Eco-design practices towards sustainable supply chain management: interpretive structural modelling (ISM) approach

Abstract

Due to increasing emphasis on sustainable practices, many organisations have attempted to leverage their supply chain performance towards balancing triple bottom line dimensions (economic, environmental and social perspectives). This paper, therefore, determines the priorities of sustainable supply chain management focusing on eco-design. Interpretive structural modelling (ISM) and Matriced’ Impacts Croise’s Multiplication Appliquée a UN Classement (MIMAC) are used to identify the hierarchical structure of the relationships among eco-design dimensions, and to analyse characteristics power of each dimension on supporting eco-design practices. The relationships and characteristics power of each dimension are used to determine indicators that are effective in enhancement of eco-design practice, evaluated through sustainable supply chain performance. Results indicate that product deployment is an important approach for improving eco-design practice towards sustainable supply chain management. This emphasises the purpose and impact of eco-design on sequential supply chain activities at deployment phase. Further research is required to make an overall assessment of eco-design practices across range of manufacturing industries, given the current research is based on inputs from a limited number of experts of selected organisations.

Keywords:

• Sustainable supply chain
• eco-design
• dimensions
• triple bottom line
• ISM

View correction statement:

Corrigendum

Introduction

In recent times, many organisations have attempted to improve sustainable supply chain practices, focusing not only on economic objectives, but also environmental perspectives and social concerns. Among various sustainable supply chain practices, eco-design plays a vital role in fostering sustainable supply chain from upstream to down-stream activities. The pursuit towards business sustainability is increasingly required as an effective business competitive strategy (Tukker et al. 2008). According to Elkington (1998), triple bottom line notion can be used as a guideline to ensure business achieving interoperable with balancing economic objectives, social and environmental issues. Supply chain management, as main mechanism driving business competitiveness must support such transformation of business towards sustainable concerns (Jayaraman, Klassen, and Linton 2007). To achieve sustainability in supply chain management, supply chain management practice must incorporate sustainable development concept into all activities (Sarkis 2012). So far, there have been various endeavours to improve sustainable development performance using multiple tools and techniques at different stages of supply chain including eco-design, green outsourcing, material selection, purchasing, transportation, logistics, disposal, reverse logistics and green marketing (Murphy and Poist 2000; Zsidisin and Siferd 2001; Cruz and Matsypura 2009). Among these approaches with green focus, eco-design has been identified as effective principle to drive sustainable supply chain management (Wilkerson 2005; Behrisch, Ramirez, and Giurco 2011). Furthermore, eco-design can play a vital role in fostering sustainable supply chain from upstream to down-stream activities.

Diverse approaches have been employed to facilitate supply chain stakeholders in practicing sustainable supply chain activities, including a comprehensive overview/analysis of definitions (Ahi and Searcy 2013), comprehensive review of interpretive structural modelling approaches and applications in investigating various relationships (Gardas, Raut, and Narkhede 2017) and review of quantitative models and modelling approaches for sustainable SCM (Seuring 2013; Brandenburg et al. 2014). Various research activities within broader subject area of green supply chain management (GSCM) have mainly focused on green supply chain practices and their influence on various performance measures. In recent times, there is increased attention to sustainable performance through green supply chain practices. There is some research work on individual sustainable supply chain practice from measurement and improvement perspectives, including investigation into integration of eco-efficiency and eco-effectiveness as important indicators in the development of sustainable industry systems and requirements for measuring sustainable supply chain performance (Erol, Sencer, and Sari 2011), influence of eco-innovation on sustainable performance and requirements of eco-design for sustainable supply chain performance (European Commission 2009) and evaluation of environmental impact during the whole life cycle of the devices in lighting industry (Principi and Fioretti 2014). Although there is some progress with the research on eco-design from various perspectives, there is limited research on investigation into effective factors in driving eco-design practice to sustainable supply chain management. This study is to construct a model that determines influential factors in eco-design process so that relationship among eco-design dimensions can be identified for developing practical guideline of eco-design within broader sustainable supply chain practice. The remainder of the paper is structured as follows. Next, literature review is presented, followed by research methodology indicating key approaches involved and methods adopted for developing an interpretive structural model and analysis of the influence of various factors on eco-design. Results of the research are later presented, followed by discussion on key findings, practical implications and limitations. Finally, conclusion is presented with future research directions.

Literature review

Eco-design plays a significant role in broader sustainable supply chain practice and has been studied from various perspectives, with a view to understanding basic principles and impact on overall supply chain performance, in particular when considering economic, environmental and social dimensions. Among various sustainable supply chain practices identified by various researchers (Walker, Di Sisto, and McBain 2008; European Commission 2009; Green et al. 2012; Laosirihongthong, Adebanjo, and Tan 2013; Tachizawa, Gimenez, and Sierra 2015), eco-design is identified as a key element and the foundation of green and sustainable supply chain practices (Zhu, Sarkis, and Lai 2008; Laosirihongthong, Adebanjo, and Tan 2013). Since the main focus of this research is on eco-design practices and the influence on supply chain performance using interpretive structural modelling approach, from sustainable supply chain management perspective, the search of articles was carried out using a number of key words, mainly eco-design, interpretive structural modelling (ISM) approach/methodology, and sustainable/green SCM. The search of research articles was in a number of scholarly databases such as Emerald Insight, ScienceDirect, Elsevier eLibrary, Taylor and Francis Online, ABI/INFORM Collection, Scopus and Inderscience Online. Initial search was restricted to articles published from 2000 to date. In addition, some articles were selected from list of references, as they are relevant for the research topic. In order to understand current level of practice and its influence on overall sustainable supply chain performance, relationship between eco-design and sustainable supply chain management as well as its theoretical basis from institutional theory perspective are important part of discussion, as presented next.

