"The cost of quality." Its a term that's widely used and widely misunderstood.

. The "cost of quality" isn't the price of creating a quality product or service. It's the cost of NOT creating a quality product or service. Every time work is redone, the cost of quality increases. Obvious examples include:

The reworking of a manufactured item. The retesting of an assembly. The rebuilding of a tool. The correction of a bank statement. The reworking of a service, such as the reprocessing of a loan operation or the replacement of a food order in a restaurant.

In short, any cost that would not have been expended if quality were perfect contributes to the cost of quality. Total Quality Costs As the figure below shows, quality costs are the total of the cost incurred by:

Investing in the prevention of nonconformance to requirements. Appraising a product or service for conformance to requirements. Failing to meet requirements.

Quality Costsgeneral description

Prevention Costs The costs of all activities specifically designed to prevent poor quality in products or services. Examples are the costs of:

Failure Costs The costs resulting from products or services not conforming to requirements or customer/user needs. Failure costs are divided into internal and external failure categories. Internal Failure Costs Failure costs occurring prior to delivery or shipment of the product, or the furnishing of a service, to the customer. Examples are the costs of:

New product review Quality planning Supplier capability surveys Process capability evaluations Quality improvement team meetings Quality improvement projects Quality education and training

Appraisal Costs The costs associated with measuring, evaluating

Scrap Rework Re-inspection

or auditing products or services to assure conformance to quality standards and performance requirements. These include the costs of:

Re-testing Material review Downgrading

External Failure Costs Failure costs occurring after delivery or shipment of the product and during or after furnishing of a service to the customer. Examples are the costs of:

Incoming and source inspection/test of purchased material In-process and final inspection/test Product, process or service audits Calibration of measuring and test equipment Associated supplies and materials

Processing customer complaints Customer returns Warranty claims Product recalls

Total Quality Costs: The sum of the above costs. This represents the difference between the actual cost of a product or service and what the reduced cost would be if there were no possibility of substandard service, failure of products or defects in their manufacture. Excerpted from the ASQ Quality Costs Committee, Principles of Quality Costs: Principles, Implementation, and Use, Third Edition, ed. Jack Campanella, ASQ Quality Press, 1999, pages 3-5. Definition and Explanation of Quality Costs: A product that meets or exceeds its design specifications and is free of defects that mar its appearance or degrade its performance is said to have high quality of conformance. Note that if an economy car is free of defects, it can have a quality of conformance that is just as high as defect-free luxury car. The purchasers of economy cars cannot expect their cars to be as opulently as luxury cars, but they can and do expect to be free of defects. Preventing, detecting and dealing with defects cause costs that are called quality costs or costs of quality. The use of the term "quality cost" is confusing to some people. It does not refer to costs such as using a higher grade leather to make a wallet or using 14K gold instead of gold plating in jewelry. Instead the term quality cost refers to all of the costs that are incurred to prevent defects or that result from defects in products. Quality costs can be broken down into four broad groups. These four groups are also termed as four (4) types of quality costs. Two of these groups are known as prevention costs and appraisal costs. These are incurred in an effort to keep defective products from falling into the hands of customers. The other two groups of costs are known as internal failure costs and external failure costs. Internal and

external failure costs are incurred because defects are produced despite efforts to prevent them therefore these costs are also known as costs of poor quality. The quality costs do not just relate to just manufacturing; rather, they relate to all the activities in a company from initial research and development (R & D) through customer service. Total quality cost can be quite high unless management gives this area special attention. Four types of quality cost are briefly explained below: Prevention Costs: Generally the most effective way to manage quality costs is to avoid having defects in the first place. It is much less costly to prevent a problem from ever happening than it is to find and correct the problem after it has occurred. Prevention costs support activities whose purpose is to reduce the number of defects. Companies employ many techniques to prevent defects for example statistical process control, quality engineering, training, and a variety of tools from total quality management (TQM). Prevention costs include activities relating to quality circles and statistical process control. Quality circles consist of small groups of employees that meet on a regular basis to discuss ways to improve quality. Both management and workers are included in these circles. Statistical process control is a technique that is used to detect whether a process is in or out of control. An out of control process results in defective units and may be caused by a miscalibrated machine or some other factor. In statistical process control, workers use charts to monitor the quality of units that pass through their workstations. With these charts, workers can quickly spot processes that are out of control and that are creating defects. Problems can be immediately corrected and further defects prevented rather than waiting for an inspector to catch the defect later. Some companies provide technical support to their suppliers as a way of preventing defects. Particularly in just in time (JIT) systems, such support to suppliers is vital. In a JIT system, parts are delivered from suppliers just in time and in just the correct quantity to fill customer orders. There are no stockpiles of parts. If a defective part is received from a supplier, the part cannot be used and the order for the ultimate customer cannot be filled in time. Hence every part received from suppliers must be free from defects. Consequently, companies that use just in time (JIT) often require that their supplier use sophisticated quality control programs such as statistical process control and that their suppliers certify that they will deliver parts and materials that are free of defects. Appraisal Costs:

