You are on page 1of 38

DESSERTATION REPORT On Study of reverse supply chain management in passenger car companies

!y A"shay !in#ra#"a A$%$%&%'%( )!A*+

To )r, Sumeet -aisial Asst, prof, Operations management Amity .usiness school

/0APTER*%1 INTROD2/TION Companies are being forced to set up reverse supply chains because of environmental regulations or consumer pressures. Beginning in 2003, for example, European Union legislation ill re!uire tire manufacturers operating in Europe to arrange for the tire they sell. "n other cases, companies are recycling of one used tire for every ne

ta#ing the initiative, seeing opportunities to reduce their operating costs by reusing products or components. $or instance, %oda# remanufactures its single&use cameras after the film has been developed. 'ver the past decade, the company has recycled more than 3(0 million cameras in more than 20 countries. $inally, some companies are using reverse supply chains as integral parts of ne remanufactured. )hether a company is establishing a reverse supply chain by choice or necessity, it ill face many challenges. "t of contact ith them, decide in general figure out ho ill have to educate customers and establish ne hich activities to outsource and points hich to do itself, and businesses. Bosch, for instance, has built a successful business selling po er hand tools that have been

to #eep costs to a minimum

hile discovering innovative

ays to recover value. "t may also have to meet stringent environmental standards. *o ma#e rational decisions about the structure of a reverse supply chain, it+s best to divide the chain into its five #ey components and analy,e options, costs, and benefits for eachProduct Ac3uisition, 'ur research suggests that this tas#.retrieving the used product.is #ey to creating a profitable chain. *he !uality, !uantity, and timing of product returns need to be carefully managed. 'ther ise, companies may find themselves flooded ith returned products of such variable !uality that efficient ill need to or# closely ith remanufacturing is impossible. Companies often

retailers and other distributors to coordinate collection. Reverse 4ogistics, 'nce collected, products need to be transported to facilities for inspection, sorting, and disposition. *here is no one /best0 design for a reverse logistics net or#1 each has to be tailored to the products involved and the economics

of their reuse. Bul#y products li#e tires, for instance, only the costs for shipping and storing but also ho ma#e sense to outsource the logistics to a specialist.

ill re!uire very different

handling than small but fragile products li#e cameras. Companies should consider not !uic#ly the value of the returned products ill decline and the need for control over the products. "n many cases, it ill

Inspection and Disposition, *he testing, sorting, and grading of returned products are labor&intensive and time consuming tas#s. But the process can be streamlined if a company sub2ects the returns to !uality standards and uses sensors, bar codes, and other technologies to automate trac#ing and testing. "n general, a business should see# to ma#e disposition decisions.based on !uality, product configuration, or other variables.at the earliest possible stage in the returns process. *hat can eliminate many logistics costs and get remanufactured products to mar#et faster. Reconditioning, Companies may capture value from returned products by extracting and reconditioning components for reuse or by completely remanufacturing the products for resale. 3econditioning and remanufacturing processes tend to be much less predictable than traditional manufacturing because there can be a large degree of uncertainty in the timing and !uality of returned products. 4gain, ma#ing smart decisions early in the chain.in particular, hen you accept and sort returns. ill help to reduce manufacturing variability and, hence, costs. Distri.ution and Sales, "f a company plans to sell a recycled product, it first needs to determine hether there is demand for it or hether a ne mar#et must be created. "f it+s the latter, the company should expect to ma#e heavy investments in consumer education and other mar#eting efforts. 5otential customers for remanufactured products or components include not 2ust the original purchasers but also ne customers in different mar#ets. *he company may, for example, customers ho cannot afford the ne products but ho buy used versions at lo er prices. ant to target ould 2ump at the chance to

"n general, the companies that have been most successful chains are those that closely coordinate them creating hat and manufacturing decisions indicate hether the motor is

ith their reverse supply

ith their for ard supply chains,

e call a closed&loop system. $or example, they ma#e product design ith eventual recycling and reconditioning in mind. hich orth reconditioning. *he technology dramatically ith reverse supply chains, for ard thin#ing pays big

Bosch is a good example. "t builds sensors into the motors of its po er tools,

reduces inspection and disposition costs, enabling the company to ma#e a profit on the remanufactured tools. Even dividends. 3emanufacturing is considered the ultimate form of recycling and is sometimes referred to as a 6 in& in& in+ situation compared to traditional manufacturing since the customer pays less, the remanufacturing companies earn more and the environment benefits from less usage of ra materials and energy. *he benefits of ho found that products, are remanufacturing have been put into figures by 7iutini and 7audette, remanufactured products cost 809 to :;9 less to produce than ne

typically 309 to 809 cheaper for the customer to buy, and save globally the energy e!uivalent of (: million barrels of crude oil annually. "n a study by <undin and =ee, it as noted that (( of (2 environmental research studies found remanufacturing as a preferable option, at least in comparison to ne manufacturing. *he remanufacturing industry has gro n recently, and to date, close to 8,000 establishments are confirmed in the U<4 alone, ith more than ((0 #no n product

areas. *he automotive part remanufacturing industry is roughly estimated to be a >?; to >(00 billion dollar industry 4 fairly ne and interesting area orld ide, here the value of the remanufactured parts as estimated to be >80 billion in the U<4 in 200@. ithin remanufacturing in the automotive industry concerns mechatronics Ae.g. po er steering systems, central loc#ing systems and anti& loc# bra#ing systemsB and electronic systems Ae.g. engine control units and distance control unitsB communicating via a Controller 4rea Cet or# AC4CB bus. Developed by Bosch in (@?3, C4C is a serial communication bus designed to provide robust, simple and efficient communication for in&vehicle net or#s. *he rapid gro th in complexity of automotive electronics in the follo ing decades made traditional point& to&point iring increasingly expensive to manufacture, install and maintain1 hence, C4C soon became adopted by all car manufacturers, resulting in a sales gro th of C4C nodes from merely ;0 million in (@@@ to more than 380 million in 2003. 7iven the many C4C nodes in a modern vehicle and the high costs associated ith replacing a malfunctioning device, ranging bet een 200 and 3,000 E, there is an economic incitement for remanufacturing such devices and an opportunity not yet exploited. *his research as part of a research pro2ect called 6C4C 3EF4C+, ith the target to as develop innovative diagnosis methods and technologies for automotive mechatronics and electronic remanufacturing. *he pro2ect, funded by the European Union, conducted by Bayreuth University A7ermanyB, =in#Gping University A< edenB, the University of 4pplied <ciences Coburg A7ermanyB, $raunhofer 5ro2ect 7roup 5rocess "nnovation A7ermanyB and eight European small and medium&si,ed enterprises A<FEsB. *he aim of this paper is to identify previously un#no n, and verify #no n, reverse logistics challenges experienced by <FEs that are about to remanufacture or trade automotive mechatronics and electronic systems communicating through the C4C bus system. Reverse logistics challenges * in theory 3emanufacturing differs from traditional one& ay manufacturing in several *hese differences are also associated ays.

ith manufacturing challenges and are

necessities to reali,e a successful remanufacturing system Ai.e. core ac!uisition, remanufacturing process and redistribution.*hese challenges have been recogni,ed by

researchers, but also have been summari,ed by =undmar# et al. "n the literature about challenges

ho also categori,ed

them according to here and hen in the remanufacturing system they occur. ithin the collection phase of the remanufacturing system, there is a lot of emphasis on a lac# of control regarding !uantity, !uality and timing of the returned products. *his lac# of control is recogni,ed by, for example, and is caused byH 3eflection of the uncertain life of a product H 5roduct life cycle stage and the rate of technological change H *he dispose behaviour, hich results in a stochastic return pattern *he lac# of control regarding !uantity, !uality and timing of the returned products AcoresB is described as the ma2or difference bet een a traditional production& distribution net or# and a product recovery net or# .*he handling of these control issues is stated as the #ey for creating profitable remanufacturing by 7uide and Ian )assenhove ."n addition, there is also uncertainty regarding the demand of the remanufactured products. *his uncertainty is caused by the follo ingH *he rate of technical development. *he demand for a product might suddenly drop due to the technical development . H Detailed forecasting is not possible to perform due to uncertainties regarding timing and !uantities of the eturned products. "n order to maximi,e profit, a remanufacturer must be able to balance the return of cores ith the demand from customers for remanufactured products. "f not, the levels of customer service A hen demand exceeds remanufacturer faces the ris# of building up excessive amounts of inventory A hen returns exceed demandB or lo supplyB . *he uncertainties in supply and demand ma#e it hard for many remanufacturing companies to balance supply and demand .4ll companies do not try to balance the supply ith demand since the uncertainties in supply and demand ma#es inventory management and control functions more complicated . *he #ind of motivation for returns could also affect the situation for the remanufacturing company since a ta#e&bac# obligation might give the remanufacturing company an abundance of used products. 4 survey conducted ith 8? remanufacturing companies by .sho ed that more than half of the companies had no control over the timing or the !uantity of the returns. *he remanufacturing companies that do not try to balance supply ith demand instead dispose excess used products on a regular basis . Excess used products might cost a lot of money, and the disposal cost might be high .*he storage

