Developing a decision framework via content analysis

Developing a decision framework via content analysis

Benjamin T. Hazen, Dianne J. Hall and Joe B. Hanna Department of Supply Chain and Information System Management, College of Business, Auburn University, Auburn, Alabama, USA

Abstract

Purpose – The purpose of this study is to identify the critical components of the reverse logistics (RL) disposition decision-making process and suggest a decision framework that may guide future investigation and practice.

Design/methodology/approach – The authors utilized a problem-driven content analysis methodology. RL literature from 2000 through 2010 was content analyzed to determine which components may impact a firm’s RL disposition decision.

Findings – The authors extrapolated seven RL disposition decision components from a compilation of 60 variables identified in the literature. Practical implications and suggestions for future research are offered, and a RL disposition decision-making framework is presented.

Research limitations/implications – Although methodological techniques were carefully followed, the nature of a content analysis may be subject to author bias. Future investigation and use of the framework presented will verify the findings presented here.

Practical implications – This study identifies seven components that should be considered when deciding which RL disposition alternative should be adopted and integrates these components into a decision-making framework. Supply chain professionals who refer to this framework during the decision process will benefit from a more comprehensive analysis of potential RL disposition alternatives.

Originality/value – Congruent with recent assertions suggesting that RL research is evolving from an operational-level focus to a holistic business process approach for maximizing value recovery, this study synthesizes operational-level research to develop a practical framework for RL disposition decision-making.

Keywords Reverse logistics, Returns, Disposition, Content analysis, Decision making

Paper type Research paper

Introduction Managing return product flow is becoming increasingly important to the success of supply chain firms, particularly as the volume of return flow substantially increases (Guide Jr et al., 2006). Because $100 billion worth of products are returned in the USA each year (Stock et al., 2002), the returns management process can be an integral part of a firm’s supply chain (Rogers et al., 2002). Accordingly, returned product disposition should not only happen quickly (Blackburn et al., 2004), but disposition decision-makers must consider a variety of decision parameters to ensure that the chosen disposition policy is the most advantageous for the organization. To date, no study has investigated what components comprise the disposition decision-making process. Thus, the purpose of this study is to identify relevant decision parameters

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

www.emeraldinsight.com/0960-0035.htm

IJPDLM 42,3

244

Received 14 October 2010 Revised 9 March 2011, 16 August 2011, 22 October 2011 Accepted 24 October 2011

International Journal of Physical Distribution & Logistics Management Vol. 42 No. 3, 2012 pp. 244-274 q Emerald Group Publishing Limited 0960-0035 DOI 10.1108/09600031211225954

 

 

and create a framework that will help guide business decisions-makers and future research regarding which disposition option to choose.

Much of the extant reverse logistics (RL) disposition literature seeks to optimize operational processes. For example, much of the literature discusses various aspects of managing returns for remanufacturing (Atasu and Cetinkaya, 2006; Inderfurth, 2005; Lu and Bostel, 2007; Teunter et al., 2006; Webster and Mitra, 2007). However, Guide Jr and Van Wassenhove (2009) posit that research in closed-loop supply chains (CLSCs) is evolving from a technical focus on operational-level activities to a holistic business process approach for maximizing value recovery. Additional research has noted the importance of understanding the factors involved in carefully examining the impact of the disposition decision on the rest of the firm (Blumberg, 1999). However, the literature is sparse in the area of RL disposition decision-making and is therefore an area in need of further study (Stock and Mulki, 2009). The current study capitalizes on the abundance of operational-level research to lay the groundwork for future decision-making research.

Background Literature on RL and the nature of decisions regarding RL processes is becoming more abundant as the area matures from primarily a step-retracing supply chain process to one that stands on its own as a necessary process to which management should pay close attention. While some research begins to investigate considerations for pursuing returns management activities in general (Rogers et al., 2002), a comprehensive understanding of the elements inherent in the disposition decision has not been developed. Without such an understanding, practitioners struggle to develop best practices and researchers cannot provide support to them. Research is needed to adequately provide an understanding of this area. The current study provides the foundation for that understanding by identifying the key components of the RL disposition decision and providing a framework for practice and future research.

