Selecting management systems for sustainable ...

Available online at www.sciencedirect.com

Expert Systems with Applications Expert Systems with Applications 36 (2009) 1444–1458 www.elsevier.com/locate/eswa

Selecting management systems for sustainable development in SMEs: A novel hybrid model based on DEMATEL, ANP, and ZOGP Wen-Hsien Tsai a,*, Wen-Chin Chou a,b a

Department of Business Administration, National Central University, Jhongli, Taoyuan 320, Taiwan, ROC b Department of Applied Economics, Yu Da College of Business, Chaochiao, Miaoli 361, Taiwan, ROC

Abstract In recent years, sustainable development strategy for enterprises has become an important issue around the globe. There are four management systems (i.e. ISO 9001, ISO 14001, OHSAS 18001, and SA 8000) that can help small and medium enterprises (SMEs) to create sustainable competitive advantages. In view of the fact that the shortage of resources – time, personnel, as well as money – rules most SMEs, this paper proposes a novel hybrid model for selecting optimal management systems under resource constraints, and illustrates the practical application of such a model through an example. This model first applies the Decision Making Trial and Evaluation Laboratory (DEMATEL) approach to construct interrelations among criteria that organizations require. The second step is to obtain the criterion weights through ANP. Lastly, ANP is integrated with a zero–one goal programming (ZOGP) model to obtain optimal alternatives with desired organizational benefits by fully utilizing limited resources. The purpose of this study is to present an integrated approach that could cope with the interdependencies among various criteria and deal with the constraints on resources, and to demonstrate how to select management systems for phased implementation. Therefore, the main contribution of this paper is to enhance the capacity of SMEs to effectively address the challenge of sustainable development through a novel model of prioritizing available management systems. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: Management system; Sustainable development; Analytic network process (ANP); Zero–one goal programming (ZOGP); Decision Making Trial and Evaluation Laboratory (DEMATEL)

1. Introduction According to the Brundtland committee’s report ‘‘Our Common Future”, sustainability is defined as the ability to ‘‘meet the needs of the present without compromising the ability of future generations to meet their own needs” (WCED, 1987). At the Rio Summit in 1992, the United Nations further expanded the above definition and adopted a set of principles to guide future sustainable development. The Declaration on Environment and Development defines the rights of people toward development, and their responsibilities to safeguard the common environment (Quaddus & Siddique, 2001). From then on, environmental and sus*

Corresponding author. Tel.: +886 3 4267247; fax: +886 3 4222891. E-mail addresses: [email protected] (W.-H. Tsai), wcchou@ ydu.edu.tw (W.-C. Chou). 0957-4174/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.eswa.2007.11.058

tainable development issues have been pushed to a higher priority on social agendas. In taking a note from the ‘‘3 Ps” of Marketing, sustainable development can be said to have its own version of the ‘‘3 Ps”, i.e. People, Planet, and Profit. All three aspects have to be satisfied before an entrepreneurial activity to be labeled as sustainable (Crals & Vereeck, 2005). Therefore, firms applying the concepts are often referred to as managing to the ‘‘triple bottom line” (Elkington, 1997). This approach to business – taking environmental, social and financial results into consideration in the development and implementation of a corporate business strategy – is a movement gaining momentum around the world. Many companies are evaluating and reporting on their social and environmental performance, in response to demands from consumers, employees and communities (Mowat, 2002).

W.-H. Tsai, W.-C. Chou / Expert Systems with Applications 36 (2009) 1444–1458

To achieve the goal of ‘‘triple bottom line of sustainability”, the implementation and certification of quality (ISO 9001), environmental (ISO 14001) and occupational health and safety (OHSAS 18001) systems has become an important activity (Zeng, Shi, & Lou, 2007). ISO 9001 has contributed to better quality, higher productivity, greater customer satisfaction, and greater profit. ISO 14001 has contributed to better environmental performance, greater eco-efficiency, greener products, and more transparency for and acceptance by external environmentally concerned stakeholders. OHSAS 18001 has contributed to safer and healthier workplaces, more efficient work processes, improved employee perceptions of the working environment, and greater recruitment attractiveness. SA 8000 has contributed to achieving higher social accountability and better employees’ quality of life (Robson et al., 2007; Rohitratana, 2002; Zwetsloot, 2003). In short, implementation of management systems would generate benefits for profit (quality), planet (environment) and people (health & safety and social accountability) to become sustainable entrepreneurs. Today, these four management systems still have great potential for the companies those have not yet implemented them. While the gains of a whole range of sustainability certificates can be substantial in terms of risk control, improvement in business relationships with large companies, and good reputation, the question is being raised regarding how small and medium enterprises (SMEs) can achieve sustainable entrepreneurs. Shortage of resources – time, manpower, as well as money – is the rule for most SMEs (Crals & Vereeck, 2005). According to the resource-based view (Penrose, 1959), differences in resources should be utilized and lead to differences in sustainable competitive advantages. However, when SMEs brand themselves as sustainable entrepreneurs, they should be willing to devote time and effort to the project and select a simple, pragmatic and effective format that is tailored to their needs and compatible with their competitive strategies (Crals & Vereeck, 2005). Under the constraints of finite resources and budgets, SMEs cannot implement all the required management systems simultaneously. The decision-making involved in selecting appropriate management systems to create sustainable competitive advantages is a very important topic, which can be formulated as a multi-criteria decision-making (MCDM) problem. There is still a lack of study regarding the integration of interdependent objectives of the SMEs and the allocation of the limited resources to selecting management systems so far; the paper thus presents a novel integrated model to solve this problem. To identify the interactions among evaluation criteria of alternative systems, the Decision Making Trial and Evaluation Laboratory (DEMATEL) approach (Fontela & Gabus, 1976) is used to construct a network structure with interdependent relationships. We could extract the mutual relationships of interdependencies among various criteria and the strength of interdependence (Tamura & Akazawa, 2005) by using this method. Since

1445

these criteria are not independent, the conventional AHP, which is assumed as criteria independence, is not suitable to evaluate an MCDM problem in the real world. The ANP (Analytic network process) was proposed by Saaty (1996) to overcome the problem of dependence and feedback among criteria or alternatives (Liou, Tzeng, & Chang, 2007). Furthermore, the ANP approach is used to decide the relative weights of the criteria. It improves the visibility of decision-making processes and generates the priorities between the decision alternatives. In order to provide a systematic approach to set priorities among multi-criteria and trade-off among objectives, ANP is applied prior to goal programming formulation. The priorities obtained through ANP are then combined with a zero–one goal programming (ZOGP) model to handle the interactions between organizational objectives and the constraints on resources. The purpose of this study is to present an integrated approach that could cope with the interdependencies among various criteria and the constraints on resources, and to demonstrate how to select management systems for phased implementation. Therefore, the main contribution of this paper is to enhance the capacity of SMEs to address the challenge of sustainable development more effectively through a novel model by prioritizing available management systems. The rest of this paper is organized as follows. Sections 2 will review the literature on four management systems. Section 3 presents an integrated model for selecting management systems. An example for application has been illustrated in Section 4. Furthermore, several scenarios for different ANP priority weights and resources conditions are taken into account to verify the effectiveness of the model. Conclusions are presented in Section 5. 2. Management systems for sustainable development Organizational sustainability is viewed as performance based on the triple bottom line (TBL) of economy, environment and social responsibility (Isaksson, 2006). The general objectives and basic principles of sustainable development may be understood through theories, but a consistent methodology to achieve sustainable development or maintain sustainability is practically difficult (Brent, Rogers, Ramabitsa-Siimane, & Rohwer, 2007). Crals and Vereeck (2005) point out that ISO 14001, EMAS, SA 8000, and AA 1000 can promote sustainable entrepreneurship in the perspective of management. Isaksson (2006) notes that total quality management (TQM) can improve sustainability. On one hand, some academic studies focus on the benefits and effectiveness of standardized management systems (Boulter & Bendell, 2002; Briscoe, Fawcett, & Todd, 2005; Petroni, 2001; Poksinska, Eklund, & Dahlgaard, 2006; Robson et al., 2007; Rohitratana, 2002; Tsim, Yeung, & Leung, 2002; Zwetsloot, 2003). For example, Poksinska et al. (2006) demonstrate that the external benefit for the implementation of ISO 9001:2000 is like improved cus-

