Sustainable supply chain management: An integrated model for optimising supply chain network design
By
Mohsen Varsei B.Sc.(double), M.Sc.
This thesis is submitted in fulfilment of the requirements for the degree of
Doctor of Philosophy School of Management University of South Australia Business School
May 2016
Dedication to Mehrnoush for 4 years patience, support and unconditional love throughout the period of exploration, focus, and often stress
Table of Contents Table of Contents ............................................................................................................................... iii List of Figures .................................................................................................................................... vi List of Tables .................................................................................................................................... vii Abstract ............................................................................................................................................ viii Declaration ......................................................................................................................................... ix Acknowledgment ................................................................................................................................ x Selected Awards, Refereed Publications and Presentations.............................................................. xii 1
2
Introduction ............................................................................................................................... 1 1.1
Outline................................................................................................................................. 1
1.2
Some definitions ................................................................................................................. 2
1.3
Research background .......................................................................................................... 3
1.4
Research questions .............................................................................................................. 6
1.5
Research significance and aims .......................................................................................... 7
1.6
An overview of the research methodology ......................................................................... 9
1.7
Ethics considerations......................................................................................................... 10
1.8
Thesis structure ................................................................................................................. 11
From supply chain to sustainable supply chain management: A literature review .......... 13 2.1
Introduction ....................................................................................................................... 13
2.2
Supply chain management ................................................................................................ 13
2.2.1 Supply chain .................................................................................................................. 14 2.2.2 Managing supply chains ................................................................................................ 16 2.3
Sustainability ..................................................................................................................... 23
2.3.1 Definition of sustainability............................................................................................... 24
iii
2.3.2 The historical background of the concept of sustainability.............................................. 25 2.4
Sustainable supply chain management ............................................................................. 32
2.4.1 The economic aspect ........................................................................................................ 35 2.4.2 The environmental aspect ................................................................................................ 37 2.4.3 The social aspect .............................................................................................................. 44 2.4.4 Analysis of the literature: Some research gaps ................................................................ 48 2.5 3
Summary and conclusion .................................................................................................. 52
Addressing the gap: A framework for sustainable supply chain design ........................... 54 3.1
Introduction ....................................................................................................................... 54
3.2
Supply chain design .......................................................................................................... 55
3.2.1 The need for the detailed analysis of supply chain design ............................................ 56 3.2.2 Facility location and supply chain design ..................................................................... 57 3.3
Analytical modelling for supply chain design: Research methodology ............................ 59
3.3.2 Trade-offs and analytical approaches............................................................................ 61 3.3.3 Optimisation and model ................................................................................................ 62 3.4
Sustainable supply chain design ....................................................................................... 65
3.4.1 Economic performance measures ................................................................................. 65 3.4.2 Environmental performance measures .......................................................................... 70 3.4.3 Social performance measures........................................................................................ 73 3.4.4 The proposed framework .............................................................................................. 77 3.5 4
Summary and conclusion .................................................................................................. 79
A model for sustainable supply chain design........................................................................ 81 4.1
Introduction ....................................................................................................................... 81
4.2
Why wine? ........................................................................................................................ 82
4.3
The wine supply chain: An overview ............................................................................... 86
4.3.1 From wine to the wine industry ...................................................................................... 86 4.3.2 A typical wine supply chain ............................................................................................ 88 4.4
Model development .......................................................................................................... 91
4.4.1 Problem statement ............................................................................................................ 92
iv
4.4.2 Assumptions, indices, parameters and decision variables................................................ 95 4.4.3 Formulation of objective functions ................................................................................ 100 4.4.4 Formulation of constraints ............................................................................................. 104 4.5 5
Summary and conclusion ................................................................................................ 109
Case study for model application: Description, results and discussion............................ 111 5.1
Introduction ..................................................................................................................... 111
5.2
Case study research for analytical modelling studies...................................................... 112
5.3
Case setting ..................................................................................................................... 115
5.3.1 Sampling, problem statement and data collection .......................................................... 115 5.3.2 Customised sustainable supply chain design model for the problem ............................. 126 5.4
Solution method, results and discussion ......................................................................... 129
5.4.1 Solution method ............................................................................................................. 129 5.4.2 Results and discussion.................................................................................................... 132 5.5 6
Summary and conclusion ................................................................................................ 140
Conclusion.............................................................................................................................. 141 6.1
Introduction ..................................................................................................................... 141
6.2
A summary and discussion.............................................................................................. 142
6.3
Contributions and managerial implications..................................................................... 148
6.3.1 A novel framework ..................................................................................................... 148 6.3.2 A novel model ............................................................................................................. 149 6.3.3 Optimising multidimensional sustainability objectives .............................................. 150 6.4
Limitations and directions for future research ................................................................ 151
6.5
Concluding remarks ........................................................................................................ 156
References ...................................................................................................................................... 158 Appendix1: The ethics approval letter ............................................................................................ 193 Appendix2: The interview protocol ................................................................................................ 193
v
List of Figures Figure 2.1:The supply chain network, adapted from Christopher (2005, p.5)………………………..15 Figure 2.2: Scope of supply chain management, adapted from Mentzer, Stank and Esper (2008, p.38)………………………………………………………………………………...………...17 Figure 2.3: Supply chain management: key elements, adapted from Lambert (2008)……..……....21 Figure 2.4: The classification of the green supply chain management literature, adapted from Srivastava (2007).………………………………………………………………………......40 Figure 2.5: The literature review method, adapted from Soni and Kodali (2011) and Winter and Knemeyer (2013).………………………………………………………………………...49 Figure 2.6: Sustainability dimensions addressed in the reviewed papers (out of 195 papers).……51 Figure 2.7: Distribution of reviewed papers based on their methods as well as social and/or environmental dimensions………………………………………………………………………………….52 Figure 3.1: The proposed framework for designing sustainable supply chains…………………….. 79 Figure 4.1: A typical wine supply chain (photos: courtesy of Google.com)…………………………..89 Figure 4.2: Decision variables in a typical wine supply chain (X: binary variables, Y: flow variables) …………………………………………………………………...94 Figure 5.1: ABC’s supply chain from winery to demand points………………………………………119 Figure 5.2: An illustration of ABC’s current supply chain network (solid lines) and the alternative scenario S5 (dashed lines)…………………………………………………………………..134
vi
List of Tables Table 2.1: Some major events which raise awareness about sustainability………………………….29 Table 2.2: Categories and aspects of the Global Reporting Initiative G4 sustainability framework (Global Reporting Initiative 2013)…..………………………………………………………...31 Table 2.3: Definitions of sustainable supply chain management……………………………………...32 Table 3.1: Summary of the relevant sustainable supply chain design studies in the literature…….68 Table 3.2: The proposed pairwise comparison scale…………………………………………………...76 Table 4.1: Top countries ranked by total grape production in 2009 (Winetitles 2013)……………...86 Table 4.2: Top countries ranked by total wine production in 2009 (Winetitles 2013)………………..87 Table 4.3: Top countries ranked by total wine exports in 2009 (Winetitles 2013)……………………87 Table 5.1: Emission factor in gram CO2-e per tonne kilometre, adapted from the Wine Carbon Calculator Protocol Version 1.2 (FIVS 2008)……………………………………....124 Table 5.2: Scores for the selected social categories…………………………………………………..125 Table 5.3: Normalised social coefficients…………………………………………………………….....126 Table 5.4: The pay-off table generated by the augmented ε-constraint method and the ABC existing scenario…………………………………………………………………………...133 Table 5.5: The Pareto optimal solutions (cost in million AUD)……………………………………….136 Table 5.6: The pay-off table based on the prevalent freight transport practice in Australia……....138
vii
Abstract This research comes at a time when many businesses and people are increasingly facing various global sustainability challenges, and when the literature is still scarce on research simultaneously incorporating three dimensions of sustainability—economic, environmental and social—in supply chain design and management. Much of the existing literature on sustainable supply chain management addresses one or two dimensions in isolation. The main aim of the research is to explore how some of the key aspects of all three dimensions of sustainability can be incorporated into a single overarching framework and analytical model for supply chain network design, which typically involves companies’ key strategic location and transportation decisions with profound implications for sustainability. These decisions often necessitate careful examination using the modelling and optimisation approaches, hence the study employs analytical mathematical research methodology. The research proposes a novel multidimensional framework for sustainable supply chain design. Using the framework, the study investigates the wine supply chain and proposes a novel multi-objective model for sustainable wine supply chain design. Supply chain design decisions in the wine industry influence company profitability and have critical environmental and social implications. To demonstrate the application of the model, the research develops a case study of a large-sized wine company located in Australia. The model is solved by a multi-objective optimisation method resulting in a set of Pareto optimal (i.e. trade-off) solutions and the associated supply chain design scenarios. The case study highlights trade-offs between the objectives which should be considered when optimising supply chain design. This study offers significant contributions and managerial implications. It could help researchers explicitly examine existing supply chains in various industries in terms of multiple sustainability indicators. It could also assist companies with designing and developing sustainable supply chains, creating the business cases for sustainability, and ultimately reshaping value chains towards a more sustainable future.