We define the RL disposition decision as leading to the establishment of an organizational policy regarding which recovery option to pursue for a specific product or line of products. Because this decision should be made in accordance with a firm’s current policies, market position and objectives, we propose that the RL disposition decision requires great consideration. RL is comprised of all functions that begin with acquiring a returned product and end when the owning firm has extracted all possible value from the item through proper disposition. This disposition process includes options ranging from simply reusing the product to properly disposing of the product.

Much literature has been devoted to identifying and describing the actual disposition alternatives, such as reuse, recycling, and remanufacturing (Blackburn et al., 2004; Carter and Ellram, 1998; Krikke et al., 2004; Thierry et al., 1995). Some of this research has also addressed a variety of design considerations for reverse channels, such as investigating the tradeoff between efficient and responsive reverse supply chains (Blackburn et al., 2004) and modularity (Krikke et al., 2004). However, research regarding which RL disposition alternative should be employed by a given firm for a given product line is markedly absent in the literature. When determining a policy regarding how to handle returns, it would behoove decision-makers to follow an established decision-making process that takes into account all relevant considerations. To date, no study has assimilated all of these considerations or attempted to create such a decision-making framework. This lack of understanding makes it difficult to fully

RL disposition decision-making

245

 

 

comprehend which issues are important when making these increasingly important RL decisions. To determine a foundation for such an understanding, this research uses content analysis of RL literature to identify RL disposition decision criteria.

The remainder of this article is structured as follows. First, we briefly review foundational RL literature, where we describe how numerous factors have been acknowledged in extant literature to affect the RL process. We then offer a brief background of the four disposition alternatives that are generally described in the literature. The methodology of the content analysis is then described, followed by a discussion of the findings. Next we present a validity check of our findings, where components derived from our analysis are compared with factors addressed in existing RL frameworks. We then offer a discussion of the practical implications of our results, which also describes ideas for future research in RL decision-making. Finally, we integrate our findings with extant decision-making literature to create a RL disposition decision-making framework.

RL defined A review of supply chain management (SCM) literature reveals that some terms often encompass numerous definitions. Notably, the terms “logistics” and “SCM” lack universal definitions as multiple conceptual perspectives exist (Larson et al., 2007; Stock and Boyer, 2009). Similarly, there is not a consensus in the literature regarding the terms used to describe the reverse processes within the supply chain (Lambert, 2008). Therefore, the term “RL” will be used in this research to encompass all returns processes and is synonymous with terms such as “closed-loop supply chain” or “returns management”kopic. For the purpose of this paper, we adopt Stock’s (1998, pp. 20-1) comprehensive definition of RL:

[. . .] from a business logistics perspective, the term refers to the role of logistics in product returns, source reduction, recycling, materials substitution, reuse of materials, waste disposal, and refurbishing, repair, and remanufacturing; from an engineering logistics perspective, it is referred to as reverse logistics management (RLM) and is a systematic business model that applies best logistics engineering and management methodologies across the enterprise in order to profitably close the loop on the supply chain.

Foundational literature A variety of internal and external forces affect a firm’s RL processes. Building upon previous marketing research (Achrol et al., 1983; Stern and Reve, 1980) and their review of the logistics literature that specifically addresses external marketing factors (Barry et al., 1993; Bronstad and Evans-Correia, 1992; Cairncross, 1992; Kopicki et al., 1993; Livingstone and Sparks, 1994; Murphy et al., 1995; Pohlen and Farris II, 1992; Stock, 1992), Carter and Ellram (1998) developed a framework that describes the forces that affect RL. Their framework posits that the task environment consists of four distinct organizational entities that affect the firm’s RL operations. They are:

(1) suppliers (input);

(2) buyers (output);

(3) government (regulatory); and

(4) competitors (competitive).

IJPDLM 42,3

246

 

 

The task environment is embedded within the overall market environment consisting of legal, economic, political, and social variables.