Any defective parts and products should be caught as early as possible in the production process. Appraisal costs, which are sometimes called inspection costs, are incurred to identify defective products before the products are shipped to customers. Unfortunately performing appraisal activates doesn't keep defects from happening again and most managers realize now that maintaining an army of inspectors is a costly and ineffective approach to quality control. Employees are increasingly being asked to be responsible for their own quality control. This approach along with designing products to be easy to manufacture properly, allows quality to be built into products rather than relying on inspections to get the defects out. Internal failure Costs: Failure costs are incurred when a product fails to conform to its design specifications. Failure costs can be either internal or external. Internal failure costs result from identification of defects before they are shipped to customers. These costs include scrap, rejected products, reworking of defective units, and downtime caused by quality problem. The more effective a company's appraisal activities the greater the chance of catching defects internally and the greater the level of internal failure costs. This is the price that is paid to avoid incurring external failure costs, which can be devastating. External Failure Costs: When a defective product is delivered to customer, external failure cost is the result. External failure costs include warranty, repairs and replacements, product recalls, liability arising from legal actions against a company, and lost sales arising from a reputation for poor quality. Such costs can decimate profits. In the past, some managers have taken the attitude, "Let's go ahead and ship everything to customers, and we'll take care of any problems under the warranty." This attitude generally results in high external failure costs, customer ill will, and declining market share and profits. External failure costs usually give rise to another intangible cost. These intangible costs are hidden costs that involve the company's image. They can be three or four times greater than tangible costs. Missing a deadline or other quality problems can be intangible costs of quality. Internal failure costs, external failure costs and intangible costs that impair the goodwill of the company occur due to a poor quality so these costs are also known as costs of poor quality by some persons.

Examples of four types of quality cost are given below: Prevention Costs Internal Failure Costs Systems development Net cost of scrap Quality engineering Net cost of spoilage Quality training Rework labor and overhead Quality circles Re-inspection of reworked products statistical process control Retesting of reworked products Supervision of prevention activities Downtime caused by quality problems Quality data gathering, analysis, and Disposal of defective products reporting Analysis of the cause of defects in Quality improvement projects production Technical support provided to suppliers Re-entering data because of keying errors Audits of the effectiveness of the quality Debugging software errors system Appraisal Costs External Failure Costs Test and inspection of incoming materials Cost of field servicing and handling Test and inspection of in-process goods complaints Final product testing and inspection Warranty repairs and replacements Supplies used in testing and inspection Repairs and replacements beyond the Supervision of testing and inspection warranty period activities Product recalls Depreciation of test equipment Liability arising from defective products Maintenance of test equipment Returns and allowances arising from Plant utilities in the inspection area quality problems Field testing and appraisal at customer Lost sales arising from a reputation for site poor quality.

Quality Costs

Percentage of Sale(a) Prevention costs:

Quality training Reliability engineering

$30,000 $79,000 Appraisal costs: $109,000 3.73%

Materials inspection

$19,000

Product acceptance Process acceptance

$10,000 $35,000 Internal failure costs: $64,000 2.19%

Scrap Rework

$40,000 $34,000 External failure costs: $74,000 2.53%

Customer complaints Warranty Repair Total quality costs

$24,000 $24,000 $15,000 $63,000 $310,000 10.62% (b) 2.16%

(a) Actual sales of $2,920,000 (b) $310,000/$2,920,000 = 10.62 percent. Difference is rounding error.