1446


tomer relations, the internal benefits most often mentioned have more structure and order in the work and standardization of organizational processes. Petroni (2001) notes that implementation of ISO 14001 and subsequent registration can facilitate progress towards increased market share, improved working climate and customer satisfaction, improved efficiency of operations and processes, and cost reduction. Robson et al. (2007) summarize evidence from thirteen studies for the effectiveness of voluntary and mandatory OHSMS interventions that indicates positive effects. They include better safety climate, more organizational action taken on OHS issues, decrease in injury rates, decrease in disability-related costs, and increase in workplace productivity. On the other hand, some studies focused on integrating two or three management systems from various viewpoints (Jørgensen, Remmen, & Mellado, 2006; Karapetrovic & Jonker, 2003; Labodova’, 2004; Low & Pong, 2003; Zeng, Tian, & Shi, 2005). The above discussion identifies a gap in the literature as to how SMEs will select ‘‘management systems” to become sustainable entrepreneurs owing to their resource limitations. It is important to bridge this gap because SMEs are concerned with this issue. A management system can be defined as the mechanism that includes organizational structure, responsibilities and procedures, and necessity to carry out certain goals. Different management systems can be put in place to achieve various goals, such as environmental care, quality assurance, and safety. The best-known standard for quality management is ISO 9001. Examples of environmental management systems are ISO 14001 and EMAS. The most widely accepted social management system is SA 8000 (Crals & Vereeck, 2005), whereas the most renowned health and safety management system is OHSAS 18001. In order to create sustainable competitive advantages, this study proposes four standardized management systems for SMEs. Four different standards for these management systems are briefly presented in the following.

2.2. Environmental management systems (ISO 14001) The ISO 14000 series are comprised of five aspects: environmental management system (EMS), environmental auditing, environmental labeling, environmental performance evaluation, and life cycle assessment. These standards are classified into two types: guidance notes and specifications. ISO 14001 is the core of the series and its adoption is voluntary (Zeng, Tam, Deng, & Tam, 2003). As a subset of ISO 14000, EMS takes a systematic approach and provides a tool to enable organizations to control the impact of their activities, products, or services on the natural environment (Orecchini, 2000). Today, ISO 14001 or ISO 9001 certified companies around the world insist that certification is a prerequisite for business relationships (Khan, 2008). 2.3. Occupational health and safety management systems (OHSAS 18001) The OHSAS 18001 aims to create and maintain a safe working environment, while protecting and maintaining good health of the workers (Zeng et al., 2007). Although it does not set out specific occupational health and safety performance criteria, nor does it give detailed specifications for the design of a management system, any organization can be OHSAS 18001 compliant by: (1) establishing an occupational health and safety management system (OHSMS) to minimize risks to its employees and other interested parties; (2) implementing, maintaining, and continually improving an OHSMS; (3) assuring itself of its conformance with its stated OHS policy; (4) demonstrating such conformance to others; (5) seeking certification/registration of its OHSMS by an external organization; (6) making a self-determination and declaration of conformance with the standard’s specifications (Pun & Hui, 2002). 2.4. Social accountability management system (SA 8000)

2.1. Quality management systems (ISO 9001) The quality management system (QMS) generally follows the Plan-Do-Check-Act (PDCA) principle and can be implemented according to ISO 9001. Business, consumer and governmental pressures led to the implementation of product quality management systems in a large number of companies. In time, such quality management systems became so important, that the lack of having implemented a certified QMS has become a trade barrier for certain countries (Labodova’, 2004). The latest ISO 9001:2000 revision is based on the following eight quality management principles: (1) customer-focused organizations; (2) leadership; (3) involvement of people; (4) process approach; (5) system approach to management; (6) continual improvement; (7) factual approach to decision-making; (8) mutually beneficial supplier relationships (ISO, 2000; Zeng et al., 2005).

The SA 8000 is a certification program that is created by the Council on Economic Priorities Accreditation Association in 1997 and revised in 2001 (CEPAA, 2001). The goal is to publish an auditable international standard for socially responsible business (Jaffe & Weiss, 2006). It specifically focuses on working conditions, workers’ rights, and child labor (Castka, Bamber, Bamber, & Sharp, 2004). Under SA 8000 guidelines, workplace conditions must conform to the following regulations (CEPAA, 2001): (1) employers must not hire children under the age of 15 years; (2) employers cannot force workers to work against their will; (3) employers must take protective measures to guarantee workers’ health and safety; (4) workers must have the freedom to bargain with employers; (5) racial and other discrimination is forbidden; (6) employers must not use or support the use of disciplinary practices; (7) employees must be paid at least the minimum wage;


(8) the working time must be limited to 48 h/week and overtime at 12 h/week; (9) the management system should be standardized. Thaler-Carter (1999) suggests that instead of viewing SA 8000 as a labor issue, it should be treated as a productivity issue. Since this standard is perceived as a component of the image and reputation of the product, importers may put pressure on their suppliers by selecting only those suppliers who can afford the standard (Rohitratana, 2002). 3. An integrated model for management systems selection Quality, environment, health & safety, and social accountability information provided by the ‘‘Management System” has become the critical foundation to enhance the capacity of enterprises’ to address the challenge of sustainable development more effectively, but SMEs are restricted by the finite resources and cannot implement all the required systems simultaneously. Therefore, we propose an integrated MCDM model that combines the methods of DEMATEL, ANP and ZOGP to solve this problem of the resource limitations. The criteria of MCDM generally include both quantitative and qualitative factors, but the quantitative criteria may be measured in incomparable units. Financial criteria frequently become the major factors in selecting management systems. However, other criteria such as customer and internal business process, which are less easy to quantify, must also be considered. Thus, this study adopts the sustainability balanced scorecard (Dias-Sardinha & Reijnders, 2005; Dias-Sardinha, Reijnders, & Antunes, 2002; Epstein & Wisner, 2001; Figge, Hahn, Schaltegger, & Wagner, 2002) to define the criteria for selecting management systems. The criteria cluster contains four evaluation factors: learning and growth perspective (LGP), internal business process perspective (IBP), customer/stakeholder perspective (CSP), and financial perspective (FP). The balanced scorecard (BSC) essentially follows a oneway linear approach to strategic performance management. Its ‘‘causal chain” contains financial outcome measures and their performance drivers, linked together in assumed cause and effect relationships (Kaplan & Norton, 1996). Very worthwhile critiques of the BSC concept indicate that there is no causal relationship between the measures from these four perspectives (Brignall, 2003; Norreklit, 2000, 2003). With regard to sustainable development strategy, there is no causal relationship between the measures from these four perspectives, either. Interdependent relations between these four perspectives for sustainable management must be constructed separately for the priority fields of action. Generally, if organizations promote the ability of employees through learning spill prevention, then they would lead to minimized severity of chemical spills, improved customer perceptions and reduced third-party liability. Similarly, in order to reduce the complaint of customers about non-green products, we will have to increase training in the life-cycle analysis and