viii
Declaration I declare that: •
this thesis presents work carried out by myself and does not incorporate without acknowledgment any material previously submitted for a degree or diploma in any university;
•
to the best of my knowledge it does not contain any materials previously published or written by another person except where due reference is made in the text; and all substantive contributions by others to the work presented, including jointly authored publications, is clearly acknowledged.
Mohsen Varsei May 2016
ix
Acknowledgment It is really difficult to express my sincere gratitude to all those who have helped in the completion of this research in just a few paragraphs. First, I cordially thank the University of South Australia and the School of Management who provided me with an international scholarship for three years (University President Scholarship from 2012 to 2014), encouraging an industrial engineer/manager, who then worked in industry in a senior position and observed many sustainability-related challenges, to take a long trip, move to Adelaide with his wife, pursue his passion, and explore the interface between supply chain management and sustainability. Second, I owe my deepest gratitude to my supervisor, Associate Professor Claudine Soosay, for her invaluable guidance and assistance throughout the study. Without her encouragement, insightful thoughts and constructive comments, this dissertation would not have been finalised. And, throughout close interaction with her, I learned a lot about how to conduct a good research project. Third, this study is to some extent multidisciplinary and I have benefited from advice given by several scholars from multiple academic communities. In particular, I thank Dr. Sergey Polyakovskiy (for his help in the multi-objective optimisation and CPLEX coding), Associate Professor Behnam Fahimnia, Professor Roger Burritt, Associate Professor Frank Neumann and Professor Joseph Sarkis. Fourth, I thank executives at a number of Australian wine, bottling and logistics businesses. They have put their precious time and provided me with data, information and valuable practical insights assisting in the analytical model development and deployment. I also express my gratitude to three executives at the South Australian Wine Industry Association, the Winemakers’ Federation of Australia and the Australian Wine Research Institute for sharing their practical insights into the wine industry, supply chain and associated sustainability concerns. Last, but not least, I would like to sincerely acknowledge the support of the University of South Australia Business School and the School of Management through organising x
several invaluable events, workshops and research seminars. These include research design and writing workshops (special thanks to Dr. Wendy Bastalich and Dr. Monica Behrend), in addition to the workshops on various research methods and management/business topics. I have learned a lot from more than many—those who I have mentioned above and many others who have genuinely shared their knowledge with me…to all: Thank You!
xi
Selected Awards, Refereed Publications and Presentations • 2015, the Winner of 2015 Outstanding Paper Award - Supply Chain Management: An International Journal (Emerald Literati Network Awards for Excellence) www.emeraldgrouppublishing.com/authors/literati/awards.htm?year=2015&journal=scm •
2015, Best Paper recognition and a Finalist for the All-Academy Carolyn Dexter Award (top 6) for the best international paper of the 2015 Academy of Management , www.aom.org/Meetings/awards/Annual-Meeting-Program-Awards.aspx
• 2015, Honourable Mention in the 2015 NBS-Impact on Practice Award: jointly by the Network for Business Sustainability (NBS) and the 2015 Academy of Management (ONE Division) ‘for the paper chosen to be both academically rigorous and also to have strong practitioner implications’. Varsei, M., Polyakovskiy, S. ‘Sustainable supply chain network design: A case of the wine industry in Australia’, Omega, The International Journal of Management Science (Impact Factor: 4.376, ABDC rank A*), available on-line, in press. Varsei, M., Christ, K. & Burritt, R. (2015), ‘Bottling Location and the Global Wine Supply Chain: Dollar, Water and Carbon Trade-offs’, Best Paper Proceedings of the 75th Annual Meeting of the Academy of Management, DOI: 10.5465/AMBPP.2015.45 [Carolyn Dexter Award Finalist (top 6), and, Honourable Mention in the NBS-Impact on Practice Award] Varsei, M. (2014), ‘Supply chain design in the wine industry’, presented at the Optimisation and Logistics Group at the University of Adelaide, Australia, 27 August. URL: http://cs.adelaide.edu.au/~optlog/seminars.php Soosay, C., Fearne, A. and Varsei, M. (2014), ‘Extending sustainable practices beyond organisations to supply chains’, pp.71-90, in Harris, H., Sandhu, S. and McKenzie, S. (eds), Linking Global and Individual Sustainability, Springer, Netherlands. DOI: 10.1007/978-94-017-9008-6_6 Varsei, M., Soosay, C., Fahimnia, B. and Sarkis, J. (2014), ‘Framing sustainability performance of supply chains with multidimensional indicators’, Supply Chain Management: An International Journal, Vol. 19 No. 3, pp. 242-257. DOI: http://dx.doi.org/10.1108/SCM-12-2013-0436 . [the Winner of journal’s 2015 Outstanding Paper Award, featured as the journal’s sample article] Varsei, M., Soosay, C. (2013), ‘A conceptual framework for the design and management of sustainable supply chains’, Proceedings of the 27th Australian and New Zealand Academy of Management , 4-6 December, Hobart, Australia. Varsei, M., Soosay, C., Fahimnia, B. (2013), ‘A multi-dimensional assessment framework for sustainable supply chains’, presented at the 73rd Annual Meeting of the Academy of Management (OM Division), 9-13 August, Orlando, USA, DOI: 10.5465/AMBPP.2013.11110.
