicle's axle and chassis when travelling along a road under standard conditions. Also in this class is a low cost alternative, the Machine for Evaluating Roughness using Low-cost INstrumentation (MERLIN) that can be used to both

Roughness surveys using a RTRRMS. When roughness measurements are required on more than a few short sections of road, a RTRRMS is recommended. The main advantages of these types of systems are their relative low cost and the high speed of data collection. The systems are capable of surveys at speeds up to 80 km/h, so many hundreds of kilometres of road can be measured in a day. The TRL Bump Integrator (BI) Unit is a response-type road roughness measuring device that is mounted in a vehicle. The instrument measures the roughness in terms of the cumulative uni-directional movement between the rear axle and the chassis of a vehicle in motion. The BI system comprises of a bump integrator unit, a counter unit with 2 displays, connection leads and an optional installation kit. The system is powered by the 12 volt battery of the vehicle.

Fitting the BI unit The BI unit is mounted in a rear-wheel drive vehicle as shown diagrammatically in Figure B3. The unit is bolted to the rear floorpan of the vehicle directly above the centre of the rear axle. A 25mm hole needs to be cut in the floorpan and a bracket or hook fixed to the centre of the differential housing of the rear axle.

45

Figure B1 Operation of the MERLIN

46

Figure B2

Road roughness estimation scale for paved roads with asphaltic concrete or surface dressed surfacings (after Sayers et al, 1986b)

47

Figure B3 Diagrammatical representation of the TRL Integrator Unit fitted to a vehicle Before each survey, the flexible metal cord from the cylindrical drum of the BI unit is passed through the hole in the floor and hooked onto the bracket on the rear axle. This cord must not touch the sides of the hole. Tension in the cord is maintained by a return spring inside the drum of the BI unit. The BI unit measures the unidirectional movement, in centimetres, between the vehicle chassis and the axle as the vehicle is driven along the road. This is displayed on a counter box, usually fixed to the front passenger fascia. turned clockwise or suddenly released after being tensioned as the internal spring mechanism could be damaged. v When measurements are being taken the vehicle should normally be driven at constant speed, avoiding acceleration, deceleration and gear changes. This is necessary because the vehicle's response to a given profile varies with speed. To reduce reproducibility errors it is best to operate the RTRRMS at a standard speed of 80 km/h. However, if this speed is unsafe for reasons of traffic, pedestrians or restrictive road geometry, a lower speed of 50 or 32 km/h can be used. It is usual to use the same operating speed for all of the surveys. Calibration must be carried out at this operating speed. vi For general surveys, it is recommended that readings are recorded at half kilometre intervals. This distance should be measured with a precision odometer. The use of the vehicle odometer or kilometre posts is not recommended for survey purposes. vii There are two counters in the recording unit, connected by a changeover switch. This allows the observer to throw the switch at the end of each measurement interval so that the reading can be manually recorded while the other counter is working. The first counter can then be re-set to zero ready for the next changeover. Software is also available which automatically records the roughness data, vehicle speed and distances in spreadsheet form. viii The type of road surfacing should also be recorded to aid future analysis of the data. On completion of the survey, the wile cord should be disconnected from the rear axle. ix After the survey, the results should be converted into vehicle response roughness values (VR). The counts measured by the BI are in units of cumulative centimetres of uni-directional movement of the rear axle. These should be converted to vehicle response roughness values using the following equation.

Survey procedure i A safe working environment should be maintained at all times. Many organisations will have on-site safety procedures which should be followed. As the vehicle may be moving slower than the majority of other traffic, it should be clearly signed and fitted with flashing lights. The vehicle should be well maintained and in good working order. The wheels should be properly balanced and the steering geometry correctly aligned. The tyres should not have flat spots or be unduly worn. Tyre pressures should be maintained precisely to the manufacturers specifications. The load in the vehicle must be constant. Ideally the vehicle should contain only the driver and observer, and no other load should be carried. The engine and suspension system should be fully warmed-up before measurements commence. This can be achieved by driving the vehicle for at least 5km before measurements start. The tension cord from the BI unit to the axle should only be connected during the survey. At all other times the cord should be disconnected to stop unnecessary wear to the BI unit. When attaching the cord to the rear axle, the cord should be pre-tensioned by turning the BI pulley 2.5 turns anti-clockwise. The wire is then wound around the pulley 2 turns in the same direction as the arrow. Note: the pulley must NOT be

ii

iii

iv

48

VR = BI count x 10 section length (km) Where VR = Vehicle Response (mnVkm) BI = No of counts per section (cm) These vehicle response roughness values should then be converted to units of estimated IRI, E[IRI], using a calibration that is unique to the RTRRMS at that time. The results of a typical survey in terms of E[IRI] are shown in Figure B4.

ii

Calibration of a RTRRMS The RTRRMS must be regularly calibrated against an instrument such as the TRL Profile Beam, the MERLIN or a rod and level survey. This calibration should preferably be carried out before the survey and checked on control' sites during the survey period to ensure that the RTRRMS remains within calibration. The calibration of the RTRRMS will need to be rechecked before any subsequent surveys or after any part of the suspension of the vehicle is replaced. The calibration exercise basically involves comparing the results from the RTRRMS and the calibration instrument over several short road sections. The relationship obtained by this comparison can then be used to convert RTRRMS survey results into units of E[IRI]. The recommended practice for roughness calibration is described below. i A minimum of eight sections should be selected with varying roughness levels that span the range of roughness of the road network. The calibration sites should be on a similar type of road (ie paved or unpaved roads) to those being surveyed. The

iii

sections should have a minimum length of 200m and should be of uniform roughness over their length. In practice it may be difficult to find long homogeneous sections on very rough roads. In this case it is better to include a shorter section than to omit high roughness sites from the calibration. The sections should be straight and flat, with adequate run-up and slow-down lengths and should have no hazards such as junctions so that the vehicle can travel in a straight course at constant speed along the whole section. The roughness of each section should be measured by the RTRRMS at the same vehicle speed that is to be used for the general survey. The value of VR (mm/km) should be the mean value of at least three test runs. The calibration instrument should measure roughness in both wheelpaths. The average of these IRI values (in m/km) is then plotted against the vehicle response for each of the test sections. The calibration equation for the RTRRMS is then derived by calculating the best fit line for the points. This relationship generally has a quadratic form but has also been found to be logarithmic depending upon the characteristics of the vehicles suspension and the levels of roughness over which the RTRRMS is being calibrated. E[IRI] = a + b VR + c VR2

Where

E[IRI] VR a, b and c

= Estimated IRI (m/km) = Vehicle Response (mm/km) = constants

The calibration equation can then be used to convert data from the RTRRMS into units of E[IRI].

Figure B4 Road roughness profile 49

Appendix C: Deflection beam measurements

The simplest method of measuring the deflection of a road pavement is to use a loaded lorry and the deflection beam, originally devised by A C Benkelman. The beam consists of a slender pivoted beam, approximately 3.7m long, supported in a low frame which rests on the road. The frame is fitted with a dial gauge for registering the movement at one end of the pivoted beam, the other end of which rests on the surface of the road (Figure C1). When making a deflection measurement, the tip of the beam is inserted between the dual rear-wheel assembly of the loaded truck. The dial gauge is set to zero and the truck then drives slowly forward. As the wheels approach the tip of the beam, the road surface deflects downwards and the movement is registered by the dial gauge. As the wheels move away from the tip of the beam, the road surface recovers and the dial gauge reading returns to approximately zero. The test procedure used by the TRL is described in detail by Smith and Jones (1980) and is summarised below.

vi The transient deflection is the average of the loading and recovery deflections. At least two tests should be carried out at each chainage and the mean value is used to represent the transient test result. If the results of the two tests do not fall within the repeatability limits described in Table C 1 then a third test should be carried out. Table C1 Repeatability of duplicate transient deflection tests

Deflection test procedure The lorry should have a capacity of at least 5 tonnes and should be fitted with twin rear wheels having a spacing of 40mnn between the tyres. The lorry is loaded to give a rear axle load of 6350 kg (ie 3175 kg on each pair of twin rear wheels). The recommended tyre size is 8.25 x 20 and the tyres should be inflated to a pressure of 585 kN/m2. ii Mark the point, in the vergeside wheelpath, at which the deflection is to be measured and position the lorry so that the rear wheels are 1.3m behind the marked point. iii Insert the deflection beam between the twin rear wheels until its measuring tip rests on the marked point. Insert a second beam between the offside wheels, if deflections are to be measured in both wheelpaths. It is helpful in positioning the lorry and aligning the beams if a pointer is fixed to the lorry 1.3m in front of each pair of rear wheels. iv Adjust the footscrews on the frame of the beam to ensure that the frame is level and that the pivoted arm is free to move. Adjust the dial gauge to zero and turn the buzzer on. Record the dial gauge reading which should be zero or some small positive or negative number. v The maximum and final reading of the dial gauge should be recorded while the lorry is driven slowly forward to a point at least 5m in front of the marked point. The buzzer should remain on until the final reading is taken. Care must be taken to ensure that a wheel does not touch the beam. If it does the test should be repeated. i

Deflection readings can be affected by a number of factors which should be taken into account before the results can be interpreted. These are the temperature of the road, plastic flow of the surfacing between the loading wheels, seasonal effects and the size of the deflection bowl. Road temperature The stiffness of asphalt surfacings will change with temperature and therefore the magnitude of deflection can also change. The temperature of the bituminous surfacing is recorded when the deflection measurement is taken, thus allowing the value of deflection to be corrected to a standard temperature. It is recommended that 35C, measured at a depth of 40mm in the surfacing, is a suitable standard temperature for roads in tropical climates. The relation between temperature and deflection for a particular pavement is obtained by studying the change in deflection on a number of test points as the temperature rises from early morning to midday (Jones and Smith, 1980). It is not possible to produce general correction curves to cover all roads found in tropical countries so it is necessary to establish the deflection/temperature relationship for each project. Fortunately, it is often found that little or no correction is required when the road surfacing is either old and age hardened or relatively thin. Plastic flow Plastic flow of new bituminous surfacings can occur during deflection testing. As the surfacing is squeezed up between the twin wheels the transient

50

Figure Cl Diagrammatic representation of the deflection beam

51

deflection obtained will be less than the true value. Plastic flow can easily be identified by high negative final readings being recorded during the transient test. Alternative test procedures such as the `rebound' deflection test (Smith and Jones, 1980) do not identify when plastic flow is occurring.

Seasonal effects In areas where the moisture content of the subgrade changes seasonally, the deflection will also change. For overlay design purposes, it is usual to use values which are representative of the most adverse seasonal conditions. It is therefore normal practice to carry out surveys just after the rainy season. If this cannot be done, an attempt should be made to correct for the seasonal effect. However, this requires a considerable data bank of deflection results and rainfall records before reliable corrections can be made.

Size of deflection bowl The size of the deflection bowl can occasionally be so large that the front feet of the deflection beam lie within the bowl at the beginning of the deflection test. If this happens, the loading and recovery deflection will differ. The simplest way to check whether the differences in loading and recovery deflection are caused by the size of the bowl is to place the tip of another beam close to the front feet of the measurement beam at the beginning of the transient test. This second beam can be used to measure any subsequent movement of the feet of the first beam as the lorry moves forward. If feet movements larger than 0.06mm are observed only the recovery part of the deflection cycle should be used for estimating the value of transient deflection.

52

Appendix D: Deflection beam survey procedure

A safe working environment should be maintained at all times. Many organisations will have on-site safety procedures which should be followed. Where there are no local safety procedures those described in Overseas Road Note 2 are recommended (TRRL, 1985). Deflection beam measurements are made in both wheelpaths of the slow lane on dual carriageways and in both lanes of a two-lane road. The following strategy is adopted. Tests are carried out on a basic pattern of 50 or 100 metre spacings. ii Additional tests should be undertaken on any areas showing atypical surface distress. iii When a deflection value indicates the need for a significantly thicker overlay than is required for the adjacent section, the length of road involved should be determined by additional tests. After all measurements have been made, they should be corrected for any temperature effect (Appendix C). It is then convenient to plot the deflection profile of the road for each lane, using the larger deflection of either wheelpath at each chainage. Any areas showing exceptionally high deflections which may need reconstruction or special treatment can then be identified. The deflection profile is then used to divide the road into homogeneous sections, in such a way as to minimise variation in deflections within each section. The minimum length of these sections should be compatible with the frequency of thickness adjustments which can sensibly be made by the paving machine, whilst still maintaining satisfactory finished levels. When selecting the sections the topography, subgrade type, pavement construction and maintenance history should all be considered. There are a number of statistical techniques that can be used to divide deflection data into homogeneous sections. One of these techniques is the cumulative sum method, where plots of the cumulative sums of deviations from the mean deflection against chainage can be used to discern the sections. The cumulative sum is calculated in the following way. Si = xi - xm + S i-l where xi = Deflection at chainage xm = Mean deflection Si = Cumulative sum of the deviations from the mean deflection at chainage i Using the cumulative sums, the extent to which the measured deflections on sections of road varies from the mean deflection of the whole road can be determined. Changes in the slope of the line connecting the cumulative sums will indicate inhomogeneity. i

The coefficient of variation (CoV = standard deviation/mean) may be used to determine the level of homogeneity using the following guidelines. CoV < 0.2 0.2 < CoV < 0.3 CoV > 0.3 good homogeneity moderate homogeneity poor homogeneity

CoVs greater than 0.3 usually indicate a highly skewed distribution produced, for example, by a number of relatively `stronger' points within a weaker section. Other authorities (NITRR, 1983) (AUSTROADS, 1992) have recommended, as a guide, that a homogeneous section is one where the deflection values have a CoV of 0.25 or less. The final stage of the procedure is to calculate the representative deflection for each homogeneous section of the road. The proposed method will tend to separate out areas of very high deflections on areas that warrant special treatment or reconstruction and therefore the distribution of the remaining deflection measurements will approximate to a normal distribution. The representative deflection, which is the 90th percentile value, can then be calculated as follows: Representative deflection = x m + 1.3 standard deviation

53

Appendix E: Falling Weight Deflectometer (FWD) test procedure

Calibration Evidence of a satisfactory absolute calibration of the deflection sensors and the load cell shall be provided by the operator of the FWD. The absolute calibration should be carried out annually, or as soon as possible after any sensor has been replaced. The calibration should be carried out by either the manufacturer or a recognised testing authority accredited by the manufacturer. In addition to the annual absolute calibration other checks need to be carried out every 6 weeks. These are the consistency check and the relative consistency check. The consistency check is used to verify whether the central deflection sensor and the load cell are giving reproducible results over a period of time. In this check five test points, in three road sections, are tested at regular 6 week intervals. The road sections selected should be representative of the pavement structures that are generally being tested, be in good condition, be lightly trafficked and be efficiently drained such that any seasonal variation in deflection is minimised. If the sections have significant layers of bituminous material then the temperature of surfacing should be recorded during the tests. Examination of the variation in deflection, normalised to a standard load and temperature, will provide an indication of any inconsistency in the equipment. The relative consistency check is used to ensure that all the deflection sensors on the FWD are in calibration with respect to each other. The relative consistency check uses a calibration tower, supplied by the manufacture, in which all the sensors are stacked vertically. The sensors are then all subject to the same pavement deflection. The procedure is fully described in the manufacturers literature or can be found in SHRP-P-652 (1993b). Test procedure A safe working environment should be maintained at all times. Many organisations will have on-site safety procedures which should be followed. The safety aspects of a FWD survey are particularly difficult to manage, as it is a mobile operation, and the supervising engineer should ensure that satisfactory procedures are followed. Where short lengths of road are being investigated they should be coned off. If measurements are being carried out over longer lengths of road then the operator, driver and traffic control personnel should always be extremely aware of both the movements of the testing equipment and other vehicles on the road. In addition to static road signs, the towing vehicle should always be fitted with

flashing lights and direction signs and all personnel should wear high visibility safety jackets. Typically tests should be carried out at intervals of 20100 metres in the vergeside wheelpath in each direction. Additional tests should be undertaken on any areas showing atypical surface distress. On flexible pavements the load level should be set at a nominal load of 50kN +/- 10%. On roads with bituminous seals, often found in the developing world, this level of load may possibly over-stress the pavement, in which case the load level should be reduced. The load should be applied through a 300mrn diameter plate and the load pulse rise time should lie between 5 and 15 milliseconds. The deflection should be measured by at least five and preferably seven deflection sensors having a resolution of one micron. The location of the sensors depends on the stiffness of the pavement structure. The stiffness of the subgrade has a major influence on the shape of the deflection bowl and therefore there should be at least two sensors at such a distance from the load centre as to enable the stiffness of the subgrade to be assessed. In the case where seven sensors are available, the recommended sensor positions are given in Table E1. Table El Recommended sensor positions

Temperature measurements When the road has an asphalt surfacing the deflection may change as the temperature of the surfacing changes. Also when the deflection bowl is to be used to estimate pavement layer moduli, the stiffness of the asphalt surfacing will need to be corrected to a standard temperature. It is therefore necessary to measure the temperature of the surfacing during testing. In temperate climates measurements taken hourly may be sufficient, however, in tropical climates the pavement temperature will rise quickly during mid-morning and can reach a temperature at which the asphalt surfacing is liable to plastic deformation during testing. This must be carefully monitored and temperature measurements at this critical time of the day may need to be taken every 15 or 20 minutes.

54

The temperature of the pavement can be measured using either a short-bulb mercury thermometer or a digital thermometer. The temperature holes should be at least 0.3m from the edge of the surfacing and, where possible, they should be pre-drilled to allow the heat to dissipate before temperatures are measured. The temperature of the surfacing should not be measured under any road markings. Glycerol or oil in the bottom of the hole will ensure a good thermal contact between the temperature probe or thermometer and the bound material. Where the asphalt surfacing is less than 150mm the temperature should be measured at a depth of 40mm. When the surfacing exceeds 150mm, it is recommended that temperatures should be recorded at two depths, 40 and 100mm.

55

Appendix F: TRL Dynamic Cone Penetrometer (DCP) test procedure

The TRL DCP uses an 8 kg hammer dropping through a height of 575mm and a 60 cone having a maximum diameter of 20mm. The instrument is assembled as shown in Figure F1. It is supplied with two spanners and a tommy bar to ensure that the screwed joints are kept tight at all times. To assist in this the following joints should be secured with a non-hardening thread locking compound prior to use: Handle/hammer shaft Coupling/hammer shaft Standard shaft/cone

will be more difficult to identify layer boundaries accurately, hence important information will be lost. When the extended version of the DCP is used the instrument is driven into the pavement to a depth of 400500mm before the extension shaft can be added. To do this the metre rule is detached from its baseplate and the shaft is split to accept the extension shaft. After re-assembly a penetration reading is taken before the test is continued. After completing the test the DCP is removed by tapping the weight upwards against the handle. Care should be taken when doing this; if it is done too vigorously the life of the instrument will be reduced. The DCP can be driven through surface dressings but it is recommended that thick bituminous surfacings are cored prior to testing the lower layers. Little difficulty is normally experienced with the penetration of most types of granular or lightly stabilised materials. It is more difficult to penetrate strongly stabilised layers, granular materials with large particles and very dense, high quality crushed stone. The TRL instrument has been designed for strong materials and therefore the operator should persevere with the test. Penetration rates as low as 0.5mm/blow are acceptable but if there is no measurable penetration after 20 consecutive blows it can be assumed that the DCP will not penetrate the material. Under these circumstances a hole can be drilled through the layer using an electric or pneumatic drill, or by coring. The lower pavement layers can then be tested in the normal way. If only occasional difficulties are experienced in penetrating granular materials, it is worthwhile repeating any failed tests a short distance away from the original test point. If, during the test, the DCP leans away from the vertical no attempt should be made to correct it because contact between the shaft and the sides of the hole can give rise to erroneous results. Research (Livneh, 1995) has shown that there can be an overestimate of subgrade strength as a result of friction on the rod caused by either tilted penetration through, or collapse of, any upper granular pavement layers. Where there is a substantial thickness of granular material, and when estimates of the actual subgrade strength are required (rather than relative values) it is recommended that a hole is drilled through the granular layer prior to testing the lower layers. If the DCP is used extensively for hard materials, wear on the cone itself will be accelerated. The cone is a replaceable part and it is recommended by other authorities that it should be replaced when its diameter is reduced by 10 per cent. However, other causes of wear can also occur hence the cone should be inspected before every test.

The instrument is usually split at the joint between the standard shaft and the coupling for carriage and storage and therefore it is not usual to use locking compound at this joint. However it is important that this joint is checked regularly during use to ensure that it does not become loose. Operating the DCP with any loose joints will significantly reduce the life of the instrument. Operation A safe working environment should be maintained at all times. Many organisations will have on-site safety procedures which should be followed. Where there are no local safety procedures those in Overseas Road Note 2 are recommended (TRRL, 1985). After assembly, the first task is to record the zero reading of the instrument. This is done by standing the DCP on a hard surface, such as concrete, checking that it is vertical and then entering the zero reading in the appropriate place on the proforma (See Figure F2). The DCP needs three operators, one to hold the instrument, one to raise and drop the weight and a technician to record the readings. The instrument is held vertical and the weight raised to the handle. Care should be taken to ensure that the weight is touching the handle, but not lifting the instrument, before it is allowed to drop. The operator must let it fall freely and not partially lower it with his hands. It is recommended that a reading should be taken at increments of penetration of about l0mm. However it is usually easier to take a reading after a set number of blows. It is therefore necessary to change the number of blows between readings, according to the strength of the layer being penetrated. For good quality granular bases readings every 5 or 10 blows are usually satisfactory but for weaker sub-base layers and subgrades readings every I or 2 blows may be appropriate. There is no disadvantage in taking too many readings, but if readings are taken too infrequently, weak spots may be missed and it

56

Interpretation of results The results of the DCP test are usually recorded on a field data sheet similar to that shown in Figure F2. The results can then either be plotted by hand, as shown in Figure F3, or processed by computer (TRRL, 1990). Relationships between DCP readings and CBR have been obtained by several research authorities (see Figure F4). Agreement is generally good over most of the range but differences are apparent at low values of CBR in fine grained materials. It is expected that for such materials the relationship between DCP and CBR will depend on material state and therefore, if more precise values are needed it is advisable to calibrate the DCP for the material being evaluated.

57

Figure Fl TRL Dynamic Cone Penetrometer

58

Figure F2 DCP test field sheet

59

Figure F3 Typical DCP test result

Figure F4 DCP-CBR relationships

60

Appendix G: Test pit procedure

Purpose The purpose of carrying out a test pit investigation is to confirm the engineers understanding of the information from surface condition, deflection and DCP surveys. It is a time consuming and expensive operation, and for this reason the location of each test pit should be carefully selected to maximise the benefit of any data collected. The condition of the road pavement and the primary purpose for the pit investigation should be recorded on the Test Pit Log (see Figure G1).

Sampling and testing Before commencing the survey in the field, the engineer should be clear as to the information required from each test pit. This will depend on the results of previous surveys, the materials specifications in use and an understanding of the pavement behaviour. Some field testing might be necessary as well as subsequent laboratory testing of samples extracted from the pit. Table GI summarises the various tests that may be required and references the relevant standards with which the tests should comply. Not all these tests will be necessary and the engineer must decide on those which are required. Procedure A safe working environment should be maintained at all times. Many organisations will have on-site safety procedures which should be followed. Where there are no local safety procedures those described in Overseas Road Note 2 are recommended (TRRL, 1985). Once it has been decided what testing is to be carried out and the location of the trial pits has been confirmed, the following procedure should be adopted. i ii Set up traffic control. Accurately locate position of test pit and record this on the Pavement Test Pit Log (see Figure G1). Usually, the position of a pit will be apparent after completion due to the patched surface. However, if long term monitoring is required, a permanent location marker should be placed at the roadside. Record any relevant details such as surrounding drainage features, road condition and weather. Define the edge of the test pit and remove surfacing. The required size of pit will depend on the sample sizes necessary for the selected tests, but it can be increased later if found to be too small. Usually an area of about 0.8m by 0.8m will be sufficient for manual excavation, and the minimum working area required for a backhoe operation will be sufficient for machine excavations. The edge of the pit can be cut with a jack hammer or pick and the surfacing `peeled' off, taking care not to disturb the surface of the roadbase. The average thickness of surfacing should be recorded. If density tests are to be performed, a smooth, clean and even surface is required. It is important for the accuracy of the test that the layer is homogeneous. For the sand replacement method, no prior knowledge is required of the layer thickness since this becomes obvious as the hole is excavated. If a nuclear density meter is used,

Labour, equipment and materials Test pits can be excavated either by machine or manually. The choice will normally be determined by the availability of plant and the test pit programme, as machine operations are usually more productive but more costly than manual methods. The following personnel are required: traffic controllers - a minimum of one at each end of the site; 2 (if machine excavation) or 3 (if manual excavation) labourers; 1 machine operator if applicable; 1 driver for vehicle; and 1 supervising technician.

Equipment and materials requirements are as follows: 1 backhoe (for machine excavation); 1 jack hammer with generator (to assist with manual excavation); 1 pick; 1 or 2 spades (a fence post hole digger can also be useful); 1 tamper or plate compactor for backfilling test pit; material to backfill and seal test pit : gravel, cement for stabilising gravel, water and cold nix for resurfacing; 1 broom to tidy area on completion; 1 chisel is often useful to assist with inspecting the wall of the test pit; equipment necessary to complete any required onsite testing; 1 tape measure and thin steel bar to span pit (to assist with depth measurements); sample bags and containers, with some means of labelling each; test pit log forms and clipboard; and sample log book.

iii

iv

61

the thickness of the layer can be estimated from the DCP results to determine the depth of testing. v On completion of any required density testing, the layer can be removed over the extent of the trial pit, a visual assessment made of the material and samples taken for laboratory testing. Care should be taken not to disturb the adjacent lower layer. The thickness of the layer and the depth at which samples are taken should be measured. All information should be recorded on the Pavement Test Pit Log. vi Continue to sample, test and excavate each pavement layer following the procedure above. Once it has been decided that there is no need to excavate further, the total depth of pit should be recorded along with any other information such as appearance of water in any of the layers. vii All samples should be clearly labelled and proposed tests for the pit materials should be logged in a sample log book to avoid later confusion in the laboratory. viii The pit should be backfilled in layers with suitable material which should be properly compacted. It is often good practice to stabilise the upper layer with cement accepting that full compaction will not be achieved. A bituminous cold mix can be used to patch the backfilled pit. ix The site should be cleared and left in a tidy and safe condition for traffic.

