Designing The

International Journal of Physical Distribution & Logistics ManagementEmerald Article: Designing the reverse supply chain: the impact of the product residual valueChiara Gobbi

Article information:

To cite this document: Chiara Gobbi, (2011),"Designing the reverse supply chain: the impact of the product residual value", International Journal of Physical Distribution & Logistics Management, Vol. 41 Iss: 8 pp. 768 - 796

Permanent link to this document: http://dx.doi.org/10.1108/09600031111166429

Downloaded on: 29-04-2012

References: This document contains references to 100 other documents

To copy this document: [email protected]

This document has been downloaded 964 times.

Access to this document was granted through an Emerald subscription provided by PONDICHERRY UNIVERSITY

For Authors: If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service. Information about how to choose which publication to write for and submission guidelines are available for all. Additional help for authors is available for Emerald subscribers. Please visit www.emeraldinsight.com/authors for more information.

About Emerald www.emeraldinsight.comWith over forty years' experience, Emerald Group Publishing is a leading independent publisher of global research with impact in business, society, public policy and education. In total, Emerald publishes over 275 journals and more than 130 book series, as well as an extensive range of online products and services. Emerald is both COUNTER 3 and TRANSFER compliant. The organization is a partner of the Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation.

*Related content and download information correct at time of download.

Designing the reverse supplychain: the impact of the product

residual valueChiara Gobbi

Department of Operations Management, Copenhagen Business School,Frederiksberg, Denmark

Abstract

Purpose – The purpose of this paper is to explore the impact of the product residual value (PRV) andthe loss of value over time of returned products in the reverse supply chain configuration. It alsoexamines whether or not the distinction of Fisher’s functional and innovative products holds for thereverse supply chain.

Design/methodology/approach – In order to identify the relevance of the Fisher model, the modelneeds to be recast in terms of PRV, which, in this context, is considered the independent variable in thereverse logistics arena. Products defined as innovative in Fisher’s taxonomy correspond to disposedproducts with high residual value, whereas functional products correspond to disposed products withlow residual value. Furthermore, the PRV and the speed at which returned products lose their valueare considered in order to determine the configuration of the reverse supply chain that allows forrecapturing most of the PRV. These notions have then been tested by analyzing two reverse supplychains with a case study research methodology.

Findings – The findings show that low PRV is associated with second-class recovery options(recycling and energy recovery) and that high PRV is associated with first-class recovery options(reconditioning and remarketing). When the recovery option is recycling, time is not relevant, theprimary objective is cost reduction (efficiency), the chain is centralized, and actors and phases of thereverse chain are determined by the specificity of the recycling process. When the recovery option isreconditioning, time is primarily relevant, tradeoffs between costs and time efficiency are necessary, thechain presents a centralized structure, and the presence of other types of actors and phases influences thestructure of the reverse supply chain.

Research limitations/implications – The focus is restricted to the industry of electrical andelectronic products.

Practical implications – Based on the outcome of the study, managers are able to determine thebasic prerequisites for the design of their reverse supply chains.

Originality/value – Previous literature suggests that when the PRV is high, early productdifferentiation is necessary, and the chain is therefore decentralized. The paper demonstrates that thisis not confirmed in the case of low returned volumes and high reconditioning quality standards.

Keywords Electronics industry, Returns, Recycling, Salvage, Reverse supply chain design,Reverse logistics, Product life cycle, Product residual value

Paper type Research paper

1. IntroductionBecause firms have historically concentrated on getting their products and services tothe market, scientific contributions and business practices for the delivery chain havebeen well explored. Yet relatively few contributions have examined the potential of thereverse chain from a holistic point of view (Spengler and Schroter, 2003; Hanafi et al.,2008; Janse et al., 2010). Although such concepts as just in time, lean production,

The current issue and full text archive of this journal is available at

www.emeraldinsight.com/0960-0035.htm

IJPDLM41,8

768

International Journal of PhysicalDistribution & Logistics ManagementVol. 41 No. 8, 2011pp. 768-796q Emerald Group Publishing Limited0960-0035DOI 10.1108/09600031111166429

mass customization, postponement, efficient customer response, vendor managedinventory, and continuous replenishment have been extensively explored from atheoretical and operational point of view, the literature contains few attempts toredeploy the same body of knowledge to investigate the reverse chain (Krikke et al.,2004). The focus of product return programs has traditionally been on costminimization (Rogers and Tibben-Lembke, 2001). Over time, the level of interest withincompanies has expanded beyond the provision of reasonable after-sales service at thelowest possible cost. Companies are currently recognizing the increasing value ofproducts and technology created at the end of the forward supply chain. This change inperspective, along with the impact of environmental legislation, has started to shifttheir focus toward various types of recovery programs. In fact, repair, refurbishment,remanufacturing and remarketing, cannibalization and recycling of raw materials areall examples of recovery options that can represent an attractive business opportunity,a positive answer to sustainable development, and a way of achieving competitiveadvantage (De Koster et al., 2002; Jack et al., 2010).

Andel (1997, p. 61) stated: “[. . .] by ignoring the efficient return and refurbishmentor disposal of products, many companies miss out a significant return on investment”.It is therefore suggested that “reverse”, if strategically managed, can provide acompetitive advantage by consolidating the market position with the overall benefit ofimproving company image.

However, prior to efficient management, it is necessary to design and implement thereverse chain that better supports the process of recovering value from disposedproducts. Grounded knowledge has been developed on several aspects of the reversesupply chain:

. Differences between the forward and reverse supply chain (Fleischmann et al.,2001, 2004).

. Differences between open loop and closed loop reverse supply chains (Guide et al.,2003; Geyer and Jackson, 2004; Krikke et al., 2004).

. Reverse supply chain types (Fleischmann et al., 2001; De Brito et al., 2003).

. Reverse supply chain drivers (Rogers and Tibben-Lembke, 2001; De Brito andDekker, 2004; Toffel, 2004; Ravi et al., 2005).

. Enabling factors (Rogers and Tibben-Lembke, 2001; Krikke et al., 2004).

. Recovery options (Thierry et al., 1995) and recovery phases (Prahinski andKocabasoglu, 2006).

. Types and status of returned products, and actors involved in the reverse chain(De Brito and Dekker, 2004; Krikke et al., 2004).

Instead, existing contributions on reverse supply chain design and configuration arelimited almost totally to quantitative research and modeling, and they fail to providebusiness rules and a comprehensive framework for designing and implementing thereverse chain that better matches the conditions of the returned product (Derimel andGokcen, 2008; Stock and Mulki, 2009).

The aim of the study is to close the existing knowledge gap by identifying the mostappropriate reverse chain structure in relation to the product residual value (PRV),considering if a centralized or decentralized structure is most appropriate,as centralization versus decentralization is a central issue in all forward supply

Designingthe reverse

supply chain

769

chain design literature (Chopra and Meindl, 2010). Furthermore, the study investigateswhether the well-known business model developed by Fisher (1997) can also be appliedto the case of reverse chains.

A multiple case study analysis within the electrical and electronic industries hasbeen conducted in order to address these questions. Both industries comprise a broadrange of products (e.g. blenders, grinders, washing machines, medical equipment,ICT equipment) and are going through significant changes in recovery management asenvironmental legislation forces producers to consider appropriate recovery strategies.In addition, the sector is characterized by major changes in technologies and by ashortening of the product life cycle (Herold, 2007).

This study focuses on End-of-Life (EoL) and End-of-Use (EoU) products that arereturned by end-users. Other types of returns such as commercial returns,manufacturing returns, product recalls, by-products, and packaging returns havebeen excluded. Even though EoU and EoL are sometimes used as synonyms in theliterature, there is a distinction between EoU and EoL returns (Krikke et al., 2004;Guide and Van Wassenhove, 2009). EoU products, as opposed to EoL products, maypreserve functionality intact, and they are usually generated either by majortechnological breakthroughs that cause obsolescence or simply by customers’ changedpreferences (e.g. replacement of mobile phones, audio-video devices, liquid crystaldisplay monitors replacing cathode ray tube monitors). The distinction is relevantbecause EoU products may flow into secondary markets, prolonging the life cycle bydelaying the moment the product becomes waste. Instead, in case of EoL returns,product functionality is lost (i.e. worn out products). Krikke et al. (2004, p. 26) definesEoL products as “the returns taken back from the market to avoid environmental orcommercial damage”; EoU products, on the other hand, are products “returned aftersome period of operations due to the end of leasing, trade-in or product replacement”.

The companies involved are manufacturers, logistics service providers, andrecycling companies. The scope is therefore limited by the choice of industry, the typeof returns, and company typology. However, focusing on an industry helps to controlextraneous variations and to define the limits for generalization (Eisenhardt, 1989).

The next section presents the literature review, followed by a section describing theresearch problem and the research proportions, a section presenting the researchmethodology and design, and a section presenting a description of the case studies andoutcomes. Then a presentation of the findings follows. The paper ends by providingconclusions, limitations, and indications for further research.

2. Literature reviewThe literature review is organized according to the definition of central terms andconcepts used in the paper, followed by a section on the research gap.

2.1 Definitions of central termsPrior to further discussion, central terms within this research need to be clarified.As one of the research objectives is the identification of the most appropriate reversechain structure in relation to the PRV and to consider whether a centralized or adecentralized structure is most adequate, three terms are central to further analysis:supply chain structure, centralized versus decentralized structure, and PRV.

IJPDLM41,8

770

The literature yields three terms that characterize the “formation” and planning ofsupply chain: supply chain structure, supply chain configuration, and supply chaindesign.

According to Lambert and Cooper (2000), supply chain structure refers to the lengthof the supply chain and the number of suppliers and customers at each level. Chen andPaulraj (2004) use the term to refer to a firm plus its suppliers and customers, where themain issues are the distribution of the tasks between the actors and the coordinationmechanism that enhance supply chain performance.