FIGURE 4 ALLISON PRODUCTS INTERIM STANDARD PERFORMANCE REPORT FOR THE YEAR ENDED MARCH 31, 20X2 Actual Costs Budgeted Costs(a) Variance Prevention costs: Quality training $30,000 $30,000 $0 Reliability engineering $79,000 $80,000 $1,000 F Total prevention $109,000 $110,000 $1,000 F Appraisal costs: Materials inspection $19,000 $28,000 $9,000 F Product acceptance $10,000 $15,000 $5,000 F Process acceptance $35,000 $35,000 $0 Total appraisal $64,000 $78,000 $14,000 F Internal failure costs: Scrap $40,000 $44,000 $4,000 F Rework $34,000 $36,500 $2,500 F

Total internal failure Fixed: Customer complaints

$74,000 External failure costs:

$80,500

$6,500 F

$24,000 Variable: Warranty $24,000 Repair $15,000 Total external failure $63,000 Total quality costs $310,000 Percentage of actual sales 10.62% (a) Based on actual sales (b) Actual sales of $2,920,000 FIGURE 5

$25,000

$1,000

$20,000 ($4,000) U $17,500 $2,500 F $62,500 ($500) U $331,000 $21,000 F 11.34% 0.72% F

ALLISON PRODUCTS QUALITY COST, ONE-YEAR TREND FOR THE YEAR ENDED MARCH 31, 20X2 Actual Costs 20x2(a) Budgeted Costs 20x1 Prevention costs: $30,000 $79,000 $109,000 Appraisal costs: $19,000 $10,000 $35,000 $64,000 Internal failure costs: $40,000 $34,000 $74,000 External failure costs: Fixed: $24,000 Variable: $24,000 $15,000 $63,000 $310,000 Variance

Quality training Reliability engineering Total prevention Materials inspection Product acceptance Process acceptance Total appraisal Scrap Rework Total internal failure

$36,000 $120,000 $156,000 $33,600 $16,800 $39,200 $89,600 $48,000 $40,000 $88,000

$6,000 F $41,000 F $47,000 F $14,600 $6,800 $4,200 $25,600 F F F F

$8,000 F $6,000 F $14,000 F

Customer complaints Warranty Repair Total external failure Total quality costs

$33,000 $23,000 $16,400 $72,400 $406,000

$9,000 F ($1,000) $1,400 $9,400 $96,000 U F F F

Percentage of actual sales 10.62% (a) Based on actual sales = $2,920,000 FIGURE 6

13.90%

3.29% F

ALLISON PRODUCTS LONG-RANGE PERFORMANCE REPORT FOR THE YEAR ENDED MARCH 31, 20X2 Actual Costs Target Costs(a) Variance Prevention costs: Quality training $30,000 $14,000 ($16,000) U Reliability engineering $79,000 $39,000 ($40,000) U Total prevention $109,000 $53,000 ($56,000) U Appraisal costs: Materials inspection $19,000 $7,900 ($11,100) U Product acceptance $10,000 $0 ($10,000) U Process acceptance $35,000 $12,000 ($23,000) U Total appraisal $64,000 $19,900 ($44,100) U Internal failure costs: Scrap $40,000 $0 ($40,000) U Rework $34,000 $0 (34,000) U Total internal failure $74,000 $0 ($74,000) U External failure costs: Fixed: Customer complaints $24,000 $0 ($24,000) U Variable: Warranty $24,000 $0 ($24,000) U Repair $15,000 $0 ($15,000) U Total external failure $63,000 $0 ($63,000) U Total quality costs $310,000 $72,900 ($237,100) U Percentage of actual sales 10.62% 2.50% -8.12% U (a) Based on actual sales of $2,920,000 These costs are value-added costs. Quality Management in Construction Projects: A Case Study of Quality Management in Construction Projects of an Oil and Gas Company Currently, although businesses involved in construction activities apply quality management systems, but in practice there are a lot of poor works that worry people and society. Consequences of poor quality will damage property and waste money of business and