1447

design process for environment, which would increase the market share. Likewise, poor financial performance from environmental regulatory fines would lead to train employee in emergency response, improve internal process and enhance customer satisfaction. Consequently, there is a network structure among criteria. The DEMATEL model could be one of the tools for formalizing such relations. The DEMATEL, used to research and solve complicated and intertwined problems, has been successfully applied in many situations, such as marketing strategies, R&D project, e-learning evaluation, managers’ competencies, control systems and airline safety problems (Chiu et al., 2006; Hori and Shimizu, 1999; Lin and Wu, 2008; Liou et al., 2007; Tzeng et al., 2007; Wu & Lee, 2007). Appendix A summarizes some definitions and properties of DEMATEL. The relationship of a network structure and the degree of interdependence are determined from the result of DEMATEL. Subsequently, we employ ANP to obtain the weight of each perspective (criterion). The ANP method may transform qualitative judgments into quantitative values, and is more appropriate for selecting management systems. This method so far has had only a few applications in academic literature. It has been applied to project selection (Lee & Kim, 2000; Meade & Presley, 2002), production planning (Karsak, Sozer, & Alptekin, 2002; Lin, Chiu, & Tsai, 2008; Meade & Sarkis, 1999), and strategic decision (Leung, Lam, & Cao, 2006; Ravi, Shankar, & Tiwari, 2005; Wu & Lee, 2007b). Appendix B summarizes some definitions and properties of ANP. The ANP method can identify the priorities for selecting management systems. However, selection of these systems involves not only management system priorities, but also resource limitation such as budget. Finally, we use zero– one goal programming (ZOGP) to handle the MCDM problem and to attain the objectives of an organization while considering restricted resources (Mathirajan & Ramanathan, 2007). Goal programming (GP) is first introduced by Charnes and Cooper in 1955. The model assigns optimal values to a set of variables in situations involving multiple and conflicting goals (Chen & Shyu, 2006). This model has been applied in a variety of ranked resource selection schemes (C ¸ ekyay, Gu¨mu¨ssoy, & Ertay, 2005; Chen & Shyu, 2006; Lee & Kim, 2000; Mathirajan & Ramanathan, 2007). It permits the consideration of resource limitations and other selection limitations that must be rigidly observed in the project selection problems (Lee & Kim, 2000). Appendix C summarizes some definitions and properties of ZOGP. In summary, we develop a combined DEMATEL, ANP and ZOGP methods for selecting management systems. The weights evaluated in ZOGP model consider the interdependent effect among criteria, using DEMATEL and ANP to solve this problem. Then, it selects the optimal solution that will maximize the benefits according to the priorities specified without violating any of the resource constraints. In the next section, we demonstrate how a combined model can be used in selecting management systems.

1448


Goal

Sustainable development (SD) strategy

Selecting management systems

Determine the decision criteria

Criteria

Financial Perspective Customer / Stakeholder Perspective

Learning & Growth Perspective Internal Business Process Perspective

Use DEMATEL to analyze the interdependent relationship among the criteria

Use ANP to calculate the weights of decision criteria

Determine the parameters of goal constraints

Zero-one goal programming formulation

Find an optimal solution

ISO 9001

ISO 14001

OHSAS 18001

SA 8000

Alternatives

Fig. 1. An overview of the proposed model (modified from Chen and Shyu, 2006).

An overview of the proposed model process in this paper is shown in Fig. 1. The ZOGP model results provide an optimal solution that can help managers to decide the most appropriate alternatives. 4. Illustrative example and discussion In this section, an example is presented to illustrate the application of the proposed method for evaluating and implementing a sustainable development (SD) strategy. 4.1. Problem description Assume that Company A is a Taiwanese firm with 180 employees. The company engages in the business of paper

and packaging production and design. The papermaking industry is highly polluted. In a liberalized market environment, compliance to international manufacturing standards can help exporters in improving their competitive position. This is so because standards such as the ISO 9001, ISO 14001, OHSAS 18001, and SA 8000 can certify for the company’s capabilities and quality standards as a manufacturing firm. Currently, a serious challenge for Company A arises from the fact that it has more than one important customer who asks for the company to meet the buyers’ certification standards (Briscoe et al., 2005). Hence, the company tries to maintain competitive advantages and brand itself as a sustainable entrepreneur. But, company A confronts much trouble in selecting a fitting SD strategy that involved several complex factors


systematically, such as the purposes, the condition of resources and capabilities, and even the preferences of company. For handling this MCDM problem of SD strategy selection, assume company A adopts our proposed method and set up an expert panel. The following shows how company A utilizes our proposed method to evaluate and select its management systems logically.

1449

Table 3 The total-relation matrix LGP

IBP

CSP

FP

D

D+R

DR

LGP IBP CSP FP

1.030 1.375 1.096 1.021

1.386 1.215 1.226 1.010

1.186 1.393 0.900 0.927

1.158 1.275 1.031 0.743

4.760 5.258 4.253 3.701

9.282 10.095 8.659 7.908

0.237 0.421 0.153 0.506

R

4.522

4.837

4.406

4.207

4.2. Applications of the proposed method The expert panel follows the DEMATEL method with the five-step procedure in Appendix A. First, they define the decision goals for selecting management systems. In step 2, the expert panel adopts the four perspectives (criteria) as evaluation factors. Also, they decide to use a scale (Table A1) for making assessments. In step 3, once the relationships between those factors are measured by the expert panel through the use of the scale, the data could be obtained. Then, using this method to compute these assessment data, the initial direct-relation matrix is produced. In step 4, based on the initial direct-relation matrix (Table 1), the normalized direct-relation matrix is obtained by DEMATEL formula (Table 2). Next, the total-relation matrix (Table 3) is acquired by using formula (A3). Then, using formulas (A4)–(A6), the impact-diagraph-map could be acquired by mapping a dataset of (D + R, D R). In this study, a threshold value (p) of 1.0 is decided on, in consultation with the expert panel. This number is the most appropriate value to acquire a suitable relationship from trying above and under this number. The value under 1.0 gains too many factors and complex relationships in the whole system; the relationship is not obvious above 1.0. Based on the above threshold value, the impact-diagraphmap is obtained as shown in Fig. 2, it is clear that the evaluation perspectives are visually divided into the dispatcher group, including LGP and IBP while the receiver group is composed of such factors as CSP and FP. Therefore, the impact-diagraph-map shows that both LGP and IBP will result in CSP and FP.

Table 1 The initial direct-relation matrix

LGP IBP CSP FP

LGP

IBP

CSP

FP

0.000 3.000 2.000 2.667

3.667 0.000 3.000 1.667

2.000 3.667 0.000 1.667

2.333 2.667 2.000 0.000

Table 2 The normalized direct-relation matrix

LGP IBP CSP FP

LGP

IBP

CSP

FP

0.000 0.321 0.214 0.286

0.393 0.000 0.321 0.179

0.214 0.393 0.000 0.179

0.250 0.286 0.214 0.000

After determining the relationship structure among the criteria of selecting the management system, the ANP method is applied to calculate the weight of each criterion. In the first step, the members of the expert panel respond to 46 questions through a series of pair-wise comparisons with Saaty’s one–nine scale and represented how much more important one element was over another. A score of one indicates equal importance, while nine represents the extreme importance of one element over another (Liou et al., 2007). This ANP model is solved using the SuperDecisions software. After computing the results of their assessments, the consistency ratio (CR) values are all acceptable and the eigenvectors displayed are appropriate to enter into the supermatrix. Table 4 summarizes the pair-wise comparison of the four criteria with respect to the overall goal (or the four criteria) and Table 5 summarizes the pair-wise comparison of the four alternatives with respect to the four criteria. Then, the eigenvectors are expressed in the form as the unweighted supermatrix. Table 6 presents the values associated with the unweighted supermatrix M. Since this unweighted supermatrix includes interactions between clusters, e.g. there is inner dependence among perspectives (criteria), not every one of the columns sums to one. A transformation is required for the columns to become column-stochastic and thus minimize the possibility for divergence to infinity or convergence to zero. One transformation process is to weigh the components according to their impact on the column of blocks. The row components of the non-zero column blocks within the supermatrix are compared according to that column block. In this example two adjustments are required to be completed (since the ‘‘Alternatives” column is already column-stochastic). Each block is weighted to the relative importance weight corresponding to the component in that row. The expert panel shall assume that all clusters (blocks) are of equal importance. The first adjustment influences the two column block weightings for the ‘‘Goal” column to be a 0.50 weighting for the ‘‘Goal” column block and 0.50 for the ‘‘Perspectives” column block. It is assumed, once again, that the two column block weights ‘‘Perspectives” and ‘‘Alternatives” for the ‘‘Perspectives” column are multiplied by 0.50. The component block weights are then multiplied to each of the respective column elements. Table 7 shows the weighted supermatrix MW. The weighted supermatrix is then raised to limiting powers to be ML to capture all the interactions and to obtain a steady-state outcome. The limit supermatrix is shown in Table 8, the

1450

W.-H. Tsai, W.-C. Chou / Expert Systems with Applications 36 (2009) 1444–1458 D–R Internal business process (10.095, 0.421)

0.4 Learning & Growth (9.282, 0.237) 0.2

D+R

0

8.0

9.0

10.0

Customer & Stakeholder (8.659, -0.153) -0.2

-0.4 Financial (7.908, -0.506)

Fig. 2. The impact-digraph-map of total relation – perspectives (p P 1.0).