Sustainable supply chain management and eco-design

Sustainable supply chain management has been recognised widely as a strategic importance contributing to business competitiveness. The challenge of leveraging supply chain performance towards sustainable goal is to ensure that the notion of sustainable development integrated into all supply chain activities from upstream to downstream (Seuring 2004; Sarkis 2012). In this context, investigating on sustainable and GSCM from a perspective of building theory using total interpretive structural modelling, Dubey, Gunasekaran, and Ali (2015) argues that product design as an enabler of sustainable manufacturing has an important role to play in the success of a sustainable manufacturing, while Dubey, Gunasekaran, and Singh (2015) identifies complex relationships among enablers for GSCM. There have been numerous attempts in recent times, on transforming diverse supply chain activities into sustainable goals. These activities include eco-design, green material selection, waste management process, environmental management, green procurement, green logistics, transportation, distribution, recycling, disposal, green marketing, green technology and supplier relationship management (Handfield et al. 2002; Dubey, Gunasekaran, and Ali 2015). Among these, eco-design has been recognised as an effective mechanism driving sustainable supply chain activities (Wilkerson 2005; Behrisch, Ramirez, and Giurco 2011). Since majority of environmental impact is caused by the product features, which are determined at the design phase (Buyukozkan and Cifci 2012), the design activity with sustainable development concern is crucial to an achievement of pursuing sustainable supply chain management (Wilkerson 2005; Wong 2012).

Primarily, eco-design is based on a holistic approach that takes into consideration the notion of product life cycle with respect to environment, health and safety objectives (Fiksel and Wapman 1994; Chen, Lai, and Wen 2006; Karlsson and Luttropp 2006; Buchert et al. 2014; Soh, Ong, and Nee 2014). Yung et al. (2012) state that eco-design practice covers the life cycle activities from raw material selection till end of product life management. Furthermore, Soh, Ong, and Nee (2014) emphasises disassembly embedded design for remanufacturing as a major aspect of life cycle engineering. From sustainable design perspective, Buchert et al. (2014) recommends allocation of design methods to different phases of the product creation process for the development of more sustainable products while Chen and Chen (2014) proposes an approach to help designers to design eco-products based on biomimetic concepts. With respect to the life cycle perspective, a product life is usually divided into procurement, manufacture, use and disposal. This creates visibility of capturing integrated view of sequential activities ranging from advance development, design, raw material selection, procurement, purchasing, production, marketing, distribution, usage and final disposal (Sonnemann, Castells, and Schuhmacher 2004). Consequently, the opportunity to predict the holistic effects of changes on the product towards sustainable development throughout supply chain can be expected. Sustainable development, from a life cycle perspective within broader supply chain context can be challenging, given broader range of activities and potentially large supply chains involved. One of the approaches for understanding complex phenomenon of organisational influences associated with sustainable development is to view eco-design from institutional theory perspective.

Eco-design and institutional force perspective

Institutional theory has been predominantly employed as lens to understand the way organisations influence and implement sustainable development practices (Sarmah, Acharya, and Goyal 2006; Zhu, Sarkis, and Lai 2008, 2013; Sarkis 2012). Institutional theory suggests that adoption of operational routines is an institutional process subject to the influence of three pressures of forces including normative – coercive (refers to the influence of regulatory authorities to influence conformity), mimetic (refers to the pressure to ‘mimic’ more successful competitors in the industry) and normative (refers to market forces usually typified by pressure from customers) (DiMaggio and Powell 1983; Zhu and Sarkis 2007; Kauppi 2013). Although coercive pressure is key to drive companies of sustainable development awareness through laws and regulations, companies may pursue with ‘mimic’ to catch up with their competitors on sustainable supply chain performance. Nonetheless, companies can embrace sustainable notion in responding to consumers’ attention and requirements. Behrisch, Ramirez, and Giurco (2011) points out that diffusions of eco-design practice are driven by such pressures including legislation (coercive), customer demand (normative) and internal performance motivation (mimetic). While pressures from legislations and customer demands drive the increase uptake of eco-design in developed countries, movements of eco-design practice in developing countries are less appearance (Behrisch, Ramirez, and Giurco 2011). Furthermore, for countries focusing on exporting manufactured products like China and Taiwan, the diffusion of eco-design in product development phase is recognised as key and central part of sustainable practice (Chen et al. 2005).

Various research studies on broader sustainable supply chain practice, from institutional theory perspective have investigated the effect of institutional pressures on performance, in particular exploring relationship between institutional pressures and environmental performance (Dubey, Gunasekaran, and Ali 2015), complex relationships among enablers of sustainable and GSCM (Dubey, Gunasekaran, and Ali 2015; Dubey, Gunasekaran, and Singh 2015), the role of supermarkets in the development of legitimate sustainable practices across the dairy supply chains (Glover et al. 2014), and institutional pressures motivating manufacturing organisations to adopt GSCM practices (Zhu, Sarkis, and Lai 2013). Dubey, Gunasekaran, and Ali (2015), based on an empirical investigation using moderation effect of institutional pressures claims that environmental pressure acts as a driver for successful GSCM practices and has a positive effect on environmental performance. In another study, Zhu and Sarkis (2007) claims that organisations initiate emergent GSCM practices, due to increasing ecological pressures from a variety of institutional players. Furthermore, moderation effect of institutional pressures on organisational performance has been studied in different context, including the moderation effect of institutional pressures on firm intention to implement internet-based supply chain (Liu et al. 2010), moderating effects of institutional pressures between GSCM drivers and practices (Wu, Ding, and Chen 2012), environmental management and green operations on manufacturing firm performance (Wong 2012) as well as significant influence of institutional pressures on the adoption of operational practices in organisations (Kauppi 2013).