62

Figure G1 Test Pit Log sheet 63

Table G1 Possible information from test pit investigation

64

Appendix H: Sand patch test

The sand patch test is described in detail in BS 598 Part 105 (1990). The method is summarised below. Apparatus i ii Measuring cylinder of 50ml volume. A spreader disc comprising a flat wooden disc 65mm in diameter with a hard rubber disc 1.5mm thick, stuck to one face. The reverse face being provided with a handle. iii Washed and dried sand, with rounded particle shape, complying with the grading given in Table H 1. Table Hl Grading of sand

Procedure i Dry the surface to be measured and, if necessary, sweep clean with a brush. ii Fill the cylinder with sand and, taking care not to compact it by unnecessary compaction, strike off the sand level with the top of the cylinder. iii Pour the sand into a heap on the surface to be tested, and spread the sand over the surface, working the disc with its face kept flat, in a rotary motion so that the sand is spread into a circular patch. The patch should be of the largest diameter which results in the surface depressions just being filled with sand to a level of the peaks. iv Measure the diameter of the sand patch to the nearest 1 nun at four diameters every 45 and calculate the mean diameter (D) to the nearest 1mm. v Calculate the texture depth to the nearest 0.01 nun from the following equation. Texture depth (mm) = 63660 / D2

Note: For surfacings having a texture depth of less than 1 mm the volume of sand will have to be reduced to 25ml or less. The texture depth is then calculated using the following equation: Texture depth (mm) = Volume of sand (ml) . 1000 Area of patch (mm2)

65

Appendix I: The portable skid-resistance tester

The portable skid-resistance tester, shown in Figure I1, was developed by the Road Research Laboratory and is described in detail in Road Note 27 (RRL, 1969). The testing procedure is summarised below. Operation of the tester i After ensuring that the road surface is free from loose grit, wet both the surface of the road and the slider. ii Bring the pointer round to its stop. Release the pendulum arm by pressing button C and catch it oft the return swing, before the slider strikes the road surface. Record the indicated value. iii Return the arm and pointer to the locked position, keeping the slider clear of the road surface by means of the lifting handle. Repeat the process, spreading water over the contact area with a hand or brush between each swing. Record the mean of five successive swings, provided they do not differ by more than three units. If the range is greater than this, repeat swings until three successive readings are constant; record this value. iv Raise the head of the tester so that it swings clear of the surface again and check the free swing for any zero error. v Sliders should be renewed when the sliding edge becomes burred or rounded. One slider edge can usually be used for at least 100 tests (500 swings). New sliders should be roughened before use by swinging several times over a dry piece of road. Temperature correction The effect of temperature on rubber resilience makes it necessary to correct the measured value of skid resistance to a standard temperature. The road temperature is measured by recording the temperature of the water after the test using a digital thermometer and surface probe. It is recommended that in tropical climates the value should be corrected to a standard temperature of 35C using the following relation (Beaven and Tubey, 1978). SRV35 Where SRV35 SRVt t = (100 + t)/135.SRVt = Skid resistance value at 35C = Measured skid resistance value = Temperature of test (C)

Setting the tester i Set the base level using the in-built spirit level and the three levelling screws on the base-frame. ii Raise the head so that the pendulum arm swings clear of the surface. Movement of the head of the tester, which carries the swinging arm, graduated scale, pointer and release mechanism is controlled by a rack and pinion on the rear of the vertical column. After unclamping the locking knob A at the rear of the column, the head may be raised or lowered by turning either of the knobs B/B1. When the required height is obtained, the head unit must be locked in position by using the clamping knob A. iii Check the zero reading. This is done by first raising the swinging arm to the horizontal release position, on the right-hand side of the tester. In this position it is automatically locked in the release catch. The pointer is then brought round to its stop in line with the pendulum arm. The pendulum arm is released by pressing button C. The pointer is carried with the pendulum arm on the forward swing only. Catch the pendulum on its return swing, and note the pointer reading. Correct the zero setting as necessary by adjusting the friction rings. iv With the pendulum arm free, and hanging vertically, place the spacer, which is attached to the base of the vertical column, under the lifting-handle settingscrew to raise the slider. Lower the head of the tester, using knob B, until the slider just touches the road surface, and clamp in position with knob A. Remove the spacer. v Check the sliding length of the rubber slider over the surface under test, by gently lowering the pendulum arm until the slider just touches the surface first one side and then the other side of the vertical. The sliding length is the distance between the two extremities where the sliding edge of the rubber touches the test surface. To prevent undue wear of the slider when moving the pendulum arm through the arc of contact, the slider should be raised off the road surface by means of the lifting handle. If necessary, adjust to the correct length by raising or lowering the head slightly. When the apparatus is set correctly the sliding length should be between 125 and 127nun as indicated by the measure provided. Place the pendulum arm in its locked position. The apparatus is now ready for testing.

At this standard temperature the corrected values will be 3-5 units lower than comparable surfaces in the UK, where results are corrected to 20C.

66

Figure I1 Portable skid-resistance tester

67

68

69

70

ISSN 0951-8797

OS-E

A guide to the pavement evaluation and maintenance of bitumen-surfaced roads in tropical and sub-tropical countries

ORN 18

Old Wokingham Road Crowthorne Berkshire RG45 6AU

Overseas Development Administration 94 Victoria Street London SW1E 5JL

OVERSEAS ROAD NOTE 31 (FOURTH EDITION)

A GUIDE TO THE STRUCTURAL DESIGN OF BITUMENSURFACED ROADS IN TROPICAL AND SUB-TROPICAL COUNTRIES

Main Subject area: Roads Theme: Road Pavement Engineering Project title: The revision and extension of Road Note 31 Project reference: R5611

This document is an output from an ODA-funded research project, carried out for the benefit of developing countries.

Crown copyright 1993. The views expressed in this report are not necessarily those of the Overseas Development Administration or the Department of Transport. Overseas Centre, TRL, 1993 ISSN 0951-8797

ACKNOWLEDGEMENTS This Note is based principally on research conducted in countries throughout the world by the Overseas Centre (Programme Director Dr J Rolt) on behalf of the Overseas Development Administration. The research has been carried out with the active collaboration of highway authorities in many countries and their help and cooperation has been essential to the success of the work. The TRL project team responsible for the main chapters consisted of J Rolt, H R Smith, T Toole and C R Jones. Additional written contributions were provided by D Newill, C J Lawrance, C S Gourley and R Robinson (now a private consultant). Chapter 7 is based on an original contribution by H E Bofinger, formerly a member of the Overseas Centre and currently a partner of Roughton & Partners. Where necessary use has also been made of other Overseas Centre publications by other authors. The draft of the Note has been reviewed by experts from the Engineering Division of ODA, universities and consulting firms and their comprehensive contributions are gratefully acknowledged.

OVERSEAS ROAD NOTES Overseas Road Notes are prepared principally for road and transport authorities in countries receiving technical assistance from the British Government. A limited number of copies is available to other organisations and to individuals with an interest in roads overseas. Enquiries should be made to: Transport Research Laboratory Crowthorne, Berkshire RG45 6AU Crown Copyright 1993 Limited extracts from the text may be reproduced, provided the source is acknowledged. For more extensive reproduction please write to the above address.

FOREWORD Road Note 31 was first published in 1962 and revised in 1966 and 1977 to take account of advances in our understanding of the behaviour of road-budding materials and their interaction in composite pavements. Many of these advances have been made by engineers and scientists working in temperate climates but a considerable amount of both fundamental and applied research has been necessary to adapt and develop the knowledge so that it can be used with confidence in tropical and subtropical regions where conditions are often very different. In addition to differences associated with climate and types of materials, problems also arise in some countries from uncontrolled vehicle loading and unreliable road maintenance. At the same time, the level of technology available for construction and maintenance can be relatively low. All this has presented a unique challenge to the highway engineer. This edition of the Road Note has drawn on the experience of TRL and collaborating organisations in over 30 tropical and sub-tropical countries. Most of this experience has been gained in carrying out a research and development programme on behalf of the Overseas Development Administration, with additional projects for other aid agencies. The new edition extends the designs of previous editions to cater for traffic up to 30 million equivalent standard axles and takes account of the variability in material properties and construction control, the uncertainty in traffic forecasts, the effects of climate and high axle loads and the overall statistical variability in road performance. The range of structures has been expanded and the chapters on the different types of materials have been enlarged to provide more detailed advice on specifications and techniques. Nevertheless, there will be situations and conditions which are not covered in the Note and there will be many examples where local knowledge can be used to refine and improve the recommendations. Indeed, the role of local engineering knowledge and the judgement of experienced engineers should never be underestimated and should always form an important part of the design process. The Note has been produced in response to a continuous demand from engineers worldwide and I am confident that the new edition will prove to be as popular with practitioners as its predecessors. John Hodges Chief Engineering Adviser Overseas Development Administration

CONTENTS
Page 1. Introduction 1.1 1.2 1.3 1.4 1.5 General Road deterioration Economic considerations Effects of climate Variability in material properties and road performance Uncertainty in traffic forecasts Basis for the design catalogue The design process 1 1 1 1 2 4.2 4.1.5 4.1.6 Compaction of embankments Site control Page 15 17 17 17 17 17 19 19 19 19 19 21 21 21

Cuttings 4.2.1 4.2.2 4.2.3 Slope stability Surveys Design and construction

2 5. Drainage and Road Shoulders 2 5.1 The drainage system Pavement cross-section Drainage of layers Shoulder materials 2 5.2 3 5.3 5 5.4 5 6. Unbound Pavement Materials 5 6.1 Roadbase materials 6.1.1 Crushed stone Naturally occurring granular materials 5 6 6.1.2 7 7 7 6.2 23 25 25 25

1.6 1.7 1.8

2. Traffic 2.1 2.2 Design life Estimating traffic flows 2.2.1 2.2.2 2.3 Baseline traffic flows Traffic forecasting

Axle loading 2.3.1 2.3.2 2.3.3 Axle equivalency Axle load surveys Determination of cumulative equivalent standard axles

Sub-bases (GS) 6.2.1 6.2.2 Bearing capacity Use as a construction platform Sub-base as a filter or separating layer

8 6.2.3 9 10 6.3 25

2.4

Accuracy

3. The Subgrade 3.1 Estimating the subgrade moisture content Determining the subgrade strength

Selected subgrade materials and capping layers (GC)

26 27 27 27 28 29 29 29 30

10 11 14 14 14 14 15

7. Cement and Lime-Stabilised Materials 7.1 7.2 7.3 7.4 Introduction The stabilisation process Selection of type of treatment Cement stabilisation 7.4.1 7.4.2 7.5 Selection of cement content Preparation of specimens

3.2

4. Embankments and Cuttings 4.1 Embankments 4.1.1 4.1.2 4.1.3 4.1.4 Introduction and survey Materials Design Construction over compressible soils

Lime stabilisation

15

Page 7.5.1 Properties of lime-stabilised materials Types of lime 10. Structure Catalogue 30 11. References 7.5.2 30 12. Bibliography 7.5.3 Selection of lime content 7.6 Pozzolans 7.7 Construction 7.7.1 7.7.2 General methodology Control of shrinkage and reflection cracks Carbonation 30 Appendix A 30 Appendix B 31 31 Appendix C 32 Appendix D 7.7.3 7.8 32 1. Quality control 33 2. 8. Bitumen-Bound Materials 8.1 8.2 8.3 Components of mix Bituminous surfacings Types of premix in common use 8.3.1 8.3.2 8.3.3 8.3.4 8.3.5 8.4 Asphaltic concrete Bitumen macadam Rolled asphalt Flexible bituminous surfacing Design to refusal density 34 3. 34 3.1 34 3.2 35 35 38 Appendix E 38 1. 38 2. 38 40 40 42 42 44 44 44 Method of operation General description The Probe Penetration Test 4. Transfer of laboratory design to compaction trials Laboratory design procedure Extended vibrating hammer compaction Extended Marshall compaction Introduction Refusal Density Design TRL Dynamic Cone Penetrometer Estimating Subgrade Moisture Content for Category 1 Conditions Applicable British Standards

Page 50 61 64 65

67

68 72 72 72 72 72

73

Possible problems with the test procedures

73 74 74 74

Bituminous roadbases 8.4.1 8.4.2 Principal mix types Sand-bitumen mixes

8.5

Manufacture and construction

9. Surface Treatments 9.1 9.2 Prime and tack coats Surface dressing 9.2.1 Single and double surface dressing Type of surface Traffic categories Chippings Binder

44 44 45 45 46 46

9.2.2 9.2.3 9.2.4 9.2.5 9.3

Slurry seals

A GUIDE TO THE STRUCTURAL DESIGN OF BITUMENSURFACED ROADS IN TROPICAL AND SUB-TROPICAL COUNTRIES
1. INTRODUCTION
1.1 GENERAL
This Road Note gives recommendations for the structural design of bituminous surfaced roads in tropical and subtropical climates It is aimed at highway engineers responsible for the design and construction of new road pavements and is appropriate for roads which are required to carry up to 30 million cumulative equivalent standard axles in one direction. The design of strengthening overlays is not covered nor is the design of earth, gravel or concrete roads. Although this Note is appropriate for the structural design of flexible roads in urban areas, some of the special requirements of urban roads, such as the consideration of kerbing, sub-soil drainage, skid resistance, etc , are not covered. For the structural design of more heavily trafficked roads, the recommendations of this Note may be supplemented by those given in the guides for the design of bituminous pavements in the. United Kingdom (Powell et al (1984)) but these are likely to require some form of calibration or adaptation to take account of the conditions encountered in the tropics. Acceptable levels of surface condition have usually been based on the expectations of road users. These expectations have been found to depend upon the class of road and the volume of traffic such that the higher the geometric standard, and therefore the higher the vehicle speeds, the lower the level of pavement distress which is acceptable. In defining these levels, economic considerations were not considered because there was insufficient knowledge of the cost trade-offs for an economic analysis to be carved out with sufficient accuracy.

1.3 ECONOMIC CONSIDERATIONS

In recent years a number of important empirical studies have shown how the costs of operating vehicles depend on the surface condition of the road. The studies have also improved our knowledge of how the deterioration of roads depends on the nature of the traffic, the properties of the road-making materials, the environment, and the maintenance strategy adopted (Parsley and Robinson (1982), Paterson (1987), Chesher and Harrison (1987), Watanatada et al (1987)). In some circumstances it is now possible to design a road in such a way that provided maintenance and strengthening can be carried out at the proper time, the total cost of the transport facility i e the sum of construction costs, maintenance costs and road user costs, can be minimised. These techniques are expected to become more widespread in the future. Also, with the introduction in many countries of pavement management systems in which road condition is monitored on a regular basis, additional information will be collected to allow road performance models to be refined. Pavement structural design could then become an integral part of the management system in which design could be modified according to the expected maintenance inputs in such a way that the most economic strategies could be adopted. Whilst these refinements lie in the future, the research has provided important guidance on structural designs suitable for tropical and sub-tropical environments and has been used, in part, in preparing this edition of Road Note 31. For the structures recommended in this Note, the level of deterioration that is reached by the end of the design period has been restricted to levels that experience has shown give rise to acceptable economic designs under a wide range of conditions. It has been assumed that routine and periodic maintenance activities are carried out to a reasonable, though not excessive, level In particular, it has been assumed that periodic maintenance is done whenever the area of road surface experiencing defects i e cracking, ravelling, etc , exceeds 15 per cent. For example, for a 10 year design period, one surface maintenance treatment is likely to be required for

1.2 ROAD DETERIORATION

The purpose of structural design is to limit the stresses induced in the subgrade by traffic to a safe level at which subgrade deformation is insignificant whilst at the same time ensuring that the road pavement layers themselves do not deteriorate to any serious extent within a specified period of time. By the nature of the materials used for construction, it is impossible to design a road pavement which does not deteriorate in some way with time and traffic, hence the aim of structural design is to limit the level of pavement distress, measured primarily in terms of riding quality, rut depth and cracking, to predetermined values. Generally these values are set so that a suitable remedial treatment at the end of the design period is a strengthening overlay of some kind but this is not necessarily so and roads can, in principle, be designed to reach a terminal condition at which mayor rehabilitation or even complete reconstruction is necessary. However, assessing appropriate remedial treatments for roads which have deteriorated beyond a certain level is a difficult task. In most design methods it is assumed that adequate routine and periodic maintenance is carved out during the design period of the road and that at the end of the design period a relatively low level of deterioration has occurred

the higher traffic levels whereas for a 15 year design period, one treatment is likely to be required for the lower traffic levels and two for the higher. These are broad guidelines only and the exact requirements will depend on local conditions.

1.4 EFFECTS OF CLIMATE

Research has shown how different types of road deteriorate and has demonstrated that some of the most common modes of failure in the tropics are often different from those encountered in temperate regions. In particular, climate related deterioration sometimes dominates performance and the research has emphasised the overriding importance of the design of bituminous surfacing materials to minimise this type of deterioration (Paterson (1987), Smith et al (1990), Strauss et al (1984)). This topic is dealt with in Chapter 8. Climate also affects the nature of the soils and rocks encountered in the tropics Soil-forming processes are still very active and the surface rocks are often deeply weathered. The soils themselves often display extreme or unusual properties which can pose considerable problems for road designers. The recent publication 'Road building in the tropics materials and methods' provides an introduction to these topics (Millard (1993)).

are taken into account, there often remains a considerable variation in performance between nominally identical pavements which cannot be fully explained. Optimum design therefore remains partly dependent on knowledge of the performance of in-service roads and quantification of the variability of the observed performance itself. Thus there is always likely to be scope for improving designs based on local experience. Nevertheless, it is the task of the designer to estimate likely variations in layer thicknesses and material strengths so that realistic target values and tolerances can be set in the specifications to ensure that satisfactory road performance can be guaranteed as far as is possible. The thickness and strength values described in this Road Note are essentially minimum values but practical considerations require that they are interpreted as lower ten percentile values with 90 per cent of all test results exceeding the values quoted. The random nature of variations in thickness and strength which occur when each layer is constructed should ensure that minor deficiencies in thickness or strength do not occur one on top of the other, or very rarely so. The importance of good practice in quarrying, material handling and stock-piling to ensure this randomness and also to minimise variations themselves cannot be over emphasised.

1.5 VARIABILITY IN MATERIAL PROPERTIES AND ROAD PERFORMANCE

Variability in material properties and construction control is generally much greater than desired by the design engineer and must be taken into account explicitly in the design process. Only a very small percentage of the area of the surface of a road needs to show distress for the road to be considered unacceptable by road users. It is therefore the weakest parts of the road or the extreme tail of the statistical distribution of 'strength' which is important in design. In well controlled full-scale experiments this variability is such that the ten per cent of the road which performs best will carry about six times more traffic before reaching a defined terminal condition than the ten per cent which performs least well. Under normal construction conditions this spread of performance becomes even greater. Some of this variability can be explained through the measured variability of those factors known to affect performance. Therefore, if the likely variability is known beforehand, it is possible, in principle, for it to be taken into account in design. It is false economy to minimise the extent of preliminary. investigations to determine this variability. In practice it is usually only the variability of subgrade strength that is considered and all other factors are controlled by means of specifications i e by setting minimum acceptable values for the key properties. But specifications need to be based on easily measurable attributes of the materials and these may not correlate well with the fundamental mechanical properties on which behaviour depends. As a result, even when the variability of subgrade strength and pavement material properties.

1.6 UNCERTAINTY IN TRAFFIC FORECASTS

Pavement design also depends on the expected level of traffic. Axle load studies and traffic counts are essential prerequisites for successful design but traffic forecasting remains a difficult task and therefore sensitivity and risk analysis are recommended. This topic is discussed in Chapter 2.

1.7 BASIS FOR THE DESIGN CATALOGUE

The pavement designs incorporated into the fourth edition of Road Note 31 are based primarily on: (a) The results of full-scale experiments where all factors affecting performance have been accurately measured and their variability quantified. (b) Studies of the performance of as-built existing road networks. Where direct empirical evidence is lacking, designs have been interpolated or extrapolated from empirical studies using road performance models (Parsley and Robinson (1982), Paterson (1987), Rolt et al (1987)) and standard analytical, mechanistic methods e.g Gerritsen and Koole (1987), Powell et al (1984), Brunton et al (1987). In view of the statistical nature of pavement design caused by the large uncertainties in traffic forecasting and the variability in material properties, climate and road behaviour, the design charts have been presented as a

catalogue of structures, each structure being applicable over a small range of traffic and subgrade strength. Such a procedure makes the charts extremely easy to use but it is important that the reader is thoroughly conversant with the notes applicable to each chart Throughout the text the component layers of a flexible pavement are referred to in the following terms (see Figure 1). Surfacing. This is the uppermost layer of the pavement and will normally consist of a bituminous surface dressing or a layer of premixed bituminous material. Where premixed materials are laid in two layers, these are known as the wearing course and the basecourse (or binder course) as shown in Figure 1. Roadbase. This is the main load-spreading layer of the pavement. It will normally consist of crushed stone or gravel, or of gravelly soils, decomposed rock, sands and sand-clays stabilised with cement, lime or bitumen. Sub-base. This is the secondary load-spreading layer underlying the roadbase. It will normally consist of a material of lower quality than that used in the roadbase such as unprocessed natural gravel, gravel-sand, or gravel-sand-clay. This layer also serves as a separating layer preventing contamination of the roadbase by the subgrade material and, under wet conditions, it has an important role to play in protecting the subgrade from damage by construction traffic. Capping layer (selected or improved subgrade). Where very weak soils are encountered, a capping layer is sometimes necessary. This may consist of better quality subgrade material imported from elsewhere or existing subgrade material improved by mechanical or chemical stabilisation. Subgrade. This is the upper layer of the natural soil which may be undisturbed local material or may be soil excavated elsewhere and placed as fill. In either case it is compacted during construction to give added strength.

This Note considers each of these steps in turn and puts special emphasis on five aspects of design that are of major significance in designing roads in most tropical countries: The influence of tropical climates on moisture conditions in road subgrades. The severe conditions imposed on exposed bituminous surfacing materials by tropical climates and the implications of this for the design of such surfacings. The interrelationship between design and maintenance. If an appropriate level of maintenance cannot be assumed. it is not possible to produce designs that will carry the anticipated traffic loading without high costs to vehicle operators through increased road deterioration. The high axle loads and tyre pressures which are common in most countries. The influence of tropical climates on the nature of the soils and rocks used in road building.

The overall process of designing a road is illustrated in Figure 2. Some of the information necessary to carry out the tasks may be available from elsewhere e.g a feasibility study or Ministry records, but all existing data will need to be checked carefully to ensure that it is both up-to-date and accurate. Likely problem areas are highlighted in the relevant chapters of this Note.

1.8 THE DESIGN PROCESS

There are three main steps to be followed in designing a new road pavement These are: (i) estimating the amount of traffic and the cumulative number of equivalent standard axles that will use the road over the selected design life; assessing the strength of the subgrade soil over which the road is to be built;

(ii)

(iii) selecting the most economical combination of pavement materials and layer thicknesses that will provide satisfactory service over the design life of the pavement (It is usually necessary to assume that an appropriate level of maintenance is also carried out).

Fig. 1 Definition of pavement layers

Fig. 2 The pavement design process

2.

TRAFFIC

2.2

ESTIMATING TRAFFIC FLOWS

The deterioration of paved roads caused by traffic results from both the magnitude of the individual wheel loads and the number of times these loads are applied. For pavement design purposes it is necessary to consider not only the total number of vehicles that will use the road but also the wheel loads (or, for convenience, the axle loads) of these vehicles. The loads imposed by private cars do not contribute significantly to the structural damage. For the purposes of structural design, cars and similar sized vehicles can be ignored and only the total number and the axle loading of the heavy vehicles that will use the road during its design life need to be considered In this context, heavy vehicles are defined as those having an unladen weight of 3000 kg or more. In some circumstances, particularly for lightly trafficked roads, construction traffic can be a significant component of overall traffic loading and the designs should take this into account.

2.2.1 Baseline traffic flows

In order to determine the total traffic over the design life of the road, the first step is to estimate baseline traffic flows. The estimate should be the (Annual) Average Daily Traffic (ADT) currently using the route, classified into the vehicle categories of cars, light goods vehicles, trucks (heavy goods vehicles) and buses. The ADT is defined as the total annual traffic summed for both directions and divided by 365. It is usually obtained by recording actual traffic flows over a shorter period from which the ADT is then estimated. For long projects, large differences in traffic along the road may make it necessary to estimate the flow at several locations.. It should be noted that for structural design purposes the traffic loading in one direction is required and for this reason care is always required when interpreting ADT figures. Traffic counts carried out over a short period as a basis for estimating the traffic flow can produce estimates which are subject to large errors because traffic flows can have large daily, weekly, monthly and seasonal variations (Howe (1972)). The daily variability in traffic flow depends on the volume of traffic. It increases as traffic levels fall, with high variability on roads carrying less than 1000 vehicles per day. Traffic flows vary more from day-to-day than from week-to-week over the year. Thus there are large errors associated with estimating average daily traffic flows (and subsequently annual traffic flows) from traffic counts of only a few days duration, or excluding the weekend. For the same reason there is a rapid decrease in the likely error as the duration of the counting period increases up to one week. For counts of longer duration, improvements in accuracy are less pronounced. Traffic flows also vary from month-to-month so that a weekly traffic count repeated at intervals during the year provides a better base for estimating the annual volume of traffic than a continuous traffic count of the same duration Traffic also varies considerably through a 24-hour period and this needs to be taken into account explicitly as outlined below. In order to reduce error, it is recommended that traffic counts to establish ADT at a specific site conform to the following practice: (i) (ii) The counts are for seven consecutive days. The counts on some of the days are for a full 24 hours, with preferably at least one 24-hour count on a weekday and one during a weekend. On the other days 16-hour counts should be sufficient. These should be grossed up to 24-hour values in the same proportion as the 16hour/24-hour split on those days when full 24-hour counts have been undertaken.

2.1 DESIGN LIFE

For most road projects an economic analysis period of between 10 and 20 years from the date of opening is appropriate, but for major projects this period should be tested as part of the appraisal process (Overseas Road Note 5, Transport and Road Research Laboratory (1988)). Whatever time period is chosen for the appraisal of a project, the road will always have some residual value at the end of this period. Choosing a pavement design life that is the same as the analysis period simplifies the economic appraisal by minimising the residual value, which is normally difficult to estimate accurately. A pavement design life of 15 years also reduces the problem of forecasting uncertain traffic trends for long periods into the future. In this context, design life does not mean that at the end of the period the pavement will be completely worn out and in need of reconstruction; it means that towards the end of the period the pavement will need to be strengthened so that it can continue to carry traffic satisfactorily for a further period. Condition surveys of bituminous pavements should be carried out about once a year as part of the inspection procedures for maintenance. These are used to determine not only the maintenance requirements but also the nature and rate of change of condition to help to identify if and when the pavement is likely to need strengthening. Stage construction consists of planned improvements to the pavement structure at fixed times through the project life. From a purely economic point of view, stage construction policies have much to commend them. However, experience has shown that budget constraints have often prevented the planned upgrading phases of stage construction projects from taking place, with the result that much of the benefit from such projects has been lost in general, stage construction policies are not recommended if there is any risk that maintenance and upgrading will not be carved out correctly or at the appropriate time.

(iii) Counts are avoided at times when travel activity is abnormal for short periods due to the payment of

wages and salaries, public holidays, etc If abnormal traffic flows persist for extended periods, for example during harvest times, additional counts need to be made to ensure this traffic is properly included. (iv) If possible, the seven-day counts should be repeated several times throughout the year.

Country-wide traffic data should be collected on a systematic basis to enable seasonal trends in traffic flows to be quantified. Unfortunately, many of the counts that are available are unreliable, especially if they have been carried out manually. Therefore, where seasonal adjustment factors are applied to traffic survey data in order to improve the accuracy of baseline traffic figures, the quality of the statistics on which they are based should be checked in the field. Classified traffic counts are normally obtained by counting manually. These counts can be supplemented by automatic counters which use either a pneumatic tube laid across the surface of the carriageway or a wire loop fixed to the carriageway surface or, preferably, buried just beneath it. Pneumatic tubes require regular maintenance and can be subject to vandalism. Buried loops are preferred, but installing a loop beneath the road surface is more difficult and more expensive than using a pneumatic tube. In their basic form, automatic counters do not distinguish between different types of vehicle and so cannot provide a classified count. Modern detector systems are now available which can perform classified vehicle counting, but such systems are expensive and not yet considered to be sufficiently robust for most developing country applications. An exception is the counter system developed specifically for developing countries by the Transport Research Laboratory.