community, and it also cause danger to people. To ensure the effective application of quality management for construction projects, it requires the strong commitment of business leaders and implementation of multiple solutions. Applying this effective quality management will bring many potential benefits for businesses. Mr. Le Van Thong made a research which main objectives were: (1) determination of quality management activities in the construction in the Oil and Gas Company; and (2) to discuss and review the effectiveness of quality management in construction of the Oil and Gas Company Conclusions Quality management system has been applied by PTSC companies in their projects. It brings success to the company for many years now, and became a famous company in the international market in oil and gas technical services industry in Vietnam. They are the useful tools to avoid and mitigate problems and to improve quality performances of the following projects. They are the principles that management use to achieve effective cost control, quality control, schedule control etc. Many civil and industry construction Corporations, companies in developed countries have been applied the successful Quality Management and got more profits and reduced lost time and cost for reworks. They recognize the importance of the Quality Management and have responsibilities to identify, assess, prevent and manage all breakdowns and risks to projects, damage to property and working environment. The quality improvement programs establish actions for achieving the objectives and targets, in line with the policy commitment of continuous improvement. When establishing targets, the following will be taken into consideration: - Who do? How? Who supervise the provision of information; as executive / coordination / control is assigned in advance and all agents involved must comply. - The system clearly and closely flexibility have created a positive consequence of works completed on schedule, ensure quality and safety. Effectiveness of the implementation quality management:

After implementation of the quality management, there are some achieved results as below: 1. Quality Assurance is to obtain completed construction that meets all contract requirements. Assurance is defined as a degree of certainty. Quality assurance personnel continually assure that the contractor's works comply with contract requirements. 2. Quality Control is the successful execution of a realistic plan to ensure that the required standards of quality construction will be met. In QC, the contractor defines procedures to manage and control his own, designer of record, consultant, architect-engineer, all subcontractor and all supplier activities so that the completed project complies with contract requirements. For design-build contracts, this includes providing and maintaining a Design Quality Control plan as a part of the overall contract QC plan. This plan, as a minimum, must assure that all documents are reviewed by a technically competent independent reviewer specifically named in the plan. This review cannot be performed by the same designers that produced the product. The design QC plan must be managed by a Design QC Manager who has verifiable engineering or architectural design experience or is a registered engineer or architect. The Design QC Manager is under the supervision of the QC Manager. Recommendations 1. The performance of Quality Management must be carried out permanently in the company to prevent all budget overrun, progress slow, reworks maybe happening from the existing projects and future projects. 2. The performance of Quality Management must be concerned from the top management to everybody in the company. 3. Examine the quality control methods being used to determine if the contractor is properly controlling design activities in design-build contracts. 4. Examine the quality control methods being used to determine if the contractor is properly controlling construction activities. 5. Make certain that the necessary changes are made in the contractor's QC system, if

excessive construction deficiencies occur. 6. Assist the contractor in understanding and implementing the contract requirements. 7. Examine ongoing and completed work. 8. Produce the quality specified in the plans and specifications and for design-build contracts in the Request for Proposal, as well as the contractor's accepted proposal. 9. Develop and maintain an effective QC system. 10. Perform all control activities and tests. 11. Prepare acceptable documentation of QC activities. His thesis abstract is copied and posted. Abstract In an organization or business will have a lot of construction projects, any project also is importance. Ensure the quality of the project is very important for each organization. Aware of that, the quality management has been applied widely in their construction projects. But how apply the successful model of quality management for a construction project that is still a big question. Organizations in Vietnam as some other countries, in fact, the signers disclaim the possibility of anticipating and detailing everything in the construction documents. Second, it is unstable for the construction environment. The complexity and size of project vary. Working conditions cannot be somewhat control. The employee is variable; its composition, motivation and size change. Cooperation between contractor and subcontractors is problematic. Those will affect the quality management of construction projects and it makes the project to delay, re-work and increase cost. So, how to manage the quality of the construction project effectively? This thesis will research