The increasing interests in investigating a range of relationships among various factors and dimensions associated with sustainable supply chain management in recent times have resulted in applications of various quantitative models (Brandenburg et al. 2014). Based on a content analysis of 134 articles through a comprehensive analysis of various methods adopted and models categorised using a hierarchy of model types, model techniques and solution method (Sasikumar and Kannan 2008), Brandenburg et al. (2014) found that most of models presented in a limited number of journals are analytically focusing on multiple criteria decision-making models (MCDM). It is evident from categorisation of various analytical methods into various structural dimensions that a large number of solution approaches including many MCDM methods have been adopted in sustainability modelling. In a similar study, Seuring (2013) using an analysis of more than 300 articles published over 15 years on the topic of green or sustainable supply chain management, identifies three dominant modelling approaches (equilibrium models, multi-criteria decision-making and analytic hierarchy process) and concludes that there has been limited empirical research to date. This suggests that MCDM approaches are very popular among many researchers when sustainable supply chain management is investigated quantitatively.

It is evident from the above discussion that there is increasing attention on sustainable and GSCM, in particular exploring relationship between institutional pressures and GSCM practices as well as quantitative modelling of sustainable performance. However, most of these research efforts focus on the overall GSCM practices rather than the effect of individual GSCM practices such as eco-design and associated key dimensions/elements on overall GSCM performance. In addition, there has been only limited empirical research (Seuring 2013). Therefore, the need for further research on investigating the impact of eco-design on overall performance, under the influence of institutional pressures is warranted, in particular providing empirical evidence for improving sustainable supply chain performance.

Eco-design dimensions

Eco-design has been the subject of many research work in recent times, in particular emphasising the need for integrating various tools for managing environment (Wong and Cote 2011), requirements for eco-design in various industries (European Commission 2009), influence of eco-innovation for sustainable performance, emphasising the green strategy, supplier relationship and supply chain performance (Bag 2016a), and positive association of eco-design practices with environmental performance measures (Sihvonen and Partanen 2016). It is evident from these research studies that eco-design is a key element of green supply chain practice and constitutes a number of dimensions (Handfield et al. 2002; Wilkerson 2005; Chen, Lai, and Wen 2006; Chiou, Chan, and Wen 2011; Buyukozkan and Cifci 2012; Kumar, Teichman, and Timpernagel 2012; Tseng et al. 2012; Wong 2012; Yung et al. 2012) for managing and improving environmental supply chain performance. For example, eco-efficiency and eco-effectiveness have been identified as important indicators in the development of sustainable industrial systems, through a case study (Wang and Cote 2011).

Apart from indicators of eco-design outlined above, there are other eco-design dimensions, factors and aspects identified in various other research activities, including environmental assessment and strategy definition identified as activities which differentiate eco-design from traditional design (Vallet et al. 2013), eco-innovation as an effective approach for decreasing environmental impact and increasing the business value, and practitioners’ perspective of using eco-design tools and techniques (Knight and Jenkins 2009). While environmental assessment and strategy definitions are influenced by eco-designer’s expertise than support from tools (Vallet et al. 2013). Knight and Jenkins (2009) concludes that eco-design techniques require some form of process-specific customisation prior to adoption, suggesting it can act as a barrier. On the other hand, eco-design implemented through integration of eco-design with traditional product realisation processes is attributed to the success of producing sustainable products in the future (Donnelly et al. 2006). Some of the common eco-design tools include environment guidelines, checklists and lifecycle assessments (Donnelly et al. 2006; Knight and Jenkins 2009). While eco-design is considered as one of the key dimensions of GSCM, eco-design encompasses design green products for reuse, recycle and recovery of material and energy as main elements (Bag 2016b).

Although the benefits of eco-design are tremendous to driving sustainable supply chain performance, there are several challenges and issues concerning with practical implementation (Baumann, Boons, and Bragd 2002). Knight and Jenkins (2009) points out that there is a significant implementation gap between theory and practice of eco-design. Eco-design is perceived primarily as an engineering focus technique, hence it is difficult to amplify its tangible benefit to other supply chain stakeholders (Zafarmand, Sugiyama, and Watanabe 2003; Karlsson and Luttropp 2006; Tukker and Tischner 2006). According to European Commission (2009), eco-design includes different interrelated parameters embedded in different phases. Such dependence can pose complexity of eco-design’ logical flows, and can cause contradictions among supply chain stakeholders (Hubner 2012). In order to reduce the complications of eco-design practice and increase the possibility of effective deployment, practical guidelines for managers play a significant role. It is beneficial to reveal a manner cognisant of practitioners towards eco-design practice (Tsai and Chang 2012). The concept of design for sustainability, introduced by Cucuzzella (2016) highlights the difference in thinking and outcome, under varying temporal and spatial approaches for sustainability.