Normal traffic. The commonest method of forecasting normal traffic is to extrapolate time series data on traffic levels and assume that growth will either remain constant in absolute terms i.e. a fixed number of vehicles per year (a linear extrapolation), or constant in relative terms i.e. a fixed percentage increase. Data on fuel sales can often be used as a guide to country-wide growth in traffic levels, although improvements in fuel economy over time should be taken into account. As a general rule it is only safe to extrapolate forward for as many years as reliable traffic data exist from the past, and for as many years as the same general economic conditions are expected to continue. As an alternative to time, growth can be related linearly to anticipated Gross Domestic Product (GDP). This is normally preferable since it explicitly takes into account changes in overall economic activity, but it has the disadvantage that a forecast of GDP is needed. The use of additional variables, such as population or fuel price, brings with it the same problem If GDP forecasts are not available, then future traffic growth should be based on time series data. If it is thought that a particular component of the traffic will grow at a different rate to the rest, it should be specifically identified and dealt with separately. For example, there may be a plan to expand a local town or open a local factory during the design life of the road, either of which could lead to different growth rates for different types of vehicle, or there may be a plan to allow larger freight vehicles on the road, in which case the growth rate for trucks may be relatively low because each truck is heavier. Whatever the forecasting procedure used, it is essential to consider the realism of forecast future levels. Few developing countries are likely to sustain the high rates of growth experienced in the past, even in the short term, and factors such as higher fuel costs and vehicle import restrictions could tend to depress future growth rates. Diverted traffic. Where parallel routes exist, traffic will usually travel on the quickest or cheapest route although this may not necessarily be the shortest. Thus, surfacing an existing road may divert traffic from a parallel and shorter route because higher speeds are possible on the surfaced road. Origin and destination surveys should be carried out to provide data on the traffic diversions likely to arise. Assignment of diverted traffic is normally done by an all-or-nothing method in which it is assumed that all vehicles that would save time or money by diverting would do so, and that vehicles that would lose time or increase costs would not transfer. With such a method it is important that all perceived costs are included. In some of the more developed countries there may be scope for modelling different scenarios using standard assignment computer programs. Diversion from other transport modes, such as rail or water, is not easy to forecast. Transport of bulk commodities will normally be by the cheapest mode, though this may not be the quickest. However, quality of service, speed and convenience are valued by intending

2.2.2 Traffic forecasting

Even with a developed economy and stable economic conditions, traffic forecasting is an uncertain process. In a developing economy the problem becomes more difficult because such economies are often very sensitive to the world prices of just one or two commodities. In order to forecast traffic growth it is necessary to separate traffic into the following three categories: (a) Normal traffic. Traffic which would pass along the existing road or track even if no new pavement were provided. (b) Diverted traffic. Traffic that changes from another route (or mode of transport) to the project road because of the improved pavement, but still travels between the same origin and destination. (c) Generated traffic. Additional traffic which occurs in response to the provision or improvement of the road.

consignors and, for general goods, diversion from other modes should not be estimated solely on the basis of door-todoor transport charges. Similarly, the choice of mode for passenger transport should not be judged purely on the basis of travel charges. The importance attached to quality of service by users has been a major contributory factor to the worldwide decline in rail transport over recent years. Diverted traffic is normally forecast to grow at the same rate as traffic on the road from which it diverted. Generated traffic. Generated traffic arises either because a journey becomes more attractive by virtue of a cost or time reduction or because of the increased development that is brought about by the road investment. Generated traffic is difficult to forecast accurately and can be easily overestimated It is only likely to be significant in those cases where the road investment brings about large reductions in transport costs. For example, in the case of a small improvement within an already developed highway system, generated traffic will be small and can normally be ignored. However, in the case of a new road allowing access to a hitherto undeveloped area, there could be large reductions in transport costs as a result of changing mode from, for example, animal-based transport to motor vehicle transport. In such a case, generated traffic could be the main component of future traffic flow. The recommended approach to forecasting generated traffic is to use demand relationships. The price elasticity of demand for transport is the responsiveness of traffic to a change in transport costs following a road investment. On inter-urban roads a distinction is normally drawn between passenger and freight traffic. On roads providing access to rural areas, a further distinction is usually made between agricultural and nonagricultural freight traffic. Evidence from several evaluation studies carried out in developing countries gives a range of between -0 6 to -2.0 for the price elasticity of demand for transport, with an average of about -1 0. This means that a one per cent decrease in transport costs leads to a one per cent increase in traffic. Calculations should be based on door-to-door travel costs estimated as a result of origin and destination surveys and not lust on that part of the trip incurred on the road under study. The available evidence suggests that the elasticity of demand for passenger travel is usually slightly greater than unity In general, the elasticity of demand for goods is much lower and depends on the proportion of transport costs in the commodity price.

(Highway Research Board (1962), Paterson (1987)). In order to determine the cumulative axle load damage that a pavement will sustain during its design life, it is necessary to express the total number of heavy vehicles that will use the road over this period in terms of the cumulative number of equivalent standard axles (esa). Axle load surveys must be carried out to determine the axle load distribution of a sample of the heavy vehicles using the road. Data collected from these surveys are used to calculate the mean number of equivalent standard axles for a typical vehicle in each class. These values are then used in conjunction with traffic forecasts to determine the predicted cumulative equivalent standard axles that the road will carry over its design life.

2.3.2 Axle load surveys

If no recent axle load data are available it is recommended that axle load surveys of heavy vehicles are undertaken whenever a major road project is being designed. Ideally, several surveys at periods which will reflect seasonal changes in the magnitude of axle loads are recommended. Portable vehicle-wheel weighing devices are available which enable a small team to weigh up to 90 vehicles per hour. Detailed guidance on carrying out axle load surveys and analysing the results is given in TRRL Road Note 40 (Transport and Road Research Laboratory (1978)). It is recommended that axle load surveys are carried out by weighing a sample of vehicles at the roadside. The sample should be chosen such that a maximum of about 60 vehicles per hour are weighed. The weighing site should be level and, if possible, constructed in such a way that vehicles are pulled clear of the road when being weighed. The portable weighbridge should be mounted in a small pit with its surface level with the surrounding area. This ensures that all of the wheels of the vehicle being weighed are level and eliminates the errors which can be introduced by even a small twist or tilt of the vehicle. More importantly, it also eliminates the large errors that can occur if all the wheels on one side of multiple axle groups are not kept in the same horizontal plane. The load distribution between axles in multiple axle groups is often uneven and therefore each axle must be weighed separately. The duration of the survey should be based on the same considerations as for traffic counting outlined in Section 2 2 1. On certain roads it may be necessary to consider whether the axle load distribution of the traffic travelling in one direction is the same as that of the traffic travelling in the opposite direction. Significant differences between the two streams can occur on roads serving docks, quarries, cement works, etc , where the vehicles travelling one way are heavily loaded but are empty on the return journey. In such cases the results from the more heavily trafficked lane should be used when converting commercial vehicle flows to the equivalent number of standard axles for pavement design. Similarly, special allowance must be made for unusual axle loads on roads which mainly serve one specific economic activity, since this can result in a particular vehicle type being predomi-

2.3
2.3.1

AXLE LOADING
Axle equivalency

The damage that vehicles do to a road depends very strongly on the axle loads of the vehicles. For pavement design purposes the damaging power of axles is related to a 'standard' axle of 8.16 tonnes using equivalence factors which have been derived from empirical studies

nant in the traffic spectrum This is often the case, for example, in timber extraction areas, mining areas and oil fields.

In order to determine the cumulative equivalent standard axles over the design life of the road, the following procedure should be followed: (i) Determine the daily traffic flow for each class of vehicle weighed using the results of the traffic survey and any other recent traffic count information that is available. Determine the average daily one-directional traffic flow for each class of vehicle.

2.3.3 Determination of cumulative equivalent standard axles

Computer programs have been written to assist with the analysis of the results from axle load surveys. These programs provide a detailed tabulation of the survey results and determine the mean equivalence factors for each vehicle type if required. If such a program is not available, standard spreadsheet programs can be used. If there are no computer facilities available the following method of analysis is recommended. The equivalence factors for each of the wheel loads measured during the axle load survey are determined using Table 2.1 or the accompanying equation to obtain the equivalence factors for vehicle axles. The factors for the axles are totalled to give the equivalence factor for each of the vehicles. For vehicles with multiple axles i e. tandems, triples etc., each axle in the multiple group is considered separately. The mean equivalence factor for each type or class of vehicle travelling in each direction must then be determined. Vehicle classes are usually defined by the number and type of axles. Note that this method of determining the mean equivalence factors must always be used; calculating the equivalence factor for the average axle load is incorrect and leads to large errors.

(ii)

(iii) Make a forecast of the one-directional traffic flow for each class of vehicle to determine the total traffic in each class that will travel over each lane during the design life (see Section 2.2.2). (iv) Determine the mean equivalence factor of each class of vehicle and for each direction from the results of this axle load survey and any other surveys that have recently been carried out. (v) The products of the cumulative one-directional traffic flows for each class of vehicle over the design life of the road and the mean equivalence factor for that class should then be calculated and added together to give the cumulative equivalent standard axle loading for each direction. The higher of the two directional values should be used for design.

TABLE 2.1

In most countries the axle load distribution of the total population of heavy vehicles using the road system remains roughly constant from year to year although there may be longterm trends resulting from the introduction of new types of vehicles or changes in vehicle regulations and their enforcement. It is therefore customary to assume that the axle load distribution of the heavy vehicles will remain unchanged for the design life of the pavement and that it can be determined by undertaking surveys of vehicle axle loads on existing roads of the same type and which serve the same function. In most developing countries the probable errors in these assumptions for a design life of 15 years are unlikely to result in a significant error in design. On dual carriageway roads and on single carriageway roads with more than two lanes, it should be assumed that the slow traffic lanes will carry all the heavy vehicles unless local experience indicates otherwise or the traffic flow exceeds about 2000 heavy vehicles per day in each direction. In the latter case, a proportion of heavy vehicles should be assigned to the slow lane according to the principles outlined in Overseas Road Note No. 6 (Transport and Road Research Laboratory (1988)). The design thickness required for the slow lane is usually applied to the whole carriageway width but there may be situations where a tapered roadbase or sub-base is appropriate. In some countries, single-lane bituminous roads are built to economise on construction costs. On such roads the traffic tends to be more channelised than on two-lane roads. The effective traffic loading in the wheelpath in one direction has been shown to be twice that for a wider road. Therefore, taking into account the traffic in both directions, the pavement thickness for these roads should be based on four times the total number of heavy vehicles that travel in one direction.

carried out to ensure that the choice of traffic class is appropriate. Formal risk analysis can also be used to evaluate the design choices as described briefly and referenced in Overseas Road Note 5 (Transport and Road Research Laboratory (1988)).

TABLE 2.2

2.4 ACCURACY
All survey data are subject to errors. Traffic data, in particular, can be very inaccurate and predictions about traffic growth are also prone to large errors. Accurate calculations of cumulative traffic are therefore very difficult to make. To minimise these errors there is no substitute for carrying out specific traffic surveys for each project for the durations suggested in Section 2.2.1. Additional errors are introduced in the calculation of cumulative standard axles because any small errors in measuring axle loads are amplified by the fourth power law relationship between the two. Fortunately, pavement thickness design is relatively insensitive to cumulative axle load and the method recommended in this. Note provides fixed structures for ranges of traffic as shown in Table 2.2. As long as the estimate of cumulative equivalent standard axles is close to the centre of one of the ranges, any errors are unlikely to affect the choice of pavement design However, if estimates of cumulative traffic are close to the boundaries of the traffic ranges then the basic traffic data and forecasts should be re-evaluated and sensitivity analyses

3. THE SUBGRADE
The type of subgrade sod is largely determined by the location of the road, but where the soils within the possible corridor for the road vary significantly in strength from place to place, it is clearly desirable to locate the pavement on the stronger soils if this does not conflict with other constraints. The strength of road subgrades is commonly assessed in terms of the California Bearing Ratio (CBR) and this is dependent on the type of soil, its density, and its moisture content. For designing the thickness of a road pavement, the strength of the subgrade should be taken as that of the soil at a moisture content equal to the wettest moisture condition likely to occur in the subgrade after the road is opened to traffic. In the tropics, subgrade moisture conditions under impermeable road pavements can be classified into three main categories: Category (1). Subgrades where the water table is sufficiently close to the ground surface to control the subgrade moisture content. The type of subgrade soil governs the depth below the road surface at which a water table becomes the dominant influence on the subgrade moisture content. For example, in non-plastic soils the water table will dominate the subgrade moisture content when it rises to within 1 m of the road surface, in sandy clays (PI<20 per cent) the water table will dominate when it rises to within 3m of the road surface, and in heavy clays (PI>40 per cent) the water table will dominate when it rises to within 7m of the road surface. In addition to areas where the water table is maintained by rainfall, this category includes coastal strips and flood plains where the water table is maintained by the sea, by a lake or by a river. Category (2). Subgrades with deep water tables and where rainfall is sufficient to produce significant changes in moisture conditions under the road. These conditions occur when rainfall exceeds evapotranspiration for at least two months of the year. The rainfall in such areas is usually greater than 250 mm per year and is often seasonal. Category (3). Subgrades in areas with no permanent water table near the ground surface and where the climate is dry throughout most of the year with an annual rainfall of 250 mm or less. Direct assessment of the likely strength or CBR of the subgrade soil is often difficult to make but its value can be inferred from an estimate of the density and equilibrium (or ultimate) moisture content of the subgrade together with knowledge of the relationship between strength, density and moisture content for the soil in question. This relationship must be determined in the

Laboratory. The density of the subgrade soil can be controlled within limits by compaction at a suitable moisture content at the time of construction. The moisture content of the subgrade soil is governed by the local climate and the depth of the water table below the road surface. In most circumstances, the first task is therefore to estimate the equilibrium moisture content as outlined in Section 3 1 below. A method of direct assessment of the subgrade strength, where this is possible, is discussed in Section 3 2 together with less precise methods of estimation which can be used if facilities for carrying out the full procedure are not available.

3.1

ESTIMATING THE SUBGRADE MOISTURE CONTENT

Category (7). The easiest method of estimating the design subgrade moisture content is to measure the moisture content in subgrades below existing pavements in similar situations at the time of the year when the water table is at its highest level. These pavements should be greater than 3m wide and more than two years old and samples should preferably be taken from under the carriageway about 0.5m from the edge. Allowance can be made for different soil types by virtue of the fact that the ratio of subgrade moisture content to plastic limit is the same for different subgrade soils when the water table and climatic conditions are similar. If there is no suitable road in the vicinity, the moisture content in the subgrade under an impermeable pavement can be estimated from a knowledge of the depth of the water table and the relationship between suction and moisture content for the subgrade soil (Russam and Croney (1960)). The test apparatus required for determining this relationship is straightforward and the method is described in Appendix B. Category (2). When the water table is not near the ground surface, the subgrade moisture condition under an impermeable pavement will depend on the balance between the water entering the subgrade through the shoulders and at the edges of the pavement during wet weather and the moisture leaving the ground by evapotranspiration during dry periods. Where the average annual rainfall is greater than 250 mm a year, the moisture condition for design purposes can be taken as the optimum moisture content given by the British Standard (Light) Compaction Test, 2.5 kg rammer method. When deciding on the depth of the water table in Category (1) or Category (2) subgrades, the possibility of the existence of local perched water tables should be borne in mind and the effects of seasonal flooding (where this occurs) should not be overlooked. Category (3). In regions where the climate is dry throughout most of the year (annual rainfall 250 mm or less), the moisture content of the subgrade under an impermeable pavement will be low. For design purposes a value of 80 per cent of the optimum moisture content obtained in the British Standard (Light) Compaction Test, 2.5 kg rammer method, should be used.

10

The methods of estimating the subgrade moisture content for design outlined above are based on the assumption that the road pavement is virtually impermeable. Dense bitumen-bound materials, stabilised soils with only very fine cracks, and crushed stone or gravel with more than 15 per cent of material finer than the 75 micron sieve are themselves impermeable (permeability less than 10-7 metres per second) and therefore subgrades under road pavements incorporating these materials are unlikely to be influenced by water infiltrating directly from above. However, if water, shed from the road surface or from elsewhere, is able to penetrate to the subgrade for any reason, the subgrade may become much wetter. In such cases the strength of subgrades with moisture conditions in Category (1) and Category (2) should be assessed on the basis of saturated CBR samples as described in Section 3.2. Subgrades with moisture conditions in Category (3) are unlikely to wet up significantly and the subgrade moisture content for design in such situations can be taken as the optimum moisture content given by the British Standard (Light) Compaction Test, 2 5 kg rammer method.

Alternatively, a more complete picture of the relationship between density, moisture content and CBR for the subgrade soil can be obtained by measuring the CBR of samples compacted at several moisture contents and at least two levels of compaction. The design CBR is then obtained by interpolation. This method is preferable since it enables an estimate to be made of the subgrade CBR at different densities and allows the effects of different levels of compaction control on the structural design to be calculated. Figure 3 shows a typical dry density/moisture content/CBR relationship for a sandy-clay soil that was obtained by compacting samples at five moisture contents to three levels of compaction: British Standard (Heavy) Compaction, 4.5 kg rammer method, British Standard (Light) Compaction, 2 5 kg rammer method, and an intermediate level of compaction. By interpolation, a design subgrade CBR of about 15 per cent is obtained if a relative density of 100 per cent of the maximum dry density obtained in the British Standard (Light) Compaction Test is specified and the subgrade moisture content was estimated to be 20 per cent. If saturated subgrade conditions are anticipated, the compacted samples for the CBR test should be saturated by immersion in water for four days before being tested. In all other cases when CBR is determined by direct measurement, the CBR samples should not be immersed since this results in over design. In areas where existing roads have been built on the same subgrade, direct measurements of the subgrade strengths can be made using a dynamic cone penetrometer (Appendix C). Except for direct measurements of CBR under existing pavements, in situ CBR measurements of subgrade soils are not recommended because of the difficulty of ensuring that the moisture and density conditions at the time of test are representative of those expected under the completed pavement. Whichever method is used to obtain the subgrade strength, each sample or each test will usually give different results and these can sometimes cover a considerable range. For design purposes it is important that the strength of the subgrade is not seriously underestimated for large areas of pavement or overestimated to such an extent that there is a risk of local failures. The best compromise for design purposes is to use the lower ten percentile value i.e. that value which is exceeded by 90 per cent of the readings. The simplest way to obtain this is to draw a cumulative frequency distribution of strength as shown in Figure 4. If the characteristics of the subgrade change significantly over sections of the route, different subgrade strength values for design should be calculated for each nominally uniform section. The structural catalogue requires that the subgrade strength for design is assigned to one of six strength classes reflecting the sensitivity of thickness design to subgrade strength. The classes are defined in Table 3 1. For subgrades with CBR's less than 2, special treatment is required which is not covered in this Road Note.

3.2

DETERMINING THE SUBGRADE STRENGTH

Having estimated the subgrade moisture content for design, it is then possible to determine the appropriate design CBR value at the specified density. It is recommended that the top 250 mm of all subgrades should be compacted during construction to a relative density of at least 100 per cent of the maximum dry density achieved in the British Standard (Light) Compaction Test, 2 5 kg rammer method, or at least 93 per cent of the maximum dry density achieved in the British Standard (Heavy) Compaction Test using the 4.5 kg rammer. With modern compaction plant a relative density of 95 per cent of the density obtained in the heavier compaction test should be achieved without difficulty but tighter control of the moisture content will be necessary. Compaction will not only improve the subgrade bearing strength but will reduce permeability and subsequent compaction by traffic. As a first step it is necessary to determine the compaction properties of the subgrade soil by carrying out standard laboratory compaction tests. Samples of the subgrade soil at the design subgrade moisture content can then be compacted in CBR moulds to the specified density and tested to determine the CBR values. With cohesionless sands, the rammer method tends to overestimate the optimum moisture content and underestimate the dry density achieved by normal field equipment. The vibrating hammer method is more appropriate for these materials. If samples of cohesive soils are compacted at moisture contents equal to or greater than the optimum moisture content, they should be left sealed for 24 hours before being tested so that excess pore water pressures induced during compaction are dissipated.

11

TABLE 3.1

If equipment for carrying out laboratory compaction and CBR tests is not available, a less precise estimate of the minimum subgrade strength class can be obtained from Table 3 2. This Table shows the estimated minimum strength class for five types of subgrade soil for various depths of water table, assuming that the subgrade is compacted to not less than 95 per cent of the maximum dry density attainable in the British Standard (Light) Compaction Test, 2.5 kg rammer method. The Table is appropriate for subgrade moisture Categories (1) and (2) but can be used for Category (3) if conservative strength estimates are acceptable. The design subgrade strength class together with the traffic class obtained in Chapter 2 are then used with the catalogue of structures to determine the pavement layer thicknesses (Chapter 10).

Fig.3 Dry density-moisture content-CBR relationships for sandy-clay soil

12

Fig.4 Distribution of subgrade strength

TABLE 3.2

13

4. EMBANKMENTS AND CUTTINGS

4.1
4.1.1

EMBANKMENTS
Introduction and survey

the presence on the slope of hollow bowl-shaped depressions with a steep head, curved in plan, which may represent the head area of old slips. They can be of any size, from a few tens of metres across to several hundred metres. Gullies that are active can put embankments at risk by bringing down debris, blocking the culvert and damming up against the embankment. An active gully carries a heavy load of material, typically of mixed sizes including sand and fines. Stable gullies generally contain only boulder and cobble sized material (the smaller sizes wash away), and may bear vegetation more than one year old in the gully floor. If the gully sides are being undercut by the stream and loose sediment moving in the gully floor, fresh debris will be brought into the gully, making the situation downstream worse. The slope below the road should be examined to ensure that it is not being undercut by the stream at the base. If this is happening, the whole embankment and road are at risk from slope failure expanding upwards from below. Evidence of slope instability is not easy to detect in trial pits because soils on steep slopes are often disturbed by slow creep under gravity, resulting in a jumbled soil profile. However, former slope movement is sometimes indicated by ancient organic horizons (buried soils) lying parallel to the present surface, or by clayey horizons lying parallel to the surface, that represent old sliding surfaces. Water often travels along these. The bedrock, too, can indicate a danger of movement. Rocks whose bedding lies parallel to the hillside, or dips out of the hillside, are prone to failure along the bedding plane, as are rocks containing joint surfaces (parallel planes of weakness) oriented this way. Weak, weathered and highly fractured rocks all represent a hazard, especially if the fissures are open, showing that the rock mass is dilating under tension. In steep side-sloping ground where the slope exceeds 1 in 6, it is normal practice to cut horizontal benches into the slope to simplify construction and to help key the embankment to the slope. At the same time, internal drainage is usually installed to remove sub-surface water from within the structure. Problems with embankments are fortunately rare but when they occur the consequences can be serious. It is therefore important that all potential problems are identified during the survey and recommendations made for more detailed investigations where necessary. Such investigations are expensive and need to be planned systematically, with additional testing and expert advice being commissioned only as required. An example of such an approach has been described by NITRR (1987a).

Embankments and cuttings will be required to obtain a satisfactory alignment on all but the lowest standard of road Embankments will be needed (i) to raise the road above flood water levels, (ii) in sidelong ground, (iii) across gullies and (iv) at the approaches to water crossings. High embankments impose a considerable load on the underlying soil and settlement should always be expected. Some transported soils are particularly susceptible, wind-blown sands and unconsolidated estuarine soils being particular examples. Organic material decays quickly above the water table and such soils are formed only in marshy areas where decay below the water level proceeds slowly. The residual soils that are widespread in the tropics are not usually compressible and any settlement which does occur is likely to be substantially complete by the time the embankment is constructed. An exception to this is the halloysitic clays developed from volcanic ash whose fragile structure can be broken down causing collapse under embankment loads. Potentially compressible soils should be identified at the survey stage which precedes new construction. During the survey it is also essential to look for evidence of water flow across the line of the road, either on the surface or at shallow depth. Temporary, perched water tables are common within residual soils and may not be readily apparent in the dry season. Drains must be installed to intercept ground water, and culverts of suitable size must be provided to allow water to cross the road alignment where necessary. It is also important to identify any areas of potential ground instability which might affect embankments. Particular care is required in gullies, which themselves may be indicative of weakness in the geological structure, but steep side-sloping ground may also be suspect and evidence of past soil movement should be sought. Evidence of past (dormant) instability is revealed by a range of slope features. On the surface, springs or patches of reeds or sedges are a sign that the slope may become saturated during the rainy season. Trees leaning at different angles (especially upslope) are a sign of disturbance by ground movement. However, it should be noted that trees leaning outwards (downslope), all at a similar angle, are usually not so much a sign of instability as a sign that the trees have grown at an angle to seek light. The age of trees can indicate former movement if they are all of a similar youthful age and there are no old trees present, this suggests that regeneration has taken place following a recent slide. The shape of the ground itself is a good indicator of past movement. The classic features of hummocky ground (irregular, pocketed surface), cracks and small ponds are signs of a deepseated landslide mass. Another sign is

4.1.2 Materials
Almost all types of soil, ranging from sandy clays through to broken rock, can be used for embankment construction, the main limitation being the ease with which the material can be handled and compacted. The embank-

14

ment material will usually be obtained from borrow areas adjacent to the road or hauled from nearby cuttings. Material of low plasticity is preferred because such material will create fewer problems in wet weather. With more plastic soils, greater care is necessary to keep the surfaces shaped and compacted so that rain water is shed quickly. If the embankment is higher than about 6 metres, it is desirable to reserve material of low plasticity for the lower layers.

pore water pressures do not reach unacceptably high levels during construction. Further precautions can be taken by installing inclinometers to detect any movement of soil which might indicate that unstable conditions exist. If necessary, consolidation can be accelerated by installing some form of vertical drainage. Sand drains consisting of columns of sand of about 500 mm diameter set at regular intervals over the area below the embankment have been used successfully but nowadays wick drains are more common. If the embankment is sufficiently stable immediately after construction, the rate of consolidation can be increased by the addition of a surcharge of additional material which is subsequently removed before the pavement is constructed. Organic soils. Organic soils are difficult to consolidate to a level where further settlement will not occur, and they provide a weak foundation even when consolidated. It is therefore best to avoid such materials altogether. If this is not possible, they should be removed and replaced If neither of these options is feasible, and provided soil suitable for embankments is available, methods of construction similar to those adopted for unconsolidated silt-clays should be used.

4.1.3

Design

Side slopes for high embankments should normally be between 1 in 1.5 and 1 in 2 (vertical: horizontal) Variations from this slope for local soils and climates are more reliably derived from local experience than from theoretical calculations. Slacker slopes are sometimes desirable for silty and clayey soils, especially in wet climates. In all cases it is important to protect the side slopes from the erosive action of rain and wind. Usually this should be done by establishing a suitable cover of vegetation (Howell et al (1991)) but granular materials will be needed in arid areas. Particular care is needed with expansive soils, especially those containing montmorillonite. If construction in such soils cannot be avoided, earthworks must be designed to minimise subsequent changes in moisture content and consequent volume changes. For example, the soil should be placed and compacted at a moisture content close to the estimated equilibrium value and it may also be advantageous to seal the road shoulders with a surface dressing. On low embankments in expansive soils, relatively shallow side slopes should be used i e 1 in 3, and these should be covered with well graded granular material. Nevertheless some volume changes must be expected with expansive soils and any cracks which develop, either in the side slopes or shoulders, should be sealed before water enters the structure. When the subgrade is a particularly expansive soil, it may be necessary to replace the expansive material with non-expansive impermeable soil to the depth affected by seasonal moisture changes.