According to Zhang et al. (2009), supply chain configuration is the instantiation of ageneric supply chain to a specific supply chain, whereas the lynchpin of supply chainconfiguration is in the coordination of product, process, and logistics decisions inrelation to a variety of customer requirements. Smirnov and Chandra (2000) affirm thatsupply chain configuration is the approach to network enterprise creation in whichenterprises are considered to be assemblies of reusable components defined on acommon knowledge domain; configuring the supply chain therefore involvesknowledge management among the chain agents. Graves and Willems (2005) and Liand Womer (2008) define supply chain configuration as the practice of selecting amongdifferent suppliers, parts, processes, and transportation options, acknowledging thateach option can vary in lead time and costs. For Akanle and Zhang (2008), configuringthe supply chain requires the identification of the best combination of resources toanswer everyday costumer orders, considering options of reconfiguration in case ofchanges in the demand pattern. Srai and Gregory (2008) provide a complete review ofthe configuration concept from a strategic management perspective.

A major part of the literature that corresponds to the quantitative and modelingresearch stream uses the term supply chain design. Supply chain design is defined asthe process of determining the number, location, and capacity of manufacturing,distribution, and storing facilities; the optimal quantity flow between them; theassignment of market demand to facilities; the selection of suppliers; the placement ofsafety stocks along the chain; and the transportation modes, production modes, anddistribution channel to deploy (Graves and Willems, 2003; Meixell and Gargeya, 2005;Shen, 2005; Bidhandi et al., 2009; Melo et al., 2009; Klibi et al., 2010; Bottani et al., 2010;Lee and Wilhelm, 2010). Even if there are different connotations to the terms supplychain structure, configuration, and design, all three are frequently found in themajority of the reviewed papers; hence we can assert that the terms are usedsynonymously in the literature. Because the purpose of this study is not to contributeto the terminology discussion, the three terms are used equivalently here.

Another important issue raised in this study is whether the reverse chain should becentralized or decentralized. Centralization versus decentralization in supply chainmanagement is discussed, for example, in Lee and Whang (1999), Bernstein andFedergruen (2005), Waller et al. (2006), Ozsen et al. (2009), and Chopra and Meindl(2010). The main tradeoffs between a centralized versus a decentralized structureconcern costs, lead time, and customer service (Abrahamsson et al., 2003). Whencentralization is pursued, the focus is on efficiency, cost reduction, and economies ofscale. When decentralization is pursued, the focus is on increased geographicalcoverage, short lead time, and increased customer service. In this study, we adopt thedefinition by Fleischmann et al. (2000, p. 660): “centralization refers to the number oflocations at which similar activities are carried out”. In a centralized structure,


supply chain

771

each activity is installed at a few locations; whereas, in a decentralized structure, thesame operation is conducted at several different locations in parallel.

Because the purpose of the reverse supply chain is to recapture value from disposedproducts (Prahinski and Kocabasoglu, 2006; Atasu et al., 2008), determining the PRV ofa dismissed product is crucial in configuring the supply chain. The PRV is considered akey concept in this study because it influences the choice of the recovery option andtherefore the supply chain structure. In finance, the residual value or salvage value isdefined as the price at which a fixed asset is expected to be sold at the end of its usefullife. In the reverse supply chain context, the residual value of a product is defined as thevalue a product retains at the end of the usage phase (Kumar et al., 2007; Atasu et al.,2008). The usage phase represents one part of the product life cycle. In this study,we refer to the product life cycle as the succession of stages a product undertakes fromdesign to disposal (design, production, distribution, usage, and ultimate disposal),which is the point of view adopted in product life cycle management theory(Tibben-Lembke, 2002). The usage phase is then the part of the product life cycle thatrefers to the use of the product by the final user: the time the product is in user’s hands.Georgiadis et al. (2006, p. 516) refer to this concept as residence time: “[. . .] the time theproduct stays with the user before it is returned”. The PRV is the value that the productretains after usage – the value remaining (preserved) after the manufacturing (valuecreation) phase and after the usage (value consumption) phase (Kumar et al., 2007;Brodin and Anderson, 2008). Recapturing value from the returned product (i.e. valuereclamation) is the aim of recovery. Kumar et al. (2007) refer to the PRV as the EoUvalue. In Prahinski and Kocabasoglu (2006) the residual value is defined as salvagevalue. In this study, we adopt the term PRV.

Furthermore, we investigate whether or not the well-known business modeldeveloped by Fisher (1997) also applies to reverse supply chains. In the Fisher (1997)model, responsive supply chains correspond to innovative products, whereas efficientsupply chains correspond to functional products. Fisher’s paper is one of the most citedin the operations management literature, and a large number of studies have beenconducted in order to validate, invalidate, or expand the model. The validity of Fisher’smodel has been tested by Wong et al. (2006), Selldin and Olhanger (2007), and Lo andPower (2010). Another group of papers expand Fisher’s model by recasting efficientand responsive supply chain configurations with lean, agile, and hybrid supply chains(Mason-Jones et al., 2000; Christopher and Towill, 2001; Huang et al., 2002; Stratton andWarburton, 2003; Goldsby et al., 2006). Childerhouse et al. (2002), Ramdas (2003), andPayne and Peters (2004) expand Fisher’s model by integrating product attributes.Another key contribution is represented by Lee’s (2002) work, in which the demanduncertainty introduced by Fisher (1997) is considered in relation to the supplyuncertainty.

2.2 Research gapExisting contributions on reverse supply chain design and configuration are limitedalmost totally to quantitative research and modeling. These studies consider specificreverse chain problems such as independent versus integrated reverse chain designwith weak or strong correlation with a pre-existing forward chain (Fleischmann et al.,2001; Beamon and Fernandes, 2004; Kara et al., 2007; Srivastava, 2008; Lee and Dong,2009; Mutha and Pokharel, 2009; Easwaran and Uster, 2010; Pishvaee et al., 2010),

IJPDLM41,8

772

facility location for reverse logistics (Shih, 2001; Bautista and Pereira, 2006;Walther et al., 2005; Queiruga et al., 2007), production planning and scheduling,capacity planning and inventory management for remanufacturing and disassembly,and reverse flow optimization (Walther and Spengler, 2005; Chanintrakul et al., 2009;de la Fuente et al., 2010).

Only a limited number of studies consider the design and implementation of thereverse chain from a more holistic point of view, providing business guidelines(Thierry et al., 1995; Rogers and Tibben-Lembke, 2001; Guide et al., 2003; Blackburn et al.,2004; Toffel, 2004; Prahinski and Kocabasoglu, 2006; Bernon and Cullen, 2007;Kocabasoglu et al., 2007; Yang and Wang, 2007; Wikner and Tang, 2008; Jack et al., 2010;Janse et al., 2010), often focusing on specific aspects of the reverse supply chain such as aspecific recovery option (Guide and Van Wassenhove, 2001), specific actors of thereverse supply chain (Hsu et al., 2009; Stock and Mulki, 2009; Simpson, 2010), or specificreturned products (Pohlen and Farris, 1992; Shih, 2001; Knemeyer et al., 2002; Toffel,2004; Hanafi et al., 2008; Grant and Banomyong, 2010).

The purpose of this paper is to fill the research gap by providing a simple holisticframework for designing and implementing the reverse chain that better matches theconditions of the returned products and to investigate if Fisher’s business model can beapplied in the reverse supply chain context.

3. Reverse supply chain design and the PRV3.1 The relationship between the PRV and the recovery optionsThierry et al. (1995) have identified the following recovery options: repair, refurbishing,remanufacturing, cannibalization, and recycling. From repair to recycling, the level ofdisassembly increases, while the possibility of preserving the product structure intactdecreases. Repair refers to the return of used products to “working order”.Refurbishing raises used products to specified quality standards. Remanufacturingaims at reconfiguring the product in order to reach the same quality standards as anew product. In cannibalization, only a small proportion is reused in the form of partsand components. Recycling aims at recovering value on a material level.

We argue that the PRV plays a central role in determining the appropriate recoveryoption (Stock and Mulki, 2009): the residual value determines whether or not thereturned product should be disposed, recycled, repaired, refurbished,or remanufactured. Indeed, it is not correct to convey products with low residualvalue to recovery options such as repair, refurbishment, or remanufacturing(in combination referred to as reconditioning), but eventually to cannibalization andrecycling (Knemeyer et al., 2002; Inderfurth, 2005; Kumar et al., 2007).

The idea of considering the PRV when designing and implementing the reversesupply chain is the main tenet of this study. This notion is important because the PRVresults to be the independent variable in the reverse arena. The PRV depends upon thereturned product conditions as well as the existence of a secondary market.Specifically, the distinction between low and high PRVs derives from a number offactors: the condition of the dismissed product (the age, quality, and intensity of usage;Guide and Van Wassenhove, 2003; Derimel and Gokcen, 2008), the demand forreconditioned products, the demand for recycled materials, the cost structure of therecovery process (Prahinski and Kocabasoglu, 2006; Kumar et al., 2007; Stockand Mulki, 2009), and the level of obsolescence (wear-out life, replacement life,


supply chain

773

technology cycle, design cycle, purchase cycle; Rose et al., 2002). All these factors arehowever exogenous, external to the reverse chain as they are all given conditions thatthe reverse chain cannot affect. They are dependent on the modality of use of theproduct, the existence of markets for reconditioned products and recycled materials,the criteria applied in product design, and the evolution of the technology embedded inthe product. Of all the factors affecting the PRV, the most important is the existence ofa secondary market that is willing to receive reconditioned products and at the price atwhich it is willing to receive them. The returned product can be in good condition, stillfunctioning, and therefore apparently preserving a high residual value. However, if thereturned product is obsolete and replaced by a newer generation of products thatprovides the same function or more functions at a lower price, the residual value is low.

These considerations lead to the first research proposition:

RP1. The PRV determines the recovery option: the higher the PRV, the higher thepossibility that the product is reconditioned (repaired, refurbished,remanufactured) and remarketed. Conversely, the lower the PRV, the higherthe probability that the product is cannibalized and recycled.

Recycling and cannibalization are classified as second-class recovery options, as theproduct structure is lost; the value is recovered uniquely in the form of components,materials, and energy generation through incineration, whereas the fraction thatcannot be recycled is disposed by landfill. Repair, refurbishment, and remanufacturingare classified, instead, as first-class recovery options because they allow therecapturing of the residual value in the form of reusable products, thereby minimizingwaste and environmental burden (Inderfurth, 2005; Simpson, 2010). This classificationfollows the Lansink’s (Figure 1) scale ladder of ecological hierarchy for productrecovery: recovery options that reuse more of the functional content of a product arepreferred from an environmental standpoint (Duflou et al., 2008).

