Green supply chain management in the electronic industry

Transcription

1 Int. J. Environ. Sci. Tech., 5 (2), , Spring 2008 ISSN: C. W. Hsu; A. H. Hu IRSEN, CEERS, IAU Green suppy chain management in the eectronic industry *C. W. Hsu; A. H. Hu Institute of Environmenta Engineering and Management, Nationa Taipei University of Technoogy,1, Sec.3, Chung-Hsiao E. Rd., Taipei 10643, Taiwan, R.O.C. Received 31 January 2008; revised 18 February 2008; accepted 29 February 2008; avaiabe onine 10 March 2008 ABSTRACT: Green suppy chain management has emerged as a proactive approach for improving environmenta performance of processes and products in accordance with the requirements of environmenta reguations. Various approaches for impementing green suppy chain management practices has been proposed and recognized in previous iteratures, yet no investigation has identified the reiabiity and vaidity of such approaches particuary in eectronic industry. This study examines the consistency approaches by factor anaysis that determines the adoption and impementation of green suppy chain management in Taiwanese eectronic industry. The fuzzy anaytic hierarchy process method is appied to prioritize the reative importance of four dimensions and twenty approaches among nine enterprises in eectronic industry. The findings indicate that these enterprises woud emphasize on suppier management performance in the crucia roe of impementing green suppy chain management. Estabishing an environmenta database of products, asing for product testing report and top management support are among the most important approaches. The resuts for the impications of green suppy chain management impementation in eectronic industry investigated in this wor generate a generic hierarchy mode for decision-maers who can prioritize those approaches for impementing green suppy chain management in Taiwan. Key words: Priority approaches, fuzzy anaytic hierarchy process, suppy chain environmenta management INTRODUCTION With increase in environmenta concerns during the past decade, a consensus is growing that environmenta poution issues accompanying industria deveopment shoud be addressed together with suppy chain management, thereby contributing to green suppy chain management (GSCM) (Sheu et a., 2005). Since the Waste Eectrica and Eectronic Equipment (WEEE), Restriction of Hazardous Substances (RoHS) and Ecodesign for Energy using Products (EuP) directives were passed by the European Union (EU), GSCM has been adopted as a proactive strategy by eading eectronic industry companies, incuding De, HP, IBM, Motoroa, Sony, Panasonic, NEC, Fujitsu and Toshiba (Zhu and Saris, 2006). Nowadays, simiar reguation has been spread throughout the word in the US, China, Japan and Korea; the emerging issue of green product seems to be quicy piced up in Asia. Some countries such Japan, Taiwan and Korea are the herads in terms of green eectronic products (Boysère and Beard, 2006). Thus, it is inferred that GSCM practice can be viewed as the primary strategy capabe of compying with the requirements of egisations and maintaining the *Corresponding Author Emai: s @ntut.edu.tw Te.: ; Fax: competitive advantage. The issue of GSCM is significant for Taiwan s eectronics industry as recent studies have shown that most of the word s manufacturing wi be reocated to Asia within the next two decades (US-AEP, 1999). Therefore, GSCM is an operationa initiative on the part of many organizations, incuding those in Asia and South Asian region which are adopting to address such environmenta issues (Rao and Hot, 2005). Taiwan is one of the most industriaized countries in the Asia-Pacific region. Most eectrica and eectronics manufacturers in Taiwan are invoved in Origina Equipment Manufacturing (OEM) and Origina Design Manufacturing (ODM). These companies pay important roes in goba marets as their products share a substantia portion in maret. Once, Taiwan was the third argest producer of information products a round the word (Chen, 2004). Nevertheess, these industries are subject to customer requests for green products and green manufacturing that compy with emerging environmenta directives (such as WEEE, RoHS and EuP). These directives, especiay the RoHS, directy impact the eectrica and eectronics industries in Taiwan and export products to the EU-exports in 2004 exceeded US$7.8 biion. More

2 C. W. Hsu; A. H. Hu than 3,000 companies were affected by the directives of the EU. These directives aso have a far-reaching infuence on suppy chain partners for mutinationa enterprises (Huang, 2005). Athough, to the best of our nowedge, various investigations have proposed different approaches to impement GSCM (Lamming and Hampson, 1996; Lippmann, 1999; US-AEP, 1999; Bowen et a., 2001; Yuang and Kieiewicz-Yuang, 2001; Rao, 2002; Evans and Johnson, 2005; Zhu et a., 2005), there have been far ess research on identifying the consistency and priority approaches to GSCM impementation with the systematic anaysis, particuary in eectronics industry. This is because the compexity of GSCM practices, customer and cost pressures and reguation uncertainty, impementing GSCM is considered as a thaness tas that increases overa product cost. For exampe, the RoHS directive acs a standardized test procedure and an updated exemption annex of chemicas. These shortcomings resut in significant probems when impementing GSCM. Furthermore, increased reguations - RoHS-EU, RoHS-Korea and RoHS-China - resut in difficuties executing GSCM practices. Hence, enterprises cannot determine whether their executive strategies conform to reguations or ensure that current management