about quality management in construction project through theory and real case that is applied in PTSC Production Services Company. Causes of Building Waste on Site Recent research indicates that about 5-10% of building materials end up aswaste on building sites. There are many contributory factors to this figure, bothhuman and mechanical, and these are outlined below: Causes of Building Waste on Site Examples ! Lack of a quality managementsystem aimed at wasteminimisatione.g. lack of waste managementplan ! Untidy construction sites e.g. waste materials are notsegregated from useful materials ! Poor handling e.g. breakage, damage, losses ! Over-sized foundations andother elementse.g. over design leads to excessexcavation and cut-offs ! Inadequate protection tofinished work e.g. finished concrete staircasesare not protected by boarding ! Limited visibility on siteresulting in damagee.g. inadequate lighting in coveredstorage area ! Poor storage e.g. pallet is not used to protectcement bags from contaminationby ground water ! Poor workmanship e.g. poor workmanship of formwork S i t e M a n a g e m e n t a n d P r a c t i c e s ! Waste generation inherited withtraditional construction methode.g. timber formwork, wet trade ! Over-ordering e.g. over ordering of concretebecomes waste ! Method of packaging e.g. inadequate protection to thematerials ! Method of transport e.g. materials drop from forklift D e l i v e r y o f p r o d u c t s ! Inadequate data regarding timeand method of deliverye.g. lack of records concerningmaterials delivery Table 3.1 Causes and examples of building waste on site

control. # Concrete WasteThe amount of concrete waste, for example, can be estimated if thematerial wastage level of concrete is known. Recent research indicated thatthe average wastage level is about 4%, which is considered the norm forthe concreting trade in this guideline. However, it could be reduced to 3%if careful material ordering and handling is applied. The amount of wastecan be estimated according to:Quantity of Concrete Works (m 3

) x Material Wastage (%) # Waste from blockwork and brickwork Inert granular waste generated by blockwork and brickwork is estimated tobe 10% of the quantity of this work required in the building project. Theestimate can be calculated according to:Quantity of Materialwork done (m 2 ) x thickness (m) x Wastage (%) # Waste from screeding and plasteringA higher wastage of 15% is given as the norm since these trades aredifficult to control. The estimate can be calculated according to:Quantity of Materialwork done (m 2 ) x thickness (m) x Wastage (%) # Waste from timber formwork Timber formwork is assumed to have been used at least 12 times beforebeing discarded. The timber waste can be estimated according to:Quantity of Formwork (m 2 ) x thickness (m) 12 (no of uses)

# Packaging WasteBuilding contractors have little control on the quantity of packaging wastesproduced, which is estimated at 5% of the volume of the materials thatrequired packaging, henceVolume of packaged construction materials x 5% # Other WastesThere are blank rows in the standard form for the provision of estimatesfor other types of wastes.Recent results for the percentage wastage by different trades for public housingprojects and private residential buildings are shown in Tables 3.2 and 3.3 forreference. Trade Material Percentage wastage Concrete Concrete 3-5%Formwork Timber broad 5%Reinforcement Steel bars 3-5%Masonry Brick and b lock 6%Dry Wall Fine aggregate 5%Wall screeding Ready-mix cement 7%Floor screeding Ready-mix cement 1 %Wall plastering Plaster 2%Ceiling plastering Plaster 2%Floor tiling Tiles 6%Wall tiling Tiles 8%Installati on of bathroom fitting Sanitary fitting 2%Installation of kitchen joinery Kitchen joinery 1% Table 3.2 Percentage wastage of materials for various trades onpublic housing projects

Trade Material Percentage wastage Concrete Concrete 4-5%Formwork Timber broad 15%Reinforcement Steel bars 1-8%Masonry Brick and block 4-8%Dry Wall Fine aggregate 6-10%Wall screeding Ready-mix cement 4-20%Floor screeding Ready-mix ce ment 4-20%Wall plastering Plaster 4-20%Ceiling plastering Plaster 4-20%Floor tiling Tiles 4-10%Wall tiling Tiles 4-10%Installation of bathroom fitting Sanitary fitting 1-5 %Installation of kitchen joinery Kitche n joinery 1-5 % Table 3.3 Percentage wastage of materials for various tradesfor private residential building Based on waste generation per GFA, it has been found that the generation rateof construction waste is in the range of 0.125m 3 to 0.25m 3 (waste index) pergross floor area GFA (m 2 ).