Table 4 Comparison of the four criteria with respect to the overall goal (or the four criteria) LGP

IBP

CSP

FP

Weights

Table 5 Comparison of the four alternatives with respect to the four criteria ISO 9001

ISO 14001

OHS 18001

SA 8000

Weights

0.141 0.455 0.263 0.141

With respect to learning & growing perspective (LGP) ISO 9001 1 1/2 2 ISO 14001 2 1 3 OHS 18001 1/2 1/3 1 SA 8000 1/2 1/3 1 CR = 0.004 (desirable value to be less than 0.100)

0.162 0.181 0.065 0.592

With respect to internal business process perspective (IBP) ISO 9001 1 2 3 2 ISO 14001 1/2 1 2 1 OHS 18001 1/3 1/2 1 1/2 SA 8000 1/2 1 2 1 CR = 0.004 (desirable value to be less than 0.100)

0.423 0.227 0.123 0.227

With respect to internal business process perspective (IBP) LGP 1 1/3 1/7 3 IBP 3 1 1/3 3 CSP 7 3 1 5 FP 1/3 1/3 1/5 1 CR = 0.089 (desirable value to be less than 0.100)

0.116 0.234 0.578 0.072

With respect to customer/stakeholder perspective (CSP) ISO 9001 1 4 3 5 ISO 14001 1/4 1 1/2 2 OHS 18001 1/3 2 1 3 SA 8000 1/5 1/2 1/3 1 CR = 0.019 (desirable value to be less than 0.100)

0.546 0.138 0.232 0.084

With respect to customer/stakerholder perspective (CSP) LGP 1 1/2 1/3 IBP 2 1 1/2 FP 3 2 1 CR = 0.009 (desirable value to be less than 0.100)

0.164 0.297 0.539

With respect to financial perspective (FP) ISO 9001 1 1/2 1 ISO 14001 2 1 2 OHS 18001 1 1/2 1 SA 8000 3 2 3 CR = 0.004 (desirable value to be less than 0.100)

0.141 0.263 0.141 0.455

With respect to financial perspective (FP) LGP 1 3 IBP 1/3 1 CR = 0.000 (desirable value to be less than 0.100)

0.750 0.250

With respect to sustainable development (SD) goal LGP 1 1/3 1/2 IBP 3 1 2 CSP 2 1/2 1 FP 1 1/3 1/2 CR = 0.004 (desirable value to be less than 0.100) With respect to learning & growing perspective (LGP) LGP 1 1 3 IBP 1 1 3 CSP 1/3 1/3 1 FP 5 3 7 CR = 0.021 (desirable value to be less than 0.100)

1 3 2 1

1/5 1/3 1/7 1

overall normalized priorities for the alternatives are given by the bottom left corner, the (3,1) block, of ML. That is wANP = (ISO 9001, ISO 14001, OHSAS 18001, SA 8000) = (0.361, 0.260, 0.156, 0.223). Therefore, the optimal

2 3 1 1

1/3 1/2 1/3 1

0.263 0.455 0.141 0.141

alternative is ‘‘ISO 9001” due to the highest priority weight of 0.361 followed by ISO 14001 system. These weights are used as priorities in the goal programming formulation. The priorities obtained from the ANP are then combined with ZOGP model to handle the constraints on resources. There exist several limitations of the available resources that must be considered in the selection from the available pool of four management systems. To complete


1451

Table 6 The unweighted supermatrix, M Goal

Perspectives (Criteria)

Alternatives

SD

LGP

IBP

CSP

FP

ISO 9001

ISO 14001

OHS 18001

SA 8000

Goal

SD

1.000

0.000

0.000

0.000

0.000

0.000

0.000

0.000

0.000

Perspectives

LGP IBP CSP FP

0.141 0.455 0.263 0.141

0.162 0.181 0.065 0.592

0.116 0.234 0.578 0.072

0.164 0.297 0.000 0.539

0.750 0.250 0.000 0.000

0.000 0.000 0.000 0.000

0.000 0.000 0.000 0.000

0.000 0.000 0.000 0.000

0.000 0.000 0.000 0.000

Alternatives

ISO 9001 ISO 14001 OHS 18001 SA 8000

0.000 0.000 0.000 0.000

0.263 0.455 0.141 0.141

0.423 0.227 0.123 0.227

0.546 0.138 0.232 0.084

0.141 0.263 0.141 0.455

1.000 0.000 0.000 0.000

0.000 1.000 0.000 0.000

0.000 0.000 1.000 0.000

0.000 0.000 0.000 1.000

Goal

Perspectives (Criteria)

Table 7 The weighted supermatrix, MW Alternatives

SD

LGP

IBP

CSP

FP

ISO 9001

ISO 14001

OHS 18001

SA 8000

Goal

SD

0.500

0.000

0.000

0.000

0.000

0.000

0.000

0.000

0.000

Perspectives

LGP IBP CSP FP

0.070 0.228 0.132 0.070

0.080 0.091 0.033 0.296

0.058 0.117 0.289 0.036

0.081 0.149 0.000 0.270

0.375 0.125 0.000 0.000

0.000 0.000 0.000 0.000

0.000 0.000 0.000 0.000

0.000 0.000 0.000 0.000

0.000 0.000 0.000 0.000

Alternatives

ISO 9001 ISO 14001 OHS 18001 SA 8000

0.000 0.000 0.000 0.000

0.132 0.228 0.070 0.070

0.212 0.113 0.061 0.114

0.273 0.069 0.116 0.042

0.070 0.132 0.070 0.228

1.000 0.000 0.000 0.000

0.000 1.000 0.000 0.000

0.000 0.000 1.000 0.000

0.000 0.000 0.000 1.000

Goal

Perspectives (criteria)

Table 8 The limit supermatrix, ML Alternatives

SD

LGP

IBP

CSP

FP

ISO 9001

ISO 14001

OHS 18001

SA 8000

Goal

SD

0.000

0.000

0.000

0.000

0.000

0.000

0.000

0.000

0.000

Perspectives

LGP IBP CSP FP

0.000 0.000 0.000 0.000

0.000 0.000 0.000 0.000

0.000 0.000 0.000 0.000

0.000 0.000 0.000 0.000

0.000 0.000 0.000 0.000

0.000 0.000 0.000 0.000

0.000 0.000 0.000 0.000

0.000 0.000 0.000 0.000

0.000 0.000 0.000 0.000

Alternatives

ISO 9001 ISO 14001 OHS 18001 SA 8000

0.361 0.260 0.156 0.223

0.269 0.376 0.143 0.212

0.402 0.236 0.146 0.216

0.415 0.216 0.188 0.181

0.221 0.302 0.143 0.334

1.000 0.000 0.000 0.000

0.000 1.000 0.000 0.000

0.000 0.000 1.000 0.000

0.000 0.000 0.000 1.000

this goal, there are four limitations: (1) a total maximum consulting fee budget of NT$700,000 is available now; (2) a total maximum certification fee budget of NT$500,000 will be available six months later; (3) a total maximum of 450 working hours of management systems training time is available now; (4) a total maximum of 1800 hours of the firm’s implementation time is available now. One flexible limitation exists. An initial consulting budget allocation is set at NT$400,000, which could vary up to but not beyond the total maximum value of NT$700,000. In Table 9, the cost and organization resource usage information for each of the four systems is presented.

Based on these data and the previously computed ANP values, we can formulate the goal constraints for this empirical problem in Table 10. This ZOGP model is solved using LINDO software. The results are summarized as follows: x1 ¼ x2 ¼ 1; x3 ¼ x4 ¼ 0 d dþ d dþ d 1 ¼ 300; 1 ¼ 0; 2 ¼ 150; 2 ¼ 0; 3 ¼ 30; þ þ d 3 ¼ 0; d 4 ¼ 0; d 4 ¼ 0; d 5 ¼ 0; d 6 ¼ 0; d 7 ¼ 1; þ d ¼ 1; d ¼ 0; d ¼ 0 8 9 9 Two management systems, ISO 9001 and ISO 14001, are chosen in the first period of project implementation according to the priorities of organizational objectives.