approaches are woring and have a ow ris. Consequenty, enterprise embraces the appropriate approaches for impementing GSCM practice and it is significant to mitigate potentia riss from green suppy chain. The centra purpose of this study is to estabish the consistency and priority approaches for impementing GSCM in response to environmenta reguations of WEEE, RoHS and EuP. The fuzzy anaytic hierarchy process (FAHP), which is appied to conduct the reative importance of different approaches, is extremey crucia, since the resuts can be used by managers impementing and adopting their own GSCM practices. MATERIAL AND METHODS To improve the AHP method and to recognize consistent strategies for impementing GSCM, this study appies the FAHP and uses trianguar fuzzy numbers to express comparative judgments of decisionmaers. A systematic approach of FAHP to identify priority approaches for GSCM impementation was adopted based on a compex and muti-criteria environment. From September to November 2006, the authors utiized FAHP to recognize the priority approaches that wi affect the impementation of GSCM practice in Taiwanese eectronics industry. The methodoogy consists of three phases, incuding (1) constructing the hierarchy of GSCM practice via factor anaysis, (2) coecting data from industria expert interviews and (3) determining the normaized weights of individua dimensions and approaches. The GSCM practices impementation and FAHP wi be discussed through a iterature review and the construction of the hierarchy of GSCM impementation wi aso be presented afterwards. Approaches to GSCM impementation A number of approaches for impementing GSCM practice have been proposed in previous iterature, in which they are aimed at mitigating the riss associated with green suppy chain interruptions or deays and protecting a company s reputation and brand image from damaging pubic controversies. Various approaches to GSCM practice have been identified by various researches; they are briefy outined beow and summarized as shown in Tabe 1. Lamming and Hampson (1996) expored the concepts of environmentay sound management (e.g. ife cyce anaysis, waste management, product stewardship and the ie) and ined them to suppy chain management practices such as vendor assessment, ean suppy, coaborative suppy strategies, estabishing environmenta purchasing poicy and woring with suppiers to enabe improvements. Lippmann (1999) proposed various critica eements for the successfu impementation of suppy chain environmenta management. Those components incude the production of written GSCM poicies, suppier meetings, training, coaborative R&D, top-eve eadership, cross-functiona integration, effective communication within companies and with suppiers, effective processes for targeting, evauating, seecting and woring with suppiers and restructuring reationships with suppiers and customers. US-AEP (1999) improved understanding of industry approaches to suppy chain environmenta management (SCEM) by focusing on seven major eectronics firms. Some of the common SCEM toos empoyed by these firms are summarized as foows: Prequaification of suppiers Environmenta requirements during the purchasing phase Suppy base environmenta performance management 206

3 Int. J. Environ. Sci. C. W. Tech., Hsu; 5 A. (2), H , Hu Spring 2008 Tabe 1: Approaches for GSCM impementation emphasized by seected authors Approach Representative references 1. Suppiers meeting Lippmann (1999), Yuang and Kieiewicz Yuang (2001), Rao (2002) 2. Environmenta auditing for suppiers Lippmann (1999), US-AEP (1999), Yuang and Kieiewicz Yuang (2001), Zhu et a. (2005) 3. Suppiers environmenta questionnaire Lamming and Hampson (1996), US-AEP (1999), Bowen et a. (2001), Rao (2002), Evans and Johnson (2005) 4. Requesting compiance statement Yuang and Kieiewicz Yuang (2001), Evans and Johnson (2005) 5. Asing for product testing report Evans and Johnson (2005) 6. Demanding Bi of materia (BOM) Evans and Johnson (2005) 7. Estabishing environmenta requirements for purchasing items Lamming and Hampson (1996), Lippmann (1999), US-AEP (1999), Evans and Johnson (2005) 8. Impementing green purchasing Yuang and Kieiewicz Yuang (2001), Rao (2002), Zhu et a. (2005) 9. Coaborative R&D with suppiers Lamming and Hampson (1996), Lippmann (1999), US-AEP (1999), Bowen et a. (2001), Rao (2002) 10. Information system Yuang and Kieiewicz Yuang (2001), Evans and Johnson (2005) 11. Joining oca recycing organizations US-AEP (1999), Bowen et a. (2001), Rao (2002) 12. Coaboration on products recycing with the same sector industry Yuang and Kieiewicz Yuang (2001) 13. Produce disassemby manuas Lamming and Hampson (1996), Rao (2002) 14. Green design US-AEP (1999), Yuang and Kieiewicz Yuang (2001), Rao (2002) 15. Environmenta education and training Lippmann (1999), Yuang and Kieiewicz Yuang (2001) 16. Top management support Lippmann (1999), US-AEP (1999), Evans and Johnson (2005) 17. Environmenta poicy for GSCM Lamming and Hampson (1996), Lippmann (1999), US-AEP (1999), Yuang and Kieiewicz Yuang (2001) 18. Cross-function integration Lippmann (1999), US-AEP (1999), Yuang and Kieiewicz Yuang (2001), Evans and Johnson (2005) 19. Manpower invovement Rao (2002) 20. Effective communication patform within companies and with suppiers Lippmann (1999) 21. Estabish a environmenta ris management system for GSCM Bowen et a. (2001) 22. Suppier evauation and seection Lamming and Hampson (1996), Yuang and Kieiewicz Yuang (2001), Rao (2002) 23. Tracing the deveopment of directives US-AEP (1999) 24. Appying LCA to carry out eco-report Lamming and Hampson (1996), US-AEP (1999), Rao (2002) 25. Estabish an environmenta database of products Lamming and Hampson (1996) 207