xii
1 1.1
Introduction
Outline
Over the past few decades, many companies in various industries have designed and expanded their supply chains into multiple locations and markets in order to gain competitive advantage (Chopra & Meindl 2013; Christopher 2005; Meixell & Gargeya 2005; Simchi-Levi, Kaminsky & Simchi-Levi 2008). This expansion is evident with the increase in ‘world merchandise exports value’ from USD 579 billion in 1973, to USD 3,676 billion in 1993, to USD 14,851 billion in 2010 (World Trade Organization 2012). Although the sole focus on economic performance has been a frequent source of interest, recent years have seen critical issues pertaining to environmental and social sustainability, such as climate change and the use of child labour, begin to permeate corporate thinking (Accenture 2012; Howard-Grenville et al. 2014; Schaltegger & Burritt 2010; Soosay, Fearne & Varsei 2014; Varsei, Christ & Burritt 2015; Varsei & Soosay 2013). Given the growing prevalence of such issues and their risks to business as well as society within which business operates, companies have been increasingly invoked to mitigate the negative environmental and social impacts of their business operations by several sources of pressure (Carter & Easton 2011; Sarkis, Zhu & Lai 2011; Soosay, Fearne & Dent 2012). As elaborated by Varsei et al. (2014), these sources include institutional norms and stakeholders such as customers, shareholders, governments, non-governmental organisations and public authorities. To minimise the negative environmental and social impacts of business, companies in various industries should carefully examine and optimise the design and management of their supply networks. Supply chain management involves major activities such as 1
materials acquisition, production, transportation and recycling, all of which may impose negative environmental and social impacts if not managed and optimised appropriately (Chaabane, Ramudhin & Paquet 2010; Howard-Grenville et al. 2014; Wisner, Tan & Leong 2008). This nexus is the main motivation for the present study. The research explores the interface between sustainability and supply chain management, particularly supply chain design (i.e. supply chain network design) given its strategic importance as a key element of supply chain management (Lambert & Cooper 2000; Owen & Daskin 1998; Simchi-Levi, Kaminsky & Simchi-Levi 2008). Because of the multidisciplinary characteristic of the sustainable supply chain management field (Winter & Knemeyer 2013), the study draws on multiple literatures including operations and supply chain management, sustainability, supply chain network design and facility location, operations research and management science, and computational science and optimisation. This chapter presents the introduction of the thesis. After outlining the study in this section, some key terms are defined in Section 1.2. Then, Section 1.3 presents a background of research, followed by outlining research significance and its principal purpose in Section 1.4, as well as research questions in Section 1.5. Section 1.6 provides an overview of the research methodology. The chapter ends by presenting the ethics consideration and thesis structure in Section 1.7 and 1.8, respectively.
1.2
Some definitions
Following are definitions for some key terms used throughout the thesis. Other definitions or explanations of these terms are presented and discussed throughout the study. These definitions may also demonstrate the complexity of the research context given several potential conflicting objectives inherited in supply chain management as well as sustainable supply chain management. Supply chain management: … a set of approaches utilized to efficiently integrate suppliers, manufacturers, warehouses and stores, so that merchandise is produced and distributed at the right quantities, to the right locations, and at the right time, in order to minimize systemwide
2
costs while satisfying service level requirements. (Simchi-Levi, Kaminsky & SimchiLevi 2008, p.1)
Sustainability, or sustainable development: … development that meets the needs of the present without compromising the ability of future generations to meet their own needs. (Brundtland 1987, p.43)
Sustainable supply chain management: … the management of supply chain operations, resources, information, and funds in order to maximize the supply chain profitability while at the same time minimizing the environmental impacts and maximizing the social well-being. (Hassini, Surti & Searcy 2012, p.70)
Model: … a representation of the essential aspects of an existing system (or a system to be constructed) which presents knowledge of that system in usable form. (Eykhoff 1974, p.1)
Considering these definitions, the next section presents a background of the study.
1.3
Research background
Supply chain management can be defined as … a set of approaches utilized to efficiently integrate suppliers, manufacturers, warehouses and stores, so that merchandise is produced and distributed at the right quantities, to the right locations, and at the right time, in order to minimize systemwide costs while satisfying service level requirements. (Simchi-Levi, Kaminsky & SimchiLevi 2008, p.1)
A typical supply chain consists of suppliers, manufacturers, distribution centres, and customers (Snyder 2011), each of which may contain complex multiple tiers such as suppliers’ suppliers or customers’ customers (Mena, Humphries & Choi 2013). Supply chain management involves a range of strategic, tactical and operational decisions linked with the firm’s business strategy and its competitive priorities. Strategic decisions, which are often interlinked, mainly deal with supply chain design, operations and supply chain strategies, supplier selection, process integration and strategic partnering in order to move products from source to final consumption efficiently and/or responsively (Chopra &
3
Meindl 2013; Fisher 1997; Kim, Sting & Loch 2014; Simchi-Levi, Kaminsky & SimchiLevi 2008). The literature highlights the strategic importance of supply chain design, which is concerned with the long-term decisions related to the number, location and capacities of various facilities in the supply chain (e.g. production plants and distribution centres), the optimal flow of material or finished products through the supply chain, and a set of suppliers or partners to select (Chopra & Meindl 2013; Meixell & Gargeya 2005; SimchiLevi, Kaminsky & Simchi-Levi 2008). Supply chain design has attracted close attention from academia and industry given its ‘strong impact on overall profitability and success’ of several companies around the world (Chopra & Meindl 2010, p.8). Research on supply chain design can be traced back to the pioneering study by Alfred Weber on facility location (Current, Min & Schilling 1990). In his book—Theory of the Location of Industries—Weber examined how to locate a production plant so as to minimise the total weighted distance (as a representation of cost) between the plant and its customers in different locations (Friedrich 1929). Since then, facility location has attracted the attention of many scholars in operations and supply chain management as well as operations research and management science (Melkote & Daskin 2001). The main research problem in facility location/supply chain design is how to select the most suitable location for a set of new facilities, or how to determine the best possible (or optimal) configuration of the supply chain (i.e. how to optimise supply chain design). This poses a challenge to the scholars given the problem’s typical computational complexity due to the associated multiple variables and conflating objectives (Boloori Arabani & Farahani 2012; Farahani, Drezner & Asgari 2009; ReVelle, Eiselt & Daskin 2008). To cope with the problem, several analytical mathematical models and optimisation approaches have been developed in the literature (Simchi-Levi, Kaminsky & Simchi-Levi 2008). These efforts are well studied by Owen and Daskin (1998), ReVelle, Eiselt and Daskin (2008) and Melo, Nickel and Saldanha-da-Gama (2009) in their comprehensive reviews of facility location/supply chain design models. Typically, there are interconnections or trade-offs between two conflicting objectives of the supply chain design problem: service level (i.e. serving customer demand responsively) and the 4