The literature review indicates that eco-product design practices which are oriented towards sustainability, can be separated into a number of elements (Handfield et al. 2002; Wilkerson 2005; Chen, Lai, and Wen 2006; Chiou, Chan, and Wen 2011; Buyukozkan and Cifci 2012; Kumar, Teichman, and Timpernagel 2012; Tseng et al. 2012; Wong 2012; Yung et al. 2012; Bag 2016a). These dimensions, outlined below form the basis for investigating effective factors in sustainable green supply chain design practice. Some aspects such as designing green products for reuse, recycle and recovery of material and energy (Bag 2016b) can be relevant to many eco-design dimensions, including raw material design, clean common and reverse logistics, depending on the type of industry and product involved.

Dimension 1: raw material design

This dimension refers to design of raw materials by avoiding and/or reducing the use of material hazard to environment through developing green materials, use of degradable materials, adjusting size and type of materials in order to reduce the amount or loss of materials used in production process and designing reused/recycle production. For example, GSCM, incorporating green procurement emphasises the use of raw materials containing no prohibited substances, making favourable environmental and financial performances for the respective companies (Chien and Shih 2007). One of the companies adopting these activities as part of raw material design is Toyota, where procurement of raw materials and parts manufacturing of Toyota are guided by selection of raw materials and components with low power consumption and/or not excessive emission of greenhouse gas, and selection of materials that are able to recycle (Wilkerson 2005; Chen, Lai, and Wen 2006).

Dimension 2: clean production

This dimension covers an approach to follow a set of principles that are development with balance, competitiveness and growth by developing production process by the green technology, waste reduction, being friendly to environment, no making pollution. It is directly connected to the production that can control and facilitate no pollution emission and environmental impact throughout the process. For example, some of those principles include design to reduce energy consumption in the production process, measures to eliminate waste by not producing out to destroy the environment, reduced waste in the production process, focusing on preventive approaches, such as reducing emissions of carbon dioxide. Apple Inc is an example of such company that conducts these activities where iPod product is a process that uses green technology completely. Another one with a quality leading company is Toshiba, where it has established production policy under RoHS standard to (i) reduce toxic substances such as lead, mercury, cadmium, hexavalent chromium, PBB and PBDE and (ii) control these substances so that they are not exposed to the environment (Kumar, Teichman, and Timpernagel 2012; Wong 2012).

Dimension 3: green packaging

Packaging that is reusable through reverse logistics including returns from external parties and customers, back to manufacturer for re-process and re-use is considered to be key element of green packaging. It can reduce impact on the environment. Green packaging is considered to be one of the key measurement items for GSCM practices implementation, in particular its connection through cooperation with customers (Zhu, Sarkis, and Lai 2008) and its impact on improving the environmental performance of an organisation and its supply chain (Sarkis 1999; Rao 2003; Rao and Holt 2005). One of the companies carrying out activities associated with green packaging is Coca Cola, the leader in packaging by promoting the green packaging innovation, such as reduction of plastic use and reuse of packaging (Chiou, Chan, and Wen 2011).

Dimension 4: transportation and distribution design

The transportation and distribution design is concerned with environmental management by consuming energy and applying technology in most effective way, also improving storage area and use of alternative energy equipment. Recently, Elhedhli and Merrick (2012), through a new network design incorporating cost of carbon emission in the overall cost of transportation and distribution concludes that emission costs should be considered when designing supply chains in jurisdictions with carbon costs. In this case, there is need for planning transportation system on friendly environmental base. Overall, aim of green transportation and distribution design is to have effective improvement in product moving route, reduction in distance to move the product, controlling variation on amount of transportation. One of the companies conducting these activities is Coca Cola, where the company has developed transportation system by using the most efficient hybrid truck, which is developed to reduce energy consumption for land transportation (Fiksel and Wapman 1994; Kumar, Teichman, and Timpernagel 2012). For instance, Unilever has established a project on reduction of greenhouse gas emission from transportation, focusing on using transportation vehicle at most benefit.

Dimension 5: product deployment design

It is design to extend usage life by optimising the product through changes to product attributes such as reduced size, light weight, low power consumption, increased maintainability (easy to install and maintain), reduced impact on environment (Handfield et al. 2002; Hubner 2012). For example, Samsung as one of the companies carrying out such activities is a leader in the industry recognised with green practices, focusing on product design and development. It made Memory card-type Green DDR3 RDIMM at capacity 32 GB, which can respond to environment preservation with the ability to save energy up to 83% comparing to the used module memory card DDR2.

Dimension 6: reverse logistics

Reverse logistics involve activities of receiving and managing products after life cycle completion and defective products during their life cycle. Such activities include receiving products from customer, delivering products to service centre or factories, disposal management, recycle system, reuse of product or parts of product and destroy methods (Chen, Lai, and Wen 2006; Chiou, Chan, and Wen 2011; Tseng et al. 2012). Coca Cola is one of the leading companies in managing effective recycle system. The company establishes the system in a factory to manage both recalling plastic and reuse of bottles. Another example of reverse logistics in eco-design practice is to create a product that is degradable without generating pollution to the environment.

Eco-design and environmental (sustainable) performance

The importance of environmental performance is well established within broader research context of green SCM practices, through a diverse range of research activities. In this context, ‘environmental performance is defined as the relationship between the organization and the environment’ (Dubey, Gunasekaran, and Ali 2015). Key elements, identified by Dubey, Gunasekaran, and Singh (2015) include environmental impacts of resources consumed, organisational process, products and services, recovery/processing of products. Recently, Bag (2016a), examining the effects of green strategy and supplier relationship building on supply chain performance, using total interpretive structural modelling concludes that green strategy and supplier relationship building are vital for enhancing supply chain performance. It is noted that improving sustainability is one of the key targets in these research works, in particular by providing a checklist for sustainable product development, in close collaboration with practitioners (Schöggl, Baumgartner, and Hofer 2017) and empirically assessing the impact of GSCM practices on performance, indicating that the adoption of GSCM practices by manufacturing organisations leads to improved environmental performance and economic performance (Green et al. 2012). Similarly, a few other research works considered environmental performance from a different perspectives, including an approach that integrates activity-based costing (ABC) and performance evaluation in a value-chain structure for optimal GSC supplier selection and flow allocation (Tsai and Hung 2009) and a conceptual framework that investigates the relationships between three dimensions of integrated green supply chain management and multiple dimensions of operational performance (Yu, Hills, and Welford 2008).