LEARNING ACTIVITY 5: PROCESS CAPABILITY

Variation to some degree exists in all manufacturing processes. And because it will always exist, it will always be a detriment to quality. Since a major goal of statistical process control is to understand a process in order to maintain or improve quality, we must be able to understand the outcome and success of that process. We must know if the process is capable of delivering the product with respect to predetermined requirements such as specification limits. This is where the study of process capability is important. Process capability is a basic step in quality improvement. A process capability study is used to analyze the state of a process, that is, the extent to which the product produced by the process conforms to the customer specification. It also gives an estimate of the quality level achieved by the process after significant causes of variation have been eliminated (i.e., after the process has been brought under control). Process capability can be determined at the beginning of the project through a pilot or preliminary study, or it can be determined and monitored on a continuing basis during manufacturing. Knowing the process capability helps us enhance the consistency of the product's performance, reduce the proportion of nonconforming products, and limit the percent of product scrap or rework generated from the process. About this course: Learning Activity 1, "Deming's Funnel Experiment," presented a demonstration designed by W. Edwards Deming that illustrated both common and special cause variation and showed the problems induced by tampering. Learning Activity 2, "Quality Improvement and Quality Tools," presented the different types of data along with concepts and tools that are commonly applied to a quality environment. Learning Activity 3, "Statistics Primer," reviewed statistical concepts and calculations that are important to the application of SPC. Learning Activity 4, "Control Charts," discussed the construction and application of control charts. This Learning Activity, Learning Activity 5, "Process Capability," explains how process capability is used to implement SPC in industry settings. Learning Activity 6, "Quality Tools and SPC Charts Questor," is an interactive quiz-like game that tests your knowledge about quality, statistics and the tools used in SPC.

Recommended activities: Note taking to organize your knowledge concerning the four scenarios that describe the state of a process. Completion of exercises to reinforce your understanding.

Objectives
Identify the state of processes (stable, unstable, capable, or incapable) from data plots on control charts. Use statistics to calculate the capability index for process data sets.

Capability and Stability

Introduction
In discussions of process capability, the term specification limits is often raised. Specification limits can also appear on control charts. However, there is often confusion between specification limits and control limits. As a manufacturing technician, you need to understand the difference between the two terms. Control limits (UCL, LCL): Are values calculated from statistics that indicate the range where an in-control process should perform. Signal when output exceeds statistical expectations. Apply only to subgroup averages. Indicate if the process is stable (in control).

Specification limits (USL, LSL): Are measurement values set by customers or product design that state requirements of the product. Limit output based on design or customer requirements. Apply to each individual product rather than subgroup averages. Indicate if the product is acceptable; any product with measurements outside of specification limits is out of spec and must be reworked or scrapped.

Control limits may change as the process improves; specification limits change only when product requirements change. Both types of limits are important to technicians because policies on how to conduct their work come from controlling processes based on these standards. Figure 5.1 shows a technical drawing of the dimensions and specifications of a pipe (the radii of the inner and outer walls). The tolerances (the specification limits of .002" for each radius) were determined by the design engineer. Process capability measures the ability of the process to meet the design tolerances.

Figure 5.1. Technical drawing of the front view of a pipe with specifications.

Four Process States

There are typically four scenarios that describe the state of a process. Each scenario is determined by stability and capability. Stability indicates whether or nor a process is in control, and capability refers to whether or not individual measurements are within specifications. The four scenarios are: Stable and Capable - The process is in an acceptable state: it is in control and producing good product within specifications. Unstable but Capable - The process is not in control (not stable) but is producing good product. This process will tend to become incapable in time. Stable but Incapable - The process is in control but not producing good product. This process needs to be improved to make it capable. Unstable and Incapable - The process is in the worst possible condition: it is not in control and not producing good product. The process must first be brought under control and then adjusted to improve product output quality.

Figure 5.2. Matrix showing the four process state scenarios.

Stable and Capable Process

When the process is stable and capable, it is in control and producing good product because no rules have been violated and no data points fall outside the specification limits. On the control chart in Figure 5.3, the upper and lower specification limits are the blue lines labeled USL and LSL, respectively. Notice that they are further from the mean than control limits (in red). This stable and capable process is currently staying within all limits. The subgroup means fluctuate about the center line but do not go beyond the control limits. Also, the common cause variation is not large enough for individual products to be out of specification.