area needed to store the excess used products is also often expensive .4nother challenge is that a remanufacturing firm typically has a large number of sources hich means that a remanufacturing firm has to bring together a large number of small volume flo s hich increases the complexity . 4ccording to =undmar# et al. the uncertainties regarding !uantity, !uality and timing of the returned products are the main challenges for the collection phase of the remanufacturing system. *he uncertainty in timing and !uantity of the returned products also ma#e the remanufacturing process less predictable than an ordinary manufacturing process .*his uncertainty ma#es production planning more difficult .*he uncertainty in !uality adds challenges to the remanufacturing process in t o different ays. * o returned products AcoresB that are identical might yield a very different set of remanufacturable parts and purchasing more difficult . 3everse supply chains comprise the activities, routes, intermediaries, etc. transporting products in the opposite hen ay compared to for ard supply chains, i.e. hich ma#es inventory planning and control

from the customer or end user, via possible intermediaries to the remanufacturer, as can be seen in $igure (. "ntermediaries can, for instance, be retailers and repair shops in the for ard supply chain and core bro#ers and scrap yards in the reverse supply chain. Jo the reverse supply chains are designed, i.e. ho used products are brought bac# to the remanufacturer, plays an essential part in the remanufacturing system as a hole since it is of most importance to receive the right cores in the right !uantities at the right time in order to be able to perform successful and profitable remanufacturing. *he designing of reverse supply chains is a delicate procedure chain. *he number and type of cores Amaterial, value, si,e, ith many variables and aspects to ta#e into consideration, and there exists no general optimal supply eight, etc.B that are to be transported highly influence the design of the reverse supply chain. "n addition, types of core suppliers, types of core ac!uisition and relationship bet een remanufacturer and supplierKcustomer have a great impact.

/ore ac3uisition 4 remanufacturer of automotive components normally has the possibility to choose from several core suppliers, as can be seen in the next section. *he choice of supplier does not, ho ever, automatically set the guidelines and rules on ho choices. "t is common that remanufacturersL differH Direct&order- in this situation, the supplier who also is the customer gives an order
for the remanufacturing of a used product. The supplier/customer sends the core to the remanufacturer, which, after being remanufactured, is sent back to the supplier/customer. Within the scope of this research, there is a tendency that this type of acquisition is common when remanufacturing relatively complex products such as engines and more likely towards end users. In addition, it is common that the customer/supplier is responsible for the transportation to the remanufacturer.

the core

ac!uisition is made but may enforce a certain ac!uisition type or open up for multiple ays to ac!uire cores vary bet een ithin the same type may different types of core suppliers, but even the ac!uisitions

H 3eman&contract- this type of transaction is some hat similar to direct&order since the supplier, hich also is the customer, gives an order for remanufacturing. 4lso, the o nership of the core and the remanufactured product remains at the customer. Jo ever, this type is guided by a contract and spans over a longer time, ith closer collaboration bet een remanufacturer and customerKsuppler and also involves greater !uantities. )hile direct&order is common to ards end users, reman&contracts are more commonly used in collaborations ith original e!uipment manufacturers A'EFsB. H Deposit&based- this means that fre!uent hen the customer buys a remanufactured product,

the customer is obligated to return a similar used product. *his type of transaction is ithin automotive remanufacturing and in particular concerning components that are cheap and often exchanged at services Ae.g. bra#e calipersB. H Credit&based- the customer receives credits for returning a core, as a discount in this case. H Buy&bac#- the remanufacturer buys the core. 'ne could say that the characteristic of this type of ac!uisition is the lac# of relationship since neither the supplier nor the hich can be used

hen buying a remanufactured product. *he supplier is also a customer

buyer has any further obligations after the transaction is made. "t is common that remanufacturers buy cores from core bro#ers or scrap yards but could possibly be end users as ell. *he supplier is, in this case, seldom the customer. Types of core suppliers *his section describes the different core supplier types that have been identified ithin this research. "t is a some hat simplified picture given since a classification is necessary in order to get a uniform vie . "t is note orthy to observe that the reverse supply chains depicted are sources of used automotive components AcoresB, not virgin spare parts, etc. "n a sense, end users normally supply all cores, but in this report, the categorisation and the type of supplier derive from the supplier closest do n in the supply chain. *he identified core suppliers are end users, scrap yards, core bro#ers, 'EFs and independent aftermar#et distributors A"4FDsB. Reverse supply chains in the /AN RE)AN pro#ect 4 !ualitative study of six <FEs, of hich their primary, if not solely, business ith, automotive mechatronic here the latter is a core segments are the remanufacturing of, or trading

devices. $our of the companies studied are 7erman remanufacturers Acompanies 4 to DB. *he remaining t o are < edish Acompanies E and $B, in the reverse supply chain. 4n overvie bro#er and hence not conducting any remanufacturing but is still an important factor of the participating companies can be seen in *able(. "n a parallel C4C 3EF4C research study on inter&organisational relationships, companies 4 to E are studied further ith the same notation . Ta.le %, +eneral characteristics of the participating companies *he follo ing sections present reverse logistics challenges identified and verified during intervie s ithin the C4C 3EF4C pro2ect. 4n overvie of the challenges experienced by the intervie ed companies can be seen in.. 4n empty space in the table means that the company does not consider the specific sub2ect as a challenge or that it has chosen not to ans er. Ta.le ', Overvie5 of reverse logistics challenges identified 5ithin /AN RE)AN /ontrolling 3uantity of cores Company 4 is sometimes forced to buy larger !uantities than needed by its customers since core suppliers re!uire minimum order !uantities. <trong competition among core ac!uisition for ne article numbers forces company B to buy cores ithout

having a concrete demand. Company D experiences difficulties ac!uiring a sufficient amount of cores for its fast&moving product variants. 4lthough the lac# of control regarding !uality is considered a challenge, both < edish companies AE and $B accept it as a part of their business. /ontrolling 3uality of cores 4ccording to company E, the challenge of controlling the !uality of the returned cores differs bet een types of products. "t claims to have total control of the !uality of the returned bra#e calipers. *here are t o reasons for that. $irstly, bra#e calipers have a simple design and are made out of a robust material but also contain fe parts, ma#e units hich orn beyond their remanufacturability infre!uent. <econdly, because of

their design, it is easy to visually determine hether a unit is remanufacturable or not. 3egarding servo pumps, the situation is different since one cannot determine a coreLs condition ithout disassembling it. "nternal parts may be bro#en beyond remanufacturing, and it is not feasible to have every single core opened before they arrive at the facility. 4nother !uality&related challenge concerns the handling of the cores. Company E estimates that :09 to M09 of the initially sound diesel particulate filters AD5$sB from one of its core suppliers are being damaged hen dismounted from the cars or hen transported to company ELs remanufacturing facility. *he damages are assumed to be caused by the mechanics at the 'EF dealers as thereby handle them in an incautious ell as by the companies responsible for the transportation of the cores & they do not recogni,e the value of the cores and ay. *his is especially critical for D5$s since rec#less handling and transportation can lead to damage not visible to the na#ed eye, hence first noticed in the remanufacturing process. *he effects are increasing scrap rates and that cores originally suited for remanufacturing can no longer be remanufactured. *his issue is also being referred to as a challenge by company B. Company C uses chec# sheets that private customers and car dealershipsKrepair shops must fill out in advance to avoid receiving non&remanufacturable cores and cores too costly to remanufacture. *his improves !uality control, but does not eliminate the need for !uality inspections at arrival.