Essentially, eco-design practice allows practitioners to integrate the notion of sustainability into all supply chain activities from raw material selection to reverse logistics. However, utilisation of eco-design is practically taking place with respect to supply chain stakeholders’ attitude, expertise, experience and goals. Hence, it is crucial to understand the relationships among eco-design dimensions. Furthermore, influential relationships among these dimensions are critical to understanding practitioner’s perception towards eco-design implementation.

While there is a plethora of research investigating (i) the relationship between organisation and performance from supply chain management perspective, (ii) inter-relationships among enablers for sustainable supply chain management, (iii) institutional pressures to pursue GSCM practices, only a few research studies have investigated relationships among eco-design practices and sustainable supply chain performance. Furthermore, almost all these research works are focusing on environmental impact from total GSCM practice perspective, except for the work of a very few research activities. These include Schöggl, Baumgartner, and Hofer (2017) that considers early phases of product design when developing a checklist for sustainable product development and Buchert et al. (2014) that proposes a holistic approach for allocation of design methods to different phases of the product creation process. This suggests that there is still lack of substantial research on investigating the impact/influence of eco-design on environmental performance.

Furthermore, to the best of our knowledge there is no study investigating the effect of eco-design on environmental performance, from the perspective of individual eco-design dimensions. In addition, there is no comparable study, on the topic of the effect of eco-design on sustainable supply chain performance using empirical validation of relationships from an emerging economy in Asia. Thus, this research is aimed at investigating the relationship between environmental performance and organisation at the root level of GSCM practice, incorporating eco-design and its dimensions as the basis for developing guidelines for improving environmental performance under GSCM practices, validated using data from an emerging economy in Asia.

Research methodology

This study employs ISM to determine the relationships between dimensions within eco-design phases. ISM is a systematic application for identifying relationships among specific items, which define a problem or an issue (Malone 1975). It is used to analyse and build the element connection model within the complicated system (Warfied 1975, 1976, 1990) and to extend existing theories to generate sustainable manufacturing framework (Dubey, Gunasekaran, and Ali 2015). The theoretical foundation of ISM is based on discrete mathematics, graph theory, social science and collective planning. The relational sequence of each element within the complex system can be analysed by ISM, and the graph of the relational structural hierarchy with the property of hierarchy can be built using quantitative methods (Attri, Dev, and Sharma 2013). The technique is reliance on data obtained from experts to determine the relationship of the involved variables. ISM has been used widely in supply chain management context. Sanjay and Ravi (2005) used ISM to identify the relationship between the issues and obstacle factors relating to IT system in supply chain management. Rajesh (2011) employed ISM to identify the basic variables of coordination on information exchange between partners in the supply chain. Sarmah, Acharya, and Goyal (2006) also used ISM and Fuzzy MICMAC to identify and classify the criteria of sharing information between variables influencing the confidence in supply chain management. Govindan et al. (2013) used ISM for analysing the relationship of GSCM practices in the electronics industry in Brazil. Recently, Dubey, Gunasekaran and Singh (2015) proposed a sustainable manufacturing framework, using total interpretive structural modelling and suggested that human agency and institutional theories can contribute to ecological modernisation theory.

It is evident from various research studies using ISM methodology outlined above that ISM is suitable technique for analysing the influence of one variable on the other variables and identifying relationships among specific elements within a system and inter-relationships among various factors (Bag 2016b). Advantages and limitations of ISM methodology are well documented (Raut, Narkhede, and Gardas 2017). Since ISM methodology can act as a tool for imposing order and direction of relationships among the variables (Sage 1977), ISM is a suitable technique for this research which aims to analyse mutual relationships among eco-design dimensions and characteristics power of each dimension on supporting eco-design practices. The various steps involved in the ISM methodology, adopted from Attri, Dev, and Sharma (2013) are as follows:

Step 1: identify and list factors

Variables affecting the system under consideration are listed (Equation (1(1) )). Those factors can be objectives, actions and individuals.(1)

In this case, factors represent eco-design dimensions, identified through a comprehensive literature review and inputs from industry experts.

Step 2: development of Structural Self-Interaction Metrics

This step involves in-depth discussions with a selected number of experts and subsequent analysis in arriving at a contextual relationship between each pair of eco-design dimensions identified in step 1. A structural self-interaction matrix shows the direction of contextual relationships among the variables. In developing Structural Self-Interaction Metrics (SSIM), four symbols (Table 1) have been used to denote the direction of relationship between two factors If the system is a set S, consisting of n variables, then . (S_i, S_j), is the ordered pair of variables S_i and S_j. Table 1 provides a list of rules that S should follow.

Table 1. Rules for forming SSIM.