4 Green management C. W. Hsu; in A. eectronic H. Hu industry Buiding environmenta considerations into product design Cooperating with suppiers to dea with end-of-pipe consumer environmenta issues Reverse ogistics Infuencing egisation to faciitate better SCEM poicies Woring with industry peers to standardize requirements (for suppiers and purchasing items) Informing suppiers of corporate environmenta concerns Promoting the exchange of information and ideas Bowen et a. (2001) conducted an exporatory anaysis of impementing patterns and inductivey derived three main types of green suppy. The first type, i.e. greening the suppy process, represents adaptations to suppier management activities, incuding coaboration with suppiers to eiminate pacaging and recycing initiatives. The second type, i.e. product-based green suppy, attempts to manage the by-products of suppied inputs such as pacing. The third type, i.e. advanced green suppy, incudes more proactive approaches such as the use of environmenta criteria in ris-sharing, evauation of buyer performance and joint cean technoogy programs with suppiers. Yuang and Kieiewicz-Yuang (2001) presented an overview of current practices in managing sustainabiity issues in suppy networs. Crossfunctiona teams, consisting of saes, environmenta personne, purchasing personne and personne from other reevant departments, can be found in organizations with the most advanced strategies for sharing sustainabiity-oriented information. Organizations mae avaiabe to their customers/ suppiers their sustainabiity purchasing poicy, goas and future targets via open days. Moreover, organizations have specific criteria as we as recognized standards (ISO14001), technica (ead-free sodering) and performance specifications that its suppiers must meet to be recognized as preferred suppiers. In addition, suppier performances can be enhanced through on-site third-party auditing or periodic sef-assessment by suppiers. Through coaboration with suppiers, training not ony is administered to companies that provide advice on sustainabiity issues in purchasing, but aso is deivered to suppiers to provide them with information on product ife cyce. Rao (2002) argues that GSCM practices shoud incude woring coaborativey with suppiers on green product designs, hoding awareness seminars, heping suppiers estabish their own environmenta programs and so on. To green the suppy chain, from the perspective of practitioners, companies have to integrate the ideas of green purchasing tota quaity management in terms of empoyee empowerment, customer focus, continuous improvement and zero waste, ife cyce anaysis and environmenta mareting. Green purchasing comprises a number of environment-based initiatives, incuding a suppier environmenta questionnaire, suppier environmenta audit and assessments, environmenta criteria for designating approved suppiers, requiring suppiers to undertae independent environmenta certification, jointy deveoping ceaner technoogy/ processes with suppiers, engaging suppiers in ecodesign and product/ process innovation. Evans and Johnson (2005) suggested that manufacturers shoud insta documented and auditabe systems to prevent non-compiant products from entering the EU. The instaation of such systems woud invove three steps. The first step is to determine the ega exposure of the company to the EU directives and senior management must support this initia exposure assessment. The next step is to assign the tas to a corporate-wide compiance team due to the compex requirements of the RoHS directive. The third step is to deveop a corporate compiance statement, which shoud incude a date for compiance and might aso outine suppier requirements, i.e. testing, documentation and so on. Through assessing the suppy chain exposure to the EU directives, companies can estabish the materia decaration process. However, companies shoud aso quaify suppiers to determine their eve of RoHS preparedness via questionnaires. Zhu et a. (2005) described a number of GSCM practices impemented by Chinese enterprises to improve their performance. Interna environmenta management is a ey to improving enterprise performance in terms of senior manager commitment and cross-functiona cooperation. Commitment of senior managers is extremey conducive to the impementation and adoption stages for GSCM, because without such upper management commitment most programs are bound to fai. A GSCM practices are integrative and require cross-functiona cooperation rather than simpy being oriented to a singe function or department. They suggested that 208

5 Int. J. Environ. Sci. C. Tech., W. Hsu; 5 (2), A. H , Hu Spring 2008 green purchasing and eco-design are two emerging approaches and companies shoud focus on the inbound or eary portions of the product suppy chain. Currenty, arge customers have exerted pressure on their suppiers to achieve better environmenta performance, resuting in greater motivation for suppiers to cooperate with customers for environmenta objectives. Fuzzy anaytic hierarchy process The anaytica hierarchy process (AHP) method, introduced by Saaty (1980), directs how to determine the priority of a set of aternatives and the reative importance of attributes in a muti-criteria decisionmaing (MCDM) probem (Saaty, 1980; Wei et a., 2005). The primary advantage of the AHP approach is the reative ease with which it handes mutipe criteria and performs quaitative and quantitative data (Kahraman et a., 2004; Meade and Saris, 1998). However, AHP is frequenty criticized for its inabiity to adequatey accommodate the inherent