Company $ considers the lac# of control of the coresL !uality as a challenge, but in contrast to the remanufacturers, @09 of the cores are visually inspected before they are bought, ma#ing the degree of control higher in this case. /ontrolling timing of cores *he lac# of control regarding the timing of the returned cores is considered a challenge by companies E and $, but only if there is a lac# of the specific core in the mar#et. Company B controls the timing of deliveries for all supplier types except for private customers, ho are harder to control. Company C lac#s control of the core deliveries from 'EFs, but the use of chec# sheets Asee previous sectionB for private customers and car dealershipsKrepair shops facilitates influence over the timing of the incoming cores from those suppliers in a positive ay. !alancing supply and demand Companies E and $ experience a lac# of control regarding balancing supply and demand, but their situation differs from each other. Company E usually gets a (2& month time hori,on on the estimated demand from the contracted 'EFs, hich are core suppliersKcustomers that company E is contracted to perform remanufacturing for. $rom these forecasts, additional cores are bought if needed from other suppliers, e.g. scrap yards and core bro#ers. <tatistics from previous years are also #ept, thereby helping to distinguish trends in demand. *he prere!uisites are similar for company B. Company $ has much shorter time hori,ons, usually only a ee# or a month ahead. 4 different aspect on the issue of balancing supply and demand is ho buys certain cores ithout having a concrete demand but believes it company $ ill rise in the

future. *he motive is t ofold1 cores that are bought prior to the actual demand are cheap, hence displaying large profit margins if later sold. "n addition, buying pre& demand cores prevents or diminishes competition the speculating comes ithin core ac!uisition. Jo ever, ith ith a price tag. 4part from the obvious cost associated

storing cores until demand arises, there is an imminent ris# that the anticipated demand never occurs, thus ma#ing the ac!uired cores less orth. *his situation especially concerns C4C bus mechatronic devices that have yet to be remanufactured.

Additional challenges 5ithin the collection phase 4 challenge identified by company E concerns the identification and sorting of cores at its suppliers, hich, in this case, are 'EF retailers. *he deliveries often contain here the personnel, even un anted parts and mechatronic devices Ae.g. turbochargers and dashboardsB. *his is believed to be caused by a lac# of routines at the retailer though having specified hich parts ere to be sent for remanufacturing, do not put ell as to store the un anted goods

the right parts in the right core bins. *he effect on company E is that arehouse space is allocated, both to store the unsorted goods as before being transported a ay for scrapping. "t is also time&consuming to perform the sorting. Jo ever, this identification issue is not considered as severe and can sometimes even provide useful information about ne the remanufacturing program. 4 challenge ac#no ledged by company $ is the decreasing number of scrap yards, hich are core suppliers to company $. 'ne plausible explanation for the decrease is that insurance companies certify fe er scrap yards no than before. *he conse!uence for company $ is t ofold. $irstly, fe er suppliers result in less competition among those remaining and hence higher core prices. <econdly, but not less important, the fe er larger suppliers remaining tend to focus their business on scrapping cars rather than dealing ith cores. *he second challenge affecting company $ is the recently increasing scrap prices, hich follo the prices of ra materials, hich, in this case, are metals. )hen scrap prices are high, scrap yards ould rather sell the dismantled cars as scrap than sell the individual mechatronic devices to core bro#ers or remanufacturers. 4 further challenge for company $ is the competition from lo &labour cost countries. )hen there are brand&ne spare parts available at a cost close to, or even cheaper, than a remanufactured part, the demand for remanufactured parts decreases. *his is also ac#no ledged by companies B and E. "n addition, company $ has experienced actions from 'EFs of ne here they have tried to part numbers currently not in

hinder competition from core bro#ers and remanufacturers, either by dumping prices parts or by clean&s eeping the mar#et from cores. <imilar actions have been here 'EFs refuse to sell spare parts needed to perform ta#en against company D,

remanufacturing of their products. Company B experiences that 'EFs delay technical information Ae.g. test parametersB about the products on purpose, hich, due to the effort, put into reverse engineering results in higher remanufacturing costs, hich is further elaborated on in . E6pected reverse logistics challenges of /AN .us mechatronic devices 4 concern for companies B, D, E and $ is the handling of cores containing C4C bus mechatronic devices. *hree perspectives on this matter have been brought up during the intervie s. *hese are the follo ingH Disassembly- it is important that 'EF retailers and scrap yards have routines for ho to remove parts ithout damaging them. H <torage- cores containing electronics are sensitive to moisture1 hence, it is important that C4C bus cores are stored in dry and preferably arm environments. H *ransportation- there is a concern that cores not being handled and stored properly during transportation bro#er. 4 challenge company $ ill be facing, and is currently facing to a certain extent, is the large gap bet een the sales price of the remanufactured product and the cost of the core. $or example, a remanufacturer buys a used ECU for >3 to >; from the core bro#er hich is then sold remanufactured for up to >;00. $or a core bro#er to sell a hich is scrap price. *his core for >3 to >;, it has to be bought for >( to ma#e profit, automotive electronics. During the intervie ith company $, it as mentioned that dealing ith electronic ill be damaged on their ay to the remanufacturer or core

is claimed to be caused by the lac# of competition among remanufacturers of

mechatronic devices ill further complicate the supply versus demand challenge since an entirely mechanical device lifetime. "n addition to the findings of this paper, another paper by $reiberger et al. N?O outlines challenges, possible solutions and technological progress for the reverse engineering process of C4C bus mechatronic devices. *hat paper includes the reverse engineering process demonstrated on an electro&hydraulic po er steering, mechatronic device used in a Iol#s agen 5olo . hich is a C4C bus ill surely fail during a carLs life, hile the fail pattern of an electronic device is much more stochastic & it might even last the vehicleLs entire

)ith expected sales of P2.; million passenger vehicles in $Q((e,"ndia+s passenger vehicle mar#et ran#s as orld+s seventh largest1 larger than mar#ets li#e United %ingdom, $rance and <pain by volume & "ndia has been one of the fe be on trac# mar#ets globally to buc# the recessionary trend and

record a strong 2;.:9 volume gro th in $Q(0. *he gro th momentum continues to ith first eleven months of $Q(( registering a gro th of 2@.?9 over the corresponding period in the previous year & <trong economic gro th, rising disposable income levels, favorable demographics, easy financing environment and relatively lo prominent gro th drivers for the industry & )hile at the one end, the gro ing domestic mar#et is attracting foreign 'EFs, on the other, established players are positioning themselves as strong contenders to offer lo &cost car manufacturing capabilities to the orld & <o far, most foreign car ma#ers, barring Jyundai have focused on the sedan and premium segment cars, shying a ay from the highly competitive small&car segment1 ith these players no launching small&cars that too designed #eeping in mind hich has so far been specifically the "ndian consumer, the small&car segment, increase ie in competitive intensity & <ome of the ne ly launched models have had good initial response and have been aggressively priced, indicating ne sacrificing profitability & =arge established incumbents in the "ndian passenger vehicle mar#et derive strength from their lo &cost manufacturing capabilities Aespecially in the small&car segmentB, strong brand recognition and can be difficult to replicate . & )e believe, hile the incumbents ill have these competitive advantage over ne er entrants, these are li#ely to diminish in the long&run as ne players ith global experience gain brand recognition and expand their net or# and product offerings ide distribution R servicing reach, something hich entrants+ strategy to grab mar#et share hile car penetration have been the

dominated by three players commanding over ?09 of the volumes is li#ely to see

& <uperior small&car portfolio, a

ide distribution and service net or# and

competitive pricing on the bac# of locally sourced auto components are going to be the #ey factors in determining the success of a foreign 'EF in the "ndian mar#et )hile competitive pressures are li#ely to intensify, e believe that strong 7D5 gro th, rising disposable income levels, easy availability of finance and more particularly "ndian consumers+ aspiration to o n cars, especially given the state of public transport, foreseeable future & )e estimate the "ndian passenger vehicle industry ill reach P8.?; million in annual sales by $Q(:, representing a gro th of (0.?9 C473 over the next five years & Cot ithstanding the strong long&term outloo#, the industry faces certain near term challenges in form of rising commodity prices, interest rates, tightening li!uidity scenario and increased competitive intensity & )e believe that rising labour costs is also li#ely to see cost increases across the supplier net or#, though it is li#ely to be mitigated by greater scale economies and higher degree of automation & )ithin the lo er priced segment AminiKcompactB, the price band is lo er priced models. *he price range may passenger vehicle mar#et *he domestic passenger vehicles industry has been on a relatively steady gro th phase over most of the last decade and has registered a (0 years C473 of (0.39 during the period. "t has been one of the fe mar#ets orld ide hich sa gro ing passenger car sales itnessed during $Q0@. Buoyant ith improving availability of during the li!uidity crisis and recessionary phase idening, ith ould ensure that the industry ill experience strong gro th in the