Symbol	Relationship between row () and column () elements
V	variable will lead to variable , not in reverse direction
A	variable will lead to variable , not in reverse direction
X	variable and will lead to each other, in both directions
O	variable and are unrelated

Step 3: reachability matrix

The reachability matrix is developed from the SSIM. Analyse the binary relationship among constituent variables within the system. The fundamental concepts of the process are ‘variable set’ and a ‘contextual relation’. The variable set is identified within some situation context, and the contextual relation is selected as a possible statement of relationship among the variables in a manner that is contextually significant for the purpose.

The SSIM has been converted into a binary matrix 1 or 0, applying the following rules:

•	If (i,j) value in the SSIM is V, (i,j) value in the reachability matrix will be 1 and (j,i) value will be 0
•	If (i,j) value in the SSIM is A, (i,j) value in the reachability matrix will be 0 and (j,i) value will be 1
•	If (i,j) value in the SSIM is X, (i,j) value in the reachability matrix will be 1 and (j,i) value will also be 1
•	If (i,j) value in the SSIM is 0, (i,j) value in the reachability matrix will be 0 and (j,i) value will also be 0

Step 4: level partitions

The final reachability matrix obtained in step 3 above is partitioned into different levels. The elements are then arranged graphically in level indicating directed links as per the relationship depicted in the reachability matrix. The diagraph is calculated from the reachability matrix. Each variable from the reachability matrix is found reachability set, antecedent set, and intersection set.

(1)	Reachability set consists of the element itself and the other elements which may impact, R denotes the adjacency reachability set.

(2)

(2)	Antecedent set consists of the element itself and the other elements which may impact it, A denotes the adjacency antecedent set.

(3)

(3)	Intersection of these sets is derived for all the variables the reachability set and antecedent set.

(4)

Method for level partitions is to select the variables for whom the reachability and the intersection sets are the same and occupy the top level in the ISM hierarchy. The top-level element in the hierarchy would not help achieve any other element above its own level. Once the top-level element is identified, it is separated out from the other elements. Then, the same process is repeated to find out the elements in the next level. This process is continued until the level of each element is found. These levels help in building the digraph and the final model. Steps of creating a directed graph based on the relationships of the final reachability matrix, converting the resulting graph into an ISM model and checking conceptual inconsistencies (Raut, Narkhede, and Gardas 2017) are carried out using analysis of mutual relationships among factors sung MICMAC method outlined next.

Analysis of mutual relationships among factors

MICMAC (Matriced’ Impacts Croise’s Multiplication Appliquée a UN Classement) can make an active or passive relationship analysis on each factor. This method is a structural analysis model presented by Godet (1979). MICMAC and ISM both apply the model of structural analysis matrix, analysing the mutual relationship among factors. Hence, these two approaches have quite similar evaluation/assessment processes. However, the ISM approach only addresses the direct binary relationship (0 or 1) between factors, and thus cannot analyse the conditions of delicate interactions between factors. However, a complex system comprising multiple factors has both direct and indirect relationships among factors. Mutual influence among factors may amplify the indirect relationships, thereby adjusting the overall system. The ISM approach first places factors in the complex system into different hierarchies. Then MICMAC approach is used to analyse the interaction status of each factor, and to consider the influence of each factor on the current and future systems.

MICMAC defines relevant variables within a system to verify the relationship between variables, and to discover the key variables. The dependence relationships among all variables are drawn as the Influence-dependence Chart. The MICMAC approach is outlined as follows:

Assume that are the elements in the key variable matrix T. The definition is as follows.

A. Influence (D) values of each row of the matrix are added(5)

denotes the sum of the influences of element i on other elements with element i as the cause. They include direct and indirect influences.

B. Dependence (R) values of each column of the matrix are added.(6)

is the summation of influences from other elements with element j being the result/outcome.

The factors have been classified into five categories, according to their clustered locations on the graph, as follows.

(1)	Influential variables

These variables are located in the top left of the graph, representing a high influence and low dependence. These variables play leading roles in constructing the entire system. Most variables in the system are in this group. These variables affect the development direction of the system, and therefore are adopted as the input parameters when developing systems.

(2)	Linkage variables

These variables are located in the top right of the graph, denoting representing high influence and high dependence. They not only influence other variables, but are also likely to be influenced by other variables, and thereby cause external factors to adjust the system. Because these variables are unstable, they are known as stake factors.

(3)	Dependent variables

These variables are located in the bottom right of the graph, denoting low influence and high dependence. They are highly sensitive to changes in influential and relay variables, and are the output factors of the entire system. Since these variables can reflect the effect of influential factors, they can be regarded as the indicator factors to evaluate the effectiveness of the whole system.

(4)	Independent variables

These variables are located in the bottom left of the graph, and represent low influence and low dependence. Because the values of these variables are distributed very close to the origin, they are also called ‘disconnected variables’, signifying that they have no influence on the overall dynamic changes to the system. However, if the values of variable were distributed close to the area of high influence, then emphasising this variable would raise the effectiveness of the system.

(5)	Regulating variables

These variables are located in the centre of the influence-dependence chart, and have the property of regulation.

The hierarchical structure graph (directed graph) is constructed to exhibit the vertical hierarchical relationships among each factor by analysing the binary interaction among design factors by ISM. The direct and indirect relationships among elements in the hierarchical structure are then analysed and defined using MICMAC. Combining the ISM and MICMAC approaches to construct factor’s hierarchical structure graph and driver power-dependence diagram creates a design pattern strategy that enables rapid and flexible adjustment in the future.