3.2 The relationship between the recovery option and the reverse chain structureAll recovery options can be linked to specific recovery operations. The following set ofoperations is recurrent: acquisition/collection, transportation, inspection, selection,separation/sorting, reconditioning, and disposal. Although acquisition andtransportation are common phases, reconditioning and remarketing are distinctive forthe reconditioning options (repair, refurbishment, and remanufacturing). In the case ofreconditioning, the reverse supply chain presents the following key processes: productacquisition, reverse logistics, inspection, reconditioning, and remarketing. In the case ofrecycling, the reverse chain presents the following key processes: product acquisition,

Figure 1.Lansink’s ladder:ecological hierarchyof recovery options

Prevention of waste

Incineration with energy recovery

Incineration without energy recovery

Landfill

Reuse of components

Material recycling

Reuse of products

–

Prio

rity

ord

er

+

IJPDLM41,8

774

reverse logistics, treatment, remarketing recycled raw materials, landfill, andincineration (Krikke et al., 2004; Prahinski and Kocabasoglu, 2006; Grant andBanomyong, 2010). The recovery option thus determines the number and characteristicsof the recovery phases and the number and type of actors involved (e.g. producers, publicinstitutions, transportation providers, logistics providers, brokers, recyclingcompanies). Consequently, the recovery option constrains the design of the reversechain (Derimel and Gokcen, 2008).

These considerations lead to the second research proposition:

RP2. The recovery option affects the reverse chain structure by determining the typeand number of phases involved, as well as the actors and their relationships.

Because the PRV erodes over time, recovery options are, at different levels, timedependent (Stock, 2001; Yang and Wang, 2007; Hsu et al., 2009; Stock and Mulki, 2009).Blackburn et al. (2004) and Guide et al. (2003) recognize that the speed at whichproducts are received, inspected, refurbished, and remarketed is crucial in recapturingmost of the value. In particular, if the returned product has a high residual value andcan be potentially reconditioned (first-class recovery options), delaying the treatmentcauses a revenue loss, because returned products are subjected to fast obsolescence anddegradation (Stock, 2001; Hsu et al., 2009).

The concept introduced by Blackburn et al. (2004) in order to consider the loss ofvalue over time is the marginal value of time (MVT), which can be expressed as:

a ¼ MVT ¼Dv

Dt

where Dv is the variation of value and Dt is the variation of time.The MVT measures the speed at which returned products lose value from the

moment they are disposed by the user. High MVT means that the time for processingthe returned product should be reduced as much as possible in order to extract most ofthe residual value rapidly. In order to reduce any delay, Blackburn et al. (2004) suggesta preponement strategy, which means to decentralize the initial phases of the recoveryprocess (i.e. testing and evaluation) in order to direct the returned products to the mostproper recovery option. It can be argued, however, that monitoring the MVT andreducing the processing time for high MVT products is relevant only if the PRV ishigh. In the case of low or no residual value, it is completely irrelevant if the returnedproduct has a high or low MVT and is processed at a high or low speed. Indeed, in thiscase, any investment made to reduce the processing time would be totally lost(Stock and Mulki, 2009). The processing time should then be monitored and eventuallyreduced when the PRV is high. In these settings, the reverse chain should have adecentralized structure, at least for the first phases of the recovery process. Not all thereturned products with high residual have a high MVT. For instance, the loss of valueover time varies across industries and product categories. Returned consumerelectronics such as personal computers (PCs) lose their value at the rate of 1 per centper week (Guide and Van Wassenhove, 2009). Products such as dishwashers andwashing machines lose their value much more slowly. Furthermore, a decentralizedstructure requires multiple investments of the same kind in multiple sites (Chopraand Meindl, 2010), which are justified only in the case of high and homogeneousvolumes (critical mass) that allow for the amortization of huge investments.


supply chain

775

These conditions are seldom verified in the reverse supply chain arena, characterizedby different returned products, different intensity of usage, high variety, and multipledispersed disposing channels (Knemeyer et al., 2002; Inderfurth, 2005; Wikner andTang, 2008; Simpson, 2010). The aim of this study is to verify whether or notdecentralization is the most convenient configuration in all cases of high PRV.

These considerations lead to the third research proposition:

RP3. Returned products with high residual value and high MVT call for adecentralized reverse chain (decentralized evaluation and testing); returnedproducts with high residual value and low MVT call for a centralized reversechain, whereas returned products with low residual value call for a centralizedstricture, whether the MVT is high or low.

By linking RP1 and RP2, we can derive that the PRV determines the recovery optionand the recovery option constrains the reverse chain design. More precisely, the PRVinfluences the choice of the recovery option that allows the most of residual value to berecaptured, and the recovery option limits the reverse supply chain configuration interms of recovery phases, processes, and actors involved in the reverse chain.Furthermore, the PRV and the speed at which the returned product loses valueconstrain the reverse supply chain configuration in another way: whenever thereturned products retain a high residual value and lose value at high speed (high MVT),the reverse chain should be decentralized (RP3).

The proposed conceptual framework that needs to be verified is shown in Figure 2.

3.3 The PRV and the efficient versus responsive reverse supply chainIf the primary objective of the reverse chain is to recover value in terms of reconditionedproducts, the chain design and configuration should guarantee a short lead time.Contrariwise, if the primary goal is to recover value in the form of recycled raw materialsand/or components, time is less relevant and cost efficiencies dominate (Krikke et al.,2004; Guide et al., 2006). This can be reinterpreted in terms of a responsive reverse chainin the first case and an efficient reverse chain in the latter case according to Fisher’sparadigm. Fisher distinguishes between functional products with stable demand and along life cycle and innovative products on the one hand with variable demand and ashort life cycle on the other hand. He then proposes two supply chain structures in which

Figure 2.Conceptual framework forreverse supply chainconfiguration

Proposition 3

Proposition 3

Proposition 2

Proposition 1

Product residual value Recovery option

Phases •ProcessesActors andrelations

?

Low product residual valuelow MVT

Reverse supply chainconfiguration

High product residual value

Centralized

High MVT

Low MVT Decentralized

••

•

•?

?

?

?

Proposition 3

IJPDLM41,8

776

an efficient supply chain needs to be designed in order to deliver products at low costsand a responsive supply chain designed for speedy response, concluding that theappropriate matching is the efficient supply chain for functional products and theresponsive supply chain for innovative products. The relevance of the Fisher’s model forreverse chains is clearly seen by recasting it in terms of PRV: products defined asinnovative in Fisher’s taxonomy correspond to disposed products with high residualvalue, whereas functional products correspond to disposed products with low residualvalue. Responsive reverse chains should therefore correspond to disposed products withhigh residual value, whereas efficient reverse supply chains should correspond todisposed products with low residual value (Table I).

4. Research methodology and designThe propositions in this study have been investigated with a qualitative researchapproach, as at least three of the four points suggested by Yin (2003) are verified:the phenomenon of interest cannot be explained in isolation from its social complexity;there is a need to consider multiple subjective perspectives; and the phenomenon canbe analyzed using concrete cases, considering their evolution within a temporalcontext. Furthermore, because the posed questions are “how” and “why” related to acontemporary set of events, over which the investigator has little control (Yin, 2003;Voss, 2009), the chosen research methodology is the case study research. Wheneverpossible, a research design that includes at least two case studies should be preferred toa single case design. Even with two cases, there is the possibility of providing forreplication (Eisenhardt, 1989). Furthermore, two cases with opposing settings areconvenient when seeking explanations rather than replication. In this type of researchdesign, if the subsequent findings support the contrasts, the results represent a strongstart toward a theoretical replication, and it vastly strengths the external validity of thefindings (Yin, 2003). In this study, two cases with opponent reverse supply drivershave been investigated: legislation and value reclamation (Herold, 2007). They areopponents because the legislation forces producers to take back dismissed producteven if the recovery process is not profitable. The aim of value reclamation chain, on theother hand, is to generate profit from recovery. Table II includes the relevantinformation on the two analyzed reverse chains.

Chain 1 is a legislation-driven reverse chain (legislation is the main driver) and Chain 2is a value-driven reverse chain (value reclamation is the main driver). In both cases, thelargest unit of analysis embeds smaller units of analysis and many different actors areinvolved. At least one interview for each type of actor was conducted, each lasting from1 to 3 h. A total of 30 interviews were performed in 23 organizations. All the interviewswere taped, transcribed, and coded. Interviews are subjected to bias (Yin, 2003); therefore,it is a good strategy to complement the data collection process with other sources.Whenever possible, site visits and direct observations were conducted along with theinterviews. Company publications were also used to collect background information.

Responsive reverse supply chain Efficient reverse supply chain

High PRV Match No matchLow PRV No match Match

Table I.The relationship between

the PRV and efficientversus responsive reverse

supply chain


supply chain

777

5. Cases descriptions5.1 Case 1: the legislation-driven reverse chainThe case under discussion concerns the implementation of the waste of electrical andelectronic equipment (WEEE) reverse chain in Denmark. The flow is financed by theproducers, which have to pay a fee according to the annual market sales. The fee isused to establish the facilities necessary for collecting and recovering the disposedproducts from the final users. In the most common situation, collective contracting isthe form of collaboration chosen to comply with the legislation; the producersparticipate in organizations called collective schemes, which support the activities ofcollection, transportation, and treatment. The collective schemes interface with themunicipal authorities, logistics service providers, and treatment companies. Therefore,types of actors involved in the WEEE system are various: governmental agencies(Ministry of Environment and WEEE System), trade associations and collectiveschemes, consumers, importers, producers, recyclers, and transportation companies,with their different levels of responsibility.