uncertainty and imprecision associated with mapping decision-maer perceptions to extract number (Kwong and Bai, 2003; Chan and Kumar, 2007). It is difficut to response to the preference of decision-maers by assigning precise numerica vaues. The FAHP, therefore, was appied to determine weight among various approaches for impementing GSCM and provided the priority of those approaches for enterprise to adopt and adjust their current GSCM practices. Some cacuation steps are essentia and expained as foows: Step 1: Estabishing the hierarchica structure Constructing the hierarchica structure with decision eements, decision-maers are requested to mae pairwise comparisons between decision aternatives and criteria using a nine-point scae. A matrices are deveoped and a pair-wise comparisons are obtained from each n decision-maer(s). Step 2: Cacuating the consistency To ensure that the priority of eements is consistent, the maximum eigenvector or reative weights and λmax is cacuated. Then, the consistency index (CI) for each matrix order n is computed by using Eq. (1). Based on the CI and random index (RI), the consistency ratio (CR) is cacuated using Eq. (2). The CI and CR are defined as foows (Saaty, 1980): λmax n C. I. = n 1 C. I. C. R. = (2) R. I. where, n is the number of items being compared in the matrix, λmax is the argest eigenvaue and RI is a random consistency index obtained from a arge number of simuation runs and varies upon the order of matrix (see Tabe 2). Step 3: Constructing a fuzzy positive matrix A decision-maer transforms the score of pair-wise comparison into inguistic variabes via the positive trianguar fuzzy number (PTFN). The fuzzy positive reciproca matrix can be defined as (Bucey, 1985) ~ [ ] ~ A = A ij where, A ~ is a fuzzy position reciproca matrix of decision-maer ; A ~ ij is the reative importance between i and j of decision eements ~ ~ A = 1, i = j, A = 1 A, i, j = 1,2,..., n ij ij Step 4: Cacuating fuzzy weights vaue According to the Lambda-Max method proposed by Csutora and Bucey (2001), the fuzzy weights of the hierarchy can be cacuated. This process is described as foows: Let á=1 to obtain the positive matrix of decision- A ~ = a. Then, appy the AHP to cacuate maer m [ ijm ] n x n weight matrixw. [ ] m W m = w im, i = 1, 2,..., n (4) Let α=0 to obtain the ower bound and upper bound A ~ = a of the positive matrix of decision-maer, [ ij ] n x n and u [ a iju ] n x n ij (1) (3) A ~ =. Then, appy the AHP to cacuate the weight matrix: W and W. u Tabe 2: Random index N R.I

6 C. W. Hsu; A. H. Hu W W u = = [ w i ], i = 1, 2,..., n [ ], i = 1, 2,..., n w iu To ensure the fuzziness of weight, two constants, i.e. S and, are cacuated as foows: = min S u w im S = min 1 i n (7) w i S u w im w iu 1 i n * The ower bound ( W ) and upper bound ( W the weight matrix are defined as: * * * W = w i, w i = S w i, i = 1, 2,..., n * u (5) (6) (8) ) of * * * W u = w iu, w iu = S u w iu, i = 1, 2,..., n (10) * * * Aggregating W, W and m W, the fuzzy weight u for decision-maer can be acquired as foows: ~ * * * W i = w i, w im, w iu, i = 1, 2,..., n (11) Appying the geometric average to incorporate the opinions of decision-maers is defined as foows: 1 ~ 1 ~ ~ n ( W W... W ) ~ 2 i = i i W where, W ~ : the fuzzy weight of decision-maers i is i incorporated with K decision-maers. W ~ : the fuzzy weight of decision eement i of i decision-maer. : number of decision-maers. i (9) (12) Constructing the hierarchy of GSCM practice via factor anaysis Through a thorough and detaied anaysis of the pertinent iterature and in-depth interviews with three senior quaity assurance and product assurance representatives, a tentative ist of 25 approaches of GSCM was deveoped as the basis for questionnaire deveopment. Respondents were ased to rate each item under a five-point Liert-type scae (e.g. 1 = not at a important, 2 = not important, 3 = moderate, 4 = important, 5 = extremey important) to indicate the extent to which each item was practiced in their respective organization. This investigation focused on samping the perceptions and experiences of GSCM-based companies in the Taiwanese eectrica and eectronics industries. Data coection invoved distributing questionnaires invoved in eectronic enterprises. Target respondents were seected from among the members of Taiwan Eectrica and Eectronics Manufacturers Association (TEEMA). Questionnaires were addressed to both the managing directors of the quaity assurance and purchasing departments of the target organizations. This was done because most of the sampe companies may ac GSCM representatives or departments. A tota of 300 questionnaires were maied out and 87 were returned, of which 84 were vaid, representing a response rate of 28%. According to the study on deveopment and vaidation of critica factors and environmenta management, its response rate is 21.9% (Wee and Quazi, 2005). Furthermore, Antony et a. (2002) aso pointed out that their research got 16.5%, which was norma and reasonabe. This impies that the response rate of this study is acceptabe and it refects the virtue of nove issue of the GSCM practice. The authors utiized factor anaysis to extract factors based upon the principa components anaysis with varimax rotation. Bartett s test of sphericity and the Kaiser-Meyer-Oin (KMO) measure of samping adequacy were empoyed to test the appropriateness of the data for factor anaysis (Kaiser, 1974; Bagozzi and Yi, 1998). The test resuts of KMO show that the compared vaue is 0.863, significanty exceeding the suggested minimum standard of 0.5 required for conducting factor anaysis (Hair et a., 1995). The authors performed factor anaysis to extract factors in accordance with the eigenvaues of discontinuity which is greater than 1 (Tabachnic and Fide, 1989) and factor oading exceeding 0.6 were necessary to choose factors (Kine, 1997). These are five variabes eiminated 210