higher priced but better value products achieving higher volumes than some of the iden further depending on the success of the 6Cano+ segment <trong gro th drivers augur favourable prospects for the "ndian

economic gro th, rising disposable income levels, favourable demographics, strong gro th from tier ""K""" cities and rural "ndia, together vehicle financing at competitive interest rates have been the #ey factors fuelling

gro th in the "ndian passenger vehicle mar#et. 4mong the emerging mar#ets, "ndia continues to have one the lo est car density, estimated at (3 cars per (,000 people compared to other mar#ets such as China A8;B, Bra,il A(:0B, and "ndonesia A82B. *he gro th has also been supported by 'EF led initiatives li#e hole host of ne model offerings from both from existing companies as ell as ne entrants in the mar#et. $urthermore, in "ndia, the car prices have remained relatively flat over the years Aad2usted for the decline in dutiesB compared to steadily rising per capita income levels. "n addition to the strong domestic demand, the 'EFs have also been positioning.<teady economic gro th and favourable demographic profile S Barring marginal blips during the last couple of years, the "ndian economy has moved into higher gro th A?.;9TB tra2ectory hich is li#ely to be sustained over the medium term. "n addition to steady economic gro th, the passenger vehicle industry is also benefitting from "ndia+s favourable demographic profile, hich is reflected by its very young population A;09 of population under the age of 2;B, steadily improving dependency ratio, gro ing urbani,ation and trend to ards smaller, nuclear families. *hese trends in turn results in higher savings and increased ability to purchase vehicles, as ell as explaining the preference for smaller&cars. "n addition to favourable demographic profile, rising per capita 7D5 levels is also resulting in improvement in vehicle affordability in "ndia, amongst the lo est hich is estimated to hile car hen compared to other ma2or automotive mar#et. "n "ndia, the

per capita 7D5 has almost doubled to U<> 3,2M0 bet een 2000 and 200@, they ere five years bac#, thereby increasing flexibility to o n cars.

prices Aad2usting for the decline in dutiesB have remained almost at the same level as

3elatively lo &penetration levels S "n terms of current mar#et si,e Aestimated at P2.; million units in $Q((eB, the "ndian passenger vehicle mar#et is relatively small compared to other emerging auto mar#ets li#e China, <outh %orea and Bra,il. Despite strong gro th itnessed for a nearly a decade, penetration of cars in "ndia continues to remain the lo est Arefer to table aboveB among emerging mar#ets. 4s gro th in passenger vehicle has been more secular in nature, supported by both ma2o rcities and tier ""K""" cities, e expect that car penetration levels ould continue to improve

mirroring the trend itnessed by some of the other mar#ets, particularly China, hich itnessed ;x increase in car density bet een 2002 and 200@. 4vailability of finance at competitive rates S )ith over :;&M09 of cars being financed in "ndia, availability of financing options at competitive rates has also been one of factors driving gro th. "n "ndia, the vehicle financing penetration has been steadily rising over the years, facilitated by competition amongst ban#ing and CB$C participants. "n comparison to China, combined here vehicle penetration rates are much lo er hich AP(0&(;9B, "ndia scores in terms of higher vehicle financing availability, strong gro th going for ard. Barring fe

ith increasing disposable income levels provides an ideal platform for instance of rise in interest rates, vehicle

financing cost has declined over a longer period of time supported by favourable interest rate regime and relatively healthy performance of the asset class amongst various consumer finance categories. *his has also encouraged lengthening of tenure of financing and =*Is, further facilitating consumer flexibility. $avourable demand scenario from smaller to ns and rural areas S "n addition to demand from urban areas, smaller to ns and rural "ndia have been incrementally driving demand for passenger vehicles in "ndia. $or instance, the share of sales from top&(0 cities has fallen to 80&8;9 from :09&:;9 over the last five&to&six years. Faruti <u,u#i, also for instance no generates nearly (@9 of its sales from non&urban areas compared to 2ust 8&;9 about five years bac#. *his has largely been prompted by rising disposable income levels in smaller to ns and rural areas, improving road connectivity and higher no. of earning members in the family. "ndustry estimates suggest that approximately :09 of the rural economy no *hat apart, substantial increase in crop prices, depends on non&agricultural income hich has been moving up over the such as trading, remittances from cities, employment in the manufacturing sector etc. past three years, has also resulted in higher disposable income. 4dditionally, the increase in land prices across the country, and the implementation of the sixth pay commission has collectively helped in supporting the gro th in the rural and semi&urban citiesKtier """ cities. *he 'EFs have also helped expand demand by targeting these mar#ets distribution reach. ith greater financing availability and better service R

/0APTER '1 4iterature revie5 "n supporting environmental sustainability, managing product returns has become a very important and challenging issue. 3esponding to this trend, researchers in many parts of the orld have conducted numerous studies in reverse logistics and reverse ere considered separately from the for ard channel of supply hen both channels are considered in an integrated supply chain, that

chains. Fean hile, there are opportunities to create added values from product returns and to improve efficiency ay, as a closed loop supply chain. 'n the other hand, remanufacturing is a very appealing option among product recovery processes that has potential to increase the value of product returns. Jo ever, the available literature and theories on closed loop supply chain aliterature revie ith remanufacturing are still limited. *his paper aims to provide that covers closed loop supply chain ith remanufacturing to or#s in this area. *he revie ill be mapped under the ill presented as a

identify its characteristics based on previous aspects

methodology iscontent analysis, classified into managerial aspects and technical ith their subse!uent topics. 3elated literatures above mentioned structures result of the revie . Birds flap their ings in oil&infested coastal aters, unable to ta#e off. *rees in a hile identifying other important aspects that have

potentials for further study. 4n outline of research opportunities

coniferous forest have lost all but a handful of their needles. 4n empty landscape is scarred by open&cast mining. *hese are 2ust some of the images that have brought home the enormous costs to the natural environmental of industrial development . 5reuss A200;, p. (B. <ustainable development has made remar#able progress in establishing environmental and social sustainability to ards operations management and the supply chain. Fanufacturers in the late (@?0+s ent beyond hat as re!uired in the legislation and ent for a greener approach in their operations systems. *he purpose of this paper is to briefly revie the literature of the green supply chain management A7r<CFB over aste the last t enty years. *he #ey themes that came out of the literature are- green operations, green design, green manufacturing, reverse logistics and

management. *his paper ill briefly discuss these issues. *his paper ill also mention the reasons hy organisations ould opt for a green supply chain. *his essay ill then conclude that there is a gap in the literature in terms of the sta#eholder vie s to ards green supply chain management. %no ing the different sta#eholders+ vie s to greening initiatives re!uires !ualitative study such as intervie s S to portray the different vie s about the concept and ho decisions. <upply chain management has traditionally been vie ed as a process herein ra this ould implicate management

materials are converted into final products, then delivered to the end&consumer ABeamon, (@@@B. *his process involves extraction and exploitation of the natural resources A<rivastava, 200MB. "t is important to note ho ever that <ince the early (@@0+s, manufacturers have been faced e live in a decade here environmental sustainability has been an important issue to business practice. ith pressure to address Environmental Fanagement AEFB in their supply chains A)u R Dann, (@@;B. *his is not an easy tas# to do ho ever. 4dding the 6green+ concept to the 6supply chain+ concept adds a ne paradigm here the supply chain head&on collision ith each other A<rivastava, 200MB. 7reen supply chain management A7r<CFB is an emerging field that strands out of the traditional supply chain perspective. *he /!uality revolution in the late (@?0+s and the supply chain revolution in the early (@@0+s0 have spar#ed businesses to become environmentally conscious A<rivastava, 200M, p. ;3B. 7r<CF has gained popularity ith both academics and practitioners to aim in reducing aste and preserving the !uality of product&life and the natural resources. Eco&efficiency and remanufacturing processes are no important assets to achieve best practice A4shley, (@@31 <rivastava, 200MB. 7lobal mar#et demands and governmental pressures are pushing businesses to become more sustainable A7uide R <rivastava, (@@?1 7ungor R 7upta, (@@@B. )alton, Jandfield and Felyny# A(@@?, p. 2B even claim that/increasing government regulation and stronger public mandates for environmental accountability have brought these issues into the executive suites, and onto strategic planning agendas.0 years are the aste management and *he #ey themes that came out in the literature over the last t enty concepts of- green design, green operations, reverse logistics, ill have a direct relation to the environment. *his is interesting because, in history, these t o paradigms ere once in

green manufacturing A7uide R <rivastava, (@@?1 U<rivastava, 200MB. *he very first green supply chain came into context in (@?@. %elle and <ilver+s A(@?@B article organisations to use to forecast products that can be potentially be reused. as the first of this literature that developed an optimal forecasting system for