Figure 3 shows the organization of the system. There are 300 municipal deliverystations located in Denmark’s three main regions of Zealand, Funen, and Jutland, whereconsumers can dispose WEEE products. The user or municipal personnel sort the WEEE

Case study 1: the legislation-driven reverse chainCase study 2: the value-driven reverse chain

Larger unit of analysis The reverse chain The reverse chainMultiple units of analysis Companies, institutional organisms, trading,

and industry associationsCompanies

Types of actor Institutional organismsIndustry/sector associations

Producers/importersLogistics providers

Producers/importersWaste management companies

Actors interviewed The Danish Ministry of Environment and theWEEE System

ALPHABETAGeodis (Germany)Geodis (Italy)DHL

Collective schemes: LWF, Elretur, Nera, RE-DK,EcodomOriginal Equipment Manufacturers (OEM): Bangand Olufsen, Bayer, Danfoss, Elgiganten, Nokia,Novo Nordisk, Zitech, CanonRecycling companies: Stena Technoworld, H. J.Hansen Elektromiljø, Electrorecycling

Product typologies Large household appliances NotebooksSmall household appliances DesktopsIT and telecommunications equipment Stationary computersConsumer equipment CRT monitorsLighting equipment LCD monitorsElectrical and electronic tools (with the exceptionof large-scale stationary industrial tools)Toys, leisure and sports equipmentMedical devicesMonitoring and control instrumentsAutomaticdispensers

ServersPrinters/scannersOvensRefrigeratorsImagining devices

Table II.Presentation of thecase studies

IJPDLM41,8

778

waste into five categories. Collection sites are geographically dispersed to facilitateproduct disposal. Consolidation occurs when transporting the disposed products fromthe delivery stations to the municipal collection stations, from the municipal collectionstation to the regional sites, and from the regional sites to the recycling plants. Therefore,the chain leads to a centralized structure. Whenever the volume reaches a level thatassures efficient truck loading, products are conveyed to the recycling plants. Recyclingconsists of removing hazardous substances and subsequent treatment. Valuablematerials are sold to refining companies that enhance the purity level of the recoveredmaterials, thereby increasing their value. All the residuals that cannot be recovered areincinerated or sent for landfill disposal. The centralized structure is aimed at minimizingprocessing and transportation costs at the expense of a long lead time, which is irrelevantin the case of this type of dismissed product considered as waste. Centralization alsoguarantees high utilization of the recycling capacity.

5.2 Case 2: the value-driven reverse chainCase 2, the value-driven reverse chain, considers a logistics service provider that offersboth forward and reverse logistics services (Geodis). Two reverse flows, representingtwo scenarios, have been analyzed:

. Scenario 1: refurbishment of PCs and mainframe equipment. Refurbishment isperformed centrally in Mainz for dismissed products collected from 15 Europeancountries. The center is almost completely dedicated to a single customer, whichis called Alphabeta for anonymity.

. Scenario 2: pretreatment/dismantling of end of utilization products in Busnago,Italy. In contrast to refurbishment, the pretreatment service is performed forcustomers located in a single country (Italy).

In both scenarios, the aim is the recovery of value in the form of products (refurbishment)or in the form of products, components, and parts (pretreatment/dismantling). The twoscenarios can be considered as two instances of the same value-driven reverse chain,

Figure 3.Representation of the

WEEE recycling process

Refining300 deliverystations

Source: Grunow and Gobbi (2009)

80 municipal collectionstations, WEEE productssorted stored in five fractions

Transportation andconsolidation Recycling

Jutland

Funen

Zealand


supply chain

779

with Geodis as the focal company. However, they do present distinctions; althoughrefurbishment is centralized, dependent on product typology (PC equipment), andcustomer-dedicated, pretreatment and dismantling is to some extent decentralized(Goedis Italy serves only the Italian market), partially product typology dependent,and not customer specific.

Scenario 1: refurbishment. The dedicated site in Mainz performs the refurbishmentof PC equipment; 200,000 units are treated yearly, with an average weight of 13 kg.Of these 200,000 units, 165,000 are refurbished and 35,000 are dismantled andscrapped, with 60,000 parts being harvested from the scrapped portion. The monthlyload is approximately 16,000 units. The Asset Recovery Center (ARC) is 99 per centdedicated to Alphabeta; therefore, the refurbishment process is highly dependent uponAlphabeta’s requirements in terms of performances and contract service-levelagreements. One of the Alphabeta’s core businesses is the leasing of IT products tobusiness users, and the ARC processes those leased products when they are returned.According to the service-level agreement, the total lead time to perform the completeprocess (picking, transportation, receiving, testing, error reporting, and refurbishment)is 12 days, counted from collection at the customer site to order completed, meaningthat testing has been performed, errors and damages have been identified andregistered, and the product has been sent to either refurbishment or dismantling. Of the12 days, seven days are for collection, transportation, and receiving, and five days areallocated to other operations (i.e. packing). Within 14 days, products must be availablein the storing area, and ready for shipment.

Geodis is in charge of collecting dismissed products whenever Alphabeta signs anew agreement with its customers. The inbound logistics must ensure collection fromAlphabeta’s customer sites in Europe, intermediate consolidation, and delivery to theRecovery Asset Center in Mainz. An integrated track and trace system allows Geodisand Alphabeta to follow the entire inbound logistics process. The transportationservice is partly performed by Geodis and partly outsourced, depending on thetransportation infrastructure deployed in each country. The consolidation phase iscrucial when the transportation service is not outsourced. The picked volumes areconsolidated in the local country hubs before being delivered to Mainz. Truck loadingoptimization must be balanced with time-dependent constraints, as returned productsare subjected to a loss of value of 2-5 per cent per month.

The main challenge for the center is to adjust capacity planning within time frames asshort as a few days. Geodis receives general information on the expected number ofpallets and product groups but not on the quantity, types of products, or conditions.At the moment of receipt, the shipment and the product serial numbers are registered.All the products are subjected to an Asset Verification Test for functionality,configuration, and cosmetics check. A fault reporting system is used to register andidentify errors and defects. After receipt, data wiping and hard disks impairment isperformed. HDDs are physically destroyed by drilling or erased by magnetic fields.Refurbishment is performed by using the parts harvested from defective machines.Of the equipment received, 20 per cent is dismantled and scrapped due to bad cosmeticconditions, obsolete technology, or denied authorization for refurbishment and due toexcessive refurbishment costs. Therefore, there is no need to supply the refurbishmentprocess with externally acquired components. On the contrary, half of the harvestedparts are used for refurbishment in Mainz and half are sent to a central warehouse

IJPDLM41,8

780

in The Netherlands and used by the European Alphabeta repair network. Whenevermachines are not refurnished, valuable parts are harvested, and the remainingcomponents are sorted into two categories: recyclable metals and harmful substances.Hazardous substances are sent to the proper disposal areas. After refurbishment,products are stored in the warehouse and finally packed. Products are always Alphabetaproperty during the overall process, and it is Alphabeta’s responsibility to remarket thereconditioned products via a brokers’ network. Reconditioned products are sold all overthe world, but primarily in Europe. Figure 4 shows the settings of Scenario 1.

Scenario 2: pretreatment and dismantling. The Geodis unit in Busnago, Milanoperates exclusively for the Italian market. The site performs the disassembly ofservers, stationary computers, and printers, mainly. It is dedicated primarily to thephases that come before recycling: dismantling, parts/components harvesting, andmaterial separation. The output can be waste, recyclable materials, and metals andreconditioned products. Waste is conferred to waste treatment plants, recyclablematerials and metals are conferred to specialized refining and melting plants, whereasreconditioned products are sold in secondary markets. The center occasionallyacquires defective lots of electrical or electronic products from the market, and theproducts are repaired and resold. The service is offered to a diversified customer panel.The total processed volume is around 3,000,000 kg per year. The volume and weight oftreated products can vary considerably, from PCs, through metal racks, to cash flowmachines. The major part of the input is recovered as:

. Refurbished machines. Returned defective lots are bought from the market, thenrepaired and remarketed.

. Remanufactured and remarketed components/parts. Remanufacturing involvesmemory cards, hard disks, monitors, and keyboards.

. Components consigned to specialized melting plants. Gold, silver, and palladiumare recovered by melting in a blast furnace.

. Material consigned to refining plants and recovered as raw material(plastic, paper, wood, glass, non-precious metals).

. The remaining part is sent for incineration and landfill.

Figure 4.Representation of the

refurbishment process atGoedis Mainz, Germany

85%

15%

100%

Customersbusiness users

Receipt Asset verification test

Data wiping

Remarketing

Dismantling

Parts harvesting

Refurbishment

Disposal

Recyclingcompanies

Brokers


supply chain

781

Inbound transportation is outsourced to several transportation providers that collectdisposed products from all over Italy and consolidate them at the local hubs in order toguarantee full truck loading. The quantity and typology of the received material isknown in advance. The receiving is organized into several steps: receiving approval andweighing, material identification, updating of bookkeeping registers, updating ofdocumentation and movement to the unloading areas, and dispatching to the processareas or preliminary depot. The process is differentiated according to the productsreceived. Products received with an order of demolition are first disassembled intovaluable parts and hazardous and non-hazardous waste. Components are processed inorder to extract valuable metals, distinguished as precious, ferrous, and non-ferrousmetals. Products that are not covered by the obligation of destruction are alsodismantled into parts, and materials and parts that are valuable are remarketed or usedfor refurbishment. The process is executed by specialized operators assigned to flexibleworking stations adapted to process any type of product. The processing capacity ismeasured in kilogram per hour, and the average capacity is 80–100 kg/h. Capacityloading is generally performed on a monthly basis, but it is recalculated for incomingspot orders. The output follows different paths according to its typology. Waste is sentto disposal facilities specializing in the treatment of hazardous or nonhazardous waste.Precious and valuable metals are conferred to refining and melting plants. Refurbishedproducts, valuable parts, and components are acquired by brokers. Figure 5 shows thesettings of Scenario 2.

6. FindingsIn RP1, it is assumed that if the disposed product retains a high residual value, it isreasonable to consider first-class recovery options with subsequent trading insecondary markets. If the product has a low residual value, second-class recoveryoptions should be considered instead.

The evidence that emerged from both cases supports the argument of RP1:. The legislation-driven reverse chain processes disposed products with low

residual value and products are conveyed to recycling. The focus is on recyclingand energy recovery in the form of incineration.

. The value-driven reverse chain processes disposed products with high residualvalue, and products are remarketed as a whole after reconditioning (Scenarios 1and 2), as valuable components/parts (Scenario 2) or as precious metals(Scenario 1). The chain concentrates on refurbishment and remanufacturing.