1452


Table 9 Resources usage information on implementing of management systems Resource usage (aij) Consulting fee (000)a Certification fee (000)a Training hours (h) Labor hours (h) a

x1

x2

x3

x4

bi

$300 $200 240 1000

$200 $150 180 800

$150 $100 150 600

$350 $300 200 400

$700 $500 450 1800

Unit is NT thousand dollars.

Table 10 ZOGP model formulation ZOGP model formulation Minimize Z= þ þ þ pl1 ðd þ 1 þ d2 þ d3 þ d4 Þ pl2 ð0:361d 5 þ 0:260d 6 þ 0:156d 7 þ 0:223d 8 Þ þ pl3 ðd 9 þ d9 Þ

Subject to þ 200x1 þ 200x2 þ 150x3 þ 350x4 þ d 1 d 1 ¼ 700 þ 200x1 þ 150x2 þ 100x3 þ 300x4 þ d 2 d 2 ¼ 500 þ 240x1 þ 180x2 þ 150x3 þ 200x4 þ d 3 d 3 ¼ 450 þ 1000x1 þ 800x2 þ 600x3 þ 400x4 þ d 4 d 4 ¼ 1800

x1 þ d 5 ¼1 x2 þ d 6 ¼1 x3 þ d 7 ¼1 x4 þ d 8 ¼1 þ 200x1 þ 200x2 þ 150x3 þ 350x4 þ d 9 d 9 ¼ 400

Goals Satisfy all limitations Select highest ANP weighted projects Use NT$400,000 consulting fee for all projects selected Avoid over-utilizing max. consulting fee Avoid over-utilizing max. certification fee Avoid over-utilizing max. training hours Avoid over-utilizing max. labor hours Select ISO 9001 system Select ISO 14001 system Select OHSAS 18001 system Select SA 8000 system Avoid over- or under-utilizing expected consulting budget

Xj = 0 or 1 for j = 1, 2, 3, 4

The OHSAS 18001 and SA 8000 systems would be implemented in the following period due to the constraints on available resources. The total consulting cost during the first implementation period is NT$400,000 for implementing the two systems, which is NT$300,000 under the total limits and not over the flexible limit. The total certification fee during the first implementation period is NT$350,000 for implementing the two systems, which is NT$150,000 under the limit. The 1800 labor hours that are permitted are fully exhausted. The two systems use 30 under training hours than the available 450 h. By combining the ANP approach and ZOGP model, Company A not only adequately considers the interdependencies among criteria, but also fully utilizes its limited resources to obtain an optimal solution.

4.3. Discussion In the illustrative example, results of DEMATEL show that ‘‘Internal business process perspective (IBP)” get the highest (D R) value and is a powerful criterion, because the value of (D + R) and (D R) is very significant. ‘‘IBP” plays the major role in management system selection, and has the greatest effect on ‘‘LGP”, ‘‘CSP”, and ‘‘FP”. On the contrary, ‘‘financial perspective (FP)” is influenced by the other criteria, because the value of (D R) is negative and lowest. That is, the company trains employees in emergency response, improves internal process, enhances customer satisfaction and community perceptions, which would lead to the lower ratio of environmental regulatory fines. As seen in Table 6, ‘‘internal business process perspective (IBP)”, with a weight of 0.455, is more important than other perspectives. Other perspectives (LGP, CSP and FP), even though having some contribution weights, are not powerful enough to have a strong impact on the final ranking. Therefore we can now realize what the most important criterion is for SMEs. It shows that sustainable entrepreneurs should keep an eye on internal development. Besides, Table 7 shows that ‘‘ ISO 9001” has more weight in ‘‘IBP” and ‘‘CSP” than others. Therefore, the results of ANP indicate that ‘‘ISO 9001” has the highest priority of alternatives, which is followed by ‘‘ISO 14001”, ‘‘SA 8000”, and ‘‘OHSAS 18001”. The final results of ZOGP shows that Company A can select only two systems (ISO 9001 and ISO 14001) under the current situation due to resource constraints. However, if the company sets up an aspired goal to achieve an ideal level, it can allocate more human resources to form a project team to build management systems and allow more employees to receive training in connection with new systems, and the results of selecting management systems may be diverse in different scenarios. Consequently, it is very important to provide a scenario analysis. This study is conducted to give an insight to decision makers. If any conditions, such as resources or the preference of experts towards the criteria, change, the proposed model can assist the management of a company in selecting proper management systems. The illustration of Company A is the scenario 1. Furthermore, we propose another nine scenarios for selecting management systems to verify the optimal solutions of this integrated model. Two sets of different ANP priority weights and five sets of different upper and flexible limitations of the available resources for management systems implementation are associated with in the 10 scenarios. This integrated model still fully utilizes those limited resources to obtain an optimal solution in selecting proper management systems. Due to the change of experts’ preference, the two sets of different ANP priority weights are (0.361, 0.260, 0.156, 0.223) and (0.305, 0.189, 0.268, 0.238), respectively. Table 11 depicts the conditions including ANP priority weights of alternatives, increased budget


and increased manpower. The last row of Table 11 shows the results of optimal selection for the 10 scenarios. In scenario 1 and scenario 2, the limitation of resources is the same. However, in scenario 1, priority weights of ISO 14001 is second highest while in scenario 2, priority weights of OHSAS 18001 is second highest; as a result, the optimal solution for scenario 1 would be to select ISO 9001 and ISO 14001. Nevertheless, scenario 2 would be to select ISO 9001 and OHSAS 18001. Therefore, the results of optimal selected systems are different according to the different ANP priority weights. In scenario 3 and scenario 4, the optimal solution would be to select ISO 9001, ISO 14001 and OHSAS 18001 due to the increase in budget and manpower. While consulting fee increases by $200,000 and certification fee increases by $100,000 compared to scenario 4, the optimal selected systems would become ISO 9001, OHSAS 18001 and SA 8000 in scenario 6. In scenario 7 and scenario 8, certification fee increases by $100,000, training hours increase by 100 h, and labor hour increase by 200 h compared to scenario 5 and scenario 6, SA 8000 would be included in the optimal selected systems compared to scenario 1 and scenario 2. The budget and manpower increase in scenario 9 and scenario 10 compared to scenario 7 and scenario 8, all of these four systems would be selected in scenario 9 and scenario 10.

1453

Table 12 presents the amounts of resources surplus or short for the 10 scenarios. The permitted consulting fee is fully exhausted in scenarios 9 and scenario 10. In scenarios 6, 9 and 10, the certification fee is fully exhausted. The permitted training hours are used under the total limits in scenarios 1–10. The permitted labor hours are also fully exhausted in scenarios 1, 3, 4, 5, 9 and 10. The optimal solutions of this model in scenarios 1–8 show the surplus resources are not enough for implementing anyone unselected systems. The results support that this integrated model has fully utilized the available resources to obtain an optimal solution in selecting management systems. 5. Conclusion Sustainable management defines a form of management, which clearly states that enhancing the value of a business is not only about continuously increasing revenues and profits, but also about reconciling the economic prosperity of a business with environmental quality and social justice (cf. Elkington, 1997). Sustainable management is the combination of management theory and the concept of sustainable development (Daub & Ergenzinger, 2005); SMEs could implement ‘‘sustainable management” through four management systems (ISO 9001, ISO 14001, OHSAS

Table 11 Scenarios analysis Resources, weights and solutions

Scenario 1

2

3

4

5

6

7

8

9

10

Consulting fee (000) Certification fee (000) Training hours (hours) Labor hours (hours) Flexible Consulting fee (000)