Carter and Ellram’s (1998) conceptual model is widely regarded as the first comprehensive RL framework. It takes into account factors that are beyond the normal scope of logisticians and illustrates the holistic nature of RL. Their work provided the foundation for further investigation, as demonstrated by Knemeyer et al. (2002) when they updated the model to account for factors recognized in more current research. Considering the theory-building work of Dowlatshahi (2000) and a review of the contemporary literature at the time, Knemeyer et al.’s (2002) model accounts for concerns at multiple levels of an organization. Similarly, Skinner et al.’s (2008) research suggests that cross-functional integration is critical to the continued success of the returns management process. Additionally, Jayaraman and Luo (2007) recognized the system-level effects of a firm’s RL policies. Their framework describes the interdisciplinary nature of the RL disposition decision and demonstrates how a firm may derive value from its returned products, thus promoting the idea that all of a firm’s activities should seek to increase profits.

Considerations regarding RL encompass issues beyond that of many other business processes. For example, De Brito and Dekker’s (2003) model emphasizes corporate citizenship, legislation, and economics as the driving forces behind RL practices. Furthermore, Rogers and Tibben-Lembke (2001) highlight the overlapping considerations between green logistics and RL, suggesting the impact that green principles may have on RL decision-making. Specifically, they describe the activities of recycling, remanufacturing, and use of reusable packing as overlapping between green logistics and RL. Conversely, Wolf and Seuring (2010) suggest that, although environmental concerns are often considered when organizations contract with third party logistics providers, those concerns are given cursory examination at best. Research also suggests that understanding customer needs regarding returns may enable organizations to develop better product placement strategies (Ofek et al., 2011). In that same vein, Jack et al. (2010) found that RL may be examined from both the viewpoint of front-end customer relationship strategies and back-end RL processes. Their research suggests that back-end processes have a positive impact on RL capabilities, which in turn increases cost savings.

The above research highlights the dynamic nature of RL and underscores the importance of identifying factors that impact the general RL process. Not specifically addressed, however, is the disposition decision. Whether or not to implement RL processes is not an issue; RL will endeavor to exist for organizations that produce or move materials through the forward supply chain. While there is some need to align the RL process with the rest of the organization’s supply chain objectives, only the disposition decision has the ability to add value to the organization when the decision is based on appropriate information (Tan and Kumar, 2006). The research reviewed thus far describes the nature of RL. Our focus now turns to describing the common RL disposition alternatives that a firm may employ.

RL disposition alternatives In this paper, the term disposition alternative is synonymous with what some authors have referred to as recovery option (Krikke et al., 2004). Deciding upon the most advantageous disposition alternative can bolster a firm’s success (Croxton et al., 2001).

RL disposition decision-making

247

 

 

However, not only do decision-makers need to understand the laws and regulations that govern proper material handling and disposal, they must also be able to recognize potential financial gains that may be realized by capitalizing on opportunities to reuse operational products, recondition damaged or used products, or recover valuable materials from products that are beyond their useful life.

Early work in regard to disposition decision models sought to identify and stratify disposition alternatives (Kopicki et al., 1993; Stock, 1992). Thierry et al.’s (1995) integrated supply chain model depicts a standard return process. Their model illustrates three separate disposition alternatives. First, a firm may employ direct reuse, which entails reusing or reselling the returned product in an as-is condition. Next, a firm may employ product recovery management, which entails processes such as repairing, refurbishing, remanufacturing, cannibalizing useable materials, and recycling materials of value. Finally, a firm may employ waste management, which entails incinerating waste or land filling.

Since the work of Thierry et al. (1995), others such as Carter and Ellram (1998), Krikke et al. (2004) and Rogers et al. (2002) have modified and stratified possible disposition alternatives. Although each study emphasized slightly differing alternatives and definitions, four common RL disposition categories seemingly emerge as comprising the core taxonomy in recent literature. In consideration of the work cited above, we propose that the following four disposition alternatives encompass the recovery options available for RL. In hierarchical order in regard to the potential residual value that can be recovered by a firm, the four alternatives are:

(1) reuse;

(2) product upgrade;

(3) material recovery; and

(4) waste management.

Reuse allows for the most value to be recovered while waste management allows for the least amount of value recovery. Although some decompose these four alternatives even further (Krikke et al., 2004), this general hierarchy is often utilized in the current literature (Blackburn et al., 2004; Prahinski and Kocabasoglu, 2006; Rogers et al., 2002; Staikos and Rahimifard, 2007). The following briefly defines each alternative and gives an example of research within each area.