Figure 5.Representation of thepretreatment/dismantlingprocess at GoedisBusnago, Italy

Business users

Consolidation hubs

Producers

Wholesalersretailers

Local institutions

Pretreatment/dismantling hub

Brokers

Recycling/refiningcompanies

IJPDLM41,8

782

The legislation-driven reverse chain processes returned products with low residualvalue and the products are conveyed to recycling. The fact that the legislation-drivenreverse chain deals with low PRV is immediately explained by considering thetypology of the returned products. The flow generates primarily historical wasteproducts that have been in use for long periods. Returned products are either brokenand repair costs more than the purchase of a new product or are obsolete – surpassedby products with enhanced functions and newer technologies. All products areconveyed to recycling, supporting the argument that low residual value (dismissedproducts are waste) matches second-class recovery options.

First-class recovery options (repair, refurbishment, remanufacturing) are carried outwhen the PRV is high. The two scenarios of the value-driven reverse chain, both therefurbishment process (the case of Geodis Germany) and the pretreatment/dismantlingprocess (the case of Geodis Italy), aim at recovering value in the form of productswhenever the residual value is high enough to justify transportation, reconditioning,and remarketing. In the first flow (Geodis Germany), the information that the PRV ishigh is well known in advance, as the dismissed products are leased, with a knownusage cycle and technology. That is why the producer Alphabeta has decided tosupport this profit generation opportunity with reconditioning and remarketing,maintaining the property of the dismissed products throughout the whole process. Inthe second flow, Geodis has invested in facilities that offer dismantling andpreliminary treatment. In this case, it is not known in advance if the PRV is high, andreconditioning is performed whenever the received product has high residual value.Once more, the first-class reconditioning options are tied to high PRV.

In RP2, it is assumed that once the most convenient recovery option is determined byassessing the PRV, the recovery option affects the reverse chain structure by determiningthe type and number of phases involved, the actors, and the actors’ relationships.

For the legislation-driven reverse chain, recycling is the recovery option. Recyclingimposes the following set of phases: collection, sorting, consolidation, transportation,recycling, remarketing recycled raw materials, and landfill and incineration ofunrecyclable fractions. The presence of actors such as recycling and treatmentcompanies and institutional governmental agencies assures the correct handling of thewaste. Because waste recycling is regulated by stringent norms, the relationshipsamong these actors are determined by specific contractual agreements.

For the value-driven reverse chain, reconditioning is the recovery option.Reconditioning comprises the following phases: collection, transportation, inspection,reconditioning, and remarketing of reconditioned products. Reconditioning imposesthe presence of specialized actors with specific assets (trained and skilled personneland equipment for testing and repair). The relationships among the actors are notregulated, as they are in the legislation-driven reverse chain, but are based upon trust,commitment, and mutual dependence. This is especially verified by ARC Germany,which is fully dedicated to service a single customer: Alphabeta. Alphabeta and Geodisare strongly linked, and they periodically work together on strategies to reduce leadtime, improve service, and enhance quality standards of the reconditioned products.

In RP3, it is assumed that returned products with high residual value and highMVT call for a decentralized reverse chain (decentralized evaluation and testing);for returned products with low or high residual value and low MVT, the reverse chainshould be centralized.


supply chain

783

The evidence emerging from both cases does not entirely support the argumentof RP3, as all phases of the recovery process are centralized in both the reverse chains,whether legislation driven or value driven. Nevertheless, the MVT is always high inScenario 1 and usually high in Scenario 2 (the value-driven reverse chain).

The legislation-driven reverse chain processes returned products with low residualvalue (waste) for which the speed of treatment is irrelevant. Recycling waste is not timesensitive; therefore the chain is structured in order to maximize cost efficiencies. In thiscase, the centralization of processing and transportation results in the most obviousconfiguration for providing economies of scale (Grunow and Gobbi, 2009). Collectionsites are geographically dispersed in order to facilitate product disposal and areimposed by the WEEE authority. But as soon as the materials are sorted, the waste isconsolidated before being transported to recycling plants. Because full utilization of thetreatment capacity is also necessary for gaining economies of scale, products are nottransferred to the recycling plants until a full truck can be loaded. Figure 6 shows thecentralized structure of the legislation-driven reverse chain.

The value-driven reverse chain processes returned products with high residualvalue. The residual value and MVT are high in the case of Scenario 1 (Geodis Mainz)because returned products are leased products that are refurbished and sold via anetwork of brokers. In Scenario 2 (Geodis Busnago), the residual value and MVT canvary according to the lots acquired from the market. Some lots present only minorfaults, and products can be easily repaired and sold. Other lots are acquired in order toextract valuable components that can be remanufactured and sold. Only a minorportion of the returned products is sent to recycling. In both Scenarios 1 and 2,therefore, the processing speed is crucial in order to extract value from the returnedproducts and gain higher profit from reselling, as the residual value and the MVT areusually high. The decentralization of the testing and repair phases accompanied by apreponement strategy is invoked in the literature in order to reduce lead time anddelays (Blackburn et al., 2004, Guide et al., 2006). This guarantees that the PRV is noteroded by unnecessary long waiting and queuing times.

The findings, however, do not support this argument. Both Scenarios 1 and 2present a centralized reverse chain structure. Neither of the two flows of thevalue-driven reverse chain (refurbishment in Mainz, pretreatment/dismantling inBusnago) supports the argument for decentralization. The first flow is clearlycentralized. The second leads toward centralization at least at the national level.

For refurbishment, the chain operated by Geodis for Alphabeta is characterized bycentralized testing and refurbishment (Figure 7). The centralization of activities hasbeen decided for a variety of reasons:

Figure 6.The centralizedlegislation-drivenreverse chain

Full truck loadingConsolidation

Decentralizedcollection Centralized

inspectionand sorting

Recycling

IJPDLM41,8

784

. Centralized operations allow tight control of quality standards for refurbishment.Local repair centers could introduce different routines that do not guarantee thesame results.

. Centralization avoids multiple investments in specialized resources. The treatedvolumes do not justify multiple local investments. Local dedicated investmentsin specialized personnel and refurbishment equipment would be reasonable onlyin the case of sizable incoming volumes.

. The refurbishment center of Geodis Germany has the capacity to serve15 European countries. Alphabeta treats the remaining volumes directly inMontpellier, France.

. Centralization allows Alphabeta to control the remarketing process and thenetwork of brokers that acquire the reconditioned products and resell them insecondary markets.

The second case, the flow operated by Geodis Italy, also presents centralized testingand dismantling (Figure 8). Centralization is on a national level, whereas Geodis is aninternational logistics service provider. During the interview with the managingdirector of the Geodis Italian Reverse Logistics Center, it emerged that the decision toserve only national customers was not dictated by considerations related to short leadtime. Given the exceeded processing capacity, the potential expansion of the marketwould allow the same service to be offered to customers outside the Italian territory.However, because national legislations require the center to acquire disposed productsas waste, complications can arise with cross-border transportation. In order to avoiddifficulties related to cross-border transportation authorizations, Geodis Italy hasdecided to maintain a national approach. The national approach is also determined

Figure 7.The centralized

value-driven reversechain of Scenario 1Recycling

Centralizedinspection

Brokers

Centralized treatment(remanufacturing/refurbishment)

Consolidation

Decentralizedcollection

Figure 8.The centralized

value-driven reversechain of Scenario 2

Decentralized collection

Consolidation

Centralized inspection Pretreatment and dismantling

Recycling

Brokers


supply chain

785

by a clear policy decided by Geodis Headquarters to avoid internal competition amongreverse logistics centers operating in other countries.

In both scenarios, even though the sorting and reconditioning activities arecentralized, the lead time is kept under tight control with an adequate reverse logisticspolicy. A short lead time, which is most important in the first scenario (ARC Mainz),is guaranteed by fast reverse logistics, which penalizes full transportation capacityutilization in order to comply with the 14 days for order completion. A necessarytradeoff occurs between speed and cost efficiencies. Despite this, as the reverse chaindeals with value lost at the rate of 3-5 per cent a month, the tradeoff is toward speed.

This study was based upon the assumption that responsive reverse supply chainscorrespond to disposed products with high residual value and that efficient reversesupply chains correspond to disposed products with low residual value. In order tocharacterize the responsive and efficient reverse supply chains, Fisher (1997) appliesthe following dimensions: the chain’s primary focus, the manufacturing focus, theinventory strategy, the lead time focus, the approach to choosing suppliers, and theproduct design strategy. In the case of a reverse supply chain, in order to determine ifthe reverse chain should be efficient or responsive, three factors are relevant:

(1) Primary purpose. The evidence that emerges from the analysis reveals that thelegislation-driven reverse chain primarily aims at recycling the incomingvolumes, gaining cost efficiencies, and respecting recovery targets, whereas thevalue-driven reverse chain primarily aims at reducing time delays andshortening lead time.

(2) The manufacturing focus that can be seen in terms of recovery focus in a reversechain. Considering the recovery process focus, the legislation-driven reversechain attempts to maintain a high utilization level of the deployed infrastructures(transportation facilities and treatment facilities), whereas the value-drivenreverse chain needs to balance high utilization rates with short lead time.

(3) The lead time focus. The legislation-driven reverse chain is not time sensitive.Returned products have low or no residual value, rendering the recovery timeirrelevant, whereas lead time is predominant in the value-driven reverse chain.

The findings are summarized in Table III.The analysis shows that the disposed products of the legislation-driven reverse

chain fulfill the characteristics of an efficient design, whereas the disposed products ofthe value-driven reverse chain fulfill the criteria for a responsive chain. Analogous to

Efficient responsive reversesupply chain Responsive reverse supply chain

Primary focus Recycle efficiently incoming volumesand maximize recovery targets

Recapture most of the PRV

Recovery process focus Maintain high average utilizationrate of the recovery infrastructures

Compromise between a high averageutilization rate of the recoveryinfrastructures and the requirementfor short lead times

Lead time focus Irrelevant to shorten lead times aslong as it does not increase costs

Investments to reduce the lead time

Table III.Efficient versusresponsive reversesupply chains

IJPDLM41,8

786

Fisher (1997), therefore, the disposed products can also be characterized as functionaland innovative products.