Data collection

This research presents guidance to adjust structure and to increase effectiveness of eco-design practice on sustainable concept basis with the ISM. Objective of this research is to set up priority of eco design towards GSCM practice. The scope of the activities is set into six dimensions that are raw material design (Eco.1), clean production (Eco.2), green packaging (Eco.3), transportation and distribution design (Eco.4), product deployment design (Eco.5) and reverse logistic (Eco.6). This study relies on inputs of the experts who are managers (Department heads) and professionals, comprising of 11 senior managers from 8 manufacturers of electronic devices, and 11 academics who have worked on the design and environment. Murry and Hammons (1995) stresses that a range of experts participated for decision is between 10 and 30 persons by method of Delphi (Chang, Wu, and Chen 2008). So, this research uses 22 experts that can be separated into 2 categories: (i) a group of experts who are from the middle management to top management in the organisation, and (ii) a group of outside experts, who are consultant, researchers and academics, experienced in designing product development process, engineering, logistics system planning and environment. Inputs from each expert are processed by ISM technique on method of calculation and analysis (Attri, Dev, and Sharma 2013). This research uses method of evaluation questionnaires by the judgement of the experts to determine the relationship. These questionnaires are assessed the value to find value of consistency index between question and measuring features (IOC) among the experts. The researchers inspect the consistency of the data prior to be analysed by creating ISM and the results of calculations for the IOC values are in acceptable threshold. So the researchers use criteria as specified in six dimensions above to assess the relationships specified in the questionnaire, based on the inputs from experts.

Using in-depth discussions with 22 experts and subsequent analysis, identification of contextual relationships between variable is developed and presented in Table 2. Then initial reachability matrix is generated by expressing the information in each cell entry into 1s and 0s (Table 3). After three iterations, the final reachability matrix is then developed by incorporating the transitivity concept which is presented in Table 4. From the final reachability matrix (Table 4), antecedent set, reachability set and intersection set are determined. Table 5 presents the common enablers identified in three levels.

Table 2. Structural Self-Interaction Metrics (SSIM).

Variable	Eco.6	Eco.5	Eco.4	Eco.3	Eco.2	Eco.1
Eco.1	V	X	O	V	X	X
Eco.2	X	A	O	V	X
Eco.3	X	X	O	X
Eco.4	O	X	X
Eco.5	X	X
Eco.6	X

Table 3. Initial Reachability matrix.

Variable	Eco.6	Eco.5	Eco.4	Eco.3	Eco.2	Eco.1
Eco.1	1	1	0	1	1	1
Eco.2	1	0	0	1	1	1
Eco.3	1	1	0	1	0	0
Eco.4	0	1	1	0	0	0
Eco.5	1	1	1	1	1	1
Eco.6	1	1	0	1	1	0

Table 4. Level partitions.

Variable	Reachability set	Antecede set	Intersection set	Level
Eco.1	1,2,5	1,2,5	1,2,5	2
Eco.2	1,2	1,2,5,6	1,2	1
Eco.3	3,5,6	1,2,3,5,6	3,5,6	1
Eco.4	4,5	4,5	4,5	2
Eco.5	1,5	1,3,4,5,6	1,5	3
Eco.6	2,3,5,6	1,2,3,5,6	2,3,5,6	1

Table 5. MICMAC analysis of the influence – dependence.

Variable	Eco.6	Eco.5	Eco.4	Eco.3	Eco.2	Eco.1	Driving power
Eco.1	1	1	0	1	1	1	5
Eco.2	1	0	0	1	1	1	4
Eco.3	1	1	0	1	0	0	3
Eco.4	0	1	1	0	0	0	2
Eco.5	1	1	1	1	1	1	6
Eco.6	1	1	0	1	1	0	4
Dependence power	5	5	2	5	4	3	24

Results and analysis

The results of structural relationship of eco-design practices (Figure 2) and influences – dependence chart (Figure 3) show that the most important variables that influence in driving the sustainable supply chain management is product deployment design (Eco.5). The second is raw material design (Eco.1) and third covers clean production (Eco.2) and reverse logistics (Eco.6). Although these two practices are influential to effectiveness of the eco-design practices at the same driving power level, but the dependence power is different. The fourth is green packaging (Eco.3). These 5 variables have linked variable character. The less influential variables to drive eco-design activity is transportation and distribution design (Eco.4). It has character to be an independent variable. This variable has no effect on change of overall system. If the variable is scattered close to the area of influence, then focus on this variable means it will increase effectiveness of the system. The results of the character on structural relationship and feature of power on driving of each variable have details as outlined below.

Figure 1. MICMAC analysis of the influence – dependence chart.

Figure 2. The interpretive structural model (ISM).

Figure 3. Graph MICMAC analysis of the influence – dependence.

The findings indicate that the efficacy of eco-design practice must begin with product deployment (Eco.5) as key activity as it can influence other eco-design activities including raw material design (Eco.1), clean production (Eco 2.), green packaging (Eco.3) and reverse logistics (Eco.6). The relationship between these activities covers the primary concept of product life cycle which begins from new product development to disposal phase. Furthermore, to specifically promote raw material design (Eco.1) it is essential to drive such activity coupling with both production deployment (Eco.5) and clean production (Eco.2) where their outputs affect each other. Similarly, to encourage more practice of clean production (Eco.2), activities including raw material (Eco.1) and reverse logistics (Eco.6) teams should work together as their performance are related. Although the primary focus of driving eco-design practice is based on the foundation relationship among raw material design (Eco.1), clean production (Eco.2) and production deployment (Eco.5). With product deployment (Eco.5) as a main influence, an encouragement of transportation and distribution (Eco. 4) strategic development that align with eco-product quality can be arranged.