/0APETER*71 Research methodology T0E OPERATIONS RESEAR/0 APPROA/0 7iven that '.3. represents an integrated frame or# to help ma#e decisions, it is important to have a clear understanding of this frame or# so that it can be applied to a generic problem. *o achieve this, the so&called '.3. approach is no detailed. *his approach comprises the follo ing seven se!uential steps- A(B 'rientation, A2B 5roblem Definition, A3B Data Collection, A8B Fodel $ormulation, A;B <olution, A:B Fodel Ialidation and 'utput 4nalysis, and AMB "mplementation and Fonitoring. *ying each of these steps together is a mechanism for continuous feedbac#1 $igure ( sho s this schematically.

8igure %- The Operations Research Approach )hile most of the academic emphasis has been on <teps 8, ; and :, the reader should bear in mind the fact that the other steps are e!ually important from a practical perspective. "ndeed, insufficient attention to these steps has been the reason hy '.3. has sometimes been mista#enly loo#ed upon as impractical or ineffective in the real orld.

Each of these steps is no

discussed in further detail. *o illustrate ho

the steps

might be applied, consider a typical scenario

here a manufacturing company is materials, capital, data

planning production for the upcoming month. *he company ma#es use of numerous resources Asuch as labor, production machinery, ra different products processing, storage space, and material handling e!uipmentB to ma#e a number of hich compete for these resources. *he products have differing profit margins and re!uire different amounts of each resource. Fany of the resources are limited in their availability. 4dditionally, there are other complicating factors such as uncertainty in the demand for the products, random machine brea#do ns, and union agreements that restrict ho the labor force can be used. 7iven this complex operating environment, the overall ob2ective is to plan next monthLs production so that the company can reali,e the maximum profit possible hile simultaneously ending up in a good position for the follo ing monthAsB. 4s an illustration of ho one might conduct an operations research study to address

this situation, consider a highly simplified instance of a production planning problem here there are t o main product lines A idgets and gi,mos, sayB and three ma2or limiting resources A4, B and C, sayB for hich each of the products compete. Each product re!uires varying amounts of each of the resources and the company incurs different costs Alabor, ra materials etc.B in ma#ing the products and reali,es different revenues hen they are sold. *he ob2ective of the '.3. pro2ect is to allocate the resources to the t o products in an optimal fashion. Orientation1 *he first step in the '.3. approach is referred to as problem orientation. *he primary ob2ective of this step is to constitute the team that ill address the problem at hand and ensure that all its members have a clear picture of the relevant issues. "t is orth noting that a distinguishing characteristic of any '.3. study is that it is done by a multifunctional team. *o digress slightly, it is also interesting that in recent years a great deal has been ritten and said about the benefits of pro2ect teams and that almost any industrial pro2ect today is conducted by multi&functional teams. Even in engineering education, team or# has become an essential ingredient of the material that is taught to students and almost all academic engineering programs re!uire team pro2ects of their students. *he team approach of '.3. is thus a very natural and desirable phenomenon.

*ypically, the team

ill have a leader and be constituted of members from various ill be affected by or have an effect upon the

functional areas or departments that

problem at hand. "n the orientation phase, the team typically meets several times to discuss all of the issues involved and to arrive at a focus on the critical ones. *his phase also involves a study of documents and literature relevant to the problem in order to determine if others have encountered the same Aor similarB problem in the past, and if so, to determine and evaluate hat as done to address the problem. *his is a point that often tends to be ignored, but in order to get a timely solution it is critical that one does not reinvent the heel. "n many '.3. studies, one actually adapts a solution procedure that has already been tried and tested, as opposed to developing a completely ne one. *he aim of the orientation phase is to obtain a clear hat should be the primary focus of the hat Aif anythingB understanding of the problem and its relationship to different operational aspects of the system, and to arrive at a consensus on pro2ect. "n addition, the team should also have an appreciation for has been done else here to solve the same Aor similarB problem. "n our hypothetical production planning example, the pro2ect team might comprise members from engineering Ato provide information about the process and technology used for productionB, production planning Ato provide information on machining times, labor, inventory and other resourcesB, sales and mar#eting Ato provide input on demand for the productsB, accounting Ato provide information on costs and revenuesB, and information systems Ato provide computeri,ed dataB. 'f course, industrial engineers or# in all of these areas. "n addition, the team might also have shopfloor ho has relationships ith several of the functional areas listed above. personnel such as a foreman or a shift supervisor and ould probably be led by a mid& level manager 4t the end of the orientation phase, the team might decide that its specific ob2ective is to maximi,e profits from its t o products over the next month. "t may also specify additional things that are desirable, such as some minimum inventory levels for the t o products at the beginning of the next month, stable desired level of machine utili,ation. Pro.lem Definition1 *his is the second, and in a significant number of cases, the most difficult step of the '.3. process. *he ob2ective here is to further refine the deliberations from the orientation phase to the point here there is a clear definition or#force levels, or some

of the problem in terms of its scope and the results desired. *his phase should not be confused ith the previous one since it is much more focused and goal oriented1 ho ever, a clear orientation aids immeasurably in obtaining this focus. Fost practicing industrial engineers can relate to this distinction and the difficulty in moving from general goals such Vincreasing productivityV or Vreducing !uality problemsV to more specific, goals. 4 clear definition of the problem has three broad components to it. *he first is the statement of an unambiguous ob2ective. 4long ith a specification of the ob2ective it is also important to define its scope, i.e., to establish limits for the analysis to follo . )hile a complete system level solution is al ays desirable, this may often be unrealistic hen the system is very large or complex and in many cases one must then focus on a portion of the system that can be effectively isolated and analy,ed. "n such instances it is important to #eep in mind that the scope of the solutions derived ill also be bounded. <ome examples of appropriate ob2ectives might be A(B Vto maximi,e profits over the next !uarter from the sales of our products,V A2B Vto minimi,e the average do ntime at customers.V *he second component of problem definition is a specification of factors that ill or#center W,V A3B Vto minimi,e total production costs at 5lant Q,V or A8B Vto minimi,e the average number of late shipments per month to ell&defined ob2ectives that ill aid in meeting these

affect the ob2ective. *hese must further be classified into alternative courses of action that are under the control of the decision ma#er and uncontrollable factors over hich he or she has no control. $or example, in a production environment, the planned production rates can be controlled but the actual mar#et demand may be unpredictable Aalthough it may be possible to scientifically forecast these ith reasonable accuracyB. *he idea here is to form a comprehensive list of all the alternative actions that can be ta#en by the decision ma#er and that Eventually, the '.3. approach optimi,es the ob2ective. *he third and final component of problem definition is a specification of the constraints on the courses of action, i.e., of setting boundaries for the specific actions that the decision&ma#er may ta#e. 4s an example, in a production environment, the ill then have an effect on the stated ob2ective. ill search for the particular course of action that

availability of resources may set limits on *his is one activity

hat levels of production can be achieved.

here the multifunctional team focus of '.3. is extremely useful ith a long list of all possible constraints hen

since constraints generated by one functional area are often not obvious to people in others. "n general, it is a good idea to start and then narro this do n to the ones that clearly have an effect on the courses of

action that can be selected. *he aim is to be comprehensive yet parsimonious specifying constraints. Continuing

ith our hypothetical illustration, the ob2ective might be to maximi,e ould be

profits from the sales of the t o products. *he alternative courses of action

the !uantities of each product to produce next month, and the alternatives might be constrained by the fact that the amounts of each of the three resources re!uired to meet the planned production must not exceed the expected availability of these resources. 4n assumption that might be made here is that all of the units produced can be sold. Cote that at this point the entire problem is stated in approach ill translate this into an analytical model. Data /ollection1 "n the third phase of the '.3. process data is collected ith the ords1 later on the '.3.

ob2ective of translating the problem defined in the second phase into a model that can then be ob2ectively analy,ed. Data typically comes from t o sources S observation and standards. *he first corresponds to the case here data is actually collected by observing the system in operation and typically, this data tends to derive from the technology of the system. $or instance, operation times might be obtained by time studies or or# methods analysis, resource usage or scrap rates might be obtained by ma#ing sample measurements over some suitable interval of time, and data on demands and availability might come from sales records, purchase orders and inventory databases. 'ther data are obtained by using standards1 a lot of cost related information tends to fall into this category. $or instance, most companies have standard values for cost items such as hourly age rates, inventory holding charges, selling prices, etc.1 these standards must then be consolidated appropriately to compute costs of various activities. 'n occasion, data may also be solicited expressly for the problem at hand through the use of surveys, !uestionnaires or other psychometric instruments.