4.1.5

Compaction of embankments

Uniformity of compaction is of prime importance in preventing uneven settlement. Although some settlement can be tolerated it is important that it is minimised, especially on the approaches to bridges and culverts where adequate compaction is essential. In the United Kingdom, compaction requirements are usually specified by means of a method specification which eliminates the need for in situ density tests (Department of Transport (1986)). In tropical countries it is more usual to use an endproduct specification. It is therefore essential that laboratory tests are carried out to determine the dry density/moisture content relationships for the soils to be used and to define the achievable densities. In the tropics the prevailing high temperatures promote the drying of soils. This can be beneficial with soils of high plasticity but, generally, greater care is necessary to keep the moisture content of the soil as close as possible to the optimum for compaction with the particular compaction plant in use. The upper 500 mm of soil immediately beneath the subbase or capping layer i e the top of the embankment fill or the natural subgrade, should be well compacted In practice this means that a minimum level of 93-95 per cent of the maximum dry density obtained in the British Standard (Heavy) Compaction Test, 4 5 kg rammer should be specified (a level of 98 per cent is usually specified for roadbases and sub-bases). The same density should also be specified for fill behind abutments to bridges and for the backfill behind culverts. For the lower layers of an embankment, a compaction level of 90-93 per cent of the maximum dry density obtained in the British Standard (Heavy) Compaction Test, 4.5 kg rammer, is suitable, or a level of 95100 per cent of the maximum density obtained in the lighter test using the 2.5

4.1.4

Construction over compressible soils

Transported soils. In the design of embankments over compressible soils, it is necessary to determine the amount of settlement which will occur and ensure that the rate of loading is sufficiently slow to prevent pore water pressures from exceeding values at which slip failures are likely to occur. A reasonably accurate estimate of total settlement can be obtained from consolidation tests as outlined by Lewis et al (1975) and TRRL (1976) but the theory usually overestimates the time required for settlement to occur. This is because most deposits of unconsolidated silt or clay soils contain horizontal lenses of permeable sandy sod which allow water to escape. High pore water pressures can be detected using piezometers set at different depths This often provides a reliable method of estimating the time required for consolidation and also provides a means of checking that

15

kg yammer. The British Standard Vibrating Hammer Test (BS 1377, Part 4 (1990)) should be used for non-cohesive soils and a level of 90-93 per cent of maximum density should be specified for the lower layers and 95 per cent for the upper layers. Compaction trials should always be carried out to determine the best way to achieve the specified density with the plant available (Parsons (1993)).

In and areas where water is either unavailable or expensive to haul, the dry compaction techniques developed by O'Connell et al (1987) and Ellis (1980) should be considered. Figure 5 illustrates that high densities can be achieved at low moisture contents using conventional compaction plant, and field trials have shown that embankments can be successfully constructed using these methods.

Fig.5 Dry density-moisture content relationships for a gravel-sand-clay

16

4.1.6

Site control

It is not easy to obtain an accurate measure of field density on site. The standard methods of measurement are tedious and not particularly reproducible Furthermore, most soils are intrinsically variable in their properties and it is difficult to carry out sufficient tests to define the density distribution. An acceptable approach to this problem is to make use of nuclear density and moisture gauges. Such devices are quicker and the results are more reproducible than traditional methods, but the instruments will usually need calibration for use with the materials in question if accurate absolute densities are required. It may also be advisable to measure the moisture contents using traditional methods but improvements in nuclear techniques are always being made and trials should be carried out for each situation. Additional site control can be exercised by making use of the Moisture Condition Test developed by Parsons (Parsons(1976), Parsons and Toombs (1987)). This test provides a rapid method of determining the moisture condition of soils and its use is described in BS 1377, Part 4 (1990).

of the slopes, the ground water regime, the type of material, the dimensions of the cut, and numerous other variables. A full investigation is therefore an expensive exercise but, fortunately, most cuttings are small and straightforward Investigations for the most difficult situations are best left to specialists and guidance on the need for this can be obtained in, for example, publications of the NITRR (1987b). Such guidance is defined by local experience and every opportunity should be taken to maintain a local data base. An integral component of a survey is to catalogue the performance of both natural and man-made slopes in the soils encountered along the length of the road and to identify the forms of failures to inform the design process and to make best use of the empirical evidence available in the area. These procedures have been described in detail by Anderson and Lloyd (1991) and considerable future advantages can be obtained if the field experience is encapsulated in the land classification procedures described by Lawrance et al (1993). Where well defined strata appear in the parent rock, it is best to locate the road over ground where the layers dip towards the hill and to avoid locating the road across hillsides where the strata are inclined in the same direction as the ground surface. During the survey, all water courses crossing the road line must be identified and the need for culverts and erosion control established.

4.2 CUTTINGS
Cuttings through sound rock can often stand at or near vertical, but in weathered rock or soil the conditions are more unstable. Instability is usually caused by an accumulation of water in the soil, and slips occur when this accumulation of water reduces the natural cohesion of the soil and increases its mass. Thus the design and construction of the road should always promote the rapid and safe movement of water from the area above the road to the area below, and under no circumstances should the road impede the flow of water or form a barrier to its movement.

4.2.3

Design and construction

4.2.1

Slope stability

Methods of analysing slope stability are usually based on measurements of the density, moisture content and strength of the soil together with calculations of the stresses in the soil using classic slip-circle analysis (Bishop (1955)). This type of analysis assumes that the soil mass is uniform. Sometimes failures do indeed follow the classic slip-circle pattern, but uniform conditions are rare, particularly in residual soils, and it is more common for slips to occur along planes of weakness in the vertical profile. Nevertheless, slope stability analysis remains an important tool in investigating the likely causes of slope failures and in determining remedial works, and such an analysis may be a necessary component of surveys to help design soil cuttings.

The angle of cutting faces will normally be defined at the survey stage. Benching of the cut faces can be a useful construction expedient enabling the cutting to be excavated in well defined stages and simplifying access for subsequent maintenance. The slope of the inclined face cannot usually be increased when benching is used and therefore the volume of earthworks is increased substantially. The bench itself can be inclined either outwards to shed water down the face of the cutting or towards the inside In the former, surface erosion may pose a problem In the latter, a paved drain will be necessary to prevent the concentration of surface water causing instability in the cutting. A similar problem applies to the use of cut-off drains at the top of the cutting which are designed to prevent run-off water from the area above the cutting from adding to the run-off problems on the cut slope itself. Unless such drains are lined and properly maintained to prevent water from entering the slope, they can be a source of weakness. Control of ground water in the cutting slopes is sometimes necessary. Various methods are available but most are expensive and complex, and need to be designed with care. It is advisable to carry out a proper ground water survey to investigate the quantity and location of sources of water and specialist advice is recommended.

4.2.2

Surveys

The construction of cuttings invariably disturbs the natural stability of the ground by the removal of lateral support and a change in the natural ground water conditions. The degree of instability will depend on the dip and stratification of the soils relative to the road alignment, the angle

17

As with embankments, it is essential that provision is made to disperse surface water from the formation at all stages of construction. Temporary formation levels should always be maintained at a slope to achieve this. Drainage is critically important because pore water pressures created by the available head of ground water in the side slopes can cause rapid distress in the pavement layers. Subsoil drains at the toe of the side slopes may be necessary to alleviate this problem. The subsequent performance, stability and maintenance of cuttings will depend on the measures introduced to alleviate the problems created by rainfall and ground water Invariably it is much more cost effective to install all the necessary elements at construction rather than to rely on remedial treatment later. Further guidance can be obtained from standard textbooks and reference books (Paige-Green (1981), Bell (1987)).

18

5. DRAINAGE AND ROAD SHOULDERS

5.1 THE DRAINAGE SYSTEM
One of the most important aspects of the design of a road is the provision made for protecting the road from surface water or ground water. If water is allowed to enter the structure of the road, the pavement will be weakened and it will be much more susceptible to damage by traffic. Water can enter the road as a result of rain penetrating the surface or as a result of the infiltration of ground water. The road surface must be constructed with a camber so that it sheds rainwater quickly and the top of the subgrade or improved subgrade must be raised above the level of the local water table to prevent it being soaked by ground water. A good road drainage system, which is properly maintained, is vital to the successful operation of a road and the road designs described in this Note are based on the assumption that the side drains and culverts associated with the road are properly designed and function correctly. Drainage within the pavement layers themselves is an essential element of structural design because the strength of the subgrade used for design purposes depends on the moisture content during the most likely adverse conditions. It is impossible to guarantee that road surfaces will remain waterproof throughout their lives, hence it is important to ensure that water is able to dram away quickly from within the pavement layers.

Damage. Shoulders should be differentiated from the carriageway e.g by the use of edge markings, different sized aggregate or different coloured aggregate. Crossfall is needed on all roads in order to assist the shedding of water into the side drains. A suitable value for paved roads is about 3 per cent for the carriageway, with a slope of about 4-6 per cent for the shoulders. An increased crossfall for the carriageway e g 4 per cent, is desirable if the quality of the final shaping of the road surface is likely to be low for any reason. There is evidence that there are also benefits to be obtained by applying steeper crossfalls to layers at successive depths in the pavement. The top of the sub-base should have a crossfall of 34 per cent and the top of the subgrade should be 4-5 per cent. These crossfalls not only improve the drainage performance of the various layers, but also provide a slightly greater thickness of material at the edge of the pavement where the structure is more vulnerable to damage. The design thickness should be that at the centre line of the pavement.

5.3 DRAINAGE OF LAYERS

Provided the crossfalls indicated above are adhered to and the bituminous surfacing and the shoulders are properly maintained, rainwater falling on the road will be shed harmlessly over the shoulders. When permeable roadbase materials are used (see Section 3 1), particular attention must be given to the drainage of this layer. Ideally, the roadbase and sub-base should extend right across the shoulders to the drainage ditches as shown in Fig 6. Under no circumstances should the `trench' type of cross-section be used in which the pavement layers are confined between continuous impervious shoulders. If it is too costly to extend the roadbase and sub-base material across the shoulder, drainage channels at 3m to 5m intervals should be cut through the shoulder to a depth of 50 mm below sub-base level. These channels should be back-filled with material of roadbase quality but which is more permeable than the roadbase itself, and should be given a fall of 1 in 10 to the side ditch. Alternatively a continuous drainage layer of pervious material of 75 mm to 100 mm thickness can be laid under the shoulder such that the bottom of the drainage layer is at the level of the top of the sub-base. This latter is by far the better of the two alternatives. In circumstances where the subgrade itself is permeable and can dram freely, it is preferable that vertical drainage is not impeded. If this can be done by ensuring that each layer of the pavement is more permeable than the layer above, then the additional drainage layer through the shoulders (layer No 7 in Fig. 6) is not required.

5.2 PAVEMENT CROSS-SECTION

The width of the carriageway and the overall geometric design of the road are dealt with in Overseas Road Note No 5 (Transport and Road Research Laboratory (1988)). For design traffic volumes in excess of about 1000 vehicles per day, carriageway widths of at least 7 metres should be used throughout and additional lanes will be needed when the capacity of a two-lane road is exceeded. Shoulders are an essential element of the structural design of a road, providing lateral support for the pavement layers. They are especially important when unbound materials are used in the pavement and for this type of construction it is recommended that shoulders should be at least 2 metres wide. For bound roadbases, shoulder width can be reduced if required If there is a large volume of non-motorised traffic, then shoulder width should be increased to a minimum of 3 metres. In order to exclude water from the road, the top of the shoulders should be impermeable and a surface dressing or other seal may need to be applied (see Chapter 9). Unsurfaced shoulders are not generally recommended because they often require considerable maintenance if satisfactory performance is to be guaranteed. Sealed shoulders also prevent the ingress of water at the edge of the pavement, which is an area particularly vulnerable to structural

5.4

SHOULDER MATERIALS

Although the ideal solution is to extend the roadbase and subbase outwards to form the shoulders, when the roadbase material is non-plastic it may lack sufficient cohesion to withstand the abrasive action of traffic unless

19

Fig.6 Cross section of road showing drainage arrangements

it is sealed with a surface dressing (see Section 9.2.1). In circumstances where extending the roadbase is not possible and the shoulder is not to be sealed, the shoulder material should be selected using the same principles as for a gravel-surfaced road or a sub-base to carry construction traffic. Thus the material should be strong enough to carry occasional vehicles and should be as cohesive as possible without being too weak when wet. The material will normally be of sub-base quality and the soaked CBR value at the specified density should exceed 30 per cent, except perhaps in and areas where the binding action of plastic fines may become a more important criterion (see Section 6 2.2, Tables 6.6 and 6 7 for general guidance).

It is also very desirable if at least the outer edge of the shoulder is able to support the growth of grasses which help to bind the surface and prevent erosion. On rural roads where the shoulders rarely need to carry traffic, excellent shoulder performance can be obtained if the whole of the shoulder is grassed In these circumstances it is necessary for this grass to be cut regularly to prevent the level of the shoulder building up above the level of the carriageway and thereby causing water to be retained at the carriageway-shoulder interface where it can penetrate the road structure and cause structural weakening.

20

6. UNBOUND PAVEMENT MATERIALS

This chapter gives guidance on the selection of unbound materials for use as roadbase, sub-base, capping and selected subgrade layers The main categories with a brief summary of their characteristics are shown in Table 6 1.

For lightly trafficked roads the requirements set out below may be too stringent and in such cases reference should be made to specific case studies, preferably for roads under similar conditions.

6.1.1

Crushed stone

6.1 ROADBASE MATERIALS

A wide range of materials can be used as unbound roadbases including crushed quarried rock, crushed and screened, mechanically stabilised, modified or naturally occurring `as dug' gravels. Their suitability for use depends primarily on the design traffic level of the pavement and climate but all roadbase materials must have a particle size distribution and particle shape which provide high mechanical stability and should contain sufficient fines (amount of material passing the 0.425 mm sieve) to produce a dense material when compacted. In circumstances where several types of roadbase are suitable, the final choice should take into account the expected level of future maintenance and the total costs over the expected life of the pavement. The use of locally available materials is encouraged, particularly at low traffic volumes (i.e. categories Ti and T2). Their use should be based on the results of performance studies and should incorporate any special design features which ensure their satisfactory performance. As a cautionary note, when considering the use of natural gravels a statistical approach should be applied in interpreting test results to ensure that their inherent variability is taken into account in the selection process.

Graded crushed stone (GB 1,A and GB1,B). Two types of material are defined in this category. One is produced by crushing fresh, quarried rock (GB1,A) and may be an all-in product, usually termed a `crusher-run', or alternatively the material may be separated by screening and recombined to produce a desired particle size distribution. The other is derived from crushing and screening natural granular material, rocks or boulders (GB1,B) and may contain a proportion of natural, fine aggregate. Typical grading limits for these materials are shown in Table 6 2. After crushing, the material should be angular in shape with a Flakiness Index (British Standard 812, Part 105 (1990)) of less than 35 per cent. If the amount of fine aggregate produced during the crushing operation is insufficient, nonplastic angular sand may be used to make up the deficiency. In constructing a crushed stone roadbase, the aim should be to achieve maximum impermeability compatible with good compaction and high stability under traffic. To ensure that the materials are sufficiently durable, they should satisfy the criteria given in Table 6.3. These are a minimum Ten Per Cent Fines Value (TFV) (British Standard 812, Part 111 (1990)) and limits on the maximum loss in strength following a period of 24 hours of soaking in water. The likely moisture conditions in the pavement are taken into account in broad terms based on climate. Other simpler tests e.g. the Aggregate Impact

TABLE 6.1

21

TABLE 6.2

TABLE 6.3

Test (British Standard 812, Part 112, 1990) may be used in quality control testing provided a relationship between the results of the chosen test and the TFV has been determined. Unique relationships do not exist between the results of the various tests but good correlations can be established for individual material types and these need to be determined locally. When dealing with materials originating from the weathering of basic igneous rocks the recommendations in Section 6.1 2 should be used. The fine fraction of a GB1,A material should be non-plastic. For GB1,B materials the maximum allowable PI is 6. When producing these materials, the percentage passing the 0.075 mm sieve should be chosen according to the grading and plasticity of the fines. For materials with non-plastic fines, the proportion passing the 0.075 mm sieve may approach 12 per cent If the PI approaches the upper limit of 6 it is desirable that the fines content be restricted to the lower end of the range. To

ensure this, a maximum Plasticity Product (PP) value of 45 is recommended where PP = PI x (percentage passing the 0.075 mm sieve) In order to meet these requirements it may be necessary to add a low proportion of hydrated lime or cement to alter the properties of the fines. Such materials are commonly referred to as modified materials Further details are given in Chapter 7. These materials may be dumped and spread by grader but it is preferable to use a payer to ensure that the completed surface is smooth with a tight finish. The material is usually kept wet during transport and laying to reduce the likelihood of particle segregation. The in situ dry density of the placed material should be a minimum of 98 per cent of the maximum dry density obtained in the British Standard (Heavy) Compaction Test, 4 5 kg rammer, or the British Standard Vibrating

22

Hammer Test (British Standard 1377, Part 4 (1990)). The compacted thickness of each layer should not exceed 200 mm. When properly constructed, crushed stone roadbases will have CBR values well in excess of 100 per cent. In these circumstances there is no need to carry out CBR tests. Dry-bound macadam (GB2,A). Dry-bound macadam is a traditional form of construction, formerly used extensively in the United Kingdom, and is comparable in performance with a graded crushed stone It has been used successfully in the tropics and is particularly applicable in areas where water is scarce or expensive to obtain It is also suitable where labourintensive construction is an economic option. The materials consist of nominal single-sized crushed stone and non-plastic fine aggregate (passing the 5.0 mm sieve). The fine material should preferably be well graded and consist of crushed rock fines or natural, angular pit sand. The dry-bound macadam process involves laying single-sized crushed stone of either 37.5 mm or 50 mm nominal size in a series of layers to achieve the design thickness. The compacted thickness of each layer should not exceed twice the nominal stone size. Each layer of coarse aggregate should be shaped and compacted and then the fine aggregate spread onto the surface and vibrated into the interstices to produce a dense layer. Any loose material remaining is brushed off and final compaction carried out, usually with a heavy smooth-wheeled roller. This sequence is then repeated until the design thickness is achieved. To aid the entry of the fines, the grading of the 37 5 mm nominal size stone should be towards the coarse end of the recommended range. Economy in the production process can be obtained if layers consisting of 50 mm nominal size stone and layers of 37 5 mm nominal size stone are both used to allow the required total thickness to be obtained more precisely and to make better overall use of the output from the crushing plant.

Water-bound macadam (GB2,B). Water-bound macadam is similar to dry-bound macadam. It consists of two components namely a relatively single-sized stone with a nominal maximum particle size of 50 mm or 37 5 mm and well graded fine aggregate which passes the 5.0 mm sieve. The coarse material is usually produced from quarrying fresh rock. The crushed stone is laid, shaped and compacted and then fines are added, rolled and washed into the surface to produce a dense material. Care is necessary in this operation to ensure that water sensitive plastic materials in the sub-base or subgrade do not become saturated. The compacted thickness of each layer should not exceed twice the maximum size of the stone The fine material should preferably be non-plastic and consist of crushed rock fines or natural, angular pit sand. Typical grading limits for the coarse fraction of GB2A or GB2B materials are given in Table 6 4. The grading of M2 and M4 correspond with nominal 50 mm and 37.5 mm single-sized roadstones (British Standard 63 (1987)) and are appropriate for use with mechanically crushed aggregate. M1 and M3 are broader specification M1 has been used for hand-broken stone but if suitable screens are available, M2, M3 and M4 are preferred. Aggregate hardness, durability, particle shape and in situ density should each conform to those given above for graded crushed stone.

6.1.2

Naturally occurring granular materials

Normal requirements for natural gravels and weathered rocks (GB3). A wide range of materials including latentic, calcareous and quartzitic gravels, river gravels and other transported gravels, or granular materials resulting from the weathering of rocks can be used successfully as roadbases. Table 6.5 contains three recommended particle size distributions for suitable materials corresponding to maximum nominal sizes of 37 5 mm, 20 mm and 10 mm Only the two larger sizes

TABLE 6.4

23

TABLE 6.5

should be considered for traffic in excess of 1 5 million equivalent standard axles. To ensure that the material has maximum mechanical stability, the particle size distribution should be approximately parallel with the grading envelope. To meet the requirements consistently, screening and crushing of the larger sizes may be required. The fraction coarser than 10 mm should consist of more than 40 per cent of particles with angular, irregular or crushed faces. The mixing of materials from different sources may be warranted in order to achieve the required grading and surface finish. This may involve adding fine or coarse materials or combinations of the two. All grading analyses should be done on materials that have been compacted. This is especially important if the aggregate fraction is susceptible to breakdown under compaction and in service. For materials whose stability decreases with breakdown, aggregate hardness criteria based on a minimum soaked Ten Per Cent Fines Value of 50kN or a maximum soaked Modified Aggregate Impact Value of 40 may be specified (British Standard 812, Part 112 (1990)). The fines of these materials should preferably be non-plastic but should normally never exceed a PI of 6 As an alternative to specifying PI, a Linear Shrinkage not exceeding 3 may be specified. If the PI approaches the upper limit of 6 it is desirable that the fines content be restricted to the lower end of the range. To ensure this, a maximum PP of 60 is recommended or alternatively a maximum Plasticity Modulus (PM) of 90 where PM = PI x (percentage passing the 0.425 mm sieve) If difficulties are encountered in meeting the plasticity criteria consideration should be given to modifying the material by the addition of a low percentage of hydrated lime or cement.

When used as a roadbase, the material should be compacted to a density equal to or greater than 98 per cent of the maximum dry density achieved in the British Standard (Heavy) Compaction Test, 4.5 kg rammer. When compacted to this density in the laboratory, the material should have a minimum CBR of 80 per cent after four days immersion in water (British Standard 1377, Part 4 (1990)). Arid and semi-arid areas. In low rainfall areas in the tropics, typically with a mean annual rainfall of less than 500 mm, and where evaporation is high, moisture conditions beneath a well sealed surface are unlikely to rise above the optimum moisture content determined in the British Standard (Heavy) Compaction Test. In such conditions, high strengths (CBR>80 per cent) are likely to develop even when natural gravels containing a substantial amount of plastic fines are used. In these situations, for the lowest traffic categories (TI, T2) the maximum allowable PI can be increased to 12 and the minimum soaked CBR criterion reduced to 60 per cent at the expected field density. Materials of basic igneous origin. Materials in this group are sometimes weathered and may release additional plastic fines during construction or in service. Problems are likely to worsen if water gains entry into the pavement and this can lead to rapid and premature failure. The state of decomposition also affects their long term durability when stabilised with lime or cement. The group includes common rocks such as basalts and dolerites but also covers a wider variety of rocks and granular materials derived from their weathering, transportation or other alteration (British Standards Institution (1975) and Weinert (1980)). Normal aggregate tests are often unable to identify unsuitable materials in this group. Even large, apparently sound particles may contain minerals that are decomposed and potentially expansive. The release of these minerals may lead to a consequent loss in bearing capacity. There are several methods of identifying unsound aggregates. These include

24

petrographic analysis to detect secondary (clay) minerals, the use of various chemical soundness tests e.g. sodium or magnesium sulphate (British Standard 812 Part 121 (1990)), the use of dye adsorption tests (Sameshima and Black (1979)) or the use of a modified Texas Ball Mill Test (Sampson and Netterberg (1989)). Indicative limits based on these tests include (i) a maximum secondary mineral content of 20 per cent, (ii) a maximum loss of 12 or 20 per cent after 5 cycles in the sodium or magnesium sulphate tests respectively (iii) a Clay Index of less than 3 and (iv) a Durability Mill Index of less than 90. In most cases it is advisable to seek expert advice when considering their use, especially when new deposits are being evaluated. It is also important to subject the material to a range of tests since no specific method can consistently identify problem materials. Materials of marginal quality. In many parts of the world, asdug gravels which do not normally meet the normal specifications for roadbases have been used successfully. They include latentic, calcareous and volcanic gravels. In general their use should be confined to the lower traffic categories (i.e. T1 and T2) unless local studies have shown that they have performed successfully at higher levels. Successful use often depends on specific design and construction features. It is not possible to give general guidance on the use of all such materials and the reader is advised to consult the appropriate source references (e g. CIRIA (1988), Lionfanga et al (1987), Netterberg and Pinard (1991), Newill et al (1987) and Rolt et al (1987)). The calcareous gravels, which include calcretes and marly limestones, deserve special mention. Typically, the plasticity requirements for these materials, all other things being equal, can be increased by up to 50 per cent above the normal requirements in the same climatic area without any detrimental effect on the performance of otherwise mechanically stable bases. Strict control of grading is also less important and deviation from a continuous grading is tolerable.

6.2.1

Bearing Capacity

A minimum CBR of 30 per cent is required at the highest anticipated moisture content when compacted to the specified field density, usually a minimum of 95 per cent of the maximum dry density achieved in the British Standard (Heavy) Compaction Test, 4.5 kg rammer. Under conditions of good drainage and when the water table is not near the ground surface (see Chapter 3) the field moisture content under a sealed pavement will be equal to or less than the optimum moisture content in the British Standard (Light) Compaction Test, 2.5 kg rammer. In such conditions, the sub-base material should be tested in the laboratory in an unsaturated state. Except in and areas (Category (3) in Chapter 3), if the roadbase allows water to drain into the lower layers, as may occur with unsealed shoulders and under conditions of poor surface maintenance where the roadbase is pervious (see Section 3 1), saturation of the sub-base is likely. In these circumstances the bearing capacity should be determined on samples soaked in water for a period of four days. The test should be conducted on samples prepared at the density and moisture content likely to be achieved in the field In order to achieve the required bearing capacity, and for uniform support to be provided to the upper pavement, limits on soil plasticity and particle size distribution may be required. Materials which meet the recommendations of Tables 6.6 and 6.7 will usually be found to have adequate bearing capacity.

6.2.2

Use as a construction platform

6.2 SUB-BASES (GS)

The sub-base is an important load spreading layer in the completed pavement. It enables traffic stresses to be reduced to acceptable levels in the subgrade, it acts as a working platform for the construction of the upper pavement layers and it acts as a separation layer between subgrade and roadbase. Under special circumstances it may also act as a filter or as a drainage layer. In wet climatic conditions, the most stringent requirements are dictated by the need to support construction traffic and paving equipment. In these circumstances the sub-base material needs to be more tightly specified In dry climatic conditions, in areas of good drainage, and where the road surface remains well sealed, unsaturated moisture conditions prevail and subbase specifications may be relaxed. The selection of sub-base materials will therefore depend on the design function of the layer and the anticipated moisture regime, both in service and at construction.

In many circumstances the requirements of a sub-base are governed by its ability to support construction traffic without excessive deformation or ravelling. A high quality sub-base is therefore required where loading or climatic conditions during construction are severe. Suitable material should possess properties similar to those of a good surfacing material for unpaved roads. The material should be well graded and have a plasticity index at the lower end of the appropriate range for an ideal unpaved road wearing course under the prevailing climatic conditions. These considerations form the basis of the criteria given in Tables 6.6 and 6.7. Material meeting the requirements for severe conditions will usually be of higher quality than the standard sub-base (GS). If materials to these requirements are unavailable, trafficking trials should be conducted to determine the performance of alternative materials under typical site conditions. In the construction of low-volume roads, where cost savings at construction are particularly important, local experience is often invaluable and a wider range of materials may often be found to be acceptable.

6.2.3

Sub-base as a filter or separating layer

This may be required to protect a drainage layer from blockage by a finer material or to prevent migration of fines and the mixing of two layers. The two functions are similar except that for use as a filter the material needs to be capable of allowing drainage to take place and therefore the amount of material passing the 0 075 mm sieve must be restricted.

25

TABLE 6.6

TABLE 6.7

6.3 SELECTED SUBGRADE MATERIALS AND CAPPING LAYERS (GC)

These materials are often required to provide sufficient cover on weak subgrades. They are used in the lower pavement layers as a substitute for a thick sub-base to reduce costs. The requirements are less strict than for sub-bases. A minimum CBR of 15 per cent is specified at the highest anticipated moisture content measured on samples compacted in the laboratory at the specified field density. This density is usually specified as a minimum of 95 per cent of the maximum dry density in the British Standard (Heavy) Compaction Test, 4 5 kg yammer In estimating the likely soil moisture conditions, the designer should take into account the functions of the overlying sub-base layer and its expected moisture condition and the moisture conditions in the subgrade. If either of these layers is likely to be saturated during the life of the road, then the selected layer should also be assessed in this state. Recommended gradings or plasticity criteria are not given for these materials. However, it is desirable to select reasonably homogeneous materials since overall pavement behaviour is often enhanced by this. The selection of materials which show the least change in bearing capacity from dry to wet is also beneficial.