7 Int. J. Environ. Sci. C. Tech., W. Hsu; 5 (2), A , H. Hu Spring 2008 Tabe 3: Trianguar fuzzy numbers (Lee et a., 2008) Linguistic variabes Positive trianguar fuzzy number Positive reciproca trianguar fuzzy number Extremey strong (9, 9, 9) (1/9, 1/9, 1/9) Intermediate (7, 8, 9) (1/9, 1/8, 1/7) Very strong (6, 7, 8) (1/8, 1/7, 1/6) Intermediate (5, 6, 7) (1/7, 1/6, 1/5) Strong (4, 5, 6) (1/6, 1/5, 1/4) Intermediate (3, 4, 5) (1/5, 1/4, 1/3) Moderatey strong (2, 3, 4) (1/4, 1/3, 1/2) Intermediate (1, 2, 3) (1/3, 1/2, 1) Equay strong (1, 1, 1) (1, 1, 1) tests and item anaysis were recacuated without those five items. Tabe 4 ists the Cronbach s apha vaues, ranging from to , after the five items were dropped. Generay, Cronbach s apha vaue exceeding 0.7 is considered to have high interna consistency of scae (Lin et a., 2002). A the Cronbach s apha vaues in our study are greater than 0.7, reveaing the high interna consistency. Content vaidity depends on how we the researchers create measurement items to cover the content domain of the variabe measured (Nunnay, Dimension Suppier management (D 1) Approaches 1. Environmenta auditing for suppiers (D 11) 2. Suppier environmenta questionnaire (D 12) 3. Requesting compiance statement (D 13) 4. Asing for product testing report (D 14) 5. Demanding bi of materia (BOM) (D 15) 6. Estabishing environmenta requirements for purchasing items (D 16) 7. Impementing green purchasing (D 17) Product recycing (D 2) 1. Joining oca recycing organization (D 21) 2. Coaboration on products recycing with the same sector industry (D 22) 3. Produce disassemby manua (D 23) Organization invovement (D 3) Life cyce management (D 4) Tabe 4: Factor anaysis resuts 1. Green design (D 31) 2. Top management support (D 32) 3. Environmenta poicy for GSCM (D 33) 4. Cross-function integration (D 34) 5. Manpower invovement (D 35) 6. Effective communication patform within companies and with suppiers (D 36) 7. Estabish an environmenta ris management system for GSCM (D 37) 8. Suppier evauation and seection (D 38) 1. Appying LCA to carry out eco-report (D 41) 2. Estabish an environmenta database of products (D 42) Item oading range Eigenvaues Cumuative percentage Cronbach s apha because their factor oadings are ess than The remaining 20 items, therefore, were re-anayzed and extracted into four dimensions, which was denominated as suppier management, product recycing, organization invovement and ife cyce management as a hierarchy of the GSCM impementation (See Fig. 1). Reiabiity concerns the extent to which an experience, test or any measuring procedure yieds the same resuts on repeated trias (Carmines and Zeer, 1979). The reiabiity of the factors needs to be determined to support any measures of vaidity that may be empoyed (Nunnay, 1978). Interna consistency anaysis was performed to measure the reiabiity of the items under each critica factor using Cronbach s apha (Wee and Quazi, 2005). Both reiabiity 1967). Content vaidity is subjectivey evauated by researchers (Yusof and Aspinwa, 2000). The content vaidity of the questionnaire in this wor is based on an exhaustive iterature review and detaied evauations by three senior quaity assurance and product assurance practitioners. Consequenty, the measures of GSCM constructed by the factor anaysis have content vaidity. Criterion-reation vaidity, sometimes caed predictive vaidity or externa vaidity, is concerned with the extent to which a measuring instrument is reated to an independent measure of the reevant criterion (Badri and Davis, 1995). No criterions were designed to expore the correation with the performance of GSCM. The resuts of this study may provide a better understanding and hep identify the opportunities of GSCM impementation. 211

8 Green management C. W. Hsu; in A. eectronic H. Hu industry Goa Dimension Approach Environmenta auditing for suppiers (D 11) Suppier environmenta questionnaire (D 12) Suppier management (D 1) Requesting compiance statement (D 13) Asing for product testing report (D 14) Demanding bi of materia (BOM) (D 15) Estabishing environmenta requirements for purchasing items (D 16) Seection of priority approaches for impementing GSCM Product recycing (D 2) Organization invovement (D 3) Impementing green purchasing (D 17) Joining oca recycing organization (D 21) Coaboration on products recycing with the same sector industry (D 22) Produce disassemby manua (D 23) Green design (D 31) Top management support (D 32) Environmenta poicy for GSCM (D 33) Cross-function integration (D 34) Manpower invovement (D 35) Effective communication patform within companies and with suppiers (D 36) Estabish an environmenta ris management system for GSCM (D 37) Suppier evauation and seection (D 38) Life cyce management (D 4) Measuring and coecting data Managers who are responsibe for impementing GSCM of nine we-nown IT enterprises from Green Pan (GP) program were invented to attend a persona interview. These companies were famiiar with the GSCM practices in their organizations; they served as evauators to determine the reative weights against a Appying LCA to carry out eco-report (D 41) Estabish an environmenta database of products (D 42) Fig.1: The hierarchy of GSCM impementation given ist of approaches affecting impementation of GSCM practice. These companies were seected as the member of GP program which was aunched by the Ministry of Economic Affairs (MOEA) at the end of 2004 for promoting GSCM practice in compying with EU s environmenta reguations as WEEE, RoHS and 212