Reuse. Direct reuse is an option that presents itself when a customer returns an unused product back to the place of purchase, thus inserting the product back into the supply chain for use. At the retailer level, once the product is no longer serviceable or requires some sort of upgrade (e.g. cleaning, replacing accessories, remanufacturing, repackaging, etc.) direct reuse is no longer an option. Generally, this option exists only if the location in which the product resides in the supply chain possesses the capability to return the product to retail condition. This process includes products that are completely unused and products that are returned after such light use that upgrade is not required in order to return the product to new status.

Assuming that the returned product is in new condition, a variety of options exist. The product can be again offered for sale by the retailer, shipped laterally to another retailer, shipped back to the distributor, or shipped to any other place within the forward or reverse supply chain where stock levels require such an item. Logisticians and retail

IJPDLM 42,3

248

 

 

managers are primarily concerned with accounting for the quantity and quality of these returns. These unknown quantities provide even more variability to a process that is already susceptible to forecasting error. This additional variability increases the bullwhip effect within the supply chain and can lead to increased inventory (Vlachos and Dekker, 2003). If returns can be adequately forecasted and properly managed, the returns that are available for direct reuse can reduce transportation, procurement, and storage costs while improving productivity as each item returned to the end of the supply chain offsets the need for another item to be pulled through the forward supply chain (Giuntini and Andel, 1995b; Mollenkopf and Closs, 2005).

Product upgrade. The product upgrade alternative is concerned with repairing, refurbishing, or remanufacturing an item in order to extend the life of and derive value from the original core unit (Krikke et al., 2004). Product upgrade becomes an option when the possibility of direct reuse is either no longer available (e.g. the product is in used condition) or not economical (e.g. there is no longer a market requirement for the product). If executed properly, product upgrade can create profitable business opportunities through recapturing value that would otherwise have been lost (Clendenin, 1997; Giuntini and Andel, 1995a). The term “upgrade” implies improving the product from its end-of-life condition to that of a condition acceptable for future use or sale. The condition and quality of upgraded products can vary greatly, depending on the upgrade technique chosen and the purpose of the upgrade.

The definitions of repair, refurbish, and remanufacture are debatable and the usage of such terms differs within the literature. However, Majumder and Groenevelt (2001) suggest that remanufacturing is the primary means of product upgrade and is the term usually assigned to any upgrade function. Remanufacturing is defined as:

[. . .] the process of disassembling used items, inspecting and repairing/reworking the components, and using these in a new product manufacture. A product is considered remanufactured if its primary components come from a used product (Majumder and Groenevelt, 2001, p. 125).

The current study adopts this definition. Material recovery. Material recovery involves recovering any portion of a returned

product that may contain value. Material recovery can entail cannibalizing entire pieces not requiring upgrade that can be reused (Krikke et al., 2004), recovering parts or pieces that may be reused (Blackburn et al., 2004), or extracting recyclable materials for reuse or to sell as a commodity. Early RL literature often focused on recycling (Guiltinan and Nwokoye, 1975; Pohlen and Farris II, 1992); thus, some scholars posit that RL has been most closely associated with recycling and environmental matters (Daugherty et al., 2002). Although the topic of sustainability is becoming popular in recent literature, determine how to extract value from returned products and ensure regulatory compliance are still prominent topics in this area (Roy et al., 2006).

Waste management. Once a firm has decided that it is no longer of value to reuse, upgrade, or recover materials from a specific product, the product then becomes waste. Lyons (2005, p. 71) defines waste as:

[. . .] something that is perceived to have either no inherent value to its owner, or the amount of effort required to access that value is greater than the expected return [. . .] waste is a residual that is discarded.

RL disposition decision-making

249

 

 

In regard to the disposition decision, relegating a product to waste entails deriving no more value from that product. Subsequently, this alternative is the least desirable disposition alternative as the business implications end at this juncture. However, waste management has become an important topic in recent years as economic and environmental forces demand environmentally-friendly and cost-effective handling of waste.