The following criteria should be used to evaluate a sub-base as a separating or filter layer. a) The ratio D15(coarse layer) should be less than 5 D85(fine layer) where D15 is the sieve size through which 15 per cent by weight of the material passes and D85 is the sieve size through which 85 per cent passes. b) The ratio D50(coarse layer) should be less than 25 D50(fine layer) For a filter to possess the required drainage characteristics a further requirement is: c) The ratio Q15(coarse layer) should lie between 5 and 40 D15(fine layer) These criteria may be applied to the materials at both the road base/sub-base and the sub-base/subgrade interfaces. Further details can be obtained in the appropriate references e.g. (NAASRA (1983)).

26

7. CEMENT AND LIMESTABILISED MATERIALS

7.1 INTRODUCTION

Cracks can reflect through the surfacing and allow water to enter the pavement structure. If carbon dioxide has access to the material, the stabilisation reactions are reversible and the strength of the layers can decrease. The construction operations require more skill and control than for the equivalent unstabilised material.

This chapter gives guidance on the manufacture and use of cement and lime-stabilised materials in roadbase, sub-base, capping and selected fill layers of pavements. The stabilising process involves the addition of a stabilising agent to the soil, intimate mixing with sufficient water to achieve the optimum moisture content, compaction of the mixture, and final curing to ensure that the strength potential is realised. The subject has been reviewed by Sherwood (1993). Many natural materials can be stabilised to make them suitable for road pavements but this process is only economical when the cost of overcoming a deficiency in one material is less than the cost of importing another material which is satisfactory without stabilisation. Stabilisation can enhance the properties of road materials and pavement layers in the following ways: A substantial proportion of their strength is retained when they become saturated with water. Surface deflections are reduced. Resistance to erosion is increased. Materials in the supporting layer cannot contaminate the stabilised layer. The effective elastic moduli of granular layers constructed above stabilised layers are increased. Lime-stabilised material is suitable for use as a capping layer or working platform when the in situ material is excessively wet or weak and removal is not economical.

Methods of dealing with these problems are outlined in Section 7.7. The minimum acceptable strength of a stabilised material depends on its position in the pavement structure and the level of traffic. It must be sufficiently strong to resist traffic stresses but upper limits of strength are usually set to minimise the risk of reflection cracking. Three types of stabilised layer have been used in the structural design catalogue and the strengths required for each are defined in Table 7.1.

7.2

THE STABILISATION PROCESS

When lime is added to a cohesive soil, calcium ions replace sodium ions in the clay fraction until the soil becomes saturated with calcium and the pH rises to a value in excess of 12 (i a highly alkaline). The quantity of lime required to satisfy these reactions is determined by the initial consumption of lime test (ICL), (British Standard 1924 (1990)). The solubility of silica and alumina in the soil increase dramatically when the pH is greater than 12 and their reaction with lime can then proceed producing cementitious calcium silicates and aluminates. Amorphous silica reacts particularly well with lime. The cementitious compounds form a skeleton which holds the soil particles and aggregates together (NITRR (1986)). The primary hydration of cement forms calcium silicate and aluminate hydrates, releasing lime which reacts with sod components, as described above, to produce additional cementitious material.

Associated with these desirable qualities are several possible problems: Traffic, thermal and shrinkage stresses can cause stabilised layers to crack.

TABLE 7.1

27

7.3 SELECTION OF TYPE OF TREATMENT

The selection of the stabiliser is based on the plasticity and particle size distribution of the material to be treated. The appropriate stabiliser can be selected according to the criteria shown in Table 7 2 adapted from NAASRA (1986). Some control over the grading can be achieved by limiting the coefficient of uniformity to a minimum value of 5. The coefficient of uniformity is defined as the ratio of the sieve size through which 60 per cent of the material passes to the sieve size through which 10 per cent passes (D60/D10 in the nomenclature of Section 6 2 3). If the coefficient of uniformity lies below this value the cost of stabilisation will be high and the maintenance of cracks in the finished road could be expensive. Except for

materials containing amorphous silica e.g some sandstones and chert, material with low plasticity is usually best treated with cement. However, reactive silica in the form of pozzolans can be added to sods with low plasticity to make them suitable for stabilisation with lime. If the plasticity of the soil is high there are usually sufficient reactive clay minerals which can be readily stabilised with lime. Cement is more difficult to mix intimately with plastic materials but this problem can be alleviated by pretreating the soil with approximately 2 per cent of lime to make it more workable. If possible, the quality of the material to be stabilised should meet the minimum standards set out in Table 7.3. Stabilised layers constructed from these materials are more likely to perform satisfactorily even if they are affected by carbonation during their lifetime (Section 7.7 3). Materials which do not comply with Table 7 3 can

TABLE 7.2

TABLE 7.3

28

7.4.2
sometimes be stabilised but more additive will be required and the cost and the risk from cracking and carbonation will increase. Some aspects of construction must also be considered in selecting the stabiliser. It is not always possible to divert traffic during construction and the work must then be carried out in halfwidths. The rate of gain of strength in the pavement layer may sometimes need to be rapid so that traffic can be routed over the completed pavement as soon as possible Under these circumstances, cement stabilisation, with a faster curing period, is likely to be more suitable than lime stabilisation. Certain types of organic compounds in soils can affect the hydration of cement and inhibit the gain in strength. It is recommended that the effects of organic matter are assessed by strength tests as outlined below

Preparation of specimens

The optimum moisture content and the maximum dry density for mixtures of soil plus stabiliser are determined according to British Standard 1924 (1990) for additions of 2, 4, 6 and 8 per cent of cement. These specimens should be compacted as soon as the mixing is completed. Delays of the order of two hours occur in practice and changes taking place within the mixed material result in changes in their compaction characteristics. To determine the sensitivity of the stabilised materials to delays in compaction, another set of tests must be conducted after two hours have elapsed since the completion of mixing. Samples for the strength tests should also be mixed and left for two hours before being compacted into 150 mm cubes at 97 per cent of the maximum dry density obtained, after a similar two hour delay, in the British Standard (Heavy) Compaction Test, 4.5 kg rammer. These samples are then moist cured for 7 days and soaked for 7 days in accordance with BS 1924. Two methods of moist curing are described in the Standard. The preferred method is to seal the specimens in wax but if this is not possible they must be wrapped in cling film and sealed in plastic bags. The specimens should be maintained at 25C during the whole curing and soaking period. When the soaking phase is completed, the samples are crushed, their strengths measured, and an estimate made of the cement content needed to achieve the target strength. If suitable moulds are not available to produce cube specimens then 200 mm x 100 mm cylinders, 115.5 mm x 105 mm cylinders or 127 mm x 152 mm cylinders may be used and the results multiplied by the following correction factors to calculate equivalent cube strengths. Sample Type 200 mm x 100 mm diameter 115.5 mm x 105 mm diameter 127 mm x 152 mm diameter Correction Factor 1.25 1 04 0 96

7.4
7.4.1

CEMENT STABILISATION
Selection of cement content

The cement content determines whether the characteristics of the mixture are dominated by the properties of the original soil or by the hydration products. As the proportion of cement in the mixture increases, so the strength increases. Strength also increases with time During the first one or two days after construction this increase is rapid. Thereafter, the rate slows down although strength gain continues provided the layer is well cured. The choice of cement content depends on the strength required, the durability of the mixture, and the soundness of the aggregate The minimum cement content, expressed as a percentage of the dry weight of soil, should exceed the quantity consumed in the initial ion exchange reactions. Until research into the initial consumption of cement (ICC) is completed it is recommended that the percentage of cement added should be equal to or greater than the ICL. If there is any possibility that the material to be stabilised is unsound e.g. weathered basic igneous materials, then the Gravel ICL Test (NITRR (1984)) is preferred. In this test the aggregate is ground up to release any active clay minerals and the total sample tested The durability of the stabilised mixture which satisfies the strength requirements for the particular layer should also be assessed. Mixtures produced from sound materials complying with the minimum requirements of Table 7.3 can be assumed to be durable if they achieve the design strength. Mixtures produced from other materials should be checked using the wetdry brushing test (ASTM (1987)) which gives a good indication of the likelihood that a stabilised material will retain adequate strength during its service life in a pavement (Paige-Green et al (1990)). Additional stabiliser is normally incorporated to take account of the variability in mixing which occurs on site. If good control is exercised over the construction operations, an extra one per cent of stabiliser is satisfactory for this purpose.

As an alternative, the strength of stabilised sub-base material may be measured by the CBR test after 7 days of moist curing and 7 days of soaking. A minimum strength of 70 CBR is recommended. When the plasticity of the soil makes it difficult to pulverise and mix intimately with the cement, the workability can be improved by first pre-treating the sod with 2 to 3 per cent of lime, lightly compacting the mixture, and leaving it to stand for 24 hours. The material is then repulvensed and stabilised with cement. If this method is used, the laboratory design procedure is modified to include the pre-treatment phase before testing as described above.

29

7.5 LIME STABILISATION

7.5.1 Properties of lime-stabilised materials

and rapid increase in the strength and trafficabdity of the wet material. In many parts of the world, lime has been produced on a small scale for many hundreds of years to make mortars and lime washes for buildings. Different types of kilns have been used and most appear to be relatively effective. Trials have been carried out by TRRL in Ghana (Elks (1974)) to determine the output possible from small kilns and to assess the suitability of lime produced without commercial process control for soil stabilisation. Small batch kilns have subsequently been used to produce lime for stabilised layers on major road projects.

When lime is added to a plastic material, it first flocculates the clay and substantially reduces the plasticity index. This reduction of plasticity is time dependent during the initial weeks, and has the effect of increasing the optimum moisture content and decreasing the maximum dry density in compaction. The compaction characteristics are therefore constantly changing with time and delays in compaction cause reductions in density and consequential reductions in strength and durability. The workability of the soil also improves as the soil becomes more friable. If the amount of lime added exceeds the ICL, the stabilised material will generally be non-plastic or only slightly plastic. Both the ion exchange reaction and the production of cementitious materials increases the stability and reduces the volume change within the clay fraction It is not unusual for the swell to be reduced from 7 or 8 per cent to 0.1 per cent by the addition of lime. The ion exchange reaction occurs quickly and can increase the CBR of clayey materials by a factor of two or three. The production of cementitious materials can continue for ten years or more but the strength developed will be influenced by the materials and the environment. The elastic modulus behaves similarly to the strength and continues to increase for a number of years. Between the ages of one month and two to three years there can be a four-fold increase in the elastic modulus.

7.5.3

Selection of lime content

The procedure for selecting the lime content follows the steps used for selecting cement content and should, therefore, be carried out in accordance with British Standard 1924 (1990). The curing period for lime-stabilised materials is 21 days of moist cure followed by 7 days of soaking. In tropical and sub-tropical countries the temperature of the samples should be maintained at 25C which is near to the ambient temperature. Accelerated curing at higher temperatures is not recommended because the correlation with normal curing at temperatures near to the ambient temperature can differ from soil to soil. At high temperatures the reaction products formed by lime and the reactive silica in the soil can be completely different from those formed at ambient temperatures.

7.6 POZZOLANS
7.5.2 Types of lime
One of the primary sources of pozzolan is the pulvensed fuel ash (PFA) collected from the boilers of coal-fired electricity generating stations. PFA is usually mixed with lime in the proportions of 1 of lime to 3 or 4 of PFA but ratios of 1 to 2 up to 1 to 10 are used. The proportion depends on the reactivity of the particular fly ash which varies substantially from source to source. Lime and fly ash treated layers have a similar performance to cement treated layers constructed from the same aggregate material. The final mixtures should be chosen after a series of laboratory tests carried out after 21 days of moist cure and 7 days of soaking to determine the optimum ratio of lime to fly ash and the optimum lime content (expressed as a percentage of dry soil). In many tropical countries there are substantial quantities of bagasse (the fibrous residue from the crushing of sugar cane) and husks from rice. Both are rich in silica. When burnt, their ash contains a substantial amount of amorphous silica which reacts with lime (Cook and Suwanvitaya (1982), Mehta (1979)). Lime and rice-husk ash mixtures gain strength quickly during the early period of curing but little additional strength is obtained after 28 days of moist curing. The long-term strength depends on the stability of the calcium The most common form of commercial lime used in lime stabilisation is hydrated high calcium lime, Ca(OH)2, but monohydrated dolomitic lime, Ca(OH)2 MgO, calcitic quick lime, CaO, and dolomitic quicklime, CaO.MgO are .also used. For hydrated high calcium lime the majority of the free lime, which is defined as the calcium oxide and calcium hydroxide that is not combined with other constituents, should be present as calcium hydroxide. British Standard 890 requires a minimum free lime and magnesia content, (CaO + MgO), of 65 per cent. Quicklime has a much higher bulk density than hydrated lime and it can be produced in various aggregate sizes. It is less dusty than hydrated lime but the dust is much more dangerous and strict safety precautions are necessary when it is used. For quicklime, British Standard 890 requires a minimum free lime and magnesia content, (CaO + MgO), of 85 per cent. Quicklime is an excellent stabiliser if the material is very wet. When it comes into contact with the wet soil the quicklime absorbs a large amount of water as it hydrates. This process is exothermic and the heat produced acts as a further drying agent for the soil. The removal of water and the increase in plastic limit cause a substantial

30

silicate hydrates. Under certain conditions lime leaching can occur and eventually the strength will be reduced, but the presence of excess lime (free lime) can stabilise the calcium silicate hydrate. Mixtures of lime and rice-husk ash in the proportions 2:3 are the most stable and have the highest strength but the durability may be improved by increasing the lime content to give a 1:1 mixture.

material. If the layer is over stressed, shear planes will be formed near the top of the layer and premature failure along this plane is likely, particularly when the layer is only covered by a surface dressing. Multi-layer construction. When two or more lifts are required to construct a thick layer of stabilised material, care must be taken to prevent carbonation at the surface of the bottom lift. It is also important that the stabiliser is mixed to the full depth of each layer. A weak band of any type can cause over stressing and premature failure of the top lift followed by deterioration of the lower section. In general, the thickness of a lift should not be greater than 200 mm or less than 100 mm. Care should be taken to reduce the density gradient in the layer because permeable material in the lower part of the layer makes it more susceptible to carbonation from below If necessary, a layer should be compacted in two parts to make the bottom less permeable. The compaction operation should be completed within two hours and the length of road which is processed at any time should be adjusted to allow this to be achieved. Curing. Proper curing is very important for three reasons: It ensures that sufficient moisture is retained in the layer so that the stabiliser can continue to hydrate. It reduces shrinkage. It reduces the risk of carbonation from the top of the layer.

7.7 CONSTRUCTION
7.7.1 General methodology

The construction of stabilised layers follows the same procedure whether the stabilising agent is cement, lime or mixtures of limepozzolan. After the surface of the layer has been shaped, the stabiliser is spread and then mixed through the layer. Sufficient water is added to meet the compaction requirements and the material mixed again. The layer must be compacted as soon as possible, trimmed, re-rolled and then cured. The effect of each operation on the design and performance of the pavement is discussed below. Spreading the stabiliser. The stabiliser can be spread manually by `spotting' the bags at predetermined intervals, breaking the bags and then raking the stabiliser across the surface as uniformly as possible. Lime has a much lower bulk density than cement and it is possible to achieve a more uniform distribution with lime when stabilisers are spread manually. Alternatively, mechanical spreaders can be used to meter the required amount of stabiliser onto the surface. Mixing. Robust mixing equipment of suitable power for the pavement layer being processed is capable of pulverising the soil and blending it with the stabiliser and water. The most efficient of these machines carry out the operation in one pass, enabling the layer to be compacted quickly and minimising the loss of density and strength caused by any delay in compaction. Multi-pass machines are satisfactory provided the length of pavement being processed is not excessive and each section of pavement can be processed within an acceptable time. Graders have been used to mix stabilised materials but they are inefficient for pulverising cohesive materials and a considerable number of passes are needed before the quality of mixing is acceptable. They are therefore very slow and should only be considered for processing lime-stabilised layers because of the greater workability of lime-stabilised materials and the subsequent diffusion of lime through the soil aggregations (Stocker (1972)). Plant pre-mixing gives the possibility of better control than inplace spreading and mixing provided that the plant is close enough to the site to overcome possible problems caused by delays in delivery. This can often be justified by the lower safety margins on stabiliser content and target layer thicknesses that are possible. Compaction. A stabilised layer must be compacted as soon as possible after mixing has been completed in order that the full strength potential can be realised and the density can be achieved without over stressing the

In a hot and dry climate the need for good curing is very important but the prevention of moisture loss is difficult. If the surface is sprayed constantly and kept damp day and night, the moisture content in the main portion of the layer will remain stable but the operation is likely to leach stabiliser from the top portion of the layer. If the spraying operation is intermittent and the surface dries from time to time (a common occurrence when this method is used), the curing will be completely ineffective. Spraying can be a much more efficient curing system if a layer of sand, 30 to 40 mm thick, is first spread on top of the stabilised layer. If this is done the number of spraying cycles per day can be reduced and there is a considerable saving in the amount of water used. After seven days, the sand should be brushed off and the surface primed with a low viscosity cutback bitumen. An alternative method of curing is to first apply a very light spray of water followed by either a viscous cutback bitumen, such as MC 3000, or a slow setting emulsion. Neither of these will completely penetrate the surface of the stabilised layer and will leave a continuous bitumen film to act as a curing membrane. It is essential that all traffic is kept off the membrane for seven days. After this time, any excess bitumen can be absorbed by sanding the surface.

31

A prime coat cannot serve as a curing membrane. Research has shown that a prime penetrates too far into the layer and insufficient bitumen is retained on the surface to provide the necessary continuous film (Bofinger et al (1978)).

pacted with pneumatic-tyred rollers rather than vibrating types. Shrinkage problems in plastic gravels can be substantially reduced if air-dry gravel is used and the whole construction is completed within two hours, the water being added as late as possible during the mixing operation. It is generally not possible to use gravel in a completely air-dry condition, but the lower the initial moisture content and the quicker it is mixed and compacted, the smaller will be the subsequent shrinkage strains. Having accepted that some shrinkage cracks are inevitable in the stabilised layer, the most effective method of preventing these from reflecting through the bituminous surfacing is to cover the cemented layer with a substantial thickness of granular material. This is the design philosophy in Charts numbered 2, 4 and 6 in Chapter 10. When cemented material is used as a roadbase (Chart 8) a flexible surfacing such as a double surface dressing is recommended. Experience in a number of countries has shown that a further surface dressing applied after 2-3 years can partially or completely seal any subsequent cracking, particularly where lime is the stabilising agent.

7.7.2

Control of shrinkage and reflection cracks

There is no simple method of preventing shrinkage cracks occurring in stabilised layers. However, design and construction techniques can be adopted which go some way to alleviating the problem. Shrinkage, particularly in cement-stabilised materials, has been shown (Bofinger et al (1978)) to be influenced by Loss of water, particularly during the initial curing period. Cement content. Density of the compacted material. Method of compaction. Pre-treatment moisture content of the material to be stabilised.

7.7.3

Carbonation

Proper curing is essential not only for maintaining the hydration action but also to reduce volume changes within the layer. The longer the initial period of moist cure the smaller the shrinkage when the layer subsequently dries. When the layer eventually dries, the increased strength associated with a high stabiliser content will cause the shrinkage cracks to form at increased spacing and have substantial width. With lower cement contents, the shrinkage cracks occur at reduced spacing and the material will crack more readily under traffic because of its reduced strength. The probability of these finer cracks reflecting through the surfacing is reduced, but the stabilised layer itself will be both weaker and less durable. In order to maximise both the strength and durability of the pavement layer the material is generally compacted to the maximum density possible. However, for some stabilised materials it is sometimes difficult to achieve normal compaction standards and any increase in compactive effort to achieve them may have the adverse effect of causing shear planes in the surface of the layer or increasing the subsequent shrinkage of the material as its density is increased. If it proves difficult to achieve the target density, a higher stabiliser content should be considered in order that an adequately strong and durable layer can be produced at a lower density. Laboratory tests have shown that samples compacted by impact loading shrink considerably more than those compacted by static loading or by kneading compaction. Where reflection cracking is likely to be a problem, it is therefore recommended that the layer should be com-

If cement or lime-stabilised materials are exposed to air, the hydration products may react with carbon dioxide thereby reducing the strength of the material by an average of 40 per cent of the unconfined compressive strength (Paige-Green et al (1990)). This reaction is associated with a decrease in the pH of the material from more than 12 to about 8.5. The presence and depth of carbonation can be detected by testing the pH of the stabilised layer with phenolphthalein indicator and checking for the presence of carbonates with hydrochloric acid (Netterberg (1984)). A reasonable indication of whether the material being stabilised will be subject to serious carbonation can be obtained from the wet/dry test for durability (Paige-Green et al (1990)). Good curing practices, as outlined in Section 7.7.1, are the best means of preventing carbonation in roadbases. The risk of carbonation can be reduced by taking the following precautions: Avoid wet/dry cycles during the curing phase. Seal as soon as possible to exclude carbon dioxide. Compact as early as possible to increase the density and to reduce the permeability. Reduce the possibility of reflection cracks.

There may be some conflict between the last two points and care should be taken not to over compact the layer. Checks should be made during construction and if the depth of carbonated material is more than 2 to 3 mm the carbonated layer should be removed by heavy brushing or grading before the surfacing is applied.

32

7.8 QUALITY CONTROL

A high level of quality control is necessary in the manufacture of cement and lime-stabilised materials, as with all other materials used in the road pavement, but several factors need special consideration. Storage and handling of stabilisers. Unless cement and lime are properly stored and used in a fresh condition the quality of the pavement layer will be substantially reduced. Cement must be stored in a solid, watertight shed and the bags stacked as tightly as possible. Doors and windows should only be opened if absolutely necessary. The cement which is delivered from the manufacturer first should also be used first. Even if cement is properly stored the following losses in strength will occur:After 3 months After 6 months After 1 year After 2 years 20% reduction 30% reduction 40% reduction 50% reduction

Lime should be packed in sealed bags, tightly stacked and stored under cover or at least under a watertight tarpaulin. If it becomes contaminated or damp, it can only be used as a filler. Lime which is older than 6 months should be discarded. Distribution of stabiliser. After the layer has been properly processed, at least 20 samples should be taken for determination of the stabiliser content. The mixing efficiency is acceptable if the coefficient of variation is less than 30 per cent. Great care is necessary in multi-layer construction to ensure that good mixing extends to the full depth of all the layers. Opening to traffic. Insufficient research has been carried out to determine the precise effects of opening a road to traffic before the completion of the curing period but it is considered that allowing traffic on the pavement during the first two days can be beneficial for some stabilised layers provided the traffic does not mark the 'green' surface and all traffic is kept off the pavement from the end of the second day until one week has elapsed (Williams (1986)). Early trafficking has a similar effect to that of pre-cracking the layer by rolling within a day or two of its construction but rolling is preferred because it ensures even coverage of the full width of the carriageway. Layers which are pre-cracked or trafficked early must be allowed to develop sufficient strength to prevent abrasion of the edges of each crack before the layer is opened to general traffic. The slab strength of these layers is effectively destroyed and it is recommended that early trafficking is only acceptable for layers of cemented roadbase type CB2.

33

8. BITUMEN-BOUND MATERIALS
This chapter describes types of bituminous materials, commonly referred to as premixes, which are manufactured in asphalt mixing plants and laid hot. In situ mixing using either labour intensive techniques or mechanised plant can also be used for making roadbases for lower standard roads but these methods are not generally recommended and are not discussed in detail here.

High resistance to deformation. High resistance to fatigue and the ability to withstand high strains i.e they need to be flexible. Sufficient stiffness to reduce the stresses transmitted to the underlying pavement layers. High resistance to environmental degradation i e. good durability. Low permeability to prevent the ingress of water and air. Good workability to allow adequate compaction to be obtained during construction.

8.1 COMPONENTS OF A MIX

The coarse aggregates used for making premix should be produced by crushing sound, unweathered rock or natural gravel. The specifications for the aggregates are similar to those for granular roadbases. The aggregate must be clean and free of clay and organic material. To obtain good mechanical interlock and good compaction the particles should be angular and not flaky. Rough-textured material is preferable. Gravel should be crushed to produce at least two fractured faces on each particle. The aggregate must be strong enough to resist crushing during mixing and laying as well as in service. Aggregates which are exposed to traffic must also be resistant to abrasion and polishing. Highly absorptive aggregates are wasteful of bitumen and give rise to problems in mix design. They should be avoided where possible but if there is no choice, the absorption of bitumen must be taken into account in the mix design procedure. Hydrophillic aggregates which have a poor affinity for bitumen in the presence of water should also be avoided They may be acceptable only where protection from water can be guaranteed The fine aggregate can be crushed rock or natural sand and should also be clean and free from organic impurities. The filler (material passing the 0.075 mm sieve) can be crushed rock fines, Portland cement or hydrated lime. Portland cement or hydrated lime is often added to natural filler (1-2 per cent by mass of total mix) to assist the adhesion of the bitumen to the aggregate. Fresh hydrated lime can help reduce the rate of hardening of bitumen in surface dressings and may have a similar effect in premixes. Suitable specifications for the coarse and fine mineral components are given in Tables 8 1 and 8.2.

The requirements of a mix which will ensure each of these characteristics are often conflicting In temperate climates it has proved possible to design mixes which possess an acceptable balance of properties giving long service lives under a range of loading and climatic conditions. In the tropics, higher temperatures and high axle loads produce an environment which is more severe thereby making the mix requirements more critical and an overall balance of properties more difficult to obtain. High temperatures initially reduce the stiffness of mixes, making them more prone to deformation, and also cause the bitumen to oxidise and harden more rapidly, thereby reducing its durability. Unfortunately the requirements for improved durability i e. increased bitumen content and lower voids, usually conflict with the requirements for higher stiffness and improved deformation resistance. As a result, the tolerances on mix specifications need to be very narrow and a high level of quality control at all stages of manufacture is essential. The requirements are so critical for wearing course mixes that different mix designs are often necessary for different conditions on the same road. For example, mixes suitable for areas carrying heavy, slow-moving traffic, such as on climbing lanes, or areas where traffic is highly channelled, will be unsuitable for flat, open terrain where traffic moves more rapidly. A mix suitable for the latter is likely to deform on a climbing lane and a mix suitable for a climbing lane is likely to possess poor durability in flat terrain. In severe locations the use of bitumen modifiers is often advantageous (Hoban (1990), M. Hizam Harun and Jones (1992)). The age hardening of the bitumen in the wearing course is much greater at the exposed surface where the effect of the environment is much more severe and it is this hardened, brittle skin that usually cracks early in the life of the surfacing (Rolt et al (1986)). In areas where the diurnal temperature range is large, for example in most desert areas, thermal stresses can significantly increase the rate at which cracking occurs. The risk of premature cracking can be greatly reduced by applying a surface dressing to the wearing course soon after it has been laid, preferably after a few weeks of trafficking by construction traffic. This provides a bitumen-rich layer with a high strain tolerance at the point of potential weakness whilst also providing a good surface texture with

8.2 BITUMINOUS SURFACINGS

The most critical layer of the pavement is the bituminous surfacing, and the highest quality material is necessary for this layer. Where thick bituminous surfacings are required, they are normally constructed with a wearing course laid on a basecourse (sometimes called a binder course) which can be made to slightly less stringent specifications. To perform satisfactorily as road surfacings, bitumen aggregate mixes need to possess the following charactenstics:-

34

TABLE 8.1

improved skid resistant properties. If such a surface dressing is used, some cost savings can often be made by using a basecourse material in place of the wearing course. For severely loaded sites, such mixes can be designed to have a high resistance to deformation and under these conditions a surface dressing is essential if early cracking is to be prevented. It has also been shown (Smith et al (1990)) that 40/50, 60/70 and 80/100 penetration grade bitumens in the surface of wearing courses all tend to harden to a similar viscosity within a short time. It is therefore recommended that 60/70 pen bitumen is used to provide a suitable compromise between workability, deformation resistance and potential hardening in service. If possible, a bitumen should be selected which has a low temperature sensitiv-

ity and good resistance to hardening as indicated by the standard and extended forms of the Rolling Thin Film Oven Test (ASTM, D2872, Dickinson (1982)).