9 Dimension Suppier management (D 1) Product recycing (D 2) Organization invovement (D 3) Life cyce management (D 4) Int. J. Environ. Sci. C. Tech., W. Hsu; 5 (2), A , H. Hu Spring 2008 Tabe 5: Loca and goba weights of each approach for impementing GSCM Loca weights a Approach Loca Goba weights weights b Raning Environmenta auditing for suppiers (D 11) Suppier environmenta questionnaire (D 12) Requesting compiance statement (D 13) Asing for product testing report (D 14) Demanding bi of materia (BOM) (D 15) Estabishing environmenta requirements for purchasing items (D 16) Impementing green purchasing (D 17) Joining oca recycing organization (D 21) Coaboration on products recycing with the same sector industry (D 22) Produce disassemby manua (D 23) Green design (D 31) Top management support (D 32) Environmenta poicy for GSCM (D 33) Cross-function integration (D 34) Manpower invovement (D 35) Effective communication patform within companies and with suppiers (D 36) Estabish an environmenta ris management system for GSCM (D 37) Suppier evauation and seection (D 38) Appying LCA to carry out eco-report (D 41) Estabish an environmenta database of products (D 42) a. Loca weight is derived from judgment with respect to a singe criterion. b. Goba weight is derived from mutipication by the weight of the criteria EUP. Many eectronics companies, particuary in assemby manufacturers, are invoved and embrace various green initiatives to green their suppy chain, incuding three tass: estabishing an information management system for green suppy chains; deveoping a recycing system and management patform for green products and generating a certification database for green parts and components (MOEA, 2005). In terms of the extracted resuts of factor anaysis, a FAHP-based questionnaire survey was designed and deivered to nine industria experts who have substantia experience impementing GSCM practices for coecting data out of pair-wise comparisons. A nine-point scae was occupied to assign reative importance to pair-wise comparisons among dimensions and approaches. Determining the normaized weights To determine the reative importance of the dimensions and approaches, a set of pair-wise comparison matrices were transated into the eigenvectors probems and then were normaized to unify the resut so as to acquire the vectors of priorities. The geometric mean is utiized to aggregate the pair-wise comparisons for a sampes. The normaized oca and goba weights of the dimensions and approaches were generated by the procedure aforementioned (see Tabes 6 to 10). The resuts suggested that the overa consistency of respondents judgments fa within the acceptabe ratio of RESULTS AND DISSCUSION Nine enterprises have participated in this study, in which a were arge and medium assemby manufacturers in Taiwanese eectronics industries. In order to determine the importance of the dimensions and approaches, the judgments coected from respondents generated the normaized oca and goba weights for approaches to impementing GSCM. The resuts of priority weights determined the reative importance of individua dimensions and approaches and in turn recognized the points on which organizations shoud put their efforts throughout the process of GSCM impementation. In addition, the resuts coud represent the genera status of GSCM impementation in Taiwanese eectronics enterprises. Three sets of normaized weights had generated to determine the importance of approaches to GSCM impementation as shown in Tabe 5. The second coumn is the oca weights with respect to the dimensions; the fourth and fifth coumns are the oca and goba weights for each approach, respectivey. The goba weights of each approach have been cacuated by mutipying the oca weights of each approach by the oca weights of each dimension. 213

10 C. W. Hsu; A. H. Hu Tabe 6: Pairwise comparison matrix and weights with respect to the goa D 1 D 2 D 3 D 4 Weights D 1 (1, 1, 1) (4.932, 5.944, 6.952) (0.956, 1.387, 2.080) (1.817, 2.190, 2.653) D 2 (0.144, 0.160, 0.203) (1, 1, 1) (0.454, 0.575, 0.747) (0.630, 0.928, 1.357) D 3 (0.481, 0.721, 1.046) (1.339, 1.738, 2.201) (1, 1, 1) (3.420, 4.481, 5.518) D 4 (0.377, 0.457, 0.550) (0.737, 1.077, 1.587) (0.181, 0.223, 0.292) (1, 1, 1) λ max = CI=0.019 CR=0.021 Tabe 7: Pairwise comparison matrix and weights with respect to suppier management dimension D 11 D 12 D 13 D 14 D 15 D 16 D 17 Weights D 11 (1, 1, 1) (6.316, 7.319, 8.320) D 12 (0.120, 0.137, 0.158) (1, 1, 1) D 13 (0.271, 0.397, (1.587, 2.466, 0.606) 3.634) D 14 (1.000, 1.710, (2.520, 3.557, 2.884) 4.762) D 15 (1.587, 2.466, (4.579, 5.593, 3.634) 6.604) D 16 (0.511, 0.630, (2.520, 3.557, 0.794) 4.762) D 17 (1.260, 1.613, (2.190, 3.175, 2.000) 4.481) λ max = CI= CR= (1.651, 2.520, 3.684) (0.275, 0.405, 0.630) (1, 1, 1) (0.500, 1.000, 2.000) (0.500, 1.000, 2.000) (0.437, 0.794, 1.587) (0.347, 0.550, 1.000) (0.347, 0.585, 1.000) (0.210, 0.281, 0.397) (0.500, 1.000, 2.000) (1, 1, 1) (0.315, 0.481, 0.794) (0.464, 0.794, 1.260) (0.368, 0.630, 1.101) (0.275, 0.405, 0.630) (0.151, 0.179, 0.218) (0.500, 1.000, 2.000) (1.260, 2.080, 3.175) (1, 1, 1) (0.303, 0.464, 0.794) (0.437, 0.794, 1.587) (1.260, 1.587, 1.957) (0.210, 0.281, 0.397) (0.630, 1.260, 2.289) (0.794, 1.260, 2.154) (1.260, 2.154, 3.302) (1, 1, 1) (0.630, 1.260, 2.289) (0.500, 0.620, 0.794) (0.223, 0.315, 0.457) (1.000, 1.817, 2.621) (0.909, 1.587, 2.714) (0.630, 1.260, 2.289) (0.437, 0.794, 1.587) (1, 1, 1) Tabe 8: Pairwise comparison matrix and weights with respect to product