7. ConclusionsMany studies in the reverse supply chain management literature take a narrowapproach, addressing specific aspects of the reverse chain or employing mathematicalmodeling. This study provides a simple framework for designing the reverse chain onthe basis of the evaluation of the PRV, which depends upon a series of factorsexogenous to the reverse chain (Rose et al., 2002; Guide and Van Wassenhove, 2003;Kumar et al., 2007).

It has been hypothesized that the reverse chain design depends on the recoveryoption, which ultimately depends on the PRV, whereas the centralized versusdecentralized structure depends on the evaluation of the MVT. RP1 has postulated thedependency of the recovery option on the PRV. RP2 has postulated the dependency ofthe reverse chain design in terms of phases, process, and actors involved. RP3 hasadded guidelines for reverse chain design in terms of centralized versus decentralizedstructure in relation to the MVT. Furthermore, it has been assumed that by recastingFisher’s model in terms of functional products corresponding to products with lowresidual value and innovative products with products with high residual value, it ispossible to determine if the reverse chain should aim at efficiency or responsiveness.

The analysis revealed that RP1 and RP2 are supported by the cases’ evidence. RP3has been partially verified, as the reverse chain results to be centralized in all cases.Instead, the application of Fisher’s model to the reverse chain results in consistency inall cases. The propositions have been tested with a case study research methodology,and two reverse chains – legislation driven and value driven – have been considered.

The legislation-driven reverse chain, which deals with disposed products with lowor no residual value, has recycling as the recovery option. Because efficiency isimperative for recycling, the chain structure is arranged in such as way as to gain costefficiencies. Central consolidation hubs are utilized in order to collect centrally all thewaste delivered to the municipal collection stations. By consolidating, the chain obtainseconomies of scale in transportation and treatment. The actors participating in thechain are recycling companies, governmental and institutional organisms, specializedlogistics providers, and producers’ associations.

The value-driven reverse chain, which deals with disposed products with highresidual value, has reconditioning as the recovery option. Because reconditioning istime sensitive and the PRV decreases as time passes, it is crucial to reduce the lead timeand eliminate delays. The chain operated by Geodis Germany for Alphabeta isextremely time dependent: 14 days is the total available lead time from collectionto completed order. Chain responsiveness is reached to the detriment of cost efficienciesin transportation. The tradeoff between cost savings and responsiveness is in favor oftime reduction. The second chain, operated by Geodis Busnago, is less time dependent,as it acquires a mix of products with different residual values. The actors participatingin the chain are logistics service providers, producers, and business users.

In summary, this study has demonstrated that:. First-class recovery options (i.e. repair, refurbishment, remanufacturing) must be

considered for returned product with high residual value; second-class recovery


supply chain

787

options (i.e. recycling and incineration) must be considered for returned productswith low or no residual value.

. The recovery option determines the phases of the recovery process, the chainconfiguration, the type of actors involved, and their relationships.

. Although the literature suggests decentralization for returned products withhigh PRV and high MVT, the decision to implement a centralized versusdecentralized reverse supply chain structure should be driven by the size ofincoming volumes and the need of expertise (dedicated tooling and specializedpersonnel) in reconditioning.

. The efficient reverse supply chain is the right match for products with lowresidual value (legislation-driven reverse chain), whereas the responsive reversesupply chain is the right match for high PRV (value-driven reverse chain).

Results obtained from the case analysis are shown in Figure 9.

7.1 Limitations and indications for further researchContrary to RP3, supposing that high residual value and high MVT correspond todecentralized testing and evaluation, the value-driven reverse chain has been shown tobe centralized in all its phases. As soon as returned products are collected, the transferto a central treatment center occurs. The managing directors of Geodis Italy andGeodis ARC Mainz explained the choice of a centralized structure primarily in relationto incoming volumes and high investments required in terms of specialized recoveryequipment and personnel. The center in Mainz benefits from a favorable condition:having a long-term contract with Alphabeta for the refurbishment of leasedIT equipment being returned from 15 European countries. The contract assures that

Figure 9.Design framework for thereverse chain dependingon PRV and the MVT

Product residualvalue (PRV)

Low PRV

High PRV

Recovery option:recycling

Recovery option:reconditioning

Main objective: reduce costs – efficientreverse chain

Phases: collection, sorting, consolidation,transportation, recycling, remarketingrecycled raw materialsActors: final users, specialized logisticsproviders, producers associations,governmental organisms, recyclingcompanies

__________________________________Chain structure: centralized

Main objective: tradeoffs between costsand time efficiencies, timeefficiency predominant – responsivereverse chainPhases: collection, transportation,inspection, reconditioning, remarketingof reconditioned productsActors: business users, logisticsproviders, producers, brokers

_________________________________Chain structure: centralized

High MVT

Low MVT

Low or highMVT irrelevant

IJPDLM41,8

788

the center has adequate incoming volumes for full utilization treatment capacity,whereas the center in Italy struggles on a daily basis to chase returned lots from themarket. It appears, therefore, that current incoming volumes do not justify theestablishment of local recovery facilities. Indeed, Alphabeta, one of the worldwidelargest producers of PCs and mainframes, has only one other recovery center inMontpellier, France; two assets recovery centers are sufficient to refurbish leasedequipment returned from all over Europe. At present, the reverse value chain analyzedin this study tolerates high transportation costs in order to guarantee short processingtime in a centralized setting. Further research could address industries or productscharacterized by larger return rates and determine if the centralized chain structurestill holds.

This study focuses on two types of returns: EoL and EoU. Manufacturing returns,product recalls, by-products, packaging returns, and commercial returns have beenexcluded from this study. Further research can address these types of returns anddetermine if findings are still valid.

In particular, commercial returns could be an interesting area to investigate in orderto enhance the understanding of the reverse supply chain, as they may confute somefindings of this study. For instance, commercial returns may support a decentralizedreverse chain structure, as they usually require only minor processing before beingplaced back on the shelf.

The main actors in this study are manufacturing companies, logistics serviceproviders, and recycling companies. In particular, logistics providers have emerged askey players in the reverse supply chain management arena (Halldorsson andSkjøtt-Larsen, 2006). Further research can be conducted to investigate the role of theseactors that can benefit from the favored position of owning the logistics infrastructure,complex knowledge of the mechanism involved in logistics, and idiosyncraticresources that can be leveraged for other activities, such as product returns. It is notmerely fortuitous that major logistics providers are investigating ways of enhancingtheir reverse logistics portfolio (Prahinski and Kocabasoglu, 2006).

This study has addressed the industry of electrical and electronic products.Although a focusing on industry level helps to control extraneous variations, it limitsthe possibility of generalizing the findings to other industries (Yin, 2003). Therefore,further research should consider other industries in order to expand the applicability ofresults. In particular, the relevance of the PRV and the MVT that have beendeterminant in this study may be shown to be irrelevant in industries unaffected byrapid technological change and rapid loss of product value.

This study is based upon qualitative research design and it utilizes case studies asresearch methodology. Two case studies with opposing drivers (legislation and value)have been considered. The choice of two cases with opponent tenets has the advantagethat results represent a strong start towards a theoretical replication, vastlystrengthening the external validity of the findings (Yin, 2003). Nevertheless, furtherresearch could aim at theoretical replication by including other drivers (e.g. socialreputation and company image), types of returns, products returned at different stagesof the product life cycle, and other cases of value-driven and legislation-driven reversechains (e.g. the reverse chain for EoL vehicles or batteries, both of which are regulatedby European directives). Further research can also be directed at refining the research


supply chain

789

propositions by employing an analytical approach. Surveys, for instance, couldrepresent a viable way of quantifying and strengthening the relevance of the findings.

This study provides a useful framework for producers wanting to determine asuccessful strategy for product returns. The starting point would be to determinewhich product return flows have a direct impact on the company and segment themaccording to the PRV. Further research could address the level of involvement of theproducers on recovery actions (Herold, 2007). A high level of involvement coulddetermine another configuration of the reverse supply chain – a more decentralizedstructure in which retailers and wholesalers are involved by sharing the benefit of aprofitable reverse chain, for instance.

Economic metrics for efficiency and effectiveness have been well explored in thesupply chain management literature. In fact, an entire research stream is devoted toperformance management systems (Morgan, 2007). Because the reverse flow has highimpact in terms of benefits guaranteed to the environment and the society, it would berelevant to combine performance indicators for economic efficiency with performanceindicators for environmental efficiency. An interesting area for further research couldbe the definition of a performance management system that combines economic andenvironmental indicators.

This study has contributed to the general understanding of the supply chainmanagement field. Supply chain management has received increasing attention over thelast 20 years, but limited contributions have evaluated product returns from a managementpoint of view. It is still an open question if it is correct to consider reverse supply chain as aseparate stream or if the reverse flow should be considered as part of the forward supplychain in the logic of closing the loop (Krikke et al., 2004; Geyer and Jackson, 2004). Thisstudy does not support the argument of closed loop supply chains, as most often the returnflow cannot be considered as part of the original supply chain. What emerges morefrequently are two distinct flows, in which the demand of the forward flow becomes thesupply of the reverse flow, the actors are different, and strategic objectives diverge. Thus,whereas the essence of supply chains is to create and provide value, reverse chains have theobjective of recovering value or complying with environmental regulations.

References

Abrahamsson, M., Alding, N. and Stahre, F. (2003), “Logistics platforms for improved strategicflexibility”, International Journal of Logistics: Research and Applications, Vol. 6 No. 3,pp. 85-106.

Akanle, O.M. and Zhang, D.Z. (2008), “Agent-based model for optimising supply-chainconfigurations”, International Journal of Production Economics, Vol. 115 No. 2, pp. 444-60.

Andel, T. (1997), “Reverse logistics: a second chance to profit”, Transportation & Distribution,Vol. 38 No. 7, pp. 61-4.

Atasu, A., Guide, V.D.R. Jr and Van Wassenhove, L.N. (2008), “Product reuse economics inclosed-loop supply chain research”, Production and Operations Management, Vol. 17 No. 5,pp. 483-96.

Bautista, J. and Pereira, J. (2006), “Modeling the problem of locating collection areas for urbanwaste management. An application to the metropolitan area of Barcelona”, Omega, Vol. 34No. 6, pp. 617-29.

Beamon, B.M. and Fernandes, C. (2004), “Supply chain network configuration for productrecovery”, Production Planning & Control, Vol. 15 No. 3, pp. 270-81.