annualised supply chain cost (i.e. a summation of the fixed cost of facilities and the variable cost of production, transportation, etc.) (Shapiro 2007; Simchi-Levi, Kaminsky & Simchi-Levi 2008). Most of the published supply chain design studies have traditionally focused on the economic dimension of the problem with cost minimisation (or profit maximisation) and service level maximisation being the predominant objectives (Fahimnia 2011; Melo, Nickel & Saldanha-da-Gama 2009; Shapiro 2007). However, the nature of trade-offs may change over time. Although many companies have designed their supply chains only in the pursuit of profits (Chaabane, Ramudhin & Paquet 2012), they are increasingly forced by various stakeholders to consider additional performance indicators pertaining to the environmental and social dimensions (Tang & Zhou 2012; Varsei et al. 2014; Wu & Pagell 2011). ‘It is clear that sustainability is becoming integral to the way to do business’(Accenture 2012, p.2). Therefore, companies should design sustainable supply chains despite the higher level complexity in terms of modelling and optimisation as well as ‘balancing priorities’ and ‘dealing with strategic trade-offs among the economic, environmental and social elements’ of sustainability (Wu & Pagell 2011, p.577). These challenges have emerged when ‘putting sustainability into supply chain management’ in order ‘to fulfil the demands of sustainability’(Beske & Seuring 2014, p.322). Sustainability was defined in the United Nations report on sustainable development, known as the Brundtland Report, as ‘development that meets the needs of the present without compromising the ability of future generations to meet their own needs’ (Brundtland 1987, p.43). The report, released in 1987, was perhaps aimed at drawing global attention to intragenerational equity (i.e. balancing the needs of the existing generation) and intergenerational equity (i.e. meeting the needs of future generations). However, the concept of sustainability can be traced as far back as the eighteenth century in a few seminal studies by political economists such as François Quesnay, Thomas Robert Malthus and Adam Smith (Whittaker 2011). These scholars shared their thoughts on and doubts about sustainable development—some of them are still valid and represented in the Brundtland Report. These challenges include environmental limits; poverty and inequality; population growth; food and water security; and, energy, industry and urbanisation (Brundtland 1987). 5
It can be argued that sustainability challenges might best be met if there is a balance between economic, environmental and social dimensions of growth, development and performance management. This approach, known as the triple bottom line, was popularised by Elkington (1999) with the aim of urging business to assess performance using the traditional goal of increasing profit as well as considering the associated social and environmental impacts. Companies are critical contributors to sustainable development (Burritt & Schaltegger 2014; Schaltegger & Wagner 2006). They and their extended supply chains are parts of a larger system (i.e. planet and society), and their managerial decisions matter (Senge 2010). This would necessitate adopting a balanced approach to decisionmaking (United Nations 2012) in various areas including supply chain design and management. However, it appears that the balanced approach has been rarely adopted, which establishes a pressing gap in the sustainable supply chain design and management literature particularly in the associated analytical modelling approaches (Brandenburg et al. 2014; Chaabane, Ramudhin & Paquet 2012; Fahimnia, Reisi et al. 2013; Ramos, Gomes & Barbosa-Póvoa 2014; Seuring & Müller 2008; Walker et al. 2014; Winter & Knemeyer 2013). The majority of publications have primarily dealt with one or two dimensions in isolation, and relatively very few of them have addressed all three aspects—economic, environmental and social—simultaneously (Brandenburg et al. 2014; Eskandarpour et al. 2015; Seuring 2013). The modelling efforts incorporating the three sustainability dimensions are ‘neglected’ in the extant literature (Brandenburg et al. 2014, p.308). This highlighted gap in the literature is linked to the following central research question.
1.4
Research questions
The central research question in this study is: How can the economic, environmental and social dimensions of sustainability be incorporated simultaneously into the supply chain design optimisation model in order to help decision-makers develop a more sustainable supply chain? This research question can subsequently be answered with the following sub-questions. •
How can the economic performance indicators be measured? 6
•
How can the environment performance indicators be measured?
•
How can the social performance indicators be measured?
•
How can these multidimensional sustainability indicators be incorporated and optimised in the integrated supply chain design model?
1.5
Research significance and aims
Urgent action must be undertaken to change the business-as-usual scenario in order to minimise the harmful impact of human activities on the environment and society (Gupta & Palsule-Desai 2011; Howard-Grenville et al. 2014). In this regard, supply chain management plays a key role since it significantly affects the extraction, production, transportation, and recycling of materials and/or products, which all have major impacts on planet and people, in addition to profit (Chen, Zhang & Delaurentis 2014; Gupta & Palsule-Desai 2011). Therefore, supply chain management needs to extend beyond traditional economic-oriented objectives and consider a broader range of objectives including environmental and social (Soosay, Fearne & Varsei 2014). Environmental issues have gained considerable attention in the literature (HowardGrenville et al. 2014). One of the issues is climate change whose main contributors are greenhouse gas (GHG) emissions, particularly carbon dioxide (CO2) (Chaabane et al. 2008). The prediction is that by the year 2100, an increase of 2 to 4 degrees Celsius in global average temperature may result in a devastating catastrophe: extinction of at least 30% of species; water and food shortage for millions of people; and more shockingly, the higher probability of mortality in many populated areas around the world, including many parts of East Asia, as a consequence of heat waves and floods (Gupta & Palsule-Desai 2011). This reality has forced governments and other authorities to take necessary actions, for example by forming legislations on GHG emissions. These legislations could impact the economic performance of companies. As a consequence, firms are forced to mitigate negative environmental impacts (Gupta & Palsule-Desai 2011) at both organisational and supply chain levels (Soosay, Fearne & Varsei 2014).
7
Similarly, companies are increasingly facing issues related to their social performance (Dreyer, Hauschild & Schierbeck 2006), and the associated commercial and reputational risks at both organisational and supply chain levels (Carter & Rogers 2008). These issues not only could threaten their brand image, but also may impact on their economic performance (Carter & Rogers 2008). There are several cases in the media related to companies such as Walmart, Nike, Gap, H&M and Mattel who have been held responsible for poor social performance (Andersen & Skjoett-Larsen 2009; Frost & Burnett 2007). For example, the use of under-aged workers in cotton harvesting in Uzbekistan forced Walmart to cease sourcing cotton and cotton materials from Uzbekistan in 2008 (Walmart 2008). Companies, thus, should consider the social impacts of their supply chains particularly at the design/redesign stage in order to avoid reputation and financial loss in the long term (Varsei et al. 2014). As mentioned earlier, much of the existing literature on sustainable supply chain management addresses one or two dimensions in isolation (Brandenburg et al. 2014; Kleindorfer, Singhal & Van Wassenhove 2005; Pagell & Wu 2009; Seuring 2013; Seuring et al. 2008). Therefore, the main aim of the research is to address this highlighted gap in the literature within the area of supply chain design (Brandenburg et al. 2014; Chaabane, Ramudhin & Paquet 2012; Seuring 2013; Seuring & Müller 2008). The study explores how all three dimensions of sustainability can be incorporated simultaneously in supply chain design. In doing so, this research is one of few studies that incorporates some of the key aspects of all three dimensions of sustainability in a single overarching framework and analytical model for supply chain design. This may help researchers investigate existing supply chain networks in various industries in terms of multiple sustainability indicators, and explicitly examine the interconnections/trade-offs between them (Varsei, Christ & Burritt 2015). This may also assist companies in designing and developing sustainable supply chains, and ultimately in ‘reshaping value chains’ towards sustainable development (Howard-Grenville et al. 2014, p.615).