The preceding four disposition alternatives and their operational definitions assimilate the extant literature in this area for the purpose of bringing understanding to the options available to a firm when making the disposition decision. As noted above, the exact terminology and definitions of the alternatives differ within the literature. Readers interested in the disposition alternatives (or recovery options) are encouraged to Blackburn et al. (2004), Carter and Ellram (1998), Kopicki et al. (1993), Krikke et al. (2004), Prahinski and Kocabasoglu (2006), Rogers et al. (2002), Staikos and Rahimifard (2007), Stock (1992) and Thierry et al. (1995).

While much literature exists in the general area of RL, and a literature base is being built in disposition alternatives, no comprehensive analysis of extant literature for the purposes of extracting or deriving decision considerations or creating a disposition decision-making framework has been conducted. Because of the relative lack of literature specifically in the area of dispositions, our study analyzes literature in RL that addresses decision-making in general to determine those variables that appear throughout the literature. Then, we synthesize those findings into higher-level, decision-making components, which are applied to the disposition decision. The method used for identifying and assimilating these components is described next.

Methodology The purpose of this study is to identify the components of the RL disposition decision and suggest a decision-making framework. To serve this purpose, the authors required a method that would uncover the variables considered in extant RL decision-making literature. In short, we needed to derive meaningful content to address our specific purpose from a large amount of textual literature. Berelson (1952) suggests that revealing the focus of attention is one of the primary uses of content analysis. In addition, Neuman (2006) asserts that content analysis is useful for three primary types of research problems:

(1) problems involving a large amount of text;

(2) problems that must be studied from afar because of either necessity or to attain the proper scope; and

(3) problems where casual observation may not reveal the proper solution.

In the case of this research, we needed to assimilate a large amount of text to thoroughly investigate relevant variables. When viewed from afar and via conscious examination, this text may reveal a solution to our problem. Accordingly, a content analysis method was adopted for this study.

Content analysis is “a research technique for making replicable and valid inferences from texts (or other meaningful matter) to the contexts of their use” (Krippendorff, 2004, p. 18). Content analysis can be strictly quantitative when used to objectively and systematically count and record symbolic content from text. Content analysis can also

IJPDLM 42,3

250

 

 

be utilized for qualitative purposes to interpret meaning from text. Indeed, many content analysis procedures employ both qualitative and quantitative elements (Holsti, 1968).

As with many research methods, content analysis encompasses a wide variety of procedures and techniques that may be employed for use in a variety of settings to solve a multitude of problems. As such, there is no one systematic checklist to follow when conducting content analysis. This lack of standardization requires that great care be taken to develop a specific method that will yield appropriate results to answer the given research question at hand. Regardless of the specific procedure employed, the techniques must result in findings that are replicable in order to attain sufficient levels of reliability. In addition, measures must be taken to enhance validity whenever possible. Although discussion of the many ways in which to bolster reliability and validity are beyond the scope of this manuscript, we will discuss the specific techniques we employed in our research later in this article.

In this research, we adapted procedures for problem-driven content analysis suggested by Krippendorff (2004) to locate relevant materials for analysis, define the units of analysis, develop recording procedures, present the findings, infer results, and demonstrate empirical validity. These steps were carefully chosen and meticulously performed to efficaciously derive meaning from the textual content while enhancing reliability and validity to the fullest extent. The procedures and outcomes of each of these steps will be discussed throughout the remainder of this article in the order in which they were completed.

Relevant materials As discussed previously, the vast majority of extant RL literature addresses operational-level concerns of an organization. With a focus on optimizing specific operational efficiencies, much of this research is aimed toward clarifying various aspects of decisions that pertain to RL. As an example, Bhattacharya et al.’s (2006) research develops a mathematical model to determine optimal order quantities; Ferguson and Toktay’s (2006) research aims toward facilitating remanufacturing decisions. As such, this literature represents a rich body of content that is relevant to our problem, and thus useful for our analysis.

The scope of the literature search was limited to articles that designed, developed, tested, or otherwise utilized a decision support system (DSS) or simulation in regard to facilitating a decision within RL. Authors of these articles go to great lengths to identify and describe any possible variable or consideration that may be used in regard to their specific RL problem in order to enhance the relevance and validity of their research. To find articles that meet our criteria, the literature from the top eight journals in SCM, management information systems (MIS), and operations management (OM), as identified by Menachof et al. (2007, p. 151), Rainer and Miller (2005) and Gorman and Kanet (2005), respectively, was first considered. In alphabetical order by discipline, these journals are shown in Table I.