Discussion and conclusion

Although various academics and practitioners agree on the benefits of eco-design in leveraging sustainable supply chain performance (Wilkerson 2005; Buyukozkan and Cifci 2012), practical use of such principle is still scarce. In order to better promote eco-design practice, it is necessary to understand the perception of practitioners towards eco-design usage. Five dimensions have been identified, using a comprehensive literature review and include raw material design, clean production, green packaging, transportation and distribution design, product deployment design and reverse logistics. ISM methodology has been used in finding contextual relationships among parameters and MICMAC analysis has also been carried out. This study supplements previous studies with visual influential relationships between eco-design practices. Hence, practitioners can determine strategic implementation of eco-design activities related to their performance effect. Product deployment design has been identified as an important approach for improving eco-design practice. Hence, it is useful to stress the purpose and impacts of eco-design on sequential supply chain activities at product development phase. The clarification of eco-design goal must embrace supply chain stakeholders who focus on sourcing, market destination, production capabilities, and product characteristics (Pujari 2006; Kurk and Eagan 2008). At a broader level, product deployment design practice must underline a design principle, which considers all attributes of components of a product’s life cycle (Sarkis 2012) with ways of translation concurrently into a description of product performance to support the entire supply chain management process (Soylu and Dumville 2011). The organisation can take further step in establishing a formal functional team to be responsible for structured design (Deutz, McGuire, and Neighbour 2013) and communicate to other units in supply chain.

Research limitations, implications and future directions

Main limitations of this research study include (i) limited scope of eco-design dimensions selected, based on literature review and subjective judgement of experts participated in the research, (ii) limitations of ISM methodology, impacting on the quality of interpretation of data and resulting relationships, and (iii) number of experts selected only from limited range of manufacturing organisations, not representing the entire manufacturing industries in Bangkok, Thailand.

This study has important academic and practitioner implications. At macro level, in country like Thailand there have been difficulties in interpreting policy implications. Although external pressures including legislations and regulations are government fundamental instruments driving sustainable movements, the success rate of practical implementation is slow due to lack of clear guidelines and misunderstanding. With the perception gained from Thai practitioners on eco-design practices prioritisation, suitable guideline can be created to allow Thai manufacturers to achieve proactive sustainable supply management strategy (Laosirihongthong, Adebanjo, and Tan 2013) and increase their sustainable performance. For academics, alternative approach of building theory of sustainable/GSCM using interpretative structural modelling can be effective way of testing and extending existing theories, in particular complex relationships exists in different situations. However, this requires overcoming of some of the limitations of ISM methodology and/or using better methods for investigating relationships.

On managerial implications, ISM methodology should be used with careful consideration of the situation, in particular when implementing a new product development with sustainability focus on the product life cycle and entire supply chain.

It is expected that future research incorporates broader spectrum of eco-design practices/dimensions, involve manufacturing organisations across most industries and very importantly have inputs from experts who possess required knowledge about ISM methodology.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This work was supported by the Higher Education Research Promotion and National Research University Project of Thailand, Office of Higher Education Commission, Thailand Research Fund (TRF), Thammasat Business School, and Thammasat University.

Notes on contributors

P. Thamsatitdej, DBA, was a Thai Royal Government Sponsorship student, graduated with Doctor of Business Administration degree in Management Science from The University of Strathclyde, the United Kingdom. After his graduation in 2006, he started working as a Public Development Officer with the Office of Public Development Commission (OPDC). In 2009, he started his academic career with College of Innovation, Thammasat University, where he provided lectures on risk, system thinking, new product development and innovation management. He was also held several management positions including Associate Dean, Director of Innovative Consulting Center and Director of Technology Management Program. He has been an organizing committee for a number of leading international academic conferences including IEEE ICMIT, IEEE ICQR, and PICMET. He is currently working at College of Management, Mahidol University, Bangkok, THAILAND.

S. Boon-itt, PhD, is an associate professor of Operations Management at Thammasat Business School, Thammasat University, Bangkok, Thailand. His research interests are supply chain management including supply chain integration, supply chain uncertainty, and supply chain risks. He has published his research in leading journals including Journal of Operations Management, International Journal of Production Economics, International Journal of Physical Distributions and Logistics Management, and Production Planning and Control.

P. Samaranayake, PhD, is a senior lecturer at the School of Business, Western Sydney University, Australia. He has around 25 years of teaching and research experience and has published a number of papers in top-ranked international journals such as International Journal of Production Research, Supply Chain Management: An International Journal, International Journal of Operations and Production Management and European Journal of Operational Research. His areas of expertise include supply chain management, production planning, business process management, and enterprise resource planning. His educational qualifications include PhD from the University of Melbourne, ME, BSc (First Class Honours) and PG Diploma (Computer Science). Recently, he completed SAP Certification on Best Practices in ERP (BPERP) and is a certified Solution Consultant of SAP ERP for small and midsize enterprises.

M. Wannakarn, was a post-graduate researcher in the Department of Industrial Engineering, Faculty of Engineering, Thammasat University, Thailand. She is currently working at Minebea-Mitsumi Company Limited as the Head of Cost Analysis Department.

T. Laosirihongthong, PhD, is a professor of Supply Chain and Operations Management in the Department of Industrial Engineering, Faculty of Engineering, Thammasat University, Thailand. His research interests are in supply chain management and manufacturing operations management. He works actively with a range of organizations in Thailand, Vietnam, and Australia. He has published his research in leading journals including Industrial Management and Data systems, Production Planning and Control, International Journal of Production Research, and Technovation. He is the corresponding author.