8
1.6
An overview of the research methodology
First, the literature related to sustainable supply chain management is reviewed. Then, in order to address the identified gap in the literature and the aforementioned research questions, the study uses the ‘analytical mathematical research methodology’ which is often referred to as analytical modelling and optimisation (Kotzab et al. 2006; Render, Stair & Hanna 2012; Shapiro 2007; Wacker 1998, p.361). Supply chain design typically involves some key strategic decisions which often necessitate careful examination using quantitative methods (Chopra & Meindl 2013; Simchi-Levi, Kaminsky & Simchi-Levi 2008). Given its computational complexity even if only one economic indicator such as supply chain cost is considered (Farahani, Drezner & Asgari 2009; Owen & Daskin 1998; Watson et al. 2013), many scholars in the field have employed this methodology to address the supply chain design problem (Brandenburg et al. 2014; Chaabane, Ramudhin & Paquet 2012; Ramos, Gomes & Barbosa-Póvoa 2014; Seuring 2013). The analytical modelling is a ‘scientific approach to managerial decision-making’ (Render, Stair & Hanna 2012, p.2). According to Render, Stair and Hanna (2012), the typical stages in the analytical modelling methodology include: 1. define a problem 2. develop a mathematical model which can represent the problem as much as possible 3. prepare artificial data or collect real data for testing the model and illustrating its applicability (i.e. model deployment) 4. find an appropriate solution approach in order to overcome the computational complexity of the model and optimise it 5. analyse the numerical results, discuss the managerial implications and disseminate the modelling approach for future improvement and development. Chapter 3 explains this methodology and also the rationale behind it in further detail. Chapter 4 develops a mathematical model. As seen in the third stage, researchers either can use artificial data or real data for model deployment. To date, the majority of quantitative modelling studies in the sustainable supply chain management literature have used ‘madeup’ artificial data (Seuring 2013). However, to enrich the empirical foundation of analytical
9
modelling research in the field (Choi & Guide 2012), scholars have increasingly recommended collecting real data and through conducting case study research (Brandenburg et al. 2014) while acknowledging the associated challenges (Watson et al. 2013). Following this recommendation, the study conducts case study research and collects real data for the sustainable supply chain design analysis of a major wine company in Australia. An efficient multi-objective optimisation method—the ‘augmented ε-constraint’ method (Mavrotas 2009)—is used in Chapter 5 to solve the model in the the IBM ILOG CPLEX Optimiser software (i.e. a popular optimisation software package), and to find the associated solutions and alternative supply chain design scenarios with various sustainability implications.
1.7
Ethics considerations
To illustrate the application of the proposed mathematical model, this research conducted a case study and collected real data associated with the parameters of the model (i.e. inputs to the mathematical model). Therefore the study entailed ethical considerations and warranted the necessary ethics clearance prior to collection of data. The main ethical concerns included ensuring participants understood that their involvement was voluntary, and that their identity and their responses would remain anonymous and kept confidential. An ethics application, including an overview of the research and the details of the data collection process, was submitted to the Human Research Ethics Committee at the University of South Australia for approval. The approval was granted on 6th January 2014 under ethics application number 0000032203. A copy of the ethics approval letter and interview protocol are attached as Appendix 1 and Appendix 2 respectively. During the data collection period, first an information sheet and an official letter specifying the study objectives and background were sent to the participants. For those who were interested in more information about the research, a summary of the study was sent and also a presentation was delivered. All participants were reminded that they could withdraw from the study at any time without the need to give any explanation. These considerations ensured that the research would be in line with the ethics considerations.
10
1.8
Thesis structure
This thesis is organised into six chapters. This introductory chapter has provided an overview of this research, including the research background, primary purpose, significance and questions. The research methodology has also been outlined. Chapter 2 presents a review of the literature on sustainable supply chain management in order to describe some important relevant features of this field and identify some research gaps. The chapter provides an overview of supply chain management and sustainability and their historical backgrounds. Following the triple bottom line approach to sustainability (Elkington 1999), the chapter discusses the economic, environmental and social aspects of sustainable supply chain management. It concludes by analysing the literature based on a systematic literature review approach (Winter & Knemeyer 2013), which helps highlight some research gaps including the gap under investigation in this study (i.e. the simultaneous incorporation of the three sustainability dimensions into the strategic supply chain design problem). Chapter 3 aims to address the gap. It proposes a novel framework for sustainable supply chain design encompassing the economic, environmental and social dimensions simultaneously (Varsei & Soosay 2013; Varsei, Soosay & Fahimnia 2013; Varsei et al. 2014). The framework is grounded in the facility location and supply chain design literature (Drezner & Hamacher 2002; Owen & Daskin 1998). The chapter details supply chain design, facility location, and the ‘analytical mathematical research methodology’ (Wacker 1998, p.361). The relevant literature on sustainable supply chain design is discussed in detail in this chapter. Chapter 4 employs the framework to propose a novel integrated multi-objective model for sustainable supply chain design in the context of the wine industry. The chapter starts with an elaboration on why this research focuses on the wine supply chain. It highlights the interface between the wine supply chain and the three dimensions of sustainability (Christ & Burritt 2013; Soosay, Fearne & Varsei 2014). The chapter provides an overview of the wine supply chain. Drawing from the literature on wine supply chain management (Garcia et al. 2012), the chapter presents a typical wine supply chain including suppliers, wineries,
11
bottling plants, distribution centres and demand points. The chapter then develops the model based on this typical wine supply chain. The objective functions, constraints, assumptions, indices, parameters (i.e. inputs) and decision variables (i.e. outputs) of the mathematical business model are presented. The model could assist wine companies in designing sustainable supply chains in terms of a set of economic, environmental and social performance indicators (Varsei, Christ & Burritt 2015), and it may shed light on the application of such models in other industries. Chapter 5 illustrates the applicability of the proposed model through conducting a case study. It elaborates case study research (McCutcheon & Meredith 1993; Seuring 2008; Stuart et al. 2002) and its application for the analytical modelling studies as recommended by Brandenburg et al. (2014) and adopted recently by Varsei, Christ and Burritt (2015). Then the chapter describes the case setting, data collection process and the mathematical problem statement. Based on the model presented in Chapter 4, it develops a customised multi-objective model. Drawing from the literature on multi-objective optimisation (Branke et al. 2008; Zhang & Reimann 2014), the chapter uses the ‘augmented ε-constraint method’ (Mavrotas 2009) to solve the model. The Pareto optimal (i.e. trade-off) results, the associated supply chain design scenarios, and their implications are discussed in detail. Chapter 6 summarises the thesis, and discusses the theoretical contributions of the study and the associated managerial implications. The limitations of this multidisciplinary study are discussed in detail along with several opportunities for future research.
12
2
2.1
From supply chain to sustainable supply chain management: A literature review
Introduction
Sustainable supply chain management has evolved as an extension of the traditional supply chain management to incorporate sustainability. This chapter provides a review of the relevant literature. The chapter presents an overview of supply chain management, sustainability and their historical backgrounds. Then different aspects (i.e. economic, environmental and social) of the emerging area of sustainable supply chain management are discussed. The chapter ends by analysing the literature based on a systematic approach, which highlights some research gaps including the gap under examination throughout this thesis: a model for sustainable supply chain design encompassing economic, environmental and social dimensions simultaneously. This chapter is organised as follows. Sections 2.2 and 2.3 present an overview of supply chain management and sustainability, respectively. The literature on sustainable supply chain management is reviewed in Section 2.4, followed by the summary and conclusion of this chapter in Section 2.5.