8.3

TYPES OF PREMIX IN COMMON USE

The main types of premix are asphaltic concrete, bitumen macadam and hot rolled asphalt. Each type can be used in surfacings or roadbases. Their general properties and specifications suitable for tropical environments are described below. A design procedure based on 'refusal density' is suggested to enhance the standard Marshall procedure (Section 8.3 5 and Appendix D).

35

TABLE 8.2

8.3.1

Asphaltic concrete

Asphaltic concrete (AC) is a dense, continuously graded mix which relies for its strength on both the interlock between aggregate particles and, to a lesser extent, on the properties of the bitumen and filler. The mix is designed to have low air voids and low permeability to provide good durability and good fatigue behaviour but this makes the material particularly sensitive to errors in proportioning, and mix tolerances are therefore very narrow (Jackson and Brien (1962), Asphalt Institute (1983), (1989) and (1991)). The particle size distributions for wearing course material given in Table 8.3 have produced workable mixes that have not generally suffered from deformation failures but they are not ideal for conditions of severe loading e.g. slow moving heavy traffic and high temperatures (see Section 8.3.5). This is because the continuous matrix of fine aggregate, filler and bitumen is more than sufficient to fill the voids in the coarse aggregate and this reduces the particle to particle contact within the coarse aggregate and lowers the resistance to deformation. A particle size distribution that conforms to the requirements for asphaltic concrete or a close graded bitumen macadam basecourse (BC1 in Table 8.3 or BC2 in Table 8.6) is recommended for use as the wearing course in severe conditions but such mixes must be sealed. It is common practice to design the mix using the Marshall Test and to select the design binder content by calculating the mean value of the binder contents for (i) maximum stability, (ii) maximum density, (iii) the mean

value for the specified range of void contents and (d) the mean value for the specified range of flow values. Compliance of properties at this design binder content with recommended Marshall criteria is then obtained (Table 8 4). A maximum air voids content of 5 per cent is recommended to reduce the potential age hardening of the bitumen but on severe sites the overriding criteria is that a minimum air voids of 3 per cent at refusal density should be achieved. This is equivalent to the condition which will arise after heavy trafficking and is designed to ensure that serious deformation does not occur. For such a mix it is unlikely that it will also be possible to reduce the air voids content at 98 per cent of Marshall density to 5 per cent and therefore it is recommended that a surface dressing is applied to the wearing course to provide the necessary protection against age hardening. It is frequently found that mixes are designed to have the highest possible stabilities. This usually means that the binder content is reduced resulting in mixes which are more difficult to compact and are less durable. It is important to note that there is a relatively poor correlation between Marshall stability and deformation in service, and durability should not be jeopardised in the belief that a more deformation resistant mix will be produced. A better method of selecting the Marshall design binder content is to examine the range of binder contents over which each property is satisfactory, define the common range over which all properties are acceptable, and then choose a design value near the centre of the common range. If this common range is too narrow, the aggregate

36

TABLE 8.3

TABLE 8.4

37

grading should be adjusted until the range is wider and tolerances less critical. To ensure that the compacted mineral aggregate in continuously graded mixes has a voids content large enough to contain sufficient bitumen, a minimum value of the voids in the mineral aggregate (VMA) is specified, as shown in Table 8.5.

with a surface dressing soon after laying is recommended for a long maintenance-free life. Slurry seals can also be used but they are best used in combination with a surface dressing to form a Cape seal. Close graded bitumen macadam mixes offer a good basis for the design of deformation resistant materials for severe sites and in these cases they should be designed on the basis of their refusal density. Recipe mixes are not recommended in these circumstances and the Marshall design criteria in Table 8 7 should be used. At the time of construction the air voids content is virtually certain to be in excess of five per cent and therefore a surface dressing should be placed soon after construction.

TABLE 8.5

8.3.3

Rolled asphalt

Rolled asphalt is a gap-graded mix which relies for its properties primarily on the mortar of bitumen, filler (<0 075 mm) and fine aggregate (0.075 - 2.36 mm). The coarse aggregate (>2.36 mm) acts as an extender but its influence on stability and density increases as the proportion of coarse aggregate in the mix increases above approximately 55 per cent. If the coarse aggregate content is less than about 40 per cent, pre-coated chippings should be rolled into the surface to provide texture for good skid resistance where necessary. Rolled asphalt has been developed in the United Kingdom to recipe specifications but can also be designed using the Marshall Test so that the physical characteristics of the fine aggregate can be taken into account (British Standard 594 (1985)). Wearing courses made to the particle size distributions in the British Standard and with filler-to-binder ratios in the range 0.8 - 1 0 have performed well in the tropics. The compositions of suitable mixes are summarised in Table 8 8. The mixes made with natural sand are more tolerant of proportioning errors than asphaltic concrete and are easier to compact. Although the air voids tend to be slightly higher than asphaltic concrete, they are discontinuous and the mixes are impermeable.

The Marshall design procedure is based on the assumption that the densities achieved in the Marshall Test samples represent those that will occur in the wheelpaths after a few years of trafficking. If in situ air voids are too high, rapid age hardening of the bitumen will ensue. Conversely, on severely loaded sites the air voids may be reduced by traffic leading to failure through plastic flow. In this latter situation the method of designing for a minimum air voids in the mix (VIM) at refusal density should be used (see Section 8 3.5).

8.3.2

Bitumen macadam

Close graded bitumen macadams (formerly called dense bitumen macadams or DBMs) are continuously graded mixes similar to asphaltic concretes but usually with a less dense aggregate structure. They have been developed in the United Kingdom (British Standard 4987 (1984)) from empirical studies over many years and are made to recipe specifications without reference to a formal design procedure. Implicit in the design is a knowledge of which materials perform satisfactorily in the United Kingdom under given climatic conditions and strictly controlled vehicle axle loads. Doubts about their suitability for different conditions and with different materials may therefore arise but, in practice, numerous materials including crushed gravels have been used successfully. The advantage of this method is that quality control testing is simplified and this should allow more intensive compliance testing to be performed. Aggregates which behave satisfactorily in asphaltic concrete will also be satisfactory in dense bitumen macadam. Suitable specifications for both wearing course and basecourse mixes are given in Table 8.6. Sealing the wearing course

8.3.4

Flexible bituminous surfacing

It is essential that the thin bituminous surfacings (50mm) recommended for structures described in Charts 3,4 and 7 of the structural catalogue are flexible. This is particularly important for surfacings laid on granular roadbases. Mixes which are designed to have good durability rather than high stability are flexible and are likely to have `sand' and bitumen contents at the higher end of the permitted ranges. In areas where the production of sand-sized material is expensive and where there is no choice but to use higher stability mixes, additional stiffening through the ageing and embrittlement of the bitumen must be prevented by applying a surface dressing.

8.3.5

Design to refusal density

Under severe loading conditions asphalt mixes must be expected to experience significant secondary compaction in the wheelpaths. Severe conditions cannot be precisely

38

TABLE 8.6

TABLE 8.7

39

TABLE 8.8

defined but will consist of a combination of two or more of the following; High maximum temperatures Very heavy axle loads Very channelled traffic Stopping or slow moving heavy vehicles

Details of the tests and their limitations are given in Appendix D.

8.4 BITUMINOUS ROADBASES

Satisfactory bituminous roadbases for use in tropical environments can be made using a variety of specifications. They need to possess properties similar to bituminous mix surfacings but whenever they are used in conjunction with such a surfacing the loading conditions are less severe, hence the mix requirements are less critical. Nevertheless, the temperatures of roadbases in the tropics are higher than in temperate climates and the mixes are therefore more prone to deformation in early life, and ageing and embattlement later.

Failure by plastic deformation in continuously graded mixes occurs very rapidly once the VIM are below 3 per cent therefore the aim of refusal density design is to ensure that at refusal there is still at least 3 per cent voids in the mix. For sites which do not fall into the severe category, the method can be used to ensure that the maximum binder content for good durability is obtained. This may be higher than the Marshall optimum but the requirements for resistance to deformation will be maintained. Where lower axle loads and higher vehicle speeds are involved, the minimum VIM at refusal can be reduced to 2 per cent. Refusal density can be determined by two methods; (a) Extended Marshall compaction (b) Compaction by vibrating hammer

8.4.1

Principal mix types

Particle size distributions and general specifications for continuously graded mixes are given in Table 8.9. No formal design method is generally available for determining the optimum composition for these materials because the maximum particle size and proportions of aggregate greater than 25 mm precludes the use of the Marshall Test. Suitable specifications for gap-graded rolled asphalt roadbases are given in Table 8.10. All these specifications are recipes which have been developed from experience and rely on performance data.

40

TABLE 8.9

TABLE 8.10

41

for their optimum adaptation to local conditions. The following principles should be adopted for all bituminous layers but are particularly important for recipe type specifications: (i) Trials for mix production, laying and compaction should be carried out to determine suitable mix proportions and procedures. Durable mixes require a high degree of compaction and this is best achieved by specifying density in terms of maximum theoretical density of the mix or, preferably, by using a modification of the Percentage Refusal Test with extended compaction time. (British Standard 598, Part 104 (1989), Powell and Leech (1982)).

TABLE 8.11

(ii)

(iii) Mixing times and temperatures should be set at the minimum required to achieve good coating of the aggregates and satisfactory compaction. (iv) The highest bitumen content commensurate with adequate stability should be used.

(1988) and MS-22 (1983), and the US Army Corps of Engineers (1991). It is normal practice to carry out preliminary design testing to determine the suitability of available aggregates and their most economical combination to produce a job-mix formula. The jobmix particle size distribution should be reasonably parallel to the specified grading envelope and is the target grading for the mix to be produced by the asphalt plant. Loss of fines may occur during the drying and heating phase and, therefore, tests on aggregates which have passed through the asphalt plant in the normal way should be used to establish a job-mix formula which meets the specified Marshall Test criteria. The importance of detailed compaction trials at the beginning of asphalt construction work cannot be over emphasised. During these trials, compaction procedures and compliance of the production-run asphalt with the job-mix formula should be established. Adjustments to the job-mix formula and, if necessary, redesign of the mix are carried out at this stage to ensure that the final job-mix satisfies the mix design requirements and can be consistently produced by the plant. Tolerances are specified for bitumen content and for the aggregate grading to allow for normal variation in plant production and sampling. Typical tolerances for single tests are given in Table 8 12. Good quality control is essential to obtain durable asphalt and the mean values for a series of tests should be very close to the job-mix formula which, in turn, should have a grading entirely within the specified envelope. Mixing must be accomplished at the lowest temperatures and in the shortest time that will produce a mix with complete coating of the aggregate and at a suitable temperature to ensure proper compaction. The ranges of acceptable mixing and rolling temperatures are shown in Table 8 13. Very little additional compaction is achieved at the minimum rolling temperatures shown in the Table and only pneumatic tyred rollers should be used at these temperatures. Rolled asphalts are relatively easy to compact but bitumen macadams and asphaltic concretes are relatively harsh and more compactrve effort is required. Heavy pneumatic tyred rollers are usually employed, the

8.4.2

Sand-bitumen mixes

For light and medium trafficked roads (defined as roads carrying less than 300 commercial vehicles per day and with mean equivalent standard axles per vehicle of 0 5 or less) and in areas lacking coarse aggregates, bitumen-stabilised sands are an alternative. Best results are achieved with well graded angular sands in which the proportion of material passing the 0 075 mm sieve does not exceed ten per cent and is non-plastic. The bitumen can range from a viscous cutback that will require heating to a more fluid cutback or emulsion that can be used at ambient temperatures. The most viscous cutbacks that can be properly mixed at ambient temperatures are RC or MC 800 or equivalents In general, the more viscous the bitumen the higher will be the stability of the mix. The use of penetration grade bitumens will produce the highest stabilities but this will necessitate heating the sand as well as the bitumen. An example has been given by Harris et al (1983). The amount of bitumen required will generally he between 3 and 6 per cent by weight of the dry sand, the higher proportions being required with the finer-grained materials. The Marshall Test can be used for determining the amount of bitumen required (Asphalt Institute, MS-2 (1988)) Design criteria are given in Table 8.11 for sand-bitumen mixes used as roadbase materials for tropical roads carrying medium to light traffic.

8.5 MANUFACTURE AND CONSTRUCTION

General guidance on the design, manufacture and testing of bitumen macadams and rolled asphalts can be found in the British Standards, BS 4987 (1988) for macadams and BS 594 (1985) and BS 598 (1985) for rolled asphalts. Similar guidance for asphalt concrete is given in the publications of the Asphalt Institute, SS-1 (1980), MS-2

42

TABLE 8.12

TABLE 8.13

kneading action of the tyres being important in orientating the particles. Vibratory compaction has been used successfully but care is needed in selecting the appropriate frequency and amplitude of vibration, and control of mix temperature is more critical than with pneumatic tyred rollers. Steel-wheeled deadweight rollers are relatively inefficient and give rise to a smooth surface with poor texture but are required to obtain satisfactory joints. Rolling usually begins near the shoulder and progresses towards the centre It is important that directional changes of the roller are made only on cool compacted mix and that each pass of the roller should be of slightly different length to avoid the formation of ridges. The number of joints to cold, completed edges should be minimised by using two pavers in echelon or a full-width paver to avoid cold joints between adjacent layers. If this

is not possible, repositioning of the paver from lane to lane at frequent intervals is another option. If a layer is allowed to cool before the adjacent layer is placed, then the Asphalt Institute method of joint formation is recommended. The edge of the first layer must be `rolled over' and thoroughly compacted. Before laying the second lane the cold joint should be broomed if necessary and tack coated. The paver screed should be set to overlap the first mat by a sufficient amount to allow the edge of the rolled over layer to be brought up to the correct level. Coarse aggregates in the material overlapping the cold joint should be carefully removed. The remaining fine material will allow a satisfactory joint to be constructed.

43

9. SURFACE TREATMENTS
9.1 PRIME AND TACK COATS

should take into account the type of existing road surface, the traffic, the available chippings and the climate.

9.2.1

Single and double surface dressings

A prime coat is a thin layer of bitumen sprayed onto the surface of an existing layer, usually of unbound or cement/lime bound material. Its purpose can be summarised as follows: It assists in promoting and maintaining adhesion between the roadbase and the bituminous surfacing by pre-coating the surface of the roadbase and by penetrating the voids near the surface. It helps to seal the surface pores in the roadbase, thus reducing the absorption of the first spray of bitumen of a surface dressing. It helps to bind the finer particles of aggregate together in the surface of the roadbase. If the application of the surfacing is delayed for some reason, it provides the roadbase with temporary protection against the detrimental effects of rainfall and light traffic.

Single surface dressings are normally adequate when applied to a bituminous layer. To be satisfactory for non-bituminous surfacings, the quality of a single seal must be very high and subsequent minor maintenance must be carried out promptly when required. It is recommended that double surface dressings are always used on non-bituminous layers. The quality of a double surface dressing will be greatly enhanced if traffic is allowed to run on the first dressing for a minimum period of 2-3 weeks (and preferably longer) before the second dressing is applied. This allows the chippings of the first dressing to adopt a stable interlocking mosaic which provides a firm foundation for the second dressing. If the trafficking results in the contamination of the first dressing with mud or soil, this should be thoroughly cleaned off before the second dressing is applied. Sand may sometimes be used as an alternative to chippings for the second dressing. Although this cannot contribute to the overall thickness of the surfacing, the combination of binder and sand provides a useful grouting medium for the chippings of the first seal and helps to hold them in place more firmly if they are poorly shaped. A slurry seal may also be used for the same purpose. A surface dressing applied to the shoulders of paved roads can be an effective method of improving drainage and reducing erosion and pavement edge damage. There are several factors which need to be considered, their importance being dependent on traffic intensity. The shoulder material must be suitable for surface dressing and be strong enough to support stationary, heavy wheel loads. Effective 'overbanding' is necessary to give a heavy duty transition seal between the shoulder and pavement surface. Shoulder maintenance by motor grader will not be possible. Age hardening of the shoulder seal will mean that maintenance reseals will be required at approximately five yearly intervals. Edge markings or the use of different coloured aggregate will be necessary to delineate the edge of the running surface. On uphill grades it may be necessary to use rumble strips on the shoulders to prevent them being used for overtaking.

Low viscosity, medium curing cutback bitumens such as MC-30, MC-70, or in rare circumstances MC-250, can be used for prime coats (alternatively low viscosity road tar can be used if this is available). The depth of penetration should be about 3-10 mm and the quantity sprayed should be such that the surface is dry within two days. The correct viscosity and application rate are dependent primarily on the texture and density of the surface being primed. The application rate is likely to lie within the range 0.3-1 1 kg/m2. Low viscosity cutbacks are necessary for very dense cement or lime-stabilised surfaces, and high viscosity cutbacks for untreated coarse-textured surfaces. It is .usually helpful to spray the surface lightly with water before applying the prime coat as this helps to suppress dust and allows the primer to spread more easily over the surface and to penetrate. Bitumen emulsions are not suitable for priming because they tend to form a skin on the surface. The primary function of a tack coat is to act as a glue to assist bonding of a new surface layer to a previously primed surface, bituminous roadbase, or basecourse that has been left exposed for some time. Tack coats should be extremely thin and it is appropriate to use a dilute bitumen emulsion spread to give less than 0.2 kg /M2 of residual bitumen with continuous cover. When temperature conditions are satisfactory, it is possible to obtain a thin layer by lightly spraying the undiluted emulsion with a handlance and then spreading it with a pneumatic tyred roller to obtain complete coverage.

9.2

SURFACE DRESSING

The design of surface dressing is described in detail in Overseas Road Note 3 (TRRL (1982)). The design

9.2.2 Type of surface

Embedment of the chippings under traffic is dependent upon the hardness of the layer to be sealed and the size of the chippings. Assessment of layer hardness can be

44

TABLE 9.1

based on descriptive definitions or measured using a simple penetration test probe. Details of surface category, penetration values, and descriptive definitions are given in Table 9.1. The probe penetration test is described in Appendix E.

TABLE 9.2

9.2.3

Traffic categories

The volume of traffic is considered in terms of the number of commercial vehicles per day in the lane under consideration. The traffic categories are defined in Table 9 2. It should be noted that these differ from the traffic classes used in the selection of the pavement structure in Chapter 10.

9.2.4

Chippings
In selecting the nominal size of chippings for double surface dressings, the size of chipping for the first layer should be selected on the basis of the hardness of the existing surface and the traffic category as indicated in Table 9 3. The nominal size of chipping selected for the

The chippings should comply in all respects with the requirements in British Standards 63, Part 2 (1987). The size of chippings should be chosen to suit the level of traffic and the hardness of the underlying surface as shown in Table 9.3.

TABLE 9.3

45

second layer should then be about half the nominal size of the first layer to promote good interlock between the layers a g. a 20 mm first layer should be followed by a 10 mm second layer, or a 14 mm first layer should be followed by a 10 mm or 6 mm second layer. In the case of a hard existing surface where little embedment of the first layer of chippings is possible, such as a newly constructed cement-stabilised roadbase or a dense crushed rock roadbase, a 'pad coat' of 6 mm chippings should be applied first followed by 10 mm or 14 mm chippings in the second layer. The first layer of small chippings will adhere well to the hard surface and will provide a 'key' for the larger stone of the second dressing. The quantity of chippings must be sufficient to cover the entire surface of the binder film after rolling. The rate at which chippings should be spread depends on their size, shape and specific gravity, but rates can be estimated using Figure 7. The least dimension of at least 200 chippings should be measured and the 'Average Least Dimension' (ALD) determined. An alternative method based on median particle size and Flakiness Index is described in Overseas Road Note 3 (TRRL 1982)). The ALD is then used in the Figure together with the line labelled AB and the approximate rate of application of chippings read from the upper scale. This rate should be used as a guide for supply purposes. The actual rate of spread should be adjusted on site when the spreading characteristics of the chippings have been observed.

road temperatures range from 20'C to 70C. The Figure indicates that the most appropriate binders are likely to be MC 3000 or the penetration grades up to 80/100. If the correct binder is not available it is sometimes possible to blend suitable materials on site (Hitch and Stewart (1987)). To determine the rate of application of binder, an appropriate factor should be selected from Table 9.4 for each of the four sets of conditions listed. The four factors are then added together to give the total weighting factor. The Average Least Dimension of the chippings and the total weighting factor obtained from the condition constants in Table 9.4 are then used with Figure 7 to obtain the rate of application of binder. Research in Kenya (Hitch, (1981)) has shown that the rate of spread of binder should be adjusted to take account of road gradient and traffic speed as indicated in the Figure. Finally the rate of spread of binder needs to be modified to allow for the different proportions of residual bitumen in the different binders. No adjustment is needed for MC3000 but for penetration grade bitumens the spray rate should be reduced by 10 per cent for 80/100 penetration grade and 5 per cent for 300 penetration grade. For emulsions it should be increased by the factor (90/ bitumen content of the emulsion in per cent).

9.3 SLURRY SEALS

Slurry seals are mixtures of fine aggregates, Portland cement filler, bitumen emulsions and additional water (ASTM D3910-84 (1990); BS 434, Parts 1 and 2, (1984)). When freshly mixed, they have a thick consistency and can be spread to a thickness of 5 to 10 mm. This method of surfacing is not normally used for new construction because it is more expensive than surface dressing, it does not provide good surface texture, and it is considerably less durable.

9.2.5

Binder

Figure 8 shows the viscosity/road temperature relationships for a wide range of binders. In the tropics, day-time

TABLE 9.4

46

Fig.7 Surface dressing design chart

47

Fig.8 Surface temperature and choice of binder for surface dressing

48

Slurry seals are often used in combination with a surface dressing to make a 'Cape seal'. In this technique the slurry seal is applied on top of a single surface dressing to produce a surface texture which is less harsh than a surface dressing alone and a surface which is flexible and durable. However, the combination is more expensive than a double surface dressing and requires careful control during construction.

Both anionic and cationic emulsions may be used in slurry seals but cationic emulsion is normally used in slurries containing acidic aggregates, and its early breaking characteristics are advantageous when rainfall is likely to occur. A suitable specification for slurry seals is given in Table 9.5.

TABLE 9.5

49

10. STRUCTURE CATALOGUE

The basis of the catalogue has been described in Section 1.7 and most of the information necessary to use it is contained in the main chapters of this Road Note. The cells of the catalogue are defined by ranges of traffic (Chapter 2) and subgrade strength (Chapter 3) and all the materials are described in Chapters 6 to 9. A summary of requirements and reference chapters relevant to each design chart is given in Table 10 1. Although the thicknesses of layers should follow the designs whenever possible, some limited substitution of materials between sub-base and selected fill is allowable based on the structural number principles outlined in the AASHTO guide for design of pavement structures (AASHTO (1986)). Where substitution is allowed, a note is included with the design chart. The charts are designed so that, wherever possible, the thickness of each lift of material is obvious. Thus, all layers less than 200 mm will normally be constructed in

one lift and all layers thicker than 300 mm will be constructed in two lifts. Occasionally layers are of intermediate thickness and the decision on lift thickness will depend on the construction plant available and the ease with which the density in the lower levels of the lift can be achieved. The thickness of each lift need not necessarily be identical and it is often better to adjust the thickness according to the total thickness required and the maximum particle size by using a combination of gradings from Table 6.2. In Charts 3, 4 and 7 where a semi-structural surface is defined, it is important that the surfacing material should be flexible (Chapter 8) and the granular roadbase should be of the highest quality, preferably GB1,A. In traffic classes T6, T7 and T8 only granular roadbases of type GB1 or GB2 should be used, GB3 is acceptable in the lower traffic classes. For lime or cementstabilised materials, the charts already define the layers for which the three categories of material may be used. The choice of chart will depend on a variety of factors but should be based on minimising total transport costs as

TABLE 10.1

50

discussed in Section 1 3 Factors that will need to be taken into account in a full evaluation include, the likely level and timing of maintenance the probable behaviour of the structure the experience and skill of the contractors and the availability of suitable plant the cost of the different materials that might be used other risk factors

It is not possible to give detailed guidance on these issues. The charts have been developed on the basis of reasonable assumptions concerning the first three of these, as described in the text, and therefore the initial choice should be based on the local costs of the feasible options If any information is available concerning the likely behaviour of the structures under the local conditions, then a simple risk analysis can also be carried out to select the most appropriate structure (e g Ellis (1975)). With more detailed information, it should be possible to calibrate one of the road investment models such as HDM-111 (Watanatada et al (1987)) or RTIM-2 (Parsley and Robinson (1982)) and then to use the model to calculate the whole life costs associated with each of the possible structures thereby allowing the optimum choice to be made. For many roads, especially those that are more lightly trafficked, local experience will dictate the most appropriate structures and sophisticated analysis will not be warranted.

51

KEY TO STRUCTURAL CATALOGUE

52

CHART 1

GRANULAR ROADBASE / SURFACE DRESSING

53

CHART 2 COMPOSITE ROAD BASE (UNBOUND & CEMENTED) / SURFACE DRESSING

54

CHART 3

GRANULAR ROADBASE / SEMI-STRUCTURAL SURFACE

55

CHART 4

COMPOSITE ROADBASE / SEMI - STRUCTURAL SURFACE

56

CHART 5

GRANULAR ROADBASE / STRUCTURAL SURFACE

57

CHART 6

COMPOSITE ROADBASE / STRUCTURAL SURFACE

58

CHART 7

BITUMINOUS ROADBASE / SEMI-STRUCTURAL SURFACE

59

CHART 8

CEMENTED ROADBASE / SURFACE DRESSING

60

11. REFERENCES
AMERICAN SOCIETY FOR TESTING AND MATERIALS (1987) Test D559, Annual Book of ASTM Standards, Vol. 4.08. Philadelphia. AMERICAN SOCIETY FOR TESTING AND MATERIALS (1990). D3910, Annual Book of ASTM Standards, Vol. 4.03 Philadelphia. ANDERSON, M G and D M LLOYD (1991) Using a combined slope hydrology-stability model to develop cut slope design charts. Proc. Institution of Civil Engineers, Part 2, 91, London ASPHALT INSTITUTE (1980). Model construction specifications for asphalt concrete and other plant-mix types. Specification Series No I (SS-1) The Asphalt Institute, College Park, Maryland. ASPHALT INSTITUTE (1983). Principles of construction of hot-mix asphalt pavements Manual Series No. 22 (MS-22). The Asphalt Institute, College Park, Maryland. ASPHALT INSTITUTE (1988) Mix design methods for asphalt concrete and other hot-mix types Manual Series No 2 (MS-2) The Asphalt Institute, College Park, Maryland. ASPHALT INSTITUTE (1989). The Asphalt Handbook. Manual Series No 4 (MS-4) The Asphalt Institute, College Park, Maryland. ASPHALT INSTITUTE (1991). Marshall Mix Design Criteria Changes The Magazine of the Asphalt Institute Vol 5, No. 3 The Asphalt Institute, College Park, Maryland. BELL, F G (1987) Ground Engineers Reference Book. Butterworth and Co Ltd, Publishers. BISHOP, A W (1955). The use of slip circles in the stability analysis of slopes. GeotechnIques, Vol 3. London. BOFINGER, H E, H O HASSAN and R I T WILLIAMS (1978) The shrinkage of fine grained soil cement. TRRL Supplementary Report SR 398 Transport and Road Research Laboratory, Crowthorne. BRUNTON, J M, S F BROWN and P S PELL (1987). Developments to the Nottingham Analytical Design , Method for Asphalt Pavements Proc 6th Int Conf. on the Structural Design of Asphalt Pavements. University of Michigan, Ann Arbor. BULLAS, J C and G WEST (1991). Specifying clean, hard and durable aggregate for bitumen macadam roadbase TRRL Research Report RR 284 Transport and Road Research Laboratory, Crowthorne.