700 500

700 500

700 500

700 500

900 600

900 600

900 700

900 700

900 750

900 750

450

450

650

650

650

650

750

750

850

850

1800 400

1800 400

2400 600

2400 600

2400 800

2400 800

2600 800

2600 800

2800 800

2800 800

ANP weights ISO9001 ISO14001 OHS18001 SA8000

0.361 0.260 0.156 0.223

0.305 0.189 0.268 0.238

0.361 0.260 0.156 0.223

0.305 0.189 0.268 0.238

0.361 0.260 0.156 0.223

0.305 0.189 0.268 0.238

0.361 0.260 0.156 0.223

0.305 0.189 0.268 0.238

0.361 0.260 0.156 0.223

0.305 0.189 0.268 0.238

Optimal solution of selecting

ISO9001 ISO14001

ISO9001 OHS18001

ISO9001 ISO14001 OHS18001

SO9001 ISO14001 OHS18001

ISO9001 ISO14001 OHS18001

ISO9001 OHS18001 SA8000

ISO9001 ISO14001 SA8000

ISO9001 OHS18001 SA8000

ISO9001 ISO14001 OHS18001 SA8000

ISO9001 ISO14001 OHS18001 SA8000

Table 12 Results of consumed resources of scenario 1–10 Surplus (short) of resources

Scenario 1

2

3

4

5

6

7

8

9

10

Consulting fee (000) Certification fee (000) Training hours (hours) Labor hours (hours) Flexible Consulting fee (000)

300 150 30 0 0

350 200 60 200 50

150 50 80 0 50

150 50 80 0 50

350 150 80 0 250

200 0 60 400 100

150 50 130 400 50

200 100 160 600 100

0 0 80 0 (100)

0 0 80 0 (100)

1454


18001, and SA 8000) to achieve their goal of sustainable development. SMEs are resource constrained and could therefore benefit from a systematic evaluation of the management systems available to them to help address sustainability challenges. Thus, this paper integrates the DEMATEL, ANP and ZOGP methods to form a new hybrid MCDM model for reaching effective problem-solving. The particular characteristic of this model is that it integrates DEMATEL and ANP to consider the interdependencies among criteria and get prioritization of projects. Just looking at priorities of the projects is not enough to select the best alternatives in a limited resource environment. For this reason, ZOGP is used to select the best alternatives within resource constraints. Depending on evaluating result, ISO 9001 and ISO 14001 systems are selected as optimal alternatives. This illustrative example demonstrates that the proposed integrated model can attain an optimal decision for selecting management systems. The process presented in this paper represents a relevant model for managing the organizational requirements and the resource constraints in order to select appropriate systems. Furthermore, a scenario analysis based on the change of experts’ preference and the increase of resources is also provided. The objective of scenario analysis is to provide an insight to decision makers when any parameters are changing. The DEMATEL method successfully computes the effects among criteria, it can effectively divide a set of complex factors into dispatcher group and receiver group, and transform into a visible structural model. The benefits of using the ANP approach for selecting the management system come from the explicit priority weights among alternative systems. Besides, the ZOGP model assists organizations with the full utilization of limited resources for optimal management systems implementation planning. Using the ZOGP model can help the organization without exceeding both their budget and the allocated time frame. The integrated model thus overcomes some of the shortcomings of AHP. While an SME can afford to invest in one ISO certification, to obtain all certificates in order to signal to the external world that the company is a sustainable entrepreneur may prove costly for an SME (Crals & Vereeck, 2005). While the advantages are many, how do SMEs address the challenge of sustainable development more effectively? They can apply a phased implementation approach under the constraints of available resources. The other important particularity of this proposed method might be translated towards the various types of decision-making. From the integral data and information collecting process, the organization could fully understand their requirements and evaluate differences between the data models. Appendix A. Brief of the DEMATEL method In a totally interdependent system, all criteria of the systems are mutually related, directly or indirectly; thus, any

Table A1 Comparison scale of the DEMATEL method Numeral

Definition

0 1 2 3 4

No influence Low influence Medium influence High influence Very high influence

interference with one of the criteria affects all the others so it is difficult to find priorities for action (Tzeng et al., 2007). The DEMATEL method, developed by the Science and Human Affairs Program of the Battelle Memorial Institute of Geneva between 1972 and 1976, can convert the relationship between the causes and effects of criteria into an intelligible structural model of the system (Hung, Chou, & Tzeng, 2006; Tzeng et al., 2007). This method is shown as follows: Step 1: Producing the direct-relation matrix Table A1 shows the pair-wise comparison scale that may be designated into four levels (Chiu et al., 2006). The result of comparison produces the direct-relation matrix. Step 2: Normalizing the direct-relation matrix On the basis of the direct-relation matrix A, the normalized direct-relation matrix M can be obtained through formulas (A1) and (A2), in which all principal diagonal elements are equal to zero (Chiu et al., 2006; Hung et al., 2006). M ¼kA 0 B k ¼ MinB @

1 max

1 n P

16i6n j¼1

; jaij j max

1 n P

16j6n i¼1

ðA1Þ

C C; A jaij j

i; j 2 f1; 2; 3; . . . ; ng

ðA2Þ

Step 3: Obtaining the total-relation matrix Once the normalized direct-relation matrix M has been obtained, the total-relation matrix S can be derived by using formula (A3), where the I is denoted as the identity matrix (Chiu et al., 2006; Hung et al., 2006) S ¼ M þ M2 þ M3 þ ¼

1 X

Mi

i¼1

¼ MðI MÞ1

ðA3Þ

Step 4: Compute dispatcher group and receiver group Using the values of D R and D + R where R is the sum of columns and also D is the sum of rows in matrix S, a level of influence to others and a level of relationship with others are defined, as shown in formulas (A4)–(A6) (Hori & Shimizu, 1999; Wu & Lee, 2007a). Some criteria having positive values of D R have higher influence


on one another and are assumed to have higher priority and are called dispatcher; others having negative values of D R receiving more influence from another are assumed to have a lower priority and are called receiver. On the other hand, the value of D + R indicates degree of relation between each criterion with others and criteria having more values of D + R have more relationship with another and those having little values of D + R have less of a relationship with others (Seyed-Hosseini, Safaei, & Asgharpour, 2005). S ¼ ½si;j nn ; n X si;j D¼

i; j 2 f1; 2; 3; . . . ; ng

ðA4Þ ðA5Þ

j¼1

R¼

n X

si;j

ðA6Þ

i¼1

computations forms a supermatrix. Finally, after the computation of the relationship of the supermatrix and the comprehensive evaluations, it is possible to derive the interdependence of each valuation criteria and options and the weighting of priorities. The higher the priority weightings, the more priority will be placed. In this manner, it is possible to select the most appropriate option. This method is shown as follows (Lin et al., 2008): Step 1: Definition of policy issues and establishment of policy-making members Step 2: Construction of network hierarchy layer structure of the problems Step 3: Questionnaire surveys and expert preference integration Step 4: Establishment of comparison matrixes Step 5: Consistency test Step 6: Computations of supermatrix Step 7: Selection of most optimal options Fig. A1 is a generalized form of a supermatrix introduced by Saaty in 1996 to deal with the interdependence characteristics among elements and components. A supermatrix is actually a partitioned matrix, where each matrix segment represents a relationship between two nodes (components or clusters) in a system (Lee & Kim, 2000; Meade & Sarkis, 1999). Fig. A2 depicts the structure and corresponding supermatrix in a network. A node represents a component (or cluster) with elements inside it; a straight line/or an arc denotes the interactions between two components; and a loop indicates the inner dependence of elements within a component (Chung, Lee, & Pearn, 2005). When the elements of a component ‘‘Goal” depend on another component ‘‘Criteria”, we represent this relation with an arrow from component ‘‘Goal” to ‘‘Criteria”. The corresponding supermatrix of the hierarchy with three levels of clusters is also shown:

C1 e11

C1

…

e11e12 ...e1m1 e21e22 ...e2 m2

Cn en1en 2 ...enmn

W11

W12

…

W1n

W21

W22

…

W2n

e1m1

Appendix B. Brief of ANP

e21

C2

e22 …

W=

…

…

…

e2 m2

…

…

en1 en 2

Cn

…

The ANP, developed by Thomas L. Saaty, provides a way to input judgments and measurements to derive ratio scale priorities for the distribution of influence among the factors and groups of factors in the decision (Saaty, 2003). It is an extension of analytic hierarchy process (AHP). In reality, the elements within the hierarchy of various rules are often interdependent. The network relationship of ANP method does not only present the relationship between rules, but also calculate the relative weightings (eigenvectors) of each rule. The result of these

e12

C2

…

Step 5: Set threshold value and obtain the impactdigraph-map To obtain an appropriate impact-digraph-map, decision-maker must set a threshold value for the influence level. Only some elements, whose influence level in matrix S are higher than the threshold value, can be chosen and converted into the impact-digraph-map. The threshold value is decided by the decision-maker or by experts through discussion (Tzeng et al., 2007). An impact-digraph-map can be acquired by mapping the dataset of (D + R, D R), where the horizontal axis D + R, and the vertical axis D R (Wu & Lee, 2007a). The purpose of the DEMATEL enquiry in this paper is the analysis components structure of each perspective and criterion for selecting SRI, the direction and intensity of direct and indirect relationships that flow between apparently welldefined components. Experts’ knowledge is checked and analyzed to contribute to a greater understanding of the component elements and the way they interrelate. The result of DEMATEL analysis can illustrate the interrelations structure of components (Tzeng et al., 2007).