2.2
Supply chain management
Supply chain management refers to the process of planning, executing and controlling the activities and operations of a supply chain efficiently to meet planned objectives (Melo, Nickel & Saldanha-da-Gama 2009). The term supply chain management was first introduced in 1982 (Oliver & Webber 1982), and, since then, it has increasingly received considerable attention in industry as well as in academia (Lambert 2008; Simchi-Levi, Kaminsky & Simchi-Levi 2008; Stadtler 2005). 13
Supply chain management has evolved in accordance with one of the new rules of modern business management: ‘individual businesses no longer compete as solely autonomous entities, but rather as supply chains’ (Lambert & Cooper 2000, p.65). Leading-edge companies have realised the strategic importance of supply chain management simply because their rivals are competing based on their supply chain capabilities (Christopher 2005). Similarly, many researchers assert that today’s competition is about supply chain versus supply chain rather than firm versus firm (Christopher 2005; Ketchen & Hult 2007; Lambert 2008; Lambert & Cooper 2000; McCarter & Northcraft 2007). Therefore, at the strategic level, it could be inferred that how to design, manage and optimise supply chains is of critical importance for firms to gain competitive advantage (Chen & Paulraj 2004; Fine 2000). The following sections review the definitions of supply chain and supply chain management, followed by a discussion of the importance of supply chain design. 2.2.1 Supply chain There are several definitions of the term supply chain in the supply chain management literature (Chopra & Meindl 2010; Lambert, Stock & Ellram 1998; Mentzer et al. 2001; Sadler 2007; Wisner, Tan & Leong 2008), nevertheless, they all address common aspects of the supply chain concept. Lambert, Stock and Ellram (1998) assert that a supply chain is the alignment of organisations through a system or a network in order to bring products or services to market. Stadtler (2005, p.576) cites Christopher (1998) who defines a supply chain as: … a network of organizations that are involved, through upstream and downstream linkages in the different processes and activities that produce value in the form of products and services in the hand of the ultimate consumer.
Similarly, Mentzer et al. (2001) view a supply chain as a set of at least three organisations involved in the upstream (i.e. supply) and downstream (i.e. distribution) flows of products, services, finances and information from a source to a customer. Another definition notes a supply chain as ‘all parties involved, directly or indirectly, in fulfilling a customer request’ (Chopra & Meindl 2010, p.2). What might be common across these definitions is that firstly, a supply chain is a network of integrated companies and secondly, companies form their supply chains to produce value in the eyes of the ultimate customer. 14
The terms supply chain, supply chain network, and supply network have been used interchangeably in the literature by many scholars to denote the configuration of a set of connected companies (Slack et al. 2009). A supply chain may consist of suppliers, manufacturers, distributers, retailers and final customers. These typical supply chain members create a network in which products, services, information and funds flow from sources to final costumers (Chopra & Meindl 2010; Simchi-Levi, Simchi-Levi & Kaminsky 2004). Some authors argue that the term supply chain is a misnomer since the chain implies a linear system, while today’s supply chains are more represented by networks (Chopra & Meindl 2010; Snyder 2011) because there are multiple tiers of suppliers and customers for firms in reality. The supply chain design of most existing focal companies is similar to a network, as illustrated in Figure 2.1, in which a focal company is in the centre of the network of suppliers and customers (Christopher 2005; Lambert 2008; Lambert, Cooper & Pagh 1998). A focal company is the company that rules or governs the supply chain (Handfield & Nichols 1998). From the above discussion, a fundamental question may arise: why do companies have increasingly formed complex networks to run their businesses? This would raise the notion of value.
Suppliers
A focal company
Customers
Members of the focal company’s supply chain
Figure 2.1 The supply chain network, adapted from Christopher (2005, p.5)
15
Many companies have expanded their business boundaries and designed supply chains (or supply chain networks) in order to gain competitive advantage and produce value for their customers (La Londe 1997; Mentzer et al. 2001; Simchi-Levi, Kaminsky & Simchi-Levi 2008). Mentzer et al. (2001) argue that this expansion has ultimately led them to generate profits for all supply chain members. Therefore, it could be established that forming a supply chain becomes an effective means of sustaining economic performance—in other words, ‘the supply chain becomes value chain’ (Christopher 2005, p.14) . Described and popularised by Porter (1985), the value chain concept implies that companies should assess their activities (e.g. procurement, logistics and operations) and investigate whether they are able to perform all of them efficiently. If they do not, they should work in partnership with other firms that can provide value advantage (Christopher 2005; Porter 1985). The formation of long-term strategic partnerships with supply chain members such as suppliers or distributors can help focal companies create customer value and ultimately competitive advantage (Mentzer et al. 2001). In other words, value is produced not only by a focal organisation, but also by its supply chain members (Christopher 2005). Hence, supply chain management plays a significant role in maximising the overall value co-created by all supply chain members (Chopra & Meindl 2013). 2.2.2 Managing supply chains The term supply chain management was introduced as a strategic approach to logistics management and these two terms have often been used synonymously (Cooper, Lambert & Pagh 1997; Gibson, Mentzer & Cook 2005; Simchi-Levi, Kaminsky & Simchi-Levi 2008; Wolf 2008). Nonetheless, the Council of Logistics Management defined logistics as a subset of supply chain management in 1998 to differentiate between these two terms (Lambert, Cooper & Pagh 1998), and in 2005 the council changed its name to the Council of Supply Chain Management Professional as the leading-edge professional association in this field (Larson, Poist & Halldórsson 2007). According to Larson, Poist and Halldórsson (2007), this name change indicates that practitioners have gradually realised that supply chain management encompasses not only logistics, but also other functions such as procurement, operations and marketing . 16
Along with supply chain management evolution in practice, academics have tried to clarify and conceptualise these two areas. Cooper, Lambert and Pagh (1997) are among the first authors who elaborate the main differences between supply chain management and logistics management by proposing a framework . They argue that business process integration across a supply chain extends beyond logistics boundaries using the product development and commercialisation processes as an example to support their argument (Cooper, Lambert & Pagh 1997). Based on this article, Lambert, Cooper and Pagh (1998), Lambert and Cooper (2000) and Lambert (2008) refer to logistics as one of the functional silos in supply chain management, along with purchasing, production, marketing, research and development (R&D) and finance functions. From another perspective, Ross (1998) explains that while logistics is more concerned with efficient transportation and inventory management, supply chain management also includes marketing and manufacturing. To elaborate the ‘cross-disciplinary concept’ of supply chain management, Mentzer, Stank and Esper (2008, p.31) argue that the scope of supply chain management expands in order to embrace not only logistics, but also production, marketing and operations management, as portrayed in Figure 2.2.