'ne of the ma2or driving forces behind the gro th of '.3. has been the rapid gro th in computer technology and the concurrent gro th in information systems and automated data storage and retrieval. *his has been a great boon, in that '.3. analysts no have ready access to data that as previously very hard to obtain. ords, hy <imultaneously, this has also made things difficult because many companies find themselves in the situation of being data&rich but information&poor. "n other information from these sources is often very difficult. *his is one of the reasons even though the data is all present Vsome hereV and in Vsome form,V extracting useful information systems specialists are invaluable to teams involved in any nontrivial '.3. pro2ect. Data collection can have an important effect on the previous step of problem definition as ell as on the follo ing step of model formulation. *o relate data collection to our hypothetical production example, based upon variable costs of production and the selling price of each of the products, it might be determined that the profit from selling one gi,mo is >(0 and one idget is >@. "t might be determined based on time and or# measurements that each gi,mo and each idget respectively re!uires MK(0 unit and ( unit of resource (, ( unit and 2K3 unit of resource 2 and (K(0 unit and (K8 unit of resource 3. $inally, based upon prior commitments and historical data on resource availability, it might be determined that in the next month there ill be :30 units of resource (, M0? units of resource 2 and (3; units of resource 3 available for use in producing the t o products. "t should be emphasi,ed that this is only a highly simplified illustrative example and the numbers here as ell as the suggested data collection methods are also vastly simplified. "n practice, these types of numbers can often be very difficult to obtain exactly, and the final values are typically based on extensive analyses of the system and represent compromises that are agreeable to everyone on the pro2ect team. 4s an example, a mar#eting manager might cite historical production data or data from similar environments and tend to estimate resource availability in very optimistic terms. 'n the other hand, a production planner might cite scrap rates or machine do ntimes and come up final estimate ith a much more conservative estimate of the same. *he ould probably represent a compromise bet een the t o that is

acceptable to most team members.

)odel 8ormulation1 *his is the fourth phase of the '.3. process. "t is also a phase that deserves a lot of attention since modeling is a defining characteristic of all operations research pro2ects. *he term VmodelV is misunderstood by many, and is therefore explained in some detail here. 4 model may be defined formally as a selective abstraction of reality. *his definition implies that modeling is the process of capturing selected characteristics of a system or a process and then combining these into an abstract representation of the original. *he main idea here is that it is usually far easier to analy,e a simplified model than it is to analy,e the original system, and as long as the model is a reasonably accurate representation, conclusions dra n from such an analysis may be validly extrapolated bac# to the original system. *here is no single VcorrectV ay to build a model and as often noted, model&building

is more an art than a science. *he #ey point to be #ept in mind is that most often there is a natural trade&off bet een the accuracy of a model and its tractability. 4t the one extreme, it may be possible to build a very comprehensive, detailed and exact model of the system at hand1 this has the obviously desirable feature of being a highly realistic representation of the original system. )hile the very process of constructing such a detailed model can often aid immeasurably in better understanding the system, the model may ell be useless from an analytical perspective since its construction ith a lot may be extremely time&consuming and its complexity precludes any meaningful analysis. 4t the other extreme, one could build a less comprehensive model of simplifying assumptions so that it can be analy,ed easily. Jo ever, the danger here is that the model may be so lac#ing in accuracy that extrapolating results from the analysis bac# to the original system could cause serious errors. Clearly, one must dra a line some here in the middle here the model is a sufficiently accurate here to dra as given representation of the original system, yet remains tractable. %no ing such a line is precisely can only come above, the #ey

hat determines a good modeler, and this is something that

ith experience. "n the formal definition of a model that

ord is Vselective.V Javing a clear problem definition allo s one to

better determine the crucial aspects of a system that must be selected for representation by the model, and the ultimate intent is to arrive at a model that captures all the #ey elements of the system hile remaining simple enough to analy,e. Fodels may be broadly classified into four categories-

5hysical Fodels- *hese are actual, scaled do n versions of the original. Examples include a globe, a scale&model car or a model of a flo line made ith elements from a toy construction set. "n general, such models are not very common in operations research, mainly because getting accurate representations of complex systems through physical models is often impossible. 4nalogic Fodels- *hese are models that are a step do n from the first category in that they are physical models as example of an analogic model automatic&computationB problem the ob2ective operations research analysis ell, but use a physical analog to describe the system, as opposed to an exact scaled&do n version. 5erhaps the most famous as the 4C*"4C model Athe acronym stood for anti& ithout even resorting to the use of a computer. "n this ay to distribute supplies at a military as used hich demonstrated that one could conduct a valid as to find the best

depot to various demand points. <uch a problem can be solved efficiently by using techni!ues from net or# flo analysis. Jo ever the actual procedure that too# a different approach. 4n anthill on a raised platform as chosen as an analog for the depot and little mounds of sugar on their o n platforms ere chosen to represent as each demand point. *he net or# of roads connecting the various nodes distance and the idth to the capacity along that lin#. 4n army of ants

constructed using bits of string ith the length of each being proportional to the actual as then released at the anthill and the paths that they chose to get to the mounds of sugar ere then observed. 4fter the model attained a steady state, it as found that the ants by virtue of their o n tendencies had found the most efficient paths to their destinationsX 'ne could even conduct some postoptimality analysis. $or instance, various transportation capacities along each lin# could be analy,ed by proportionately varying the idth of the lin#, and a scenario here certain roads ere unusable could be hile analy,ed by simply removing the corresponding lin#s to see hat the ants ould then do. *his illustrates an analogic model. Fore importantly, it also illustrates that '.3. is typically identified ith mathematical analysis, the use of an innovative model and problem&solving procedure such as the one 2ust described is an entirely legitimate ay to conduct an '.3. study. Computer <imulation Fodels- )ith the gro th in computational po er these models have become extremely popular over the last ten to fifteen years. 4 simulation model

is one

here the system is abstracted into a computer program. )hile the specific

computer language used is not a defining characteristic, a number of languages and soft are systems have been developed solely for the purpose of building computer simulation models1 a survey of the most popular systems may be found in '3KF< *oday A'ctober (@@M, pp. 3?&8:B. *ypically, such soft are has syntax as built&in constructs that allo ell as for easy model development. Iery often they also have

provisions for graphics and animation that can help one visuali,e the system being simulated. <imulation models are analy,ed by running the soft are over some length of time that represents a suitable period hen the original system is operating under steady state. *he inputs to such models are the decision variables that are under the control of the decision&ma#er. *hese are treated as parameters and the simulation is run for various combinations of values for these parameters. 4t the end of a run statistics are gathered on various measures of performance and these are then analy,ed using standard techni!ues. *he decision&ma#er then selects the combination of values for the decision variables that yields the most desirable performance. <imulation models are extremely po erful and have one highly desirable feature- they can be used to model very complex systems simplifying assumptions and one has to be very careful validation is necessary ithout the need to ma#e too many ithout the need to sacrifice detail. 'n the other hand, ith simulation models because it is also easy to misuse ith simulation.

simulation. $irst, before using the model it must be properly validated. )hile ith any model, it is especially important ith ho <econd, the analyst must be familiar to use a simulation model correctly,

including things such as replication, run length, armup etc1 a detailed explanation of these concepts is beyond the scope of this chapter but the interested reader should refer to a good text on simulation. *hird, the analyst must be familiar ith various statistical techni!ues in order to analy,e simulation output in a meaningful fashion. $ourth, constructing a complex simulation model on a computer can often be a challenging and relatively time consuming tas#, although simulation soft are has developed to the point here this is becoming easier by the day. *he reason these issues are emphasi,ed here is that a modern simulation model can be very flashy and attractive, but its real value lies in its ability to yield insights into very complex problems. Jo ever, in order to obtain such insights a considerable level of technical s#ill is re!uired.