recycing dimension D 21 D 22 D 23 Weights D 21 (1, 1, 1) (2.000, 2.924, 4.160) (0.397, 0.693, 1.260) D 22 (0.240, 0.342, 0.500) (1, 1, 1) (0.198, 0.251, 0.347) D 23 (0.794, 1.442, 2.520) (2.884, 3.979, 5.040) (1, 1, 1) λ max = CI= CR= Tabe 9: Pairwise comparison matrix and weights with respect to organization invovement dimension D 31 D 32 D 33 D 34 D 35 D 36 D 37 D 38 Weights D 31 (1, 1, 1) (0.191, 0.317, (0.138, 0.160, (0.138, 0.160, (0.347, 0.441, (0.173, 0.212, (0.693, 1.000, (0.630, 0.894, 0.347) 0.191) 0.191) 0.550) 0.275) 1.442) 1.260) (2.884, 3.979, (1.000, 1.442, (1.000, 1.442, (2.000, 2.621, (2.884, 4.160, (1.687, 2.080, (2.000, 2.621, D 32 (1, 1, 1) 5.241) 2.080) 2.080) 3.557) 5.130) 2.466) 3.557) (5.241, 6.257, (0.303, 0.481, (0.693, 1.145, (0.794, 1.442, (0.437, 0.794, (0.500, 1.000, (0.693, 1.145, D 33 (1, 1, 1) 7.268) 0.794) 2.000) 2.520) 1.587) 2.000) 2.000) (5.241, 6.257, (0.481, 0.693, (0.500, 0.874, (1.000, 1.817, (0.585, 0.909, (0.368, 0.630, (0.630, 1.000, D 34 (1, 1, 1) 7.268) 1.000) 1.442) 2.884) 1.387) 1.101) 1.587) (1.817, 2.268, (0.281, 0.382, (0.397, 0.693, (0.347, 0.550, (0.347, 0.585, (0.368, 0.630, (0.437, 0.737, D 35 (1, 1, 1) 2.884) 0.500) 1.260) 1.000) 1.000) 1.101) 1.145) (3.634, 4.718, (0.195, 0.240, (0.630, 1.260, (0.721, 1.101, (1.000, 1.710, (0.437, 0.737, (1.000, 1.817, D 36 (1, 1, 1) 5.769) 0.347) 2.289) 1.710) 2.884) 1.145) 2.884) (0.693, 1.000, (0.405, 0.481, (0.500, 1.000, (0.909, 1.587, (0.909, 1.587, (0.874, 1.357, (0.550, 0.000, D 37 (1, 1, 1) 1.442) 0.593) 2.000) 2.714) 2.714) 2.289) 0.000) (0.794, 1.119, (0.281, 0.382, (0.500, 0.874, (0.630, 1.000, (0.874, 1.357, (0.347, 0.550, (0.874, 1.357, D ) 0.500) 1.442) 1.587) 2.289) 1.000) 2.289) (1, 1, 1) λ max = CI= CR= Tabe 10: Pairwise comparison matrix and weights with respect to ife cyce management dimension Incorporating the anaysis of FAHP evidences, the oca weights for each dimension demonstrate that suppier management (0.4337) and organizationa invovement (0.3234) are the two most important dimensions for impementing GSCM practice, foowed D 41 D 42 Weights D 41 (1, 1, 1) (0.232, 0.329, 0.481) D 42 (2.080, 3.037, 4.309) (1, 1, 1) λ max = CI= CR=

11 Int. J. Environ. Sci. C. Tech., W. Hsu; 5 (2), A , H. Hu Spring 2008 by ife cyce management (0.1228). Product recycing (0.1161) reveas to be the dimension with the owest importance as shown in Tabe 5. The approaches of product testing reports (0.2026), produce disassemby manua (0.5067), top management support (0.2670) and estabish an environmenta database of products (0.7500) revea the highest importance with regard to each dimension in sequence of suppier management, product recycing, organization invovement and ife cyce management, respectivey. Considering the goba weights in Tabe 5, it is evident that the ten prioritized approaches for impementing GSCM in Taiwanese eectrica and eectronics industry foow this order: estabish an environmenta database of products (0.0925), product testing report (0.0882), top management support (0.0867), bi of materia (BOM) (0.0835), environmenta auditing for suppiers (0.0708), compiance statement (0.0630), green purchasing (0.0609), produce disassemby manua (0.0591), estabishing environmenta requirements for purchasing items (0.0513) and environmenta poicy for GSCM (0.0493). Moreover, the resuting goba weights have aso shown that a great majority of ten prioritized approaches was the dimension of suppier management. The findings affirmed the suppier management performs in the crucia roe of impementing GSCM. In addition, the priority approaches of GSCM impementation show the respondents perceptions about the importance of them and assisted organizations recognize their strengths to move towards continuous improvement. Recognizing the consistency and priority approaches for impementing GSCM is important because of the uncertainties in current environmenta reguations so that enterprises cannot ensure whether current management approaches can compy with the requirements of reguations. Athough the previous iterature has contributed to recognize various approaches in greening the suppy chain, itter is nown about the consistency and priority approaches, particuary in eectronics industry. The main strengths of this paper, hence, are two-fods: It recognizes the consistency approaches and provides a method for raning approaches. This study proposed the use of FAHP to ran different approaches of GSCM impementation in Taiwanese eectronics industry. Despite focusing on the Taiwanese eectronics industry, the resuts of this study provide an insight into recognizing and prioritizing the approaches for impementing GSCM. It aso proposed a generic hierarchy mode for assessing the reative importance of identified approaches that woud affect the GSCM impementation and the deveopment of GSCM strategy and practice. Different organizations can mae use of the mode in accordance with their specific situations and needs. In addition, the mode aso can hep managers improve their understanding of GSCM practices and enabes decision maers to assess the perception of GSCM in their organization. Athough the sampe in this study was insufficient, it is hoped that it can serve as a base for further research on exporing the impications of GSCM for different industry sectors and regions. The approaches for impementing GSCM in eectronics industry shoud be transformed to accompany changes of environmenta reguations and customers in future research. In addition, it is aso worth mentioning that suppier management pays the crucia part of impementing GSCM, the buyer-suppier reationships affect GSCM impementation to