Figure 5.3. A stable and capable process

Unstable but Capable Process

An unstable but capable process produces subgroup means that fall outside the control limits but products that fall within specification limits. The process shown on this control chart is unstable because subgroup 9 violated Western Electric Rule 1 (one point is outside of a 3-sigma control limit). If the process were incapable (producing out-of-spec product), at least one value point would plot outside of the upper or lower blue line. Because the unstable process can easily stray out of specification limits, process engineers should investigate it in order to stabilize it. Otherwise, quality will begin to go down and eventually out-of-spec product will be manufactured.

Figure 5.4. An unstable and capable process

Stable but Incapable Process

A process that is stable but incapable shows subgroup means fluctuating around the centerline but not beyond the control limits. However, the common cause variation is too large, or the center line is off the target value, resulting in products being made that are outside of the specification limits. (Notice that the USL and LSL are "tighter" than the control limits; this is because specification limits can be set anywhere according to customer demands and design specifications set by the company.) Still, no Western Electric Rules are being violated here. Since there is product that does not meet specifications, it will need rework or must be scrapped. There is a temptation to make small changes to the process to push the product into spec. However, no attempt should be made to tamper with the process in an attempt to bring it back into the capable realm. Instead, a systematic exploration should be undertaken into the cause of incapability to determine how to improve the process.

Figure 5.5. A stable but incapable process

Unstable and Incapable Process

An unstable and incapable process produces subgroup means that are outside control limits and produces products outside of the specification limits. Such a process must be stopped immediately and brought under control before production can resume. The out-of-spec problem may disappear when the process is brought under control, because the variance will decrease. (High variance in the process outcome may be the sole reason that the out-of-spec subgroups occurred.) However, if the problem persists, an investigation must take place to find the cause of out-of-spec product. On the control chart in Figure 5.6, two rules have been violated. Rule 1 has been violated by subgroups 7, 13, 14 and 19; Rule 3 has been violated by subgroups 16-20. Also, the process has not stayed within specification limits consistently.

Figure 5.6. An unstable and incapable process

Capability Index
A capability index is a summary statistic that quantifies information on the specification limits of a product and the inherent variability of the process. There are many different capability indices. Most are a ratio of some form of allowable variation. The index that looks at the natural spread of the process is called the process capability ratio (PCR), symbolized Cp. The allowable variation is divided by the total variation expected in the process (3 sigma above and 3 below the process mean).

Here, s is the estimate of standard deviation of the individual observations (not the standard deviation

of

as on the control chart); USL is the upper specification limit; LSL is the lower specification limit.

Calculating Capability Indices

Consider the task of monitoring oxide thickness of a wafer: s =10 angstroms USL = 175 angstroms LSL = 140 angstroms

The purpose of the index is to compare the allowable spread of product variability (USL LSL) with the natural spread (common cause variation) that would normally occur (two 3-sigma lines = 6s or 6 standard

deviations).

The allowable spread (USL LSL = 35) is less than the natural or existing spread (6s = 60). This means that the process often produces product with out of spec oxide thickness. In summary, a Cp less than one suggest that a process will produce out of spec product. When the Cp is one or greater than one, the process is potentially acceptable and can produce in spec product if properly centered. Further investigation using other capability indices will provide more information about the capability of the process.

Summary
Four scenarios describe the state of a process. Each scenario is determined by stability and capability. Stability indicates whether or nor a process is in control, and capability refers to whether or not individual measurements are within specifications. The four scenarios are: Stable and Capable - The process is in an acceptable state: it is in control and producing good product within specifications. Unstable but Capable - The process is not in control (not stable) but is producing good product. This process will tend to become incapable in time. Stable but Incapable - The process is in control but not producing good product. This process needs to be improved to make it capable. Unstable and Incapable - The process is in the worst possible condition: it is not in control and not producing good product. The process must first be brought under control and then adjusted to improve product output quality.

The capability index provides information about the performance, or capability, of a process by quantifying information on a product's specification limits and the inherent variability of the process. Usually, the

capability is presented as a ratio that can be determined with through a formula such

as