CHESHER, A and R HARRISON (1987) Vehicle operating costs: evidence from developing countries. The Highway Design and Maintenance Standards Series. The International Bank for Reconstruction and Development, Washington, DC. CIRIA (1988) Laterite in road pavements. Special Publication 47. Construction Industry Research and Information Association, London, UK. COOK, D J and P SUWANVITAYA (1982) Properties and behaviour of lime-rice husk ash cements. UNCIV Report No 8208. University of New South Wales, Australia. DEPARTMENT OF TRANSPORT (1986) Specification for Highway Works. Part 2, Earthworks and Part 3, Pavements. Her Majesty's Stationery Office, London. DICKINSON, E J (1982). The performance of thin bituminous pavement surfacings in Australia. Proc. 11th ARRB Conf 11(3) pp. 35-51. ELLIS, C 1 (1974). Village-scale production of lime in Ghana TRRL Supplementary Report SR 42 Transport and Road Research Laboratory, Crowthorne. ELLIS, C 1 (1975). Risk and pavement design decisions in developing countries. TRRL Laboratory Report LR 667 Transport and Road Research Laboratory, Crowthorne. ELLIS, C I (1980). Sod compaction at low moisture contents. Field trials in the Sudan. Proc. 7th Regional Conf for Africa on Sod Mechanics and Foundation Engineering, Accra, Ghana. A A Balkema, Rotterdam. GERRITSEN, A H and R C KOOLE (1987) Seven years experience with the structural aspects of the Shell Pavement Design Manual Proc. 6th Int Conf. on the Structural Design of Asphalt Pavements. University of Michigan, Ann Arbor. HARRIS, V A P, L S HITCH and J M JOWETT (1983) Bituminous stabilisation of fine sands construction of the Baiomori - Gashua road, Nigeria. Proc. Institution of Civil Engineers, Part 1, 74 pp 277-300. London. HIGHWAY RESEARCH BOARD (1962) The AASHO Road Test. Report 5, Pavement Research Highway Research Board Special Report No 61 E National Research Council, Washington, DC. HITCH, L S (1981). Surface dressing in developing countries research in Kenya. TRRL Laboratory Report LR 1019, Transport and Road Research Laboratory, Crowthorne. HITCH, L S and M STEWART (1987) The preparation of cut-back bitumens to ASTM specifications by blending readily available constituents TRRL Research Report RR 104, Transport and Road Research Laboratory, Crowthorne.

61

M HIZAM HARUN and C R JONES(1992) The performance of polymer modified asphaltic concrete on climbing lanes in Malaysia. Proc 16th Australian Road Research Board Conf. ARRB, Melbourne. HOBAN, T (1990). Modified bitumen binders for surface dressing. Chemistry and Industry (J. Society of the Chemical Industry), No 17, pp 538-542. London. HOWE, J D G F (1972). A review of rural traffic counting methods in developing countries. RRL Report LR 427, Road Research Laboratory, Crowthorne. HOWELL, J H, J E CLARKE, C J LAWRANCE and I SUNWAR (1991). Vegetation structures for stabilising highway slopes. A manual for Nepal. Transport and Road Research Laboratory, Crowthorne. JACKSON, G P and D BRIEN (1962) Asphaltic concrete. Shell International Petroleum Company Ltd., London. LAWRANCE, C J, R J BYARD and P J BEAVEN (1993) The Terrain Evaluation Manual Transport Research Laboratory State of the Art Review Her Majesty's Stationery Office, London. LEWIS, W A, R T MURRAY and I F SYMONS (1975). Settlement and stability of embankments constructed on soft alluvial sods. Proc. Institution of Civil Engineers, Part 2 London. LIONJANGA, A V, T TOOLE and D NEWILL (1987). The development of specifications for the use of low-grade calcretes in lightly trafficked roads in Botswana Transportation Research Record 1106, Vol 1, pp 281-304. National Research Council, Washington, DC. MEHTA, P K (1979). The chemistry of cements made from rice husk ash. Proc. of Workshop on Production of Cement-Like Materials from Agro-Wastes UNIDOESCAP-RCTT-PCSIR. Peshawar, Pakistan. MILLARD, R S (1993). Roadmaking in the tropics: materials and methods. Transport Research Laboratory State of the Art Review. Her Majesty's Stationery Office, London. NATIONAL ASSOCIATION OF AUSTRALIAN STATE ROAD AUTHORITIES (1983). Guide to the control of moisture in roads NAASRA, Sydney. NATIONAL ASSOCIATION OF AUSTRALIAN STATE ROAD AUTHORITIES (1986). Guide to stabilisation in roadworks. NAASRA, Sydney. NATIONAL INSTITUTE FOR TRANSPORT AND ROAD RESEARCH (1978). The Construction of road embankments. Technical Recommendations for Highways TRH 9. Council for Scientific and Industrial Research, Pretoria. NATIONAL INSTITUTE FOR TRANSPORT AND ROAD RESEARCH (1984). Determination of the initial consumption of lime in sod (gravel ICL test). Test method CA 3483. Council for Scientific and Industrial Research, Pretoria.

NATIONAL INSTITUTE FOR TRANSPORT AND ROAD RESEARCH (1986) Cementitious stabilisers inroad construction Technical Recommendations for Highways TRH 13. Council for Scientific and Industrial Research, Pretoria. NATIONAL INSTITUTE FOR TRANSPORT AND ROAD RESEARCH (1987a). The design of road embankments. Technical Recommendations for Highways TRH 10 Council for Scientific and Industrial Research, Pretoria. NATIONAL INSTITUTE FOR TRANSPORT AND ROAD RESEARCH (1987b) The investigation, design, construction and maintenance of road cuttings Technical Recommendations for Highways TRH 18. Council for Scientific and Industrial Research, Pretoria. NETTERBERG, F (1984). Rapid field test for carbonation of lime or cement-treated materials. National Institute for Transport and Road Research Report RSl2/84. Council for Scientific and Industrial Research, Pretoria. NETTERBERG, F and M I PINARD (1991) Derivation of interim performance-related specifications for coarselygraded plastic calcrete basecourse gravels Proc. 10th Regional Conference for Africa on Soil Mechanics and Foundation Engineering, Maseru. NEWILL, D, R ROBINSON and KASSAYE AKLILU (1987) Experimental use of cinder gravel on roads in Ethiopia. Proc 9th Regional Conference for Africa on Sod Mechanics and Foundation Engineering, Lagos, Nigeria A A Balkema, Rotterdam. O'CONNELL, M J, J H G WAMBURA and D NEWILL (1987) Soil compaction at low moisture content in dry areas in Kenya Proc. 9th Regional Conf. for Africa on Soil Mechanics and Foundation Engineering, Lagos, Nigeria. A A Balkema, Rotterdam. PAIGE-GREEN, P (1981). Current techniques in groundwater control applied to cut slopes. Trans Geol Soc. S. Africa, Vol 84, pp 161-167 PAIGE-GREEN, P, F NETTERBERG and L R SAMPSON (1990). The carbonation of chemically stabilised road construction materials A guide to its identification and treatment. Division of Roads and Transport Technology. Research Report DPVT - 123 Council for Scientific and Industrial Research, Pretoria. PARSLEY, L L and R ROBINSON (1982). The TRRL Road Investment Model for developing countries (RTIM 2) TRRL Laboratory Report LR 1057, Transport and Road Research Laboratory, Crowthorne. PARSONS, A W (1976). The rapid measurement of moisture conditions of earthwork material TRRL Laboratory Report LR 750. Transport and Road Research Laboratory, Crowthorne. PARSONS, A W (1993) Compaction of soils and granular materials a review of research performed at the Transport Research Laboratory. Her Majesty's Stationery Office, London.

62

PARSONS, A W and A F TOOMBS (1987) The precision of the moisture condition test TRRL Research Report RR 90 Transport and Road Research Laboratory, Crowthorne. PATERSON, W D O (1987) Road deterioration and maintenance effects: models for planning and management. The Highway Design and Maintenance Standards Series. The International Bank for Reconstruction and Development, Washington, DC. POWELL, W D, J F POTTER, H C MAYHEW and M E NUNN (1984). The structural design of bituminous roads TRRL Laboratory Report LR 1132. Transport and Road Research Laboratory, Crowthorne. POWELL, W D and D LEECH (1982) Standards for compaction of dense roadbase macadam. TRRL Supplementary Report SR 717 Transport and Road Research Laboratory, Crowthorne. BOLT, J, S G WILLIAMS, C R JONES and H R SMITH (1987) Performance of a full scale pavement design experiment in Jamaica. Transportation Research Record 1117. Transportation Research Board, Washington DC. ROLT, J, H R SMITH and C R JONES (1986). The design and performance of bituminous overlays in tropical environments Proc 2nd Int Conf. on the Bearing Capacity of Roads and Airfields. Plymouth, UK. RUSSAM, K and D CRONEY (1960) The moisture conditions beneath ten overseas airfields. Conf on Civil Engineering Problems Overseas, London, July 1960, Paper 9. Institution of Civil Engineers, London pp 199206 and discussion pp 217-234. SAM ESHIMA, T and P M BLACK (1979). Clay index - a simple method of assessing the quality and performance of roading aggregate Proc New Zealand Roading Symposium, A3, pp 1-10. SAMPSON, L R and F NETTERBERG (1989). The durability mill. A new performance-related durability text for basecourse aggregates. The Civil Engineer In South Africa, September 1989, pp 287-294. SHERWOOD, P T (1993) Soil stabilisation with lime and cement. Transport Research Laboratory State of the Art Review. Her Majesty's Stationery Office, London. SMITH, H R, J BOLT and J H G WAMBURA (1990), The durability of bituminous overlays and wearing courses in tropical environments. Proc. 3rd Int Conf on the Bearing Capacity of Roads and Airfields. The Norwegian Institute of Technology, Trondheim, Norway. STOCKER, P T (1972). Diffusion and diffuse cementation in lime and cement-stabilised clayey soils Special Report No 8 Australian Road Research Board, Victoria, Australia.

STRAUSS, P J, V P SERVAS and G P MARAIS (1984) Unexpected surface cracking of asphaltic wearing courses. Proc 4th Conf. on Asphalt Pavements for Southern Africa, Cape Town CAPSA, South Africa. TRANSPORT AND ROAD RESEARCH LABORATORY (1976) Settlement and stability of earth embankments on soft foundations TRRL Supplementary Report SR 399. TRRL, Crowthorne. TRANSPORT AND ROAD RESEARCH LABORATORY (1978) Guide to the measurement of axle loads in developing countries using a portable weighbridge TRRL Road Note No 40 Her Majesty's Stationery Office, London. TRANSPORT AND ROAD RESEARCH LABORATORY (1982) A guide to surface dressing in tropical and sub-tropical countries Overseas Road Note No 3 TRRL, Crowthorne. TRANSPORT AND ROAD RESEARCH LABORATORY (1988). A guide to road project appraisal Overseas Road Note No 5, TRRL, Crowthorne. TRANSPORT AND ROAD RESEARCH LABORATORY (1988) A guide to geometric design Overseas Road Note No 6. TRRL, Crowthorne US ARMY CORPS OF ENGINEERS (1991) Hot-mix asphalt paving handbook US Army Corps of Engineers and National Asphalt Pavement Association. Vicksberg and Maryland, USA. WATANATADA, T. C G HARRAL, W D O PATERSON, A M DHARESHWAR, A BHANDARI and K TSUNOKAWA (1987) The Highway Design and Maintenance Standards Model Volume 1, Description of the HDM-111 Model Volume 2, User's manual for the HDM-III Model Highway Design and Maintenance Standards Series, The International Bank for Reconstruction and Development, Washington, DC. WEINERT, H H (1980) The natural road construction materials of Southern Africa Academica, Pretoria. WHITEOAK, D (1990) The Shell Bitumen Handbook. Shell Bitumen UK, London. WILLIAMS, R I T (1986) Cement treated pavements Elsevier Applied Science Publishers, London.

63

12.

BIBLIOGRAPHY

CHEN, F H (1988) Foundations on expansive soils. Pub Elsevier. CRONEY, D and P CRONEY (1991). The design and performance of road pavements. Second Edition. McGraw Hill International. Maidenhead, UK. DICKINSON, E J (1985). Bituminous roads in Australia. Australian Road Research Board, Victoria. HEAD, K H (1984). Manual of soil laboratory testing. ELE International Ltd. Pub. Pentech Press, London and Plymouth INGLES, O G AND J B METCALF (1972) Soil stabilization - Principles and practice. Butterworths, Sydney, Melbourne and Brisbane. LIME STABILISATION (1987) State of the Art Report No 5. Transportation Research Board, Washington, "DC. LIME STABILISATION MANUAL (1986). British Aggregate and Construction Materials Industries, London. NATIONAL ASSOCIATION OF AUSTRALIAN STATE ROAD AUTHORITIES (1987). Pavement design. A guide to the structural design of road pavements. NAASRA, Sydney. SHELL INTERNATIONAL PETROLEUM CO. (1978). Shell Pavement Design Manual London.

64

APPENDIX A: APPLICABLE BRITISH STANDARDS

65

66

APPENDIX B: ESTIMATING SUBGRADE MOISTURE CONTENT FOR CATEGORY 1 CONDITIONS

The subgrade moisture content under an impermeable road pavement can increase after construction where a water table exists close to the ground surface. This ultimate moisture content can be predicted from the measured relationship between soil suction and moisture content for the particular soil and knowledge of the depth of water table. Measuring the complete relationship between suction and moisture content is time consuming and a simpler, singlemeasurement procedure can be used. A small sample of soil, compacted to field density and moisture content, is placed within suitable laboratory equipment that can apply a pressure equivalent to the 'effective depth' of the water table (e g. a pressure plate extractor). The 'effective depth' of the water table for design purposes comprises the actual depth from the subgrade to the water table plus an apparent depression of the water table due to the pressure of the overlying pavement. This apparent depression varies with soil type and an approximate correction factor is given in Table B1.

To calculate the effective depth D which is used to determine the applied suction in the pressure plate extractor, the following equation is used: D = WT + (SF x t) Where WT = depth of water table below subgrade (at its highest expected seasonal level), correction factor from Table B1, pavement thickness, with consistent units for WT, t, D

SF t

= =

When equilibrium is attained in the pressure plate extractor, the sample is removed and its moisture content measured. This moisture content is the value at which the CBR for design should be estimated following standard soil tests as outlined in Section 3 2.

TABLE B1

67

APPENDIX C: TRL DYNAMIC CONE PENETROMETER

The TRL Dynamic Cone Penetrometer (DCP), shown in Figure C1, is an instrument designed for the rapid in situ measurement of the structural properties of existing road pavements with unbound granular materials. Continuous measurements can be made to a depth of 800 mm or to 1200 mm when an extension rod is fitted. The underlying principle of the DCP is that the rate of penetration of the cone, when driven by a standard force, is inversely related to the strength of the material as measured by, for example, the California Bearing Ratio (CBR) test (see Figure C2). Where the pavement layers have different strengths, the boundaries between the layers can be identified and the thickness of the layers determined. A typical result is shown in Figure C3. The DCP needs three operators, one to hold the instrument, one to raise and drop the weight and a technician to record the results. The instrument is held vertical and the weight carefully raised to the handle. Care should be taken to ensure that the weight is touching the handle, but not lifting the instrument, before it is allowed to drop and that the operator lets it fall freely and does not lower it with his hands. If during the test the DCP tilts from the vertical, no attempt should be made to correct this as contact between the shaft and the sides of the hole will give rise to erroneous results. If the angle of the instrument becomes worse, causing the weight to slide on the hammer shaft and not fall freely, the test should be abandoned. It is recommended that a reading should be taken at increments of penetration of about 10 mm. However it is usually easier to take readings after a set number of blows It is therefore necessary to change the number of blows between readings according to the strength of the layer being penetrated. For good quality granular roadbases readings every 5 or 10 blows are normally satisfactory but for weaker sub-base layers and subgrade readings every 1 or 2 blows may be appropriate. Little difficulty is normally experienced with the penetration of most types of granular of weakly stabilised materials. It is more difficult to penetrate strongly stabilised layers, granular materials with large particles and very dense, high quality crushed stone. The TRL instrument has been designed for strong materials and therefore the operator should persevere with the test. Penetration rates as low as 0.5 mm/blow are acceptable but if there is no measurable penetration after 20 consecutive blows it can be assumed that the DCP will not penetrate the material. Under these circumstances a hole can be drilled through the layer using either an electric or pneumatic drill or by coring. The lower layers of the pavement can then be tested in the normal way. DCP results are conveniently processed by computer and a program has been developed (TRRL (1990)) that is designed to assist with the interpretation and presentation of DCP data.

68

Fig.C1 TRL Dynamic cone penetrometer

69

Fig.C2 DCP-CBR relationships

70

Fig.C3 DCP test result

71

APPENDIX D: REFUSAL DENSITY DESIGN

1 INTRODUCTION

3
3.1

EXTENDED VIBRATING HAMMER COMPACTION

Laboratory design procedure

Some authorities have adopted a procedure of extended Marshall compaction to design asphalts which will retain a required minimum voids in the mix (VIM) after secondary compaction by traffic. An alternative method based on an extended form of the compaction procedure used in the Percentage Refusal Density (PRD) Test (BS 598 Part 104 (1989)) uses a vibrating hammer for compaction. These methods are appropriate for sites which are subject to severe loading where research shows that it is desirable to retain a minimum VIM of three per cent to minimise the risk of failure by plastic deformation. Neither method exactly reproduces the mode of compaction which occurs under heavy traffic but the latter procedure is both quicker and more representative. There are no national or international standards for these procedures and therefore they are both likely to be subject to further development.

In the vibrating hammer method the samples are compacted in 152-153 mm diameter moulds to a thickness approximately the same as will be laid on the road. The BS 598 compaction procedure for the PRD test is repeated if necessary to achieve an `absolute' refusal density. The electric vibrating hammer should have a power consumption of 750 watts or more and operate at a frequency of 20 to 50 Hz. Two tamping feet are used, one with a diameter of 102 mm and the other of 146 mm. Samples should be mixed so that they can be compacted immediately afterwards at an initial temperature of 140 YC for 80/100 penetration grade bitumen or 145 5''C for 60/70 penetration grade bitumen. The moulds and tamping feet must be pre-heated in an oven before starting the test. Cooling of the sample by as much as 15 to 20"C during compaction should not prevent achievement of the correct refusal density. The small tamping foot is used for most of the compaction sequence. The hammer must be held firmly in a vertical position and moved from position to position in the prescribed order, i e using the points of a compass. To identify the position, the order should be N,S,W,E,NW,SE,SW,NE or equivalent. At each point, compaction should continue for between 2 and 10 seconds, the limiting factor being that material should not be allowed to `push up' around the compaction foot. The compaction sequence is continued until a total of 2 minutes 5 seconds of compaction time has been reached. The large tamping foot is then used to smooth the surface of the sample. A spare base-plate, previously heated in the oven, is placed on top of the mould which is then turned over. The sample is driven to the new base plate with the hammer and large tamping foot. The compaction sequence is then repeated. The free base plate should be returned to the oven between compaction cycles. This is the standard PRD compaction procedure but to ensure that the refusal density is reached, it may be necessary to repeat this procedure a second time. It is suggested that trial mixes with a bitumen content which corresponds to approximately 6 per cent VIM in the Marshall test, are used to (i) determine the mass of material required to give a compacted thickness of approximately the same thickness as for the layer on the road determine the number of compaction cycles which will ensure that absolute refusal density is achieved.

EXTENDED MARSHALL COMPACTION

For severe sites, the basecourse gradings, BC1 and BC2, given in Table 8.3 and Table 8.6 are likely to be the most appropriate. The normal Marshall design procedure using 75 blows on each face should be completed first to provide an indication that Marshall design parameters will be met. The binder content corresponding to 6 per cent VIM obtained in the Marshall test should be noted and additional test samples prepared at each of three binder contents, namely the binder content corresponding to 6 per cent VIM and also binder contents which are 0 5 per cent above and 0.5 per cent below this value. These samples must be compacted to refusal. The number of blows required to produce a refusal condition will vary from one mix to another It is preferable to conduct a trial using the lowest binder content and to compact using an increasing number of blows, say 200, 300, 400, etc. until no further increase in density occurs. Usually 500 blows on each face is found to be sufficient. By plotting a graph of VIM at the refusal density against binder content the design binder content which corresponds to a VIM of 3 per cent can be determined. This value should be obtained by interpolation, not by extrapolation If necessary, the binder content range should be extended upwards or downwards, as appropriate, to permit this

(ii)

72

After these tests have been completed, samples are made with bitumen contents starting at the Marshall optimum and decreasing in 0 5 per cent steps until the bitumen content at which 3 per cent voids is retained at absolute refusal density can be determined.

It is essential to provide a surface dressing for the type of basecourse mixes which are best suited to these severe conditions. This protects the mix from severe age hardening during the period when secondary compaction occurs in the wheelpaths, and also protects those areas which will not be trafficked and are likely to retain air voids above 5 per cent

3.2

Transfer of laboratory design to compaction trials

POSSIBLE PROBLEMS WITH THE TEST PROCEDURES

After the standard PRD compaction cycle, test samples of basecourse or roadbase which have been compacted from the loose state can be expected to have densities between 1 5 and 3 per cent lower than for the same material compacted in the road but cored out and subjected to the PRD test. This is an indication of the effect of the different compaction regime and is caused by a different resultant orientation of particles. The differences between the densities for laboratory and field samples after refusal compaction should be measured to confirm whether this difference occurs. A minimum of three trial lengths should be constructed with bitumen contents at the laboratory optimum for refusal density (3 per cent VIM) and at 0.5 per cent above and 0 5 per cent below the optimum. These trials should be used to (i) determine the rolling pattern required to obtain a satisfactory density establish that the mix has satisfactory workability to allow a minimum of 93 per cent of PRD (standard compaction (BS598.1989)) to be achieved after rolling obtain cores so that the maximum binder content which allows at least 3 per cent VIM to be retained at refusal density can be confirmed.

Multi-blow Marshall compaction and vibratory compaction may cause breakdown of aggregate particles. If this occurs to a significant extent then the test is unlikely to be valid. Because of the time taken to complete the Marshall procedure, considerable care must be taken to prevent excessive cooling of the sample during compaction. It is important to note that the different particle orientation produced by these compaction methods, in comparison with that produced by roller compaction, limits the use of samples prepared in these tests to that of determining VIM at refusal It would be unwise to use samples prepared in this way for fatigue or creep tests.

(ii)

(iii)

For a given aggregate and grading, cores cut from the compacted layer can be expected to give a constant value of voids in the mineral aggregate (VMA) at the refusal density, irrespective of bitumen content. This will allow a suitable binder content to be chosen to give a minimum of 3 per cent VIM at refusal density. A minimum of 93 per cent and a mean value of 95 per cent of the standard PRD density is recommended as the specification for density on completion of compaction of the layer. From these trials and the results of laboratory tests, it is then possible to establish a job mix formula. This initial procedure is time consuming, but is justified by the long term savings in extended pavement service life that can be obtained. After this initial work, subsequent compliance testing based on analysis of mix composition and refusal density should be quick, especially if field compaction can be monitored with a nuclear density gauge.

73

APPENDIX E: THE PROBE PENETRATION TEST

1 GENERAL DESCRIPTION

This test utilises a modified sod assessment cone penetrometer originally designed by the UK Military Engineering Experimental Establishment for the assessment of in situ soil strength. The standard cone normally used with this penetrometer is replaced by a 4 mm diameter probe rod with a hemispherical tip made of hardened steel. The probe is forced into the road surface under a load of 35 kgf (343N) applied for 10 seconds and the depth of penetration is measured by a spring loaded collar that slides up the probe rod. The distance the collar has moved is measured with a modified dial gauge. The temperature of the road surface is recorded and a graphical method is used to correct the probe measurements to their equivalents at a standard temperature of WC.

METHOD OF OPERATION

All measurements are made in the nearside wheelpath of each traffic lane where maximum embedment of chippings can be expected. A minimum of ten measurements are required at each location. These should be evenly spaced along the road at intervals of 0 5m with any recently repaired or patched areas being ignored. For convenience, the measurement points can be marked with a chalk cross. The probe tip should not be centred on any large stones present in the road surface. Before each measurement the collar is slid down the probe rod until it is flush with the end of the probe. The probe is then centred on the measurement mark and a pressure of 35 kgf is applied for 10 seconds, care being taken to keep the probe vertical. The probe is then lifted clear and the distance the collar has slid up the probe is recorded in millimetres. Sometimes the point selected for test is below the general level of the surrounding road surface. It is then necessary to deduct the measurement of the initial projection of the probe tip from the final figure. The road surface temperature should be measured at the same time that the probe is used and the tests should not be made when the surface temperature exceeds 35C. In many tropical countries this will limit probe testing to the early morning. The probe readings are corrected to a standard temperature of 30C using Figure E1, and the mean of ten probe measurements is calculated and reported as the mean penetration at 30=C. Categories of road surface hardness and the corresponding ranges of penetration values are shown in Chapter 9 2.

74

Fig.E1 Correction of road surface hardness to the standard test temperature of 30 C

75

ISSN 0951-8797

OS-E

A guide to the structural design of bitumen-surfaced roads in tropical and sub-tropical countries

ORN 31

Transport and Road Research Laboratory Department of the Environment Department of Transport

Road Note 40 A guide to the measurement of axle loads in developing countries using a portable weighbridge

Crown copyright 1978 First published 1978 Third impression 1995

Extracts from the text may be reproduced except for commercial purposes, provided the source is acknowledged

NOTICE TO READERS THIS OVERSEAS ROAD NOTE IS CURRENTLY BEING REVISED AND WILL BE REISSUED IN ITS NEW VERSION DURING 2001

Contents

Foreword 1 2 3 4 5 6 7 8 Introduction Survey sites

Page

4 5 5 7 8 10 11 17 18

Survey equipment and installation Traffic counting Axle weighing Analysis of results References Appendix 1. Weighbridge operating instructions Appendix 2. Calibration of the weighbridge

19

Foreword

In many developing countries road traffic is growing rapidly both in volume and in the size and weight of the vehicles using the roads. As a consequence highway engineers concerned with designing new roads or the strengthening of existing roads in developing countries require reliable information about the distribution of the axle loads of existing traffic, and when possible, information on national or regional axle load trends. This information is required so that better forecasts can be made of the axle loads that will traverse particular roads in the future. The importance of reliable information on axle loads for pavement design purposes is emphasised by the now widely accepted assumption that the degree of pavement damage caused by an axle load is proportional to approximately the fourth or fifth power of the axle load. This implies that in many typical situations the heaviest ten per cent of axle loads in the traffic stream can produce more pavement damage than all the rest of the traffic combined. To provide adequate information on axle load distributions road-side surveys of axle loads are required. Such surveys can conveniently be made using portable wheel-weighing devices. This note describes a simple and convenient procedure for making axle load surveys in developing countries using portable weighing platforms. The procedure is based on experience gained by staff of the Overseas Unit who have undertaken a considerable number of axle load surveys in several developing countries.

A Silverleaf Director TRANSPORT AND ROAD RESEARCH LABORATORY DEPARTMENT OF THE ENVIRONMENT DEPARTMENT OF TRANSPORT 1978

Note:

Some of the equipment described may no longer be available. However suitable alternatives can be obtained from the same manufacturer.