Supply Chain Management
Operations Management
Logistics Management
Marketing Management
Production Management
The Firm
Figure 2.2 Scope of supply chain management, adapted from Mentzer, Stank and Esper (2008, p.38)
17
Mentzer, Stank and Esper (2008) propose key elements of logistics, marketing, production and operations management and compare the conceptual boundaries of these areas with the boundary of supply chain management in order to clarify its domain. They argue that logistics, marketing and production management fall under the functional decision scope that is more related to planning, controlling and executing activities ‘within the sphere of each individual functional area’( Mentzer, Stank & Esper 2008, p.33), while operations management applies to operations research and management science (OR/MS) quantitative methods (e.g. modelling and optimisation approaches) to ‘improve business processes that cross internal functional boundaries’ (Mentzer, Stank & Esper 2008, p.36). In contrast, supply chain management aims to ‘improve business processes that cross organisational boundaries’ which necessitates collaborative efforts among various organisations (Mentzer, Stank & Esper 2008, p.38). In an attempt to adopt a different approach to the supply chain management scope, Wisner, Tan and Leong (2008) argue that supply chain management requires that a company effectively deals with purchasing, operations, distribution and integration across its supply chain. Integration enables firms to link their internal processes with their suppliers and customers (Cooper, Lambert & Pagh 1997; Omar et al. 2012), yet it is the most difficult to implement according to Wisner, Tan and Leong (2008). Those processes that should be integrated include customer relationship and service management, demand management, order fulfilment, manufacturing flow management, supplier relationship management, product development and returns management (Wisner, Tan & Leong 2008). After the discussion on the supply chain management scope, it is worth briefly reviewing several definitions proposed by scholars (Chopra & Meindl 2010; Cooper, Lambert & Pagh 1997; Lambert 2008; Mentzer et al. 2001; Mentzer, Stank & Esper 2008; Sahin & Robinson 2002; Stadtler 2005; Wisner, Tan & Leong 2008). Cooper, Lambert and Pagh (1997) review definitions and describe some commonalities among them, such as involving intra- and inter-organisational integration and providing high customer value with an appropriate allocation of resources. In another attempt to review the definitions of supply chain management, Stock and Boyer (2009) examine 166 definitions in the literature. They analyse each definition in terms of 18
three identified themes (i.e. activities, benefits, and constituents) and propose their definition of the term with aim of capturing most of the themes: The management of a network of relationships within a firm and between interdependent organizations and business units consisting of material suppliers, purchasing, production facilities, logistics, marketing, and related systems that facilitate the forward and reverse flow of materials, services, finances and information from the original producer to final customer with the benefits of adding value, maximizing profitability through efficiencies, and achieving customer satisfaction. (Stock & Boyer 2009, p. 706)
While there are ‘too many definitions’ of supply chain management in the literature (Stock & Boyer 2009, p.691) and ‘consensus is lacking on the definition of the term’ (Burgess, Singh & Koroglu 2006, p.703), it can be seen that most definitions promote a holistic approach towards integration in a supply chain to synchronise capabilities in order to maximise efficiency and customer value. For the purpose of this research, supply chain management is defined as: … a set of approaches utilized to efficiently integrate suppliers, manufacturers, warehouses and stores, so that merchandise is produced and distributed at the right quantities, to the right locations, and at the right time, in order to minimize systemwide costs while satisfying service level requirements. (Simchi-Levi, Kaminsky & SimchiLevi 2008, p.1)
Given this definition, supply chain management aims to examine the impacts of supply chain members (e.g. suppliers) on the overall supply chain cost and customer satisfaction. In addition, the efficient integration of suppliers, manufacturers, distributers and stores would mean that supply chain management should address decisions at the strategic, tactical and operational levels. These decisions generally deal with several key issues such as supply chain design, distribution strategies, strategic partnering and inventory management (Simchi-Levi, Kaminsky & Simchi-Levi 2008). The definition also emphasises the importance of adopting a system-wide approach to supply chain management which may involve conflicting objectives (Shapiro 2007). For example, there is a trade-off between minimising supply chain cost and maximising customer service, because customer satisfaction might be enhanced when a focal company establishes more distribution centres in a market zone, which in turn, may need more investments (Shapiro 2007; Shen 2007). The trade-offs between objectives should be 19
evaluated carefully by means of analytical methods in order to optimise the efficient allocation of resources (Mentzer, Stank & Esper 2008; Shen 2007). While many studies focus on the aforementioned two objectives (Melo, Nickel & Saldanha-da-Gama 2009), some authors such as Lee (2004, p.1), Ketchen and Hult (2007) and Hult et al. (2006) suggest that supply chain management may need to achieve more objectives. According to Lee (2004, p.1), these include agility (i.e. the ability to ‘respond to short-term changes in demand or supply quickly’), flexibility (i.e. the ability to ‘adjust supply chain design to accommodate market changes’) and alignment (i.e. the ability to establish ‘incentives for supply chain partners to improve performance of the entire chain’). The growing importance of supply chain management, both in academia and practice, has been attributed to the success of many companies around the world (Chopra & Meindl 2010; Hult et al. 2006; Ketchen & Hult 2007; Lee 2004; Mentzer et al. 2001). Wal-Mart is an example of one of those companies that has benefited largely from supply chain management (Chopra & Meindl 2010; Ketchen & Hult 2007; Lee 2004). The marked increase in annual revenues from USD 1 billion in 1980 to USD 379 billion in 2008 indicates a significant growth of this company (Chopra & Meindl 2010). Wal-Mart has worked collaboratively with its suppliers to improve product availability and quality while reducing cost. The company has also designed its supply chain in several regions in such a way that inventory replenishment is managed efficiently, therefore Wal-Mart has cut costs relating to redundant stocks (i.e. extra inventory) (Chopra & Meindl 2010). In other words, Wal-Mart has benefited from the efficient design of its supply chain. According to Chopra and Meindl (2010, p.8), supply chain design has ‘a strong impact on overall profitability and success’. Supply chain design, structure, configuration and network design are terms that have been used interchangeably in the literature (Ballou 2001) to refer to the shape of a supply chain and to denote a key element of supply chain management. Lambert (2008), Lambert, Cooper and Pagh (1998) and Cooper, Lambert and Pagh (1997) suggest that key elements of supply chain management are: 1) supply chain design, 2) supply chain management processes, and 3) supply chain management components, as depicted in Figure 2.3. These authors suggest that a focal company should firstly design the best possible supply chain 20
and select the most suitable supply chain members. Then, the focal company needs to decide on which processes should be linked with each supply chain member. Finally, plans are made on the level of integration for each process. Given that the supply chain design is critical to the success of supply chains, it plays a key role in supply chain management. According to Chaabane, Ramudhin and Paquet (2012), supply chain design aims to maximise an organisation’s long-term economic performance through defining its best possible supply chain configuration.
Supply chain design
Supply chain management components
Supply chain management processes
Figure 2.3 Supply chain management: key elements, adapted from Lambert (2008)
Supply chain network design ‘is considered a strategic decision level problem that provides an optimal platform for the effective and efficient supply chain management’ (Tiwari et al. 2010, p.95). Typically, different decision levels in supply chain management exist, depending on the where on the time horizon a decision has impact upon the overall performance of a supply chain (Chopra & Meindl 2010). These levels are mainly classified as strategic, tactical and operational (Chopra & Meindl 2010; Mula et al. 2010; Shen 2007; Simchi-Levi, Simchi-Levi & Kaminsky 2004; Snyder 2011). Strategic level decisions are for the long-term, i.e. several years (Chopra & Meindl 2010). These decisions may include supplier selection and resource allocation, information system utilisation, as well as the decisions related to the, number, location and capacity of 21
production and warehousing facilities (Chopra & Meindl 2010; Simchi-Levi, Simchi-Levi & Kaminsky 2004). Since strategic level decisions are very expensive to alter after being made (Chopra & Meindl 2010), focal companies should take a closer look at these important issues and their potential consequences on tactical level decisions. Tactical level decisions’ effects on a supply chain mainly span from three months to a year (Simchi-Levi, Simchi-Levi & Kaminsky 2004). It is concerned with monthly demand forecasting in different market regions, inventory replenishment strategies in different locations, transportation capacity planning and its optimised frequency and the production volume of manufacturing facilities (Chopra & Meindl 2010; Mula et al. 2010; Snyder 2011). Tactical level decisions help focal companies to define operational level decisions related to weekly or daily supply chain planning and operations (Chopra & Meindl 2010; Snyder 2011). It is fair to state that successful companies, such as Wal-Mart and Dell, have benefited enormously through cascading strategies and decisions from strategic to tactical level and from tactical to operational level, which in turn, leads to effective flow in the chain from sourcing to consumption (Chopra & Meindl 2010). In other words, the importance of strategic level decisions such as supply chain design lies in the fact that they have a significant impact on tactical and operational level decisions (Shen & Qi 2007; Tiwari et al. 2010). From the above discussion, it could be inferred that efficient designing (and managing) supply chain networks has enabled leading companies such as Wal-Mart to enhance their economic performance (Khiewnavawongsa 2011). Despite Wal-Mart’s impressive financial success, there have also been criticisms about the social and environmental misconducts occurring within its supply chain upstream (Andersen & Skjoett-Larsen 2009; Greenhouse 2013b; Irwin & Clark 2006; Wells 2011) which negatively impacted its image and economic performance. Wal-Mart has invested and launched several environmental and social initiatives to manage its supply chain (Walmart 2008). Along with Wal-Mart, several other Fortune 500 top firms have increasingly published sustainability reports, set goals and launched initiatives to demonstrate their commitment to become more sustainable both at the organisational level and supply chain level (Khiewnavawongsa 2011). This might mean that organisations have realised that a combination of economic, 22
social and environmental objectives should be considered in supply chain management decisions, including those related to supply chain design (Pagell & Wu 2009; Wu & Pagell 2011). Sustainable supply chain management has emerged in the literature to shed light on the link between sustainability and supply chain management. Before reviewing this aspect and in order to clarify the different dimensions of sustainability, a summary of sustainability and its evolution are presented in the next section.