1455

Wn1

Wn2

…

enmn Fig. A1. Supermatrix (Saaty, 2001, p. 87).

Wnn

1456


(a) A hierarchy

The zero–one GP model can handle the MCDM problem and attain the objectives of an organization while considering restricted resources. The model is described as follows (Lee & Kim, 2000):

(b) A network

Goal Goal

Goal Goal

ANP Minimize Z ¼ P k ðwANP dþ di Þ i ; wj j

W21 Criteria Criteria

W32

Alternatives Alternatives

0

W22

Criteria Criteria

W32

0 Wh = W21

þ Subject to : aij xj þ d i d i ¼ bi for i ¼ 1; 2; . . . ; m;

W21

Alternatives Alternatives

0 0

0 0

0 0 Wn = W21 W22

0 0

W32

I

0

I

W32

Fig. A2. (a) Linear hierarchy and (b) nonlinear network (Saaty, 2001, pp. 255–257).

where W21 is a vector that represents the impact of the ‘‘Goal” on the ‘‘Criteria” (Saaty, 2001); and W32 is a matrix that represents the impact ‘‘Criteria” of the on each element of the ‘‘Alternatives”. W22 would indicate the interdependency, and the supermatrix of the elements in a component or between two components. Since there usually is interdependence among clusters in a network, the columns of a supermatrix usually sum to more than one. The supermatrix must be transformed first to make it stochastic, that is, each column of the matrix sums to unity. A recommended approach by Saaty is to determine the relative importance of the clusters in the supermatrix with the column cluster (block) as the controlling component, the result is known as the weighted supermatrix. To achieve a convergence on the importance weights, the weighted supermatrix is raised to the power of 2k + 1; where k is an arbitrarily large number, and this new matrix is called the limit supermatrix. By normalizing each block of this supermatrix, the final priorities of all the elements in the matrix can be obtained (Chung et al., 2005). Appendix C. Brief of zero–one goal programming GP is a well-known multiple-objective programming technique (Chen & Shyu, 2006; Liu & Hsiao, 2006). Unlike linear programming, the GP model does not optimize (maximize/minimize) the objectives directly. Instead, it attempts to minimize the deviations between the desired goals and the realized results. Also, these goals must be prioritized in a hierarchy of importance. The over and under achievements of goals is measured in GP using the so called deviation variables (Mathirajan & Ramanathan, 2007).

j ¼ 1; 2; . . . ; n

xj þ d i ¼ 1 for i ¼ m þ 1; m þ 2; . . . ; m þ n; dþ i P 0; d i P 0 xj ¼ 0 or 1

j ¼ 1; 2; . . . ; n for 8i for 8j

where Z denotes the sum of the deviation from m goals considered; n is the pool of projects from which the optimal set is selected; Pk represents a preemptive priority (P1 > P2 > P3 o Pk) for goal k; d þ i and d i are the positive or negative deviation variables for the selection criterion represents the ANP mathematical weight (resource) i; wANP j on the jth project; aij is parameter j of selection resource i; bi denotes the available resource or limitation factors that must be considered in the selection decision, and xj represents the binary variable (Lee & Kim, 2000). The ZOGP model bases the selection of the projects xj on the ANP determined weights of wANP for corresponding j ANP . The larger the w , the more likely the corresponding d j i project will be selected (Lee & Kim, 2000). References Boulter, L., & Bendell, T. (2002). How can ISO 9000:2000 help companies achieve excellence? What the company think?. Measuring Business Excellence 6(2), 37–41. Brent, A. C., Rogers, D. E. C., Ramabitsa-Siimane, T. S. M., & Rohwer, M. B. (2007). Application of the analytical hierarchy process to establish health care waste management systems that minimise infection risks in developing countries. European Journal of Operational Research, 181(1), 403–424. Brignall, S. (2003). The unbalanced scorecard: A social and environmental critique. Available from: http://www.environmental-center.com/articles/article1327/article1327.htm. Briscoe, J. A., Fawcett, S. E., & Todd, R. H. (2005). The implementation and impact of ISO9000 among small manufacturing enterprises. Journal of Small Business Management, 43(3), 309–330. Castka, P., Bamber, C. J., Bamber, D. J., & Sharp, J. M. (2004). Integrating corporate social responsibility (CSR) into ISO management systems – in search of a feasible CSR management system framework. The TQM Magazine, 16(3), 216–224. C ¸ ekyay, B., Gu¨mu¨ssoy, C. A., & Ertay, T. (2005). IS project selection based on fuzzy-ANP and ZOGP. In 35th International conference on computers and industrial engineering. Istanbul, Turkey, June 19–22. CEPAA (2001). Overview of SA 8000. Available from: http://www.saintl.org/index.cfm?fuseaction=Page.viewPage&pageId=473. Chen, R. S., & Shyu, J. Z. (2006). Selecting a weapon system using zero– one goal programming and analytic network process. Journal of Information and Optimization Sciences, 27(2), 379–399. Chiu, Y. J., Chen, H. C., Tzeng, G. H., & Shyu, J. Z. (2006). Marketing strategy based on customer behaviour for the LCD-TV. International Journal and Decision Making, 7(2/3), 143–165. Chung, S. H., Lee, A. H. I., & Pearn, W. L. (2005). Analytic network process (ANP) approach for product mix planning in semiconductor fabricator. International Journal of Production Economics, 96, 15–36.