These journal listings yielded a total of 21 unique journals because of the interdisciplinary nature of both Management Science and Harvard Business Review. Logistics research spans a multitude of disciplines (Stock, 1997). However, the three selected disciplines encompass the vast majority of literature on the specific topic of RL decision-making and are therefore thought to appropriately limit the scope of the search. Although searching only the top journals in a field may not render exhaustive

RL disposition decision-making

251

 

 

results (Webster and Watson, 2002), a comprehensive interdisciplinary analysis of this nature requires a limited scope in the preliminary search of literature. Furthermore, this listing served more as a beginning reference than as a definitive boundary. Investigation into the literature revealed additional journal titles that pertained to this topic and were subsequently explored.

This review examined all applicable literature from 2000 through 2010. In their review of RL literature, Carter and Ellram (1998) propose that the first academic work in the field was not published until the early 1990s (Kopicki et al., 1993; Stock, 1992). Their review also notes that the majority of literature throughout the 1990s was exploratory in nature, offering little theoretical grounding. Accordingly, the vast majority of RL literature is published after the year 2000, thus presenting a logical limit to the scope of this review. The authors’ objective was to determine which variables are currently being utilized in the literature. This dictated that the review reach back far enough to provide an appropriate number of articles, but not so far as to lose contemporary relevance. The year 2010 was chosen as an upper limit so as to facilitate a comprehensive search of a selected period, thus limiting the possibility of inadvertently omitting newly published literature within the stated scope of the review.

All selected journals are searchable via electronic format and were thus accessed electronically. Broad keyword searches of each journal were used to generate a large number of search results that the authors were then able to evaluate more closely. The keywords used were: CLSC, end-of-life, return, disposition, RL, decision, model, and simulation. Titles and abstracts were then reviewed to find any literature that developed a DSS or simulation within RL. Specific articles not available in electronic format were requested and received through inter-library loan. The research process yielded 73 articles that the authors were able to read and analyze for adherence to the established criteria of utilizing a DSS or simulation within RL for the purpose of decision-making. Rudimentary citation analysis of the original articles directed the authors toward other journals which, although not encompassed within the original search, were deemed to be highly applicable to this study. These additional journals are discussed below.

Supply chain management Management information systems Operations management

Harvard Business Review Communications of the ACM IIE Transactions International Journal of Logistics Management

Decision Sciences Journal of Scheduling

International Journal of Physical Dist & Logistics Management

Decision Support Systems Manufacturing and Service Operations Management

Journal of Business Logistics Harvard Business Review Mathematics of Operations Research

Journal of Operations Management

Information Systems Research Management Science

Management Science Journal of Management Information Systems

Operations Research

Supply Chain Management Review

Management Science Production and Operations Management

Transportation Journal MIS Quarterly Transportation Science

Table I. Top journals by field

IJPDLM 42,3

252

 

 

Much of the literature in this area is in the OM field, which prompted further exploration into more of these top journals. Thus, we searched the additional journals found in Gorman and Kanet’s (2005) ranked listing of OM journals. The same search procedure was performed on the additional journals; this yielded a total of 25 additional articles. In total, 98 articles that potentially met our criteria were found and acquired via inter-library loan or downloaded in full from an electronic database. However, upon close examination of the 98 articles, only 62 met the specific criteria and were reviewed. A listing of the 62 articles used for content analysis can be found in the Appendix.

Unit of analysis and recording Krippendorff’s (2004, p. 83) first component of content analysis involves unitizing, which he defined as “the systematic distinguishing of segments of text – images, voices, and other observables – that are of interest to an analysis”. In this study, we are interested in the considerations used in RL literature. Accordingly, our unit of analysis was the individual variable or parameter (we will refer to these as “considerations”) utilized in the research.

In regard to content analysis, a primary purpose of recording is to transform original texts into analyzable representations (Krippendorff, 2004). In respect to maintaining consistency, this stage of the content analysis was conducted by only one of the authors. The author began the recording process by searching each article for the considerations used in the simulation. The author investigated model explanations, discussions of variables, parameters, assumptions, and other applicable areas of the selected literature to extract and tabulate all considerations addressed in the respective article. In sum, the author simply recorded the variables considered in each study, as stated explicitly in the journal article. Each consideration and the article of origin were then recorded on a spreadsheet. Articles had as few as five and as many as 18 considerations each. A master listing of findings was compiled on a spreadsheet document such that the authors could analyze the findings.