IJPDLM41,8

790

Bernon, M. and Cullen, J. (2007), “An integrated approach to managing reverse logistics”,International Journal of Logistics: Research and Applications, Vol. 10 No. 1, pp. 41-56.

Bernstein, F. and Federgruen, A. (2005), “Decentralized supply chains with competing retailersunder demand uncertainty”, Management Science, Vol. 51 No. 1, pp. 18-29.

Bidhandi, H.M., Yusuff, R.M., Ahmad, M.M.H.M. and Bakar, M.R.A. (2009), “Development of anew approach for deterministic supply chain network design”, European Journal ofOperational Research, Vol. 198 No. 1, pp. 121-8.

Blackburn, J.D., Guide, V.D., Souza, G.C. and Wassenhove, L.N.V. (2004), “Reverse supply chainsfor commercial returns”, California Management Review, Vol. 46 No. 2, pp. 6-22.

Bottani, E., Montanari, R. and Volpi, A. (2010), “The impact of RFID and EPC network on thebullwhip effect in the Italian FMCG supply chain”, International Journal of ProductionEconomics, Vol. 124 No. 2, pp. 426-32.

Brodin, M.H. and Anderson, H. (2008), “Recycling calls for revaluation”, Supply ChainManagement: An International Journal, Vol. 13 No. 1, pp. 9-15.

Chanintrakul, P., Coronado Mondragon, A.E., Lalwani, C. and Won, C.Y. (2009), “Reverselogistics network design: a state-of-the art literature review”, International Journal ofBusiness Performance and Supply Chain Modelling, Vol. 1 No. 1, pp. 61-81.

Chen, I.J. and Paulraj, A. (2004), “Towards a theory of supply chain management: the constructsand measurements”, Journal of Operations Management, Vol. 22 No. 2, pp. 119-50.

Childerhouse, P., Aitken, J. and Towill, D.R. (2002), “Analysis and design of focused demandchains”, Journal of Operations Management, Vol. 20 No. 6, pp. 675-89.

Chopra, S. and Meindl, P. (2010), Supply Chain Management Strategy, Planning, and Operation,Prentice-Hall, Upper Saddle River, NJ.

Christopher, M. and Towill, D.R. (2001), “An integrated model for the design of agile supplychains”, International Journal of Physical Distribution & Logistics Management, Vol. 31No. 4, pp. 235-46.

De Brito, M.P. and Dekker, R. (2004), “A framework for reverse logistics”, in Dekker, R.,Fleischmann, M., Inderfurth, K. and Van Wassenhove, L.N. (Eds), Reverse Logistics:Quantitative Models for Closed-loop Supply Chains, Springer, Heidelberg, pp. 3-27.

De Brito, M.P., Flapper, S.D.P. and Dekker, R. (2003), “Reverse logistics: a review of case studies”(Report Series Research in Management ERS-2003-012-LIS), Erasmus UniversityRotterdam, Rotterdam.

de la Fuente, M.V., Ros, L. and Ortiz, A. (2010), “Enterprise modelling methodology for forwardand reverse supply chain flows integration”, Computers in Industry, Vol. 61 No. 7,pp. 702-10.

De Koster, R., De Brito, M.P. and Van de Vendel, M. (2002), “Return handling: an exploratorystudy with nine retailer warehouses”, International Journal of Retail & DistributionManagement, Vol. 30 No. 8, pp. 407-21.

Derimel, N.O. and Gokcen, H. (2008), “A mixed integer programming model for remanufacturingin reverse logistics environment”, The International Journal of Advanced ManufacturingTechnology, Vol. 39 Nos 11/12, pp. 1197-206.

Duflou, J.R., Seliger, G., Kara, S., Umeda, Y., Ometto, A. and Willems, B. (2008), “Efficiency andfeasibility of product disassembly: a case-based study”, CIRP Annals – ManufacturingTechnology, Vol. 57 No. 2, pp. 583-600.

Easwaran, G. and Uster, H. (2010), “A closed-loop supply chain network design problem withintegrated forward and reverse channel decisions”, IIE Transactions, Vol. 42 No. 11,pp. 779-92.


supply chain

791

Eisenhardt, K. (1989), “Building theories from case study research”, Academy of ManagementReview, Vol. 14 No. 4, pp. 532-50.

Fisher, M. (1997), “What is the right supply chain for your products?”, Harvard Business Review,Vol. 75 No. 2, pp. 105-16.

Fleischmann, M., Beullens, P., Bloemhof-Ruwaard, I.M. and Van Wassenhove, L.N. (2001),“The impact of product recovery on logistics network design”, Production and OperationManagement, Vol. 10 No. 2, pp. 156-73.

Fleischmann, M., Bloemhof-Ruwaard, J., Beullens, P. and Dekker, R. (2004), “Reverselogistics network design”, in Dekker, R., Fleischmann, M., Inderfurth, K. andVan Wassenhove, L.N. (Eds), Reverse Logistics: Quantitative Models for Closed-loopSupply Chains, Springer, Berlin, pp. 65-94.

Fleischmann, M., Krikke, H.R., Dekker, R. and Flapper, S.D.P. (2000), “A characterisation oflogistics network for product recovery”, Omega, Vol. 28 No. 6, pp. 653-66.

Georgiadis, P., Vlachos, D. and Tagaras, G. (2006), “The impact of product lifecycle on capacityplanning of closed-loop supply chains with remanufacturing”, Production and OperationsManagement, Vol. 15 No. 4, pp. 514-27.

Geyer, R. and Jackson, T. (2004), “Supply loops and their constrains: the industrial ecology ofrecycling and reuse”, California Management Review, Vol. 46 No. 2, pp. 55-73.

Goldsby, T.J., Griffis, S.E. and Roath, A.S. (2006), “Modeling lean, agile, and leagile supply chainstrategies”, Journal of Business Logistics, Vol. 27 No. 1, pp. 57-80.

Grant, D.B. and Banomyong, R. (2010), “Design of closed-loop supply chain and product recoverymanagement for fast-moving consumer goods: the case of a single-use camera”, Asia PacificJournal of Marketing and Logistics, Vol. 22 No. 2, pp. 232-46.

Graves, S.C. and Willems, S.P. (2003), “Supply chain design: safety place stock placementand supply chain configuration”, in de Kok, A.G. and Graves, S.C. (Eds), Handbooks inOR & MS, Vol. 11, Elsevier, Amsterdam.

Graves, S.C. and Willems, S.P. (2005), “Optimizing the supply chain configuration for newproducts”, Management Science, Vol. 51 No. 8, pp. 1165-80.

Grunow, M. and Gobbi, C. (2009), “Designing the reverse network for WEEE in Denmark”,CIRP Annals – Manufacturing Technology, Vol. 58 No. 1, pp. 391-4.

Guide, V.D.R. Jr and Van Wassenhove, L. (2001), “Managing product returns forremanufacturing”, Production and Operations Management, Vol. 10 No. 2, pp. 142-56.

Guide, V.D.R. Jr and Van Wassenhove, L. (2003), Business Aspects of Closed-loop Supply Chains,Carnegie Mellon University Press, Pittsburgh, PA.

Guide, V.D.R. Jr and Van Wassenhove, L.N. (2009), “The evolution of closed-loop supply chainsresearch”, Operation Research, Vol. 57 No. 1, pp. 10-18.

Guide, V.D.R. Jr, Harrison, T.P. and Van Wassenhove, L.N. (2003), “The challenge of closed-loopsupply chains”, Interfaces, Vol. 33 No. 6, pp. 3-6.

Guide, V.D.R. Jr, Souza, G.C., Van Wassenhove, L.N. and Blackburn, J.D. (2006), “Time value ofcommercial returns”, Management Science, Vol. 52 No. 8, pp. 1200-14.

Halldorsson, A. and Skjøtt-Larsen, T. (2006), “Role of intermediaries in reverse supply chains:third party logistics providers vs. dedicated operators”, The Proceedings of the 18thAnnual Conference for Nordic Researchers in Logistics Research Network (NOFOMA),Department of Strategy and Logistics, Norwegian School of Management, Oslo, Norway.

Hanafi, J., Kara, S. and Kaebernick, H. (2008), “Reverse logistics strategies for end-of-lifeproducts”, The International Journal of Logistics Management, Vol. 19 No. 3, pp. 367-88.

IJPDLM41,8

792

Herold, M. (2007), “A multinational perspective to managing end-of-life electronics”, doctoraldissertation series 2007/1, Laboratory of Industrial Management, Helsinki University ofTechnology.

Hsu, H.S., Alexander, C.A. and Zhu, Z. (2009), “Understanding the reverse logistics operationsof a retailer: a pilot study”, Industrial Management & Data Systems, Vol. 109 No. 4,pp. 515-31.

Huang, S.H., Uppal, M. and Shi, J. (2002), “A product driven approach to manufacturing supplychain selection”, Supply Chain Management: An International Journal, Vol. 7 Nos 3/4,pp. 189-99.

Inderfurth, K. (2005), “Impact of uncertainties on recovery behavior in a remanufacturingenvironment: a numerical analysis”, International Journal of Physical Distribution &Logistics Management, Vol. 35 No. 5, pp. 318-36.

Jack, E.P., Powers, T.L. and Skinner, L. (2010), “Reverse logistics capabilities: antecedents andcost savings”, International Journal of Physical Distribution & Logistics Management,Vol. 40 No. 3, pp. 228-46.

Janse, B., Schurr, P. and De Brito, M.P. (2010), “A reverse logistics diagnostic tool: the case of thecostumer electronics industry”, International Journal of Advanced ManufacturingTechnology, Vol. 47 Nos 5-8, pp. 495-513.

Kara, S., Rugrungruang, F. and Kaebernick, H. (2007), “Simulation modelling ofreverses logistics networks”, International Journal of Production Economics, Vol. 106No. 1, pp. 61-9.

Klibi, W., Martel, A. and Guitouni, A. (2010), “The design of robust value-creating supply chainnetworks: a critical review”, European Journal of Operational Research, Vol. 203 No. 2,pp. 283-93.