address the reated issues. Furthermore, the appication of anaytica too in determining weights for various approaches of GSCM practice is suggested to utiize anaytic networ process (ANP) in terms of feedbac systematic and interdependencies property. REFERENCES Antony, J.; Leung, K.; Knowes, G.; Gosh, S., (2002). Critica success factors of TQM impementation in Hong Kong industrias. Int. J. Qua. Reiab. Manage., 19 (5), Badri, M. A.; Davis, D., (1995). A study of measuring the critica factors of quaity management. Inter. J. Qua. Reiab. Manage., 12 (2), Bagozzi, R. P.; Yi, Y., (1998). On the evauation of structure equation modes. Acad. Mar. Sci., 16 (1), Bowen, F. E.; Cousine, P. D.; Lamming, R. C.; Faru, A. C., (2001). Expaining the gap between the theory and practice of green suppy. Greener Manage. Int., 35, Boysère, J.; Beard, A., (2006). Haogen-free aminates: wordwide trend, driving forces and current status. Circuit Word, 32 (2), Bucey, J. J., (1985). Fuzzy hierarchica anaysis. Fuzzy Set. Syst., 17, Carmines, E.; Zeer, R., (1979). Reiabiity and vaidity assessment, series: quantitative appications in socia science. Sage Pubications: Newbury Par. Chan, F. T. S.; Kumar, N., (2007). Goba suppier deveopment considering ris factors using fuzzy extended AHP-based approach. Int. J. Manage. Sci., Omega, 35 (4), Chen, S. H., (2004). Taiwanese IT firms offshore R&D in china and the connection with the goba innovation networ. Res. Poicy., 33 (2), Csutora, R.; Bucey, J. J., (2001). Fuzzy hierarchica anaysis: 215

12 Green management C. W. Hsu; in A. eectronic H. Hu industry the Lambda-Max method. Fuzzy Set. Syst., 120 (2), Evans, H.; Johnson, J., (2005). 10 Steps toward RoHS directive compiance. Cir. Assemby, 16, Hair, J.; Anderson, R.; Tatham, R.; Bac, W., (1995). Mutivariate data anaysis. 4 th. Ed. Engewood Ciffs: Prentice- Ha. Huang, J., (2005). The ministry of economic affairs estabishes a RoHS service corps to hep the eectrica and eectronic industries buid green suppy chains. Taiwan Cean. Prod. Newsetter, 1 (1), 2-3. Kahraman, C.; Cebeci, U.; Ruan, D., (2004). Muti-attribute comparison of catering service companies using fuzzy AHP: The case of Turey. Int. J. Prod. Econ., 87 (2), Kaiser, H. F., (1974). An index of factoria simpicity. Psychometria, 39, Kine, P., (1997). An easy guide to actor anaysis. Routedge: London. Kwong, C. K.; Bai, H., (2003). Determining the important weights for the customer requirement in QFD using a fuzzy AHP with an extent anaysis approach. IIE Trans., 35 (7), Lamming, R.; Hampson, J., (1996). The environment as a suppy chain management issue. Brit. J. Manage., 7 (Specia issue 1), S45-S62. Lee, H. I.; Chen, W. C.; Chang, C. J., (2008). A fuzzy AHP and BSC approach for evauating performance of IT department in the manufacturing industry in Taiwan. Expert Syst App., 34 (1), Lin, C.; Tan, B.; Chung, S., (2002). The critica factors for technoogy absorptive capacity. Ind. Manage. Data Syst., 102 (6), Lippmann, S., (1999). Suppy chain environmenta management: eements of success. Corp. Environ. Strategy, 6 (2), Meade, L.; Saris, J., (1998). Strategic anaysis of ogistics and suppy chain management systems using anaytic networ process. Trans. Res., 34 (3), MOEA, (2005). Greener Project (G-Project) officiay activated and ceremony for estabishing GP user group. Sustainabe Ind. Dev. Newsetter, 6, 7-8. Nunnay, J., (1967). Psychometric theory. McGraw-Hi: New Yor. Nunnay, J., (1978). Psychometric theory. 2 nd ed. McGraw- Hi: New Yor. Rao, P., (2002). Greening the suppy chain a new initiative in south East Asia. Int. J. Oper. Prod. Manage., 22 (6), Rao, P.; Hot, D., (2005). Do green suppy chains ead to competitiveness and economic performance?. Int. J. Oper. Prod. Manage., 25 (9), Saaty, T.L., (1980). The Anaytic Hierarchy Process. McGraw- Hi: New Yor. Sheu, J. B.; Chou, Y. H.; Hu, C. C., (2005). An integrated ogistics operationa mode for green-suppy chain management. Trans. Res., 41 (4), Tabachnic, B. G.; Fide, L. S., (1989). Using mutivariate statistic. HarperCoins: London. US-AEP, (1999). Suppy chain environmenta managementessons for eader in the eectronic industry. Cean Technoogy Environmenta Management (CTEM) Program. US-Asia Environmenta Partnership. Wee, Y. S.; Quazi, H. A., (2005). Deveopment and vaidation of critica factors of environmenta management. Ind. Manage. Data Syst., 105 (1), Wei, C. C.; Chien, C. F.; Wang, M. J., (2005). An AHP-based approach to ERP system seection. Int. J. Prod. Econ., 96 (1), Yuang, A.; Kieiewicz-Yuang, A., (2001). Sustainabe suppy networ management. Corp. Environ. Manage., 8 (3), Yusof, S. M.; Aspinwa, E. M., (2000). Critica success factors in sma and medium enterprises: survey resuts. Tota Qua. Manage., 11 (4-6), S448-S462. Zhu, O.; Saris, J.; Geng, Y., (2005). Green suppy chain management in china: pressures, practices and performance. Int. J. Oper. Prod. Manage., 25 (5), Zhu, Q.; Saris, J., (2006). An inter-sectora comparison of green suppy chain management in China: drivers and practices. J. Cean. Prod., 14 (5), AUTHOR (S) BIOSKETCHES Hsu, C. W., Ph.D., Candidate, Institute of Environmenta Engineering and Management, Nationa Taipei University of Technoogy, 1, Sec.3, Chung-Hsiao E. Rd., Taipei 10643, Taiwan. Emai: s @ntut.edu.tw Hu, A. H., Ph.D., Associate Professor at the Institute of Environmenta Engineering and Management, Nationa Taipei University of Technoogy, 1, Sec.3, Chung-Hsiao E. Rd., Taipei 10643, Taiwan. Emai: aenhu@ntut.edu.tw This artice shoud be referenced as foows: Hsu, C. W.; Hu, A. H., (2008). Green suppy chain management in the eectronic industry. Int. J. Environ. Sci. Tech., 5 (2),