A guide to the measurement of axle loads in developing countries using a portable weighbridge

1. Introduction
The deterioration of paved roads caused by traffic results from both the magnitude of the individual wheel loads and the number of times these loads are applied. For pavement-design purposes, it is therefore necessary to consider not only the total number of commercial vehicles that will use the road, but also the wheel loads (or, for convenience, the axle loads) of these vehicles. To do this, the axle-load distribution of a typical sample of the commercial vehicles using a road must be known. Then, from factors derived from 1 the AASHO Road Test which enable the damaging power of axle loads of different magnitudes to be expressed in terms of an equivalent number of 'standard' 8160 kg (18 000 lb) axle loads, the total number of commercial vehicles that will use the road during its design life can be expressed as an equivalent number of standard axles. It is not sufficient to determine the typical or average axle load, as the heavier axles do a disproportionate amount of damage. Thus it is p055ible for two roads of the same initial strength and trafficked by the same average axle load to deteriorate at very different rates, as the damaging power of traffic is very sensitive to variations in the number and weight of heavily laden vehicles in the traffic stream. The only effective way to compare the damaging effect of traffic on different roads is to measure the complete spectrum of axle loads and to calculate the appropriate equivalence factor. When any road project is being designed, or even appraised at the feasibility study stage, it is recommended that axle-load surveys of commercial vehicles are undertaken, ideally at several times of the year to reflect seasonal changes in the magnitude of the loads. Portable wheel-weighing devices are available which enable a small team to weigh up to 90 vehicles per hour at the roadside. By using the results of such surveys, more economic pavement thickness 2 designs can be produced . In many countries, the axle-load distribution of the total 'population' of commercial vehicles using the road system remains roughly constant from year to year. It is therefore customary to assume that the axle-load distribution of the commercial vehicles will remain unchanged for the design life of the road pavement and, if a design life of ten years is chosen, the probable errors caused by this assumption are generally acceptable. There are cases, however, where the introduction of a fleet of new and different vehicles can radically alter the axle-load distribution on a particular route in a short time. Such events cannot normally be forecast, and hence extrapolations from surveys of existing axle loads cannot provide for this sort of eventuality. Nevertheless, the error involved will certainly be less if an axle-load survey is undertaken than if no survey is carried out at all. The axle-load survey data may also be used to provide other information about the vehicle population such as the types of commodities being carried on certain routes, their origin and destination, and the amount of damage that the transport of different commodities does to road pavements. The effectiveness of vehicle loading regulations can also be studied. The Overseas Unit of the Transport and Road Research Laboratory has developed a portable 3 weighbridge * and has gained experience with it in 4, 5, 6, 7, axle-load surveys in many parts of the world 8 . The results of this experience are incorporated into this note, which gives guidelines for carrying out axle-load surveys, using the Overseas Unit weighbridge, on paved roads in developing countries.

2. Survey Sites
The success of an axle-load survey and the ease with which it can be carried out will depend very largely on the choice of site. The site must be selected to make it possible to sample the traffic easily and safely. Ideally the survey point should be sited on a clear stretch of road with good visibility as it is important that traffic is aware of the survey well in advance to give ample time to slow down and stop. Advance warning signs may be helpful, but experience shows that in some developing countries these are likely to be ignored by drivers. It is often useful to site the survey point at the crest of a hill where, provided the approaches on both sides have good visibility, the heavy vehicles being surveyed will have to slow down anyway to cope with the gradient Sites should always be positioned on stretches of road with no junctions or other turnings. Good layouts for survey sites are shown in Figures 1 and 2. In Figure 1, use is made of a slip

*The equipment is manufactured by: T. Deakin, Ascot Court, White Horse Business Park, Trowbridge, Wilts, BA14 OXA Fax no: 01225 762751

Fig. 1 LAYOUT OF SURVEY SITE - A

Fig. 2 LAYOUT OF SURVEY SITE - B

Fig. 3 LAYOUT OF SURVEY SITE C

Fig. 4 LAYOUT OF SURVEY SITE - D

or service road parallel to the road being surveyed. This is a particularly convenient arrangement since vehicles being weighed are isolated completely from the main traffic flow and do not create a safety hazard. However, slip roads will not often be conveniently sited for carrying out a survey and the situation shown in Figure 2 will be more commonly used. Here, the shoulder will usually need widening so that a vehicle being weighed is kept well clear of other traffic on the road. It is important that the shoulder is levelled so that all the wheels of the vehicle being weighed are in the same plane. It is also important that the shoulder is widened and levelled over a length of at least twice the length the of longest vehicle and trailer to be weighed. With either layout it is not necessary for the two weighing bays to be exactly opposite each other. However, the two survey points must be between the same pair of junctions, to ensure that the same sampling conditions apply for both directions of traffic. If sufficient equipment and manpower resources are available, vehicles travelling in both directions can be weighed simultaneously. Alternatively, vehicles travelling in one direction can be weighed on one side of the road for a period and vehicles travelling in the opposite direction can be weighed subsequently on the other side of the road. For traffic flows of over 30 commercial vehicles per hour in both directions, survey site layouts such as those shown in Figures 1 and 2 should be used. For lower flows, or where turningvehicle conflicts do not present a hazard, layouts such as those shown in Figures 3 and 4 may be used. Again use is made of either a slip road or a weighing bay constructed by widening and levelling a shoulder. However, in these cases, vehicles travelling in both directions are surveyed at the same time and weighed at the same survey point.

3. Survey Equipment and Installation

The weighing equipment described in this note was developed by the Overseas Unit of the Transport and Road Research Laboratory specifically for use in developing countries. The equipment consists of an aluminium alloy weighing platform (weighbridge), shown in Plate 1, a readout unit, shown in Plate 2, and a 12-volt car battery. The dimensions of the weighbridge are 700 x 500 x 90 mm and it is 44 kg in weight The complete system has a measuring range of 0-10 000 kg and, under field conditions of use, has an overall accuracy to within + 2% of full scale. The equipment is not adversely affected by high temperatures or humidities. A more detailed 3 description of the apparatus is given in LR 391 . The site selected for the weighbridge should be firm and level, with no 'high spots' and no risk of subsidence during weighing. The weighbridge should be installed in a pit with its top face level with the surrounding road surface. The design of a typical pit is shown in Figure 5. Timber formwork is employed to form the edges of the pit and a level concrete base, with soakaway to drain off any water that may accumulate, should be constructed. A layer of sand may be placed on the base of the concrete to facilitate the positioning of the weighing platform which should be levelled with a spirit level. It is also useful to construct a channel to the edge of the pit to carry the cable connecting the weighbridge to the readout unit; this prevents it being damaged if a vehicle inadvertently drives over it. The pit should be made sufficiently wide to allow the platform to be moved laterally, by about 200 mm, to enable large vehicle and trailer combinations to align their wheels more easily. Construction of a typical pit is shown in Plate 3 and a vehicle being weighed on an installed weighbridge is shown in Plate 4. It is helpful when siting the weighbridge pit to place it on the driver's side

TABLE 1 Vehicle classification

of the vehicle to make it easier for him to position his vehicle correctly. A white line painted along the road also helps in this respect (Plate 4). 4. Traffic Counting The first step in the survey must be to determine the volume of all traffic and the number of commercial vehicles travelling over the road. This is needed in order to determine whether it will be possible to weigh all commercial vehicles, excluding light commercial vehicles, or whether a sample will have to be taken. For pavement design purposes, only medium goods vehicles, heavy goods vehicles and buses (as defined in Table 1) are significant; hence it is not necessary to measure the axle loads of private cars and light goods vehicles. Traffic must be classified for counting and the details of the actual classification will depend on the local conditions and the exact reason for the survey. For general use, the classification given in Table 1 is suggested. Figure 6 shows a form which can be used for recording classified traffic counts. The form covers a period of 12 hours and, as each vehicle passes the observer, a box in the appropriate column is ticked. When traffic flows are high, it is easier to record the

Type of vehicle Passenger cars

Description Includes passenger cars seating not more than nine persons, estate cars, hire cars and taxis. Goods vehicles less than 1500 kg unladen weight or payload below 750 kg, including minibuses and Land-Rover type vehicles. Two-axled vehicles with twintyred rear wheels not exceeding 8.5 tonnes gross vehicle weight. Vehicles with more than two axles (trailers being included as part of the vehicle) or exceeding 8.5 tonnes gross vehicle weight. All regular passenger service vehicles and coaches.

Light goods

Medium goods

Heavy goods

Buses

Fig. 5 SUGGESTED CONSTRUCTION OF WEIGHBRIDGE PIT

Fig. 6 FORM FOR RECORDING CLASSIFIED MANUAL TRAFFIC COUNT

passing of vehicles on hand counters (one for each class of vehicle) and to transfer the totals to the field sheet at the end of each hour. Classified traffic counts prior to the axle weighing should be carried out on the same stretch of road as will be used for the axle-load survey. The counts should be made for 24 hours a day on at least seven consecutive days to enable estimates to be made of the average daily traffic, the proportion of heavy vehicles and the variations in flow throughout the day and night. The counts are best carried out by a team of three working eight-hour shifts, say from 0600-1400, 1400-2200 and 2200-0600. It is important to provide shelter for the enumerator in such a way as to give a clear view of the road and of the traffic. Where possible, to eliminate bias due to seasonal variations in traffic, it is desirable to obtain traffic flow data at different times of the year. Normally, information of this nature will only be available where a system of regular traffic censuses exists. Classified traffic counts should also be carried out during the axle-load survey. This will enable any differences in traffic flow that have taken place to be noted. During the axle-load survey, the enumerator should be sited near, but not adjacent to, the team carrying out the weighing and interviewing operations, so that the enumerator's attention is not distracted by these activities.

TABLE 2
Sites and sample required for Vehicle weighing

Maximum hourly flow of heavy and medium goods vehicles

Type of site required

Sample of heavy and medium goods vehicles to be weighed

0-30 31-60 61-120 121-180 181-240 etc

Figs 3 or 4 Figs 1 or 2 Figs 1 or 2 Figs 1 or 2 Figs 1 or 2 etc

All All Alternate Vehicles One in three One in four etc

5.

Axle Weighing

Damage caused to road pavements by passenger cars and light vehicles is negligible compared with that caused by heavy and medium goods vehicles. Thus it is not necessary to weigh light vehicles during an axle-load survey. Buses often have quite large axle loads and these should be surveyed. However, many buses operating on regular routes will pass the survey station many times during the day and their payloads will be fairly consistent. In these cases, it is often sufficient to weigh only a sample of these vehicles and to extrapolate the results accordingly. A reasonable average rate of vehicle weighing is about 60 vehicles/hour, although, with an experienced survey team, up to about 90 vehicles/hour can be weighed at peak times. From the preliminary traffic counts, an estimate must be made of the maximum hourly flow of heavy and medium goods vehicles. Using this information, the type of weighing site needed (as shown in Figures 1-4) and the sample of vehicles to be weighed may be determined from Table 2. Axle-load surveys should be carried out for seven consecutive days for 24 hours a day. Only if the preliminary traffic counts show a negligible vehicle flow at night, or if local difficulties make night working impossible, should the survey period be reduced but, even then, vehicles should be surveyed for at least 16 hours a day. Surveys of less than seven days are not recommended.

Experience shows that on most roads the axleload distribution of the traffic travelling in one direction is seldom the same as that of traffic travelling in the opposite direction. Significant differences between the two streams can occur, particularly on roads serving certain docks, quarries, cement works, etc, where vehicles travelling in one return journey. Similarly, on some routes, specialist vehicles may be in regular use, for example in timber extraction area, mining areas, and oil fields, where again, load factors in opposite directions are likely to be quite different. It is therefore important that in all surveys vehicles travelling in both directions are weighed. As in the traffic counting, it is recommended that the axle weighing team works in three eight hour shifts. Each team should consist of at least four people. One person is required to control traffic on the road and to direct vehicles into the weighing area according to the sample. A second person is required to direct the vehicle within the weighing are to drive slowly on to the weighing form. A third person is needed to record the wheel loads and a fourth to carry out a classified illustrated in Figures 3 and 4, an additional traffic controller is required. It is less tiring if the members of the survey team rotate jobs periodically. The traffic controller has the important job of stopping all the vehicles that are to be weighed. This can be a difficult task and it is particularly helpful if the controller has governmental authority to stop vehicles and also if he wears a distinctive

10

uniform. Often it is possible to obtain the services of the traffic police for this task and this has the added advantages that police are used to stopping vehicles and drivers are used to being stopped by them. However, it is important that no vehicles being surveyed are fined by the police for being overloaded as this can discredit the purpose of the survey and, more importantly, can affect the results obtained during subsequent days of the survey when operators might reduce their loads to avoid being fined. The controller also has the task of sampling vehicles as in Table 2, if this is required. The traffic controller(s) should stand in the road where they are clearly visible to the oncoming traffic as shown in Figures 1-4. Experience has shown that the use of a whistle and a fluorescent baton are helpful in controlling traffic. In addition, it is strongly recommended that the controller wears a reflective/fluorescent safety vest particularly at night. The job of the controller will be made very much easier if traffic is slowed down about 100 m from the survey point. This can be done with the help of an additional controller placed in this position. In addition, advance traffic warning signs and cones placed down the middle of the road may help. While vehicles are being stopped to be weighed, it is possible to carry out other surveys as well (Plate 5). Form A, shown in Figure 7, can be used to record details of the vehicle characteristics for calculating vehicle operating costs in a feasibility studv9. In addition, details about the vehicle trips and loads can be recorded for use in an origin-and-destination survey. This is usually carried out by the person who directs the vehicle on to the weighbridge platform. Wheel loads can be logged (Plate 6) on Form B, shown in Figure 8. Note that wheel loads and not axle loads are recorded. It may be assumed that on average, the axle load is twice the wheel load. It is also important to record traffic travelling in opposite directions on separate sheets. It is convenient, when recording the axle configuration, to use a simple code. Each axle is represented by a digit, usually a 1 or a '2' depending on how many wheels are on the end of the axle. Tandem axles are indicated by recording the digits directly after each other. A decimal point is placed between the code for a vehicle's front and back wheels. The code for trailers is recorded in the same way as for trucks but is separated from the truck code by a 'plus' sign. Semi-trailers or articulated trailers are separated by a 'minus' sign. Some examples of these codes are given in Figure 9.

Since the survey must continue uninterrupted during inclement weather, it is important that shelter is provided for the survey teams. This can be provided by a caravan, a large tent or, if nothing else is available, a vehicle. It should be positioned near enough to the survey point for the readout unit and battery to be placed under cover if necessary. The weighbridge unit itself is not affected by wet conditions. Calibration of the weighbridge is carried out by the manufacturer using a calibrated proving ring. It is useful for the User to check the calibration from time to time. Particular attention should be given to the tightening of the six bolts holding the two platform plates together. This is described in Appendix 1. The calibration procedure is described in Appendix 2. At the start of the survey, the weighbridge should be coupled to the readout unit using the sixway lead and also to the 12 volt battery with the twoway lead (blue lead negative, brown lead positive). The readout unit should then be switched on and the whole system allowed to 'warm up' for about five minutes. The 'zero' of the readout unit should then be set using a screwdriver. The voltage-check button is then depressed and the dial-gauge should give a reading greater than the 'VC' which is marked on the scale. If it does not the battery must be replaced by one that is fully charged. The calibration-check button is now depressed and the dial-gauge should indicate exactly the mark at 'CAL'. If it does not, the calibration knob must be adjusted. The 'zero' and 'calibration' must both be adjusted until the dial-gauge gives exact readings for both. The battery, zero and calibration must also be checked during the survey itself at least every half-hour, and more often if any sign of drift of the zero is observed. This procedure is summarized in Appendix 1. 6. Analysis of Results A computer program has been written to analyse the results from axle-load surveys. The program is easy to use and enables a detailed analysis to be carried out of the damage caused to the road by the transport of different commodities and by different vehicle types. It provides a detailed tabulation of the survey results and shows where vehicles exceed the permitted axle-load regulations. It also determines mean values for gross vehicle weight payload, axle load and also mean equivalence factor. The analysis can be carried out for sub-periods within the survey's length and for the complete survey as well. If the necessary computer facilities are available, this method of analysis is recommended. It is described fully in the User Manual for the program (ALSA)10. Program ALSA is available, for a nominal charge, from the Department of Transport Highway Engineering Computer Branch, St Christopher House, Southwark Street, London SEl, United Kingdom.

An automatic data-recorder unit has been developed to work in conjunction with the portable weighbridge but its use is not recommended for surveys such as those described here because the traffic levels involved do not warrant the use of equipment of this complexity.

11

Fig. 7 FORM A FOR RECORDING VEHICLE SURVEY DATA

12

Fig. 8 FORM B FOR RECORDING WHEEL LOADS

13

Fig. 9 EXAMPLE OF AXLE CONFIGURATIONS AND CODES

14

TABLE 3 give the equivalence factor for each of the vehicles. This sum can be carried out on Form C which is shown in Figure 10. Note that it is wheel loads and not axle loads which are given on Form B. The mean equivalence factor for all vehicles travelling in each direction can then be determined by adding up the equivalence factors and dividing by the numbers of vehicles. Note that this method of determining the mean equivalence factor must always be used; it is not sufficient to find the equivalence factor of the mean axle load as this will be a serious underestimate of the true value.

Equivalence factors for different wheel loads

Wheel load (kg) 1500 1750 2000 2250 2500 2750 3000 3250 3500 3750 4000 4250 4500 4750 5000 5250 5500 5750 6000 6250 6500 6750 7000 7250 7500 7750 8000 8250 8500 8750 9000 9250 9500 9750 10000

Axle load (kg) 3000 3500 4000 4500 5000 5500 6000 6500 7000 7500 8000 8500 9000 9500 10000 10500 11000 11500 12000 12500 13000 13500 14000 14500 15000 15500 16000 16500 17000 17500 18000 18500 19000 19500 20000

Equivalence factor 0.01 0.02 0.04 0.07 0.11 0.17 0.25 0.36 0.50 0.68 0.91 1.20 1.56 2.00 2.52 3.15 3.89 4.76 5.78 6.96 8.32 9.88 11.66 13.68 15.96 18.53 21.41 24.62 28.24 32.18 36.58 41.44 46.79 52.66 59.09

The equivalence factors in Table 3 have been 11 derived using the following formula : EF = (axle load in kg/8160)
4.5 5

This formula is based, in turn, upon the analysis of the 12 AASHO Road Test carried out by Liddle . Liddle also derived a separate relationship for the equivalence factors of tandem axles. However, it is not 11 ,6 recommended that this is used ; instead, all axles should be treated individually with equivalence factors calculated using the above equation, ie using Table 3. In order to determine the axle loading for one complete year, the following procedure should be followed. i) Determine the daily traffic flow for each class of vehicle weighed using the results of this traffic survey and any other recent traffic count information that is available. ii) Determine the average daily one-directional traffic flow for each of these classes of vehicles. iii) Determine the mean equivalence factor for each direction from the results of this survey and any other surveys that have recently been carried out and choose the higher of these directional values. iv) Determine the average daily number of equivalent standard axles for those classes of vehicles that have been weighed. v) Determine the annual axle loading.

The procedure is now described in more detail. i) Daily traffic flow The daily traffic flow of those vehicles that have been weighed is determined separately for each class of vehicle from the 7-day 24-hour traffic counts by taking the total number of vehicles in each class in both directions during the week and dividing by seven. As counts will have been carried out for at least two full weeks, the flow can be found from the total number of vehicles of each class during the whole counting period divided by

If computer facilities are not available, the following method of analysis is recommended. The equivalence factors for each of the wheel loads on Form B (Figure 8) are determined using Table 3 to give the equivalence factors for each axle on each vehicle. The factors for the axles are totalled to

15

Fig. 10 FORM C RECORDING EQUIVALENCE FACTORS FOR AXLES AND VECHICLES

16

the number of days that counts were carried out. A better estimate of flow will be obtained if counts are carried out at different times of the year to reflect seasonal changes, or if use can be made of figures from a national traffic census which includes the road being studied. In these cases, the flow is given by the total number of vehicles in each class counted in all surveys divided by the total number of days that counting has taken place. In all cases, the flow should be determined separately for each class of vehicle and should be the total of vehicles travelling in both directions.

7. References
1. HIGHWAY RESEARCH BOARD. The AASHO Road Test Report 7. Summary Report Highway Research Board Special Report No 61G. Washington, DC, 1962 (National Research Council). TRANSPORT AND ROAD RESEARCH LABORATORY. A guide to the structural design of bitumen-surfaced roads in tropical and sub-tropical countries. Department of the Environment, Department of Transport Road Note 31. London, 1977 (H M Stationery Office), 3rd Edition. POTOCKI, F P. A portable wheel-weighing unit and data recorder. Department of the Environment RRL Report LR 391. Crowthorne, 1971 (Road Research Laboratory). ELLIS, C I. Axle loadings on roads in West Malaysia. Proc. 4th Conf. Aust. Rd Res. Bd, 1968, 4 (Part 2), 1112-20. ELLIS, C I and F P POTOCKI. Axle load distribution on roads overseas: Abu Dhabi and Qatar 1970-71. Department of the Environment TRRL Report LR 572. Crowthorne, 1973 (Transport and Road Research Laboratory). JONES, T E. Axle-loads on paved roads in Kenya. Department of the Environment Department of Transport TRRL Report LR 763. Crowthorne, 1977 (Transport and Road Research Laboratory). JONES, TE and R ROBINSON. 1975 Turkey traffic survey (Ankara-Istanbul Expressway): axle loading. Department of the Environment TRRL Report LR 713. Crowthorne, 1976 (Transport and Road Research Laboratory). JONES, T E and GIZAW GULILAT. The commercial traffic loading on five trunk roads in Ethiopia. ERA and TRRL (UK) Joint Road Research Project. JRRP Report No 10. Addis Ababa, 1977 (Ethiopian Road Authority). ABAYNAYAKA, S W, H HIDE, G MOROSIUK and R ROBINSON. Tables for estimating vehicle operating costs on rural roads in developing countries. Department of the Environment, TRRL Report LR 723. Crowthorne, 1976 (Transport and Road Research Laboratory).

2.

ii)

Daily one-directional traffic 3.

The flow determined above is divided by two to give the daily one-directional traffic for each class of vehicle. 4. iii) Mean equivalence factor The mean equivalence factor for each direction is determined using the method described earlier, ie the equivalence factor is determined from Table 3 for each axle on the vehicle; these are added to give the equivalence factor for each vehicle; the factors for all vehicles in one direction are added and divided by the number of vehicles to give the mean equivalence factor for that direction; this is then repeated for the other direction. If other surveys have recently been carried out these can be used to supplement the main survey by providing information on seasonal variations, etc. In this case the equivalence factors for all vehicles travelling in one direction should be added and divided by the number of vehicles to give the mean equivalence factor for that direction. The same analysis is repeated for the other direction. It is usual to use the higher of these two mean equivalence factors in subsequent analyses as this represents the most heavily laden direction.

5.

6.

7.

8.

iv) Average daily equivalent standard axles The daily one-directional traffic, as determined in ii) above, for those classes of vehicles which contribute to pavement damage are added together (this should be the same classes of vehicle that have been weighed and would normally consist of 'Medium goods', 'Heavy goods' and 'Buses' classes of vehicle in Table 1). This daily total number of pavementdamaging vehicles in one direction is now multiplied by the mean equivalence factor, as determined in iii) above, to give the average daily number of equivalent standard axles.

9.

10. ROBINSON, R. ALSA: a program for the analysis of axle-load survey data. Department of the Environment TRRL Report LR 721. Crowthorne, 1976 (Transport and Road Research Laboratory).

v)

Annual axle loading

The annual axle loading is determined by multiplying the average daily number of equivalent standard axles, as determined in iv) above, by 365.

17

11.

HODGES, J W, J ROLT and T E JONES. The Kenya road transport cost study: research on road deterioration. Department of the Environment, TR R L Laboratory Report 673. Crowthorne, 1975 (Transport and Road Research Laboratory). LIDDLE, W J. Application of AASHO Road Test results to the design of flexible pavement structures. First International Conference on the Structural Design of Asphalt Pavements, University of Michigan, 20-24 August 1962. Ann Arbor, 1962 (University of Michigan).

2.

Connect readout unit to 12-volt battery using the 2-way lead (blue lead negative, brown lead positive). Switch on readout unit Leave equipment switched on for about five minutes to 'warm up'. Adjust dial-gauge reading to zero using the 'set zero' control. Battery voltage should then be checked by pressing the 'Check volts' push-button. When the corresponding reading falls below the red mark 'VC' on the dial-gauge, the battery must be exchanged for one that is fully charged. Press 'Check cal' push-button and adjust the reading to the 'Cal' mark on the dial-gauge scale using the 'Set cal' control. Repeat 5 and 7 until both the zero and calibration readings remain correct.

3. 4.

12.

5.

6.

8.

Appendix 1
7.

Weighbridge Operating Instructions Adjustment of tension in weighbridge fixing bolts The tension in the bolts holding the two plates of the weighing platform together should be checked periodically, at least once a week when in continuous use. This is done in the following way: 1. 2. 3. Place the platform upside down. Slacken-off the six bolts. Switch on the readout unit and set reading on dialgauge to mid-scale using the 'Set zero' control. Tighten each of the six bolts in turn to 150 kg each (1 small divisions on the dial-gauge) using a 19 mm box spanner, ie 900 kg in all. The order in which the bolts should be tightened is shown in Fig. 11.

8.

The equipment is now ready for use, but some occasional readjustment of the zero and calibration (as in 5 and 7) may be necessary. Warning light on readout unit Some readout units are equipped with a warning light which comes into operation when a heavy wheel load is over a long edge of the weighing platform. The wheel position should then be adjusted until central over the platform before the reading is taken. Protection of equipment It is advisable to: a. Keep the readout unit out of the sun. Protect the readout unit from the rain. Place the equipment on foam rubber or some other soft material during transportation to prevent damage due to vibration.

4.

Setting weighbridge up on site 1. Connect weighing platform to readout unit using the 6-way lead.

b. c.

The dial-gauge is normally nitrogen filled and fully sealed, and in case of damage should be sent back to the manufacturer. It is still operational with the glass broken, but will deteriorate with time because of internal condensation. Replacement of amplifier board In cases of failure of the readout unit replacement amplifier boards can be obtained from the manufacturer and should be installed in the following way. Fig. 11 ORDER OF BOLT TIGHTENING AS REFERRED TO IN APPENDIX 1 (start from a in alphabetical order) 1. 2. 3. Remove 4 screws at the bottom of readout unit. Take chassis out of wooden box. Remove 2 screws holding board to a bracket.

18

4. 5.

Slide board out of an edge connector. Place new board in position by repeating the above procedure in reversed order.

9. Appendix 2
Calibration of the weighbridge The weighbridge equipment, as supplied by the manufacturer, is calibrated against a proving ring and any further re-calibrations by the user are useful but not essential, providing a periodic check of the sensitivity is made using the in-built calibration signal, which simulates a known load. When a reliable and accurate loading device is available, the calibration can be carried out in the following manner: a) Follow the setting up procedure, including adjustment of tension in fixing bolts as described in Appendix 1. Place a plywood or hardboard sheet (700 x 500 mm and about 3 mm thick) on the platten of the testing machine. Place the platform of the weighbridge on the plywood sheet. Apply load through a 25 cm x 38 cm steel plate or block of wood (ie the area equivalent to a tyre contact area at 10 tonnes wheel load) in increments of 1000 kg, to the full value of 10,000 kg and note the corresponding readings on the weighbridge indicator.

b)

c)

d)

The calibration line is then obtained by plotting the load measured by the indicator against the corresponding value of the load applied by the machine.

19

Neg.no.E55/72

Plate 1 THE WEIGHING PLATFORM

Neg.no.B322/78

Plate 2 THE READOUT UNIT

Neg.no.CR888/75/6

Plate 3 CONSTRUCTING WEIGHING PIT

Neg.no.CR894/75/10

Plate 4 WEIGHBRIDGE INSTALLED IN PIT

Neg.no.CR891/75/12

Plate 5 TRAFFIC SURVEY AT WEIGHBRIDGE SITE

Neg.no.CR894/75/2

Plate 6 LOGGING WHEEL LOADS

NOTES

NOTES

Printed in the United Kingdom for The Stationery Office J38534 C4 2/98 10170