Knemeyer, A., Ponzurick, T. and Logar, C. (2002), “A qualitative examination of factors affectingreverse logistics systems for end-of-life computers”, International Journal of PhysicalDistribution & Logistics Management, Vol. 32 No. 6, pp. 455-79.

Kocabasoglu, C., Prahinski, C. and Klassen, R.D. (2007), “Linking forward and reverse supplychain investments: the role of business uncertainty”, Journal of Operations Management,Vol. 25 No. 6, pp. 1141-60.

Krikke, H., le Blanc, I. and van de Velde, S. (2004), “Product modularity and the design ofclosed-loop supply chains”, California Management Review, Vol. 46 No. 2, pp. 23-39.

Kumar, V., Shirodkar, P.S., Camelio, J.A. and Sutherland, J.W. (2007), “Value flowcharacterization during product lifecycle to assist in recovery decisions”, InternationalJournal of Production Research, Vol. 45 Nos 18/19, pp. 4555-72.

Lambert, D.M. and Cooper, M.C. (2000), “Issues in supply chain management”, IndustrialMarketing Management, Vol. 29 No. 1, pp. 65-83.

Lee, C. and Wilhelm, W. (2010), “On integrating theories of international economics in thestrategic planning of global supply chains and facility location”, International Journal ofProduction Economics, Vol. 124 No. 1, pp. 225-40.

Lee, D.H. and Dong, M. (2009), “Dynamic network design for reverse logistics operations underuncertainty”, Transportation Research Part E: Logistics and Transportation Review,Vol. 45 No. 1, pp. 61-71.

Lee, H. (2002), “Aligning supply chain strategies with product uncertainties”,California Management Review, Vol. 44 No. 3, pp. 105-19.

Lee, H. and Whang, S. (1999), “Decentralized multi-echelon supply chains: incentives andinformation”, Management Science, Vol. 45 No. 5, pp. 633-40.


supply chain

793

Li, H. and Womer, K. (2008), “Modeling the supply chain configuration problem withresource constraints”, International Journal of Project Management, Vol. 26 No. 6,pp. 646-54.

Lo, S.M. and Power, D. (2010), “An empirical investigation of the relationship between productnature and supply chain strategy”, Supply Chain Management: An International Journal,Vol. 15 No. 2, pp. 139-53.

Mason-Jones, R., Naylor, B. and Towill, D.R. (2000), “Engineering the leagile supply chain”,International Journal of Agile Management Systems, Vol. 2 No. 1, pp. 54-61.

Meixell, M.J. and Gargeya, V.B. (2005), “Global supply chain design: a literature review andcritique”, Transportation Research Part E: Logistics and Transportation Review, Vol. 41No. 6, pp. 531-50.

Melo, M.T., Nickel, S. and Saldanha-da-Gama, F. (2009), “Facility location and supply chainmanagement – a review”, European Journal of Operational Research, Vol. 196 No. 2,pp. 401-12.

Morgan, C. (2007), “Supply chain performance measurement: future challenges?”,The International Journal of Logistics Management, Vol. 18 No. 2, pp. 255-73.

Mutha, A. and Pokharel, S. (2009), “Strategic network design for reverse logistics andremanufacturing using new and old product modules”, Computers & IndustrialEngineering, Vol. 56 No. 1, pp. 334-46.

Ozsen, L., Daskin, M.S. and Coullard, C.R. (2009), “Facility location modeling andinventory management with multisourcing”, Transportation Science, Vol. 43 No. 4,pp. 455-72.

Payne, T. and Peters, M.J. (2004), “What is the right supply chain for your products?”,The International Journal of Logistics Management, Vol. 15 No. 2, pp. 77-92.

Pishvaee, M.S., Farahani, R.Z. and Dullaert, W. (2010), “A memetic algorithm for bi-objectiveintegrated forward/reverse logistics network design”, Computers & Operations Research,Vol. 37 No. 6, pp. 1100-12.

Pohlen, T. and Farris, M.T. II (1992), “Reverse logistics in plastics recycling”, InternationalJournal of Physical Distribution & Logistics Management, Vol. 22 No. 7, pp. 35-47.

Prahinski, C. and Kocabasoglu, C. (2006), “Empirical research opportunities in reverse supplychains”, Omega, Vol. 34 No. 6, pp. 519-32.

Queiruga, D., Walther, G., Gonzalez-Benito, J. and Spengler, T. (2007), “Evaluation of sitesfor the location of WEEE recycling plants in Spain”, Waste Management, Vol. 28 No. 1,pp. 181-90.

Ramdas, K. (2003), “Managing product variety: an integrative review and research directions”,Production and Operations Management, Vol. 12 No. 1, pp. 79-101.

Ravi, V., Shankar, R. and Tiwari, M.K. (2005), “Analysing alternatives in reverse logistics forend-of-life computers: ANP and balanced scorecard approach”, Computers & IndustrialEngineering, Vol. 48 No. 2, pp. 327-56.

Rogers, D.S. and Tibben-Lembke, R. (2001), “An examination of reverse logistics practices”,Journal of Business Logistics, Vol. 22 No. 2, pp. 129-48.

Rose, C., Stevels, A. and Ishii, K. (2002), “Method for formulating product end-of-lifestrategies for electronics industry”, Journal of Electronics Manufacturing, Vol. 11 No. 2,pp. 185-96.

Selldin, E. and Olhanger, J. (2007), “Linking products with supply chains: testing Fisher’s model”,Supply Chain Management: An International Journal, Vol. 12 No. 1, pp. 42-51.

IJPDLM41,8

794

Shen, A.J.M. (2005), “A multi-commodity supply chain design problem”, IIE Transactions, Vol. 37,pp. 753-62.

Shih, L.H. (2001), “Reverse logistics system planning for recycling electricalappliances and computer in Taiwan”, Resources, Conservation and Recycling, Vol. 32No. 1, pp. 55-72.

Simpson, D. (2010), “Use of supply relationships to recycle secondary materials”, InternationalJournal of Production Research, Vol. 48 No. 1, pp. 227-49.

Smirnov, A. and Chandra, C. (2000), “Ontology-based knowledge management for co-operativesupply chain configuration”, Proceedings of the 2000 AAAI Spring Symposium onBringing Knowledge to Business Processes, Stanford, CA, USA, pp. 85-92.

Spengler, T. and Schroter, M. (2003), “Strategic management of spare parts in closed-loop supplychain – a system dynamics approach”, Interfaces, Vol. 33 No. 6, pp. 7-17.

Srai, J.S. and Gregory, M. (2008), “A supply network configuration perspective on internationalsupply chain development”, International Journal of Operations & ProductionManagement, Vol. 28 No. 5, pp. 386-411.

Srivastava, S. (2008), “Value recovery network design for product returns”, International Journalof Physical Distribution & Logistics Management, Vol. 38 No. 4, pp. 311-31.

Stock, J. (2001), “The 7 deadly sins of reverse logistics”, Material Handling Management, Vol. 56No. 3, pp. 5-11.

Stock, J. and Mulki, J. (2009), “Products returns processing: an examination of practices ofmanufactures, wholesales/distributors, and retailers”, Journal of Business Logistics, Vol. 30No. 1, pp. 33-62.

Stratton, R. and Warburton, R.D.H. (2003), “The strategic integration of agile and lean supply”,International Journal of Production Economics, Vol. 85 No. 2, pp. 183-98.

Thierry, M., Salomon, M., Van Nunen, J. and Van Wassenhove, L. (1995), “Strategic issuesin product recovery management”, California Management Review, Vol. 37 No. 2,pp. 114-35.

Tibben-Lembke, R.S. (2002), “Life after death: reverse logistics and the product life cycle”,International Journal of Physical Distribution & Logistics Management, Vol. 32 No. 3,pp. 223-44.

Toffel, M. (2004), “Strategic management of product recovery”, California Management Review,Vol. 46 No. 2, pp. 120-41.

Voss, C. (2009), “Case research in operations management”, in Karlsson, C. (Ed.), ResearchingOperations Management, Taylor & Francis, New York, NY, pp. 162-92.

Waller, M.A., Cassady, C.R. and Ozment, J. (2006), “Impact of cross-docking on inventory in adecentralized retail supply chain”, Transportation Research Part E, Vol. 42 No. 5,pp. 359-82.

Walther, G. and Spengler, T. (2005), “Impact of the WEEE-directive on reverse logistics inGermany”, International Journal of Physical Distribution & Logistics Management, Vol. 35No. 5, pp. 337-61.

Walther, G., Spengler, T. and Queiruga, D. (2005), “Facility location and planning for treatment oflarge household appliances in Spain”, International Journal of Environmental Technologyand Management, Vol. 8 No. 4, pp. 405-25.

Wikner, J. and Tang, O. (2008), “A structural framework for closed-loop supply chains”,The International Journal of Logistics Management, Vol. 19 No. 3, pp. 344-66.


supply chain

795

Wong, C.Y., Arlbjørn, J.S., Hvolby, H.H. and Johansen, J. (2006), “Assessing responsiveness of avolatile and seasonable supply chain: a case study”, International Journal of ProductionEconomics, Vol. 104 No. 2, pp. 709-21.

Yang, H.L. and Wang, C.H. (2007), “Integrated framework for reverse logistics”, New Trends inApplied Artificial Intelligence Lecture Notes in Computer Science, Vol. 45, pp. 501-10.

Yin, R. (2003), Case Study Research. Design and Methods, Sage, Thousand Oaks, CA.

Zhang, L.L., You, X., Jiao, J.R. and Helo, P. (2009), “Supply chain configuration with co-ordinatedproduct, process and logistics decisions: an approach based on petri nets”, InternationalJournal of Production Research, Vol. 47 No. 23, pp. 6681-706.

About the authorChiara Gobbi holds a PhD from the Technical University of Denmark and is Assistant Professorin the Department of Operations Management at Copenhagen Business School. Her researchfocus is reverse supply chain management and collaborative procurement in healthcare.Chiara Gobbi can be contacted at: [email protected]

IJPDLM41,8

796

To purchase reprints of this article please e-mail: [email protected] visit our web site for further details: www.emeraldinsight.com/reprints

Designing The

Documents

loop supply chains

de la fuente

reverse supply chains

loop supply chain

copenhagen business school

control extraneous variations

physical distribution amp

reverse supply chain