Findings The content analysis yielded a total of 60 considerations used by the authors of the selected literature. These considerations are shown in Table II, in order of most frequently utilized (top left) to least frequently utilized (bottom right). The number next to each consideration represents how many different times the consideration was identified during the analysis of the articles.

Inferred results Abductively inferring contextual phenomena from textual data bridges the gap between descriptive accounts of the text and what the data actually mean by pointing to unobserved phenomena of interest to the analysts (Krippendorff, 2004). At this point in the content analysis, we had compiled and counted a listing of considerations without assigning meaning to the data. Subsequently, we next reviewed this data to extrapolate autocorrelation functions. Upon reaching agreement, seven general factors were extrapolated from the listing of 60 variables. These factors are: supply chain capabilities, costs of RL, profit from RL, environmental impact of RL, regulation, market considerations, and customer behavior.

RL disposition decision-making

253

 

 

In the literature used for analysis, supply chain capabilities encompass a myriad of variables that address whether or not a firm is prepared to commence various RL operations from a logistical perspective. Examples of such variables include a firm’s remanufacturing and inventory capacity. Therefore, we define supply chain capabilities as a firm’s existing resource capacity available for RL activities.

Cost of RL was represented in the literature in a variety of ways, to include labor and administrative costs. As such, we define cost of RL as the costs incurred to facilitate effective RL activities. Similarly, profit from RL was represented with variables such as revenues from recycling and profit margins from RL activities. We therefore define profit from RL as any profit realized via the employment of RL activities.

We found several instances where the environmental impact of RL was considered in the literature (Aksen et al., 2009; Haas et al., 2003; Staikos and Rahimifard, 2007). We define environmental impact as any consequence (positive or negative) of the practice of RL activities to the natural environment. Although regulation is often developed to address environmental concerns, environmental issues are not the only drivers of regulation. As such, we consider regulation to be separate from environmental

Customer Demand (40) Manufacturing Capacity (13) Outsourcing (3PL/4PL) (6) Product Return Volume/Rate (38) Wholesale Price (12) Environmental

Considerations (6) Remanufacturing Costs (32) Number of OEMs (Monopoly,

etc.) (9) Length of Time Customer Holds Product (5)

Cost of Acquiring Returned Product (30)

Existing Logistics Infrastructure (9)

Cost of Capital (5)

Management Strategy/Policy (23) Salvage Value (12) Safety stock (4) Inventory Costs (23) Supply of Parts Required for

Remanufacture (12) Stocking points (4)

Disposal Considerations/Scrap Costs (23)

Market Size (11) Service Level (3)

Leadtime (22) Quality of Remanufactured Item (11)

Factory Location (3)

Retail Price (22) Disassembly (Cost/Time) (11) Number of Remanufacturers (3)

Transportation Costs (21) Remanufactured Item Inventory Level (11)

RL Administrative Costs (3)

Manufacturing Costs (20) New Item Inventory Level (9) Processing Times (2) Inspection Costs (20) Delays (9) Packaging (2) Remanufacturing Capacity (18) Recycling Costs (8) Discontinuation Price (2) Sales Lost/Backorder Costs (18) Demand for Remanufactured

Part (8) Customer Segment (2)

Remanufactured Item Sales Price (17)

Revenue from Recycling (8) Value of Time (2)

Profit Margin of Remanufacturing (15)

Lot/Batch Size (7) Penalty Costs of Uncollected Returns (2)

Product Lifecycle (15) Labor Cost (7) Sorting Policy (1) Fixed Costs (15) Quantity Recycled (7) Total Quantity of Items in

Supply Chain (1) Return Quality (14) Pattern of Recovery (Collection

Location) (7) Total Cost of RL (1)

Total Serviceable Item Levels (13) Legal Considerations (7) Forecast (1)

Table II. Variables used in quantitative RL literature

IJPDLM 42,3

254

 

 

impact and found that regulation is often addressed in the literature in regard to legal constraints and compliance with environmental law. We define regulation as any law or directive imposed by a governing body that influences RL.