W.-H. Tsai, W.-C. Chou / Expert Systems with Applications 36 (2009) 1444–1458 Crals, E., & Vereeck, L. (2005). The affordability of sustainable entrepreneurship certification for SMEs. International Journal of Sustainable Development and World Ecology, 12(2), 173–184. Daub, C. H., & Ergenzinger, R. (2005). Enabling sustainable management through a new multi-disciplinary concept of customer satisfaction. European Journal of Marketing, 39(9/10), 998–1115. Dias-Sardinha, I., & Reijnders, L. (2005). Evaluating environmental and social performance of large Portuguese companies: A balanced scorecard approach. Business Strategy and the Environment, 14, 73–91. Dias-Sardinha, I., Reijnders, L., & Antunes, P. (2002). From environmental performance evaluation to eco-efficiency and sustainability balanced scorecards. Environmental Quality Management, 12(2), 51–64. Elkington, J. (1997). Cannibals with forks: The triple bottom line of 21st century business. Gabriola Island, Canada: New Society Publishers. Epstein, M. J., & Wisner, P. (2001). Using a balanced scorecard to implement sustainability. Environmental Quality Management, 11(2), 1–10. Figge, F., Hahn, T., Schaltegger, S., & Wagner, M. (2002). The sustainability balanced scorecard – linking sustainability management to business strategy. Business Strategy and the Environment, 11(5), 269–284. Fontela, E., & Gabus, A. (1976). The DEMATEL Observer, DEMATEL 1976 Report. Switzerland, Geneva, Battelle Geneva Research Center. Available from: http://homepage.sunrise.ch/mysunrise/agabus/ eff’endi/dematel/dematel.html. Hori, S., & Shimizu, Y. (1999). Designing methods of human interface for supervisory control systems. Control Engineering Practice, 7(11), 1413–1419. Hung, Y. H., Chou, S. C. T., & Tzeng, G. H. (2006). Using a fuzzy group decision approach-knowledge management adoption. In APRU DLI 2006 conference, University of Tokyo, Japan, 8–10 November. Available from: http://apru2006.dir.u-tokyo.ac.jp/pdf/3b-3.pdf, accessed 28 January 2007. Isaksson, R. (2006). Total quality management for sustainable development: Process based system models Business. Process Management Journal, 12(5), 632–645. ISO (2000). Quality management principles. Available from: http:// www.iso.org/iso/en/iso9000-14000/understand/qmp.html. Jaffe, N. R., & Weiss, J. D. (2006). The self-regulating corporation: How corporate codes can save our children? Fordham Journal of Corporate and Financial Law, 11(4), 893–923. Jørgensen, T. H., Remmen, A., & Mellado, M. D. (2006). Integrated management systems – three different levels of integration. Journal of Cleaner Production, 14(8), 713–722. Kaplan, R. S., & Norton, D. P. (1996). The balanced scorecard: Translating strategy into action. Boston: Harvard Business School Press. Karapetrovic, S., & Jonker, J. (2003). Integration of standardized management systems: Searching for a recipe and ingredient. Total Quality Management and Business Excellence, 14(4), 451–459. Karsak, E. E., Sozer, S., & Alptekin, S. E. (2002). Product planning in quality function deployment using a combined analytic network process and goal programming approach. Computers and Industrial Engineering, 44, 171–190. Khan, Z. (2008). Cleaner production: An economical option for ISO certification in developing countries. Journal of Cleaner Production, 16(1), 22–27. Labodova’, A. (2004). Implementing integrated management systems using a risk analysis based approach. Journal of Cleaner Production, 12(6), 571–580. Lee, J. W., & Kim, S. H. (2000). Using analytic network process and goal programming for interdependent information system project selection. Computers and Operations Research, 27, 367–382. Leung, L. C., Lam, K. C., & Cao, D. (2006). Implementing the balanced scorecard using the analytic hierarchy process & the analytic network process. Journal of the Operational Research Society, 57(6), 682–691.

1457

Lin, C. J., & Wu, W. W. (2008). A causal analytical method for group decision-making under fuzzy environment. Expert Systems with Applications, 34(1), 205–213. Lin, Y. H., Chiu, C. C., & Tsai, C. H. (2008). The study of applying ANP model to assess dispatching rules for wafer fabrication. Expert Systems with Applications, 34(3), 2148–2163. Liou, J. J. H., Tzeng, G. H., & Chang, H. C. (2007). Airline safety measurement using a novel hybrid model. Journal of Air Transport Management, 13(4), 243–249. Liu, E., & Hsiao, S. W. (2006). ANP-GP approach for product variety design. International Journal of Advanced Manufacturing Technology, 29, 216–225. Low, S. P., & Pong, C. Y. (2003). Integrating ISO 9001 and OHSAS 18001 for construction. Journal of Construction Engineering and Management, 129(3), 338–347. Mathirajan, M., & Ramanathan, R. (2007). A (0–1) goal programming model for scheduling the tour of a marketing executive. European Journal of Operational Research, 179(2), 554–566. Meade, L. M., & Presley, A. (2002). R&D project selection using the analytic network process. IEEE Transactions on Engineering Management, 49(1), 59–66. Meade, L. M., & Sarkis, J. (1999). Analyzing organizational project alternatives for agile manufacturing processes: An analytical network approach. International Journal of Production Research, 37(2), 241–261. Mowat, D. (2002). The VanCity difference – A case for the triple bottom line approach to business. Corporate Environmental Strategy, 9(1), 24–29. Norreklit, H. (2000). The balance on the balanced scorecard – A critical analysis of some of its assumptions. Management Accounting Research, 11, 65–88. Norreklit, H. (2003). The balanced scorecard: What is the score? A rhetorical analysis of the balanced scorecard. Accounting, Organizations and Society, 28, 591–619. Orecchini, F. (2000). The ISO 14001 certification of a machine-process. Journal of Cleaner Production, 8(1), 61–68. Petroni, A. (2001). Developing a methodology for analysis of benefits and shortcomings of ISO 14001 registration: Lessons from experience of a large machinery manufacturer. Journal of Cleaner Production, 9(4), 351–364. Penrose, E. T. (1959). The theory of the growth of the firm. Oxford, UK: Basil Blackwell Publishers. Poksinska, B., Eklund, J. A. E., & Dahlgaard, J. J. (2006). ISO 9001:2000 in small organizations – Lost opportunities, benefits and influencing factors. International Journal of Quality and Reliability Management, 23(5), 490–512. Pun, K. F., & Hui, I. K. (2002). Integrating the safety dimension into quality management systems: A process mode. Total Quality Management, 13(3), 373–391. Quaddus, M. A., & Siddique, M. A. B. (2001). Modelling sustainable development planning: A multicriteria decision conferencing approach. Environment International, 27(1), 89–95. Ravi, V., Shankar, R., & Tiwari, M. K. (2005). Analyzing alternatives in reverse logistics for end-of-life computers: ANP and balanced scorecard approach. Computers and Industrial Engineering, 48, 327–356. Robson, L. S., Clarke, J. A., Cullen, K., Bielecky, A., Severin, C., Bigelow, P. L., et al. (2007). The effectiveness of occupational health and safety management system interventions: A systematic review. Safety Science, 45, 329–353. Rohitratana, K. (2002). SA 8000: A tool to improve quality of life. Managerial Auditing Journal, 17(1/2), 60–64. Saaty, T. L. (1996). Decision making with dependence and feedback: The analytic network process (1st ed.). Pittsbrugh, USA: RWS Publications. Saaty, T. L. (2001). Decision making with dependence and feedback: The analytic network process (2nd ed.). Pittsburgh, USA: RWS Publications.

1458


Saaty, R. W. (2003). Decision making in complex environments. SuperDecisions. Available from: http://www.superdecisions.com/ index_tables.php3. Seyed-Hosseini, S. M., Safaei, N., & Asgharpour, M. J. (2005). Reprioritization of failures in a system failure mode and effects analysis by decision making trial and evaluation laboratory technique. Reliability Engineering and System Safety, 91(8), 872–881. Tamura, H., & Akazawa, K. (2005). Structural modeling and systems analysis of uneasy factors for realizing safe, secure and reliable society. Journal of Telecommunication and Information Technology, 3, 64–72. Thaler-Carter, R. (1999). Social accountability 8000. HR Magazine, 44(6), 106–112. Tsim, Y. C., Yeung, V. W. S., & Leung, E. T. C. (2002). An adaptation to ISO 9001:2000 for certified organizations. Managerial Auditing Journal, 17(5), 245–250. Tzeng, G. H., Chiang, C. H., & Li, C. W. (2007). Evaluating intertwined effects in e-learning programs: A novel hybrid MCDM model based on factor analysis and DEMATEL. Expert Systems with Applications, 32(4), 1028–1044.

Wu, W. W., & Lee, Y. T. (2007a). Developing global managers’ competencies using the fuzzy DEMATEL method. Expert Systems with Applications, 32(2), 499–507. Wu, W. W., & Lee, Y. T. (2007b). Selecting knowledge management strategies by using the analytic network process. Expert Systems with Applications, 32(3), 841–847. World Commission on Environment and Development (WCED) (1987). Our Common Future. Oxford, UK: Oxford University Press. Zeng, S. X., Tam, C. M., Deng, Z. M., & Tam, V. W. Y. (2003). ISO 14000 and the construction industry: Survey in China. Journal of Management in Engineering, 19(3), 107–113. Zeng, S. X., Shi, J. J., & Lou, G. X. (2007). A synergetic model for implementing an integrated management system: An empirical study in China. Journal of Cleaner Production, 15(8), 1760–1767. Zeng, S. X., Tian, P., & Shi, J. J. (2005). Implementing integration of ISO 9001 and ISO 14001 for construction. Managerial Auditing Journal, 20(4), 394–407. Zwetsloot, G. I. J. M. (2003). From management systems to corporate social responsibility. Journal of Business Ethics, 44(2/3), 201–207.