4 final point to #eep in mind one experiments

ith simulation is that it does not provide one

ith an

indication of the optimal strategy. "n some sense it is a trial and error process since ith various strategies that seem to ma#e sense and loo#s at the ob2ective results that the simulation model provides in order to evaluate the merits of each strategy. "f the number of decision variables is very large, then one must necessarily limit oneself to some subset of these to analy,e, and it is possible that the final strategy selected may not be the optimal one. Jo ever, from a practitioner+s perspective, the ob2ective often is to find a good strategy and not necessarily the best one, and simulation models are very useful in providing a decision&ma#er good solutions. Fathematical Fodels- *his is the final category of models, and the one that traditionally has been most commonly identified ith '.3. "n this type of model one captures the characteristics of a system or process through a set of mathematical relationships. Fathematical models can be deterministic or probabilistic. "n the former type, all parameters used to describe the model are assumed to be #no n Aor estimated ith a high degree of certaintyB. )ith probabilistic models, the exact values for some of the parameters may be un#no n but it is assumed that they are capable of being characteri,ed in some systematic fashion Ae.g., through the use of a probability distributionB. 4s an illustration, the Critical 5ath Fethod AC5FB and the 5rogram Evaluation and 3evie *echni!ue A5E3*B are t o very similar '.3. techni!ues used in the area of pro2ect planning. Jo ever, C5F is based on a deterministic mathematical model that assumes that the duration of each pro2ect activity is a #no n constant, hile 5E3* is based on a probabilistic model that assumes that each activity duration is random but follo s some specific probability distribution Atypically, the Beta distributionB. Iery broadly spea#ing, deterministic models tend to be some hat easier to analy,e than probabilistic ones1 ho ever, this is not universally true. Fost mathematical models tend to be characteri,ed by three main elements- decision variables, constraints and ob2ective functionAsB. Decision variables are used to model specific actions that are under the control of the decision&ma#er. 4n analysis of the model ill see# specific values for these variables that are desirable from one or more perspectives. Iery often S especially in large models S it is also common to define additional VconvenienceV variables for the purpose of simplifying the model or for ith

ma#ing it clearer. <trictly spea#ing, such variables are not under the control of the decision&ma#er, but they are also referred to as decision variables.Constraints are used to set limits on the range of values that each decision variable can ta#e on, and each constraint is typically a translation of some specific restriction Ae.g., the availability of some resourceB or re!uirement Ae.g., the need to meet contracted demandB. Clearly, constraints dictate the values that can be feasibly assigned to the decision variables, i.e., the specific decisions on the system or process that can be ta#en. *he third and final component of a mathematical model is the ob2ective function. *his is a mathematical statement of some measure of performance Asuch as cost, profit, time, revenue, utili,ation, etc.B and is expressed as a function of the decision variables for the model. "t is usually desired either to maximi,e or to minimi,e the value of the ob2ective function, depending on profits and minimi,e changes in hat it represents. Iery often, one may simultaneously have more than one ob2ective function to optimi,e Ae.g., maximi,e or#force levels, sayB. "n such cases there are t o options. $irst, one could focus on a single ob2ective and relegate the others to a secondary status by moving them to the set of constraints and specifying some minimum or maximum desirable value for them. *his tends to be the simpler option and the one most commonly adopted. *he other option is to use a techni!ue designed specifically for multiple ob2ectives Asuch as goal programmingB. "n using a mathematical model the idea is to first capture all the crucial aspects of the system using the three elements 2ust described, and to then optimi,e the ob2ective function by choosing Afrom among all values for the decision variables that do not violate any of the constraints specifiedB the specific values that also yield the most desirable Amaximum or minimumB value for the ob2ective function. *his process is often called mathematical programming. 4lthough many mathematical models tend to follo this form, it is certainly not a re!uirement1 for example, a model may be one or more variables are affected by changes in constructed to simply define relationships bet een several variables and the decision& ma#er may use these to study ho fall into this category. Before concluding this section on model formulation, e return to our hypothetical the values of others. Decision trees, Far#ov chains and many !ueuing models could

example and translate the statements made in the problem definition stage into a

mathematical model by using the information collected in the data collection phase. *o do this e define t o decision variables 7 and ) to represent respectively the idgets to be made and sold next month. *hen the ob2ective is hich should ensure that the production ould be translated into the follo ing number of gi,mos and

to maximi,e total profits given by (07T@). *here is a constraint corresponding to each of the three limited resources, of 7 gi,mos and ) idgets does not use up more of the corresponding resource than is available for use. *hus for resource (, this mathematical statement 0.M7T(.0) Y :30, 4dditionally, here the left&hand&side of the ine!uality

represents the resource usage and the right&hand&side the resource availability. e must also ensure that each 7 and ) value considered is a nonnegative integer, since any other value is meaningless in terms of our definition of 7 and ). *he completely mathematical model isFaximi,e Z5rofit [ (07T@)\, sub2ect to
o o o o

0.M7T(.0) Y :30 (.07TA2K3B) Y M0? 0.(7T0.2;) Y (3; 7, ) ] 0 and integers.

*his mathematical program tries to maximi,e the profit as a function of the production !uantities A7 and )B, hile ensuring that these !uantities are such that the corresponding production is feasible ith the resources available. )odel Solution1 *he fifth phase of the '.3. process is the solution of the problem represented by the model. *his is the area on analy,ing a hich a huge amount of research and development in '.3. has been focused, and there is a plethora of methods for ide range of models. "t is impossible to get into details of these various of techni!ues in a single introductory chapter such as this1 ho ever, an overvie

some of the more important methods can be found else here in this handboo#. 7enerally spea#ing, some formal training in operations research is necessary in order to appreciate ho many of these methods or# and the interested reader is urged to orth mentioning that in recent years a hich Aat least in theoryB are Vblac# peruse an introductory text on '.3.1 the section on V$urther 3eadingV at the end of the chapter lists some good boo#s. "t is also number of soft are systems have emerged

boxesV for solving various models. Jo ever, some formal education in '.3. methods is still re!uired Aor at least strongly recommendedB before using such systems. $rom the perspective of the practitioner, the most important thing is to be able to recogni,e hich of the many available techni!ues is appropriate for the model constructed. Usually, this is not a hard tas# for someone ith some rudimentary training in operations research. *he techni!ues themselves fall into several categories. 4t the lo est level one might be able to use simple graphical techni!ues or even trial and error. Jo ever, despite the fact that the development of spreadsheets has made this much easier to do, it is usually an infeasible approach for most nontrivial problems. Fost '.3. techni!ues are analytical in nature, and fall into one of four broad categories. $irst, there are simulation techni!ues, hich obviously are used to analy,e simulation models. 4 significant part of these are the actual computer programs that run the model and the methods used to do so correctly. Jo ever, the more interesting and challenging part involves the techni!ues used to analy,e the large volumes of output from the programs1 typically, these encompass a number of statistical techni!ues. *he interested reader should refer to a good boo# on simulation to see ho these t o parts fit together. *he second category comprises techni!ues of mathematical analysis used to address a model that does not necessarily have a clear ob2ective function or constraints but is nevertheless a mathematical representation of the system in !uestion. Examples include common statistical techni!ues such as regression analysis, statistical inference and analysis of variance, as optimum&see#ing techni!ues, ell as others such as !ueuing, Far#ov chains and decision analysis. *he third category consists of hich are typically used to solve the mathematical programs described in the previous section in order to find the optimum Ai.e., bestB values for the decision variables. <pecific techni!ues include linear, nonlinear, dynamic, integer, goal and stochastic programming, as ell as various net or#&based methods. 4 detailed exposition of these is beyond the scope of this chapter, but there are a number of excellent texts in mathematical programming that describe many of these methods and the interested reader should refer to one of these. *he final category of techni!ues is often referred to as heuristics. *he distinguishing feature of a heuristic techni!ue is that it is one that does not guarantee that the best solution ill be found, but at the same time is not as complex as an optimum&see#ing techni!ue. 4lthough heuristics could be simple, common&sense, rule&of&thumb type techni!ues,