Market conditions in the literature include the size of the market and number of competing firms. We narrow the definition of market conditions to entail the competitive forces realized via the existence and actions of industry competitors. In respect to Porter’s (1980) five forces, this includes the threat of new entrants, threat of substitute products, the bargaining power of suppliers, and rivalry among existing firms. Notably, this does not include the impact of customers. Our definition of market conditions differs in scope from existing conceptualizations in the supply chain and marketing literature and represents a fundamental difference between the forward and RL process. Extant supply chain literature suggests that market conditions are largely affected by customer demand (Fisher, 1997; Pagh and Cooper, 1998). In contrast, in a CLSC, we posit that customers encompass a role beyond that of simply placing a demand on the supply chain or wielding bargaining power. For example, in the CLSC literature, we found that the length of time a consumer holds a product and the condition in which the product is returned are important considerations (Geyer et al., 2007; Toktay et al., 2000; Vorasayan and Ryan, 2006). Thus, we separate customer behavior into its own category apart from market conditions in the RL context. We define customer behavior as any action taken by a customer that impacts a firm’s RL activities. For example, customer demand for reused, recycled, or remanufactured products and the willingness to return used or unused products to the supply chain constitute important customer behaviors.

These analytical constructs defined above (which we will refer to as components of the disposition decision) were extrapolated from the literature and used as the basis for categorization of variables based upon our knowledge and experience with the established theories of SCM. The extrapolation of these seven components suggests that these factors embody the considerations required for RL disposition decision-making. However, in order to qualify these results, empirical validity was tested.

Empirical validity In order to measure how much these components converge with and are discriminate from current RL literature, we evaluated current RL frameworks and compared the findings of the content analysis with the factors addressed in each framework. In this section, we discuss our literature search procedures for finding these RL frameworks, our evaluation of each framework, and the conclusions we drew regarding the validity of our findings.

Literature search The criterion for a usable article was simple and explicit: it must contain an RL framework. Although a rather broad parameter, an exhaustive search for such literature yielded very few results. Again using listings provided via the research of Menachof et al. (2007), Rainer and Miller (2005) and Gorman and Kanet (2005) as a starting point, the top SCM, MIS, and OM journals were searched. Keyword search terms were: closed-loop, end-of-life, return, disposition, RL, decision, model, and framework. The scope of this review in regard to range of dates was limited only by the start date, which was 1998. This date was chosen as this was the year of Carter and Ellram’s (1998) initial

RL disposition decision-making

255

 

 

framework publication. Although numerous articles were retrieved and reviewed for adherence to criteria, this process yielded just seven usable articles. In an attempt to find more literature, we conducted a broad search of the Business Source Premier and ABI/Inform databases. The same keyword searches generated a variety of additional articles. However, only four additional articles met the criteria of this search.

Evaluating frameworks The exhaustive search detailed above yielded 11 articles which met our criteria. In order to validate our results, the frameworks from these articles were evaluated to determine percentage of agreement with the considerations identified by our content analysis. Each of the three authors evaluated the frameworks and identified the areas addressed in each framework. Krippendorf’s a for reliability of this process was calculated to be 0.85, which is considered sufficient to draw meaningful conclusions (Krippendorff, 2004). The results of this comparison are shown in Table III, which indicates the considerations encompassed by each framework.

Correlative validity analysis Convergent validity is “the extent to which results correlate with variables known to measure the same phenomena and considered valid” (Krippendorff, 2004, p. 319). In this study, we were concerned with the validity of the components that we extrapolated from the content analysis. Subsequently, we tested the convergent validity of our findings by comparing each framework against the seven components identified by our analysis. If each component was represented in at least one other framework, then it would follow that it was relevant to RL. Percent agreement was calculated by dividing the number of frameworks that utilized each component by the total number of frameworks evaluated and is shown in the bottom row of Table III. As the results demonstrate, all components have been utilized in previous frameworks. This suggests adequate convergence, which indicates that our findings are relevant to the context of RL.

"Order a similar paper and get 100% plagiarism free, professional written paper now!"

Order Now