they are typically methods that exploit specific problem features to obtain good results. 4 relatively recent development in this area are so&called meta&heuristics Asuch as genetic algorithms, tabu search, evolutionary programming and simulated annealingB hich are general purpose methods that can be applied to a number of different problems. *hese methods in particular are increasing in popularity because of their relative simplicity and the fact that increases in computing po er have greatly increased their effectiveness. "n applying a specific techni!ue something that is important to #eep in mind from a practitionerLs perspective is that it is often sufficient to obtain a good solution even if it is not guaranteed to be the best solution. "f neither resource&availability nor time ere an issue, one ould of course loo# for the optimum solution. Jo ever, this is rarely the case in practice, and timeliness is of the essence in many instances. "n this context, it is often more important to !uic#ly obtain a solution that is satisfactory as opposed to expending a lot of effort to determine the optimum one, especially hen the marginal gain from doing so is small. *he economist Jerbert <imon uses the term VsatisficingV to describe this concept & one searches for the optimum but stops along the ay hen an acceptably good solution has been found. 4t this point, some ords about computational aspects are in order. )hen applied to a nontrivial, real& orld problem almost all of the techni!ues discussed in this section re!uire the use of a computer. "ndeed, the single biggest impetus for the increased use of '.3. methods has been the rapid increase in computational po er. 4lthough there are still large scale problems hose solution re!uires the use of mainframe computers or po erful or#stations, many big problems today are capable of being solved on des#top microcomputer systems. *here are many computer pac#ages Aand their number is gro ing by the dayB that have become popular because of their ease of use and that are typically available in various versions or si,es and interface seamlessly ith other soft are systems1 depending on their specific needs end&users can select an appropriate configuration. Fany of the soft are vendors also offer training and consulting services to help users specific techni!ues for ith getting the most out of the systems. <ome hich commercial soft are implementations are available

today include optimi,ationK mathematical programming Aincluding linear, nonlinear, integer, dynamic and goal programmingB, net or# flo s, simulation, statistical

analysis, !ueuing, forecasting, neural net or#s, decision analysis, and 5E3*KC5F. 4lso available today are commercial soft are systems that incorporate various '.3. techni!ues to address specific application areas including transportation and logistics, production planning, inventory control, scheduling, location analysis, forecasting, and supply chain management. <ome examples of popular '.3. soft are systems include C5=EW, ="CD', '<=, F5=, <4<, and <"F4C, to name 2ust a fe . )hile it ould clearly be impossible to describe herein the features of all available soft are, maga,ine such as '3KF< *oday and "E <olutions regularly publish separate surveys of various categories of soft are systems and pac#ages. *hese publications also provide pointers to different types of soft are available1 as an example, the December (@@M issue of '3KF< *oday Apages :(&M;B provides a complete resource directory for soft are and consultants. Updates to such directories are provided periodically. *he main point here is that the ability to solve complex modelsKproblems is far less of an issue today than it as a decade or t o ago, and there are plenty of readily available resources to address this issue. )e conclude this section by examining the solution to the model constructed earlier for our hypothetical production problem. Using linear programming to solve this model yields the optimal solution of 7[;80 and )[2;2, i.e., the production plan that maximi,es profits for the given data calls for the production of ;80 gi,mos and 2;2 idgets. *he reader may easily verify that this results in a profit of >M::? and fully uses up all of the first t o resources hile leaving (? units of the last resource unused. Cote that this solution is certainly not obvious by 2ust loo#ing at the mathematical model & in fact, if one ere VgreedyV and tried to ma#e as many gi,mos as possible Asince they yield higher profits per unit than the yield 7[M0? and )[0 Aat idgetsB, this ould hich point all of the second resource is used upB.

Jo ever, the resulting profit of >M0?0 is about ?9 less than the one obtained via the optimal plan. *he reason of course, is that this plan does not ma#e the most effective use of the available resources and fails to ta#e into account the interaction bet een profits and resource utili,ation. )hile the actual difference is small for this hypothetical example, the benefits of using a good '.3. techni!ue can result in very significant improvements for large real& orld problems.

9alidation and Analysis1 'nce a solution has been obtained t o things need to be done before one even considers developing a final policy or course of action for implementation. *he first is to verify that the solution itself ma#es sense. 'ftentimes, this is not the case and the most common reason is that the model used as not accurate or did not capture some ma2or issue. *he process of ensuring that the model is an accurate representation of the system is called validation and this is something that A henever possibleB should be done before actual solution. Jo ever, it is sometimes necessary to solve the model to discover inaccuracies in it. 4 typical error that might be discovered at this stage is that some important constraint as ignored in the model formulation & this ill lead to a solution that is clearly recogni,ed as being infeasible and the analyst must then go bac# and modify the model and re&solve it. *his cycle continues until one is sure that the results are sensible and come from a valid system representation. *he second part of this step in the '.3. process is referred to as postoptimality analysis, or in laypersonLs terms, a V hat&ifV analysis. 3ecall that the model that forms the basis for the solution obtained is AaB a selective abstraction of the original system, and AbB constructed using data that in many cases is not (009 accurate. <ince the validity of the solution obtained is bounded by the modelLs accuracy, a natural !uestion that is of interest to an analyst is- VJo robust is the solution parameters used to construct it^V *o illustrate this ith respect to deviations in the assumptions inherent in the model and in the values of the ith our hypothetical production ish to as# are, AaB V)ill ith idgets ere problem, examples of some !uestions that an analyst might

the optimum production plan change if the profits associated 2 could be purchased at a premium, or AcB V"f machine unreliability hat effect ould it be

overestimated by ;9, and if so ho ^V or AbB V"f some additional amount of 3esource orth buying and if so, ho much^V ere to reduce the availability of 3esource 3 by ?9, ho live in an uncertain orld, and one of

ould this have on the optimal policy^V <uch !uestions are especially of

interest to managers and decision&ma#ers

the most important aspects of a good '.3. pro2ect is the ability to provide not 2ust a recommended course of action, but also details on its range of applicability and its sensitivity to model parameters.

Before ending this section it is

orth emphasi,ing that similar to a traditional

"ndustrial Engineering pro2ect, the end result of an '.3. pro2ect is not a definitive solution to a problem. 3ather, it is an ob2ective ans er to the !uestions posed by the problem and one that puts the decision&ma#er in the correct Vball&par#.V 4s such it is critical to temper the analytical solution obtained a sound, sensible and ith common sense and sub2ective reasoning before finali,ing a plan for implementation. $rom a practitionerLs standpoint or#able plan is far more desirable than incremental improvements in the !uality of the solution obtained. *his is the emphasis of this penultimate phase of the '.3. process. Implementation and )onitoring1 *he last step in the '.3. process is to implement the final recommendation and establish control over it. "mplementation entails the constitution of a team hose leadership ill consist of some of the members on the original '.3. team. *his team is typically responsible for the development of operating procedures or manuals and a time&table for putting the plan into effect. 'nce implementation is complete, responsibility for monitoring the system is usually turned over to an operating team. $rom an '.3. perspective, the primary responsibility of the latter is to recogni,e that the implemented results are valid only as long as the operating environment is unchanged and the assumptions made by the study remain valid. *hus hen there are radical departures from the bases used to ith our or#force causes a drastic reduction in develop the plan, one must reconsider oneLs strategy. 4s a simple example production problem, if a sudden stri#e by the

the availability of labor A3esource (, sayB, one must reconsider the plan completely to derive an alternative course of action. 4s a final ord on implementation, it should be emphasi,ed that a ma2or responsibility of the operations research analyst is to convey the results of the pro2ect to management in an effective fashion. *his is something that is unfortunately not emphasi,ed sufficiently, and there are many instances of a successful study not being implemented because the details and the benefits are not conveyed effectively to management. )hile this is of course true of any pro2ect in general, it is especially significant bac#ground. ith '.3. because of its mathematical content and ithout a strong !uantitative its potential to not be fully understood by a manager

You might also like