2.3
Sustainability
Sustainability has been ‘a major theme’ of public debate and political discussion (Moscardo 2013, p.2), and has become an important term in today’s business management lexicon (Banerjee 2003). According to Wells (2011), although its definition and scope are diverse among authors, there is a general consensus that it includes at least three interdependent aspects: economic, social and environmental dimensions. Many studies indicate that sustainability concerns should be addressed in firms’ operational and competitive strategies (Matos & Hall 2007), due to various forces driven by stakeholders including governments, non-governmental organisations, professional and industry organisations, business partners and competitors, investors and risk assessors, and more importantly, customers and the wider community (Moscardo 2013). Wells (2011) argues that the publicity of sustainability concerns related to major international companies has intensified public debate in this area further. This is in addition to major issues related to climate change, carbon tax, pollution, resource scarcity, poverty and financial stability. Some examples of these events include the Exxon Valdez oil spill in Prince William Sound, Alaska, and its long-term impact on the environment and local communities (Carson et al. 2003); Nike Inc.’s use of sweatshops by its suppliers located in developing countries (Gini 2005); BHP Billiton’s Ok Tedi mine in Papua New Guinea and consequent major water pollution; and the recent ocean oil spill in the Gulf of Mexico which is one of the worst pollution events in the history of the United States (Wells 2011). More recently, the eight-storey Rana Plaza clothing factory building in Bangladesh collapsed and killed more than a thousand workers in April 2013 (Green 2013). This 23
disaster in Bangladesh with tremendous loss of lives has been challenging to several companies and brands such as Wal-Mart, PVH (the parent company of Calvin Klein and Tommy Hilfiger), Gap, Mango and H&M because the factory with poor working conditions was the supplier to these companies (Greenhouse 2013a; Manik & Yardley 2013). In general, it could be argued that there is no legal restriction on partnership with suppliers located in countries in which the risk of social misconducts is higher and whose governments are reluctant to take acceptable steps toward social and environmental sustainability (Manik & Yardley 2013). However, focal companies should go beyond legal requirements because they are increasingly being held accountable for their supply chains’ misconducts, which might threaten their economic outcomes. Focal companies may face damage to their stock value, financial repercussions from consumers, or sustained public protests if they do not adopt stricter codes of conduct (Greenhouse 2013a). While it is generally agreed that organisations must be financially viable to survive (Banerjee 2008, 2011; Wells 2011), it could be established that they should embed environmental and social initiatives to maintain their brand image (Stewart 2005). Du Pisani (2006, p.92) argues that organisations should draw a distinction between economic ‘growth (quantitative change)’ and ‘development (qualitative change)’ and should emphasise the latter which is sensitive to the needs of society and environment. Otherwise they may jeopardise what has been called the license to operate: a set of social rules which are both explicit (in regulations) and implicit (in values and norms) and together define the acceptable operating conditions of a business in a society (Wells 2013). Wells (2011) argues that organisations should redefine their business models to incorporate sustainability principles in order to maintain competitive advantage. In the next few sections, the definition and the historical development of sustainability are presented to clarify its scope and principles. 2.3.1 Definition of sustainability There are several definitions of the term ‘sustainability’ in the literature. One definition involves maintaining a balance between three components: profits, environment and people. Another refers to increasing the total quality of life and maintaining ecological processes, on which human life depends (Giddings, Hopwood & O’Brien 2002; Moscardo 24
2013). The ‘most well-adopted and most often quoted definition of sustainability’ (Carter & Rogers 2008, p.363) is the one provided by the United Nations report on sustainability in 1987, known as the Brundtland Report: … development that meets the needs of the present without compromising the ability of future generations to meet their own needs. (Brundtland 1987, p.43)
According to Wells (2013), the Brundtland Report equally emphasises both intragenerational equity (i.e. balancing the needs of the generation living now) and intergenerational equity (i.e. meeting the needs of future generations). Therefore it draws global attention to sustainability challenges which include the concept of environmental limits; poverty; inequality; population growth and food security; energy, industry and urbanisation; biodiversity and ecosystem integrity; a need for the institutional change for sustainable development; and a need for long-term orientation (Brundtland 1987). Based on this report and the mentioned challenges, it appears that sustainability has three main areas of focus, namely economic, environmental and social, also known as the triple bottom line. This approach was popularised by John Elkington (1999) in his book: Cannibals with forks: the triple bottom line of 21st century business. He encouraged businesses to consider their performance using not only the traditional goals of profits but also their social and environmental impacts (Elkington 1999, p.vii). The triple bottom line approach has played a crucial role in shaping several sustainability initiatives and reporting frameworks such as the Global Reporting Initiative framework (Moscardo 2013). 2.3.2 The historical background of the concept of sustainability Although some authors mention that the concept of sustainability originated from the Brundtland Report, it appears that the idea of sustainability can be traced back to the eighteenth century in the pursuit of economic growth (Wolfe 2012). The discussion about the historical background of the concept of sustainability can be categorised into two time periods: before and after 1900. Sustainability and the associated challenges have origins in a few seminal studies by some political economists (Lumley & Armstrong 2004). As early as 1758, François Quesnay, a French political economist, considered the health of society and nature in his book Tableau
25
économique (economic table) and then Thomas Robert Malthus, a British political economist, published An Essay on the Principle of Population in 1798 (Moscardo 2013; Paul 2008). Malthus articulated a controversial idea about the economics of population growth and advocated ‘limits to growth’, arguing that the world would face famine and disaster because population, when unchecked, rises in a geometrical ratio due to economic growth, but subsistence increases in an arithmetical ratio (Elliott 2005; Paul 2008). This continued to be a debate thereafter. Although Malthus’s prediction failed to materialise due to many technological advancements, his concern resurfaced many years later and influenced the approach to sustainability, particularly regarding the depletion of the earth’s non-renewable resources (Baumol, Litan & Schramm 2007; Elliott 2005), which can be seen in the study by Meadows, Randers and Meadows (2005). While Malthus forecasted ‘a very black world’ (Elliott 2005, p.267), Adam Smith, one the most influential economists, expounded how economic growth provides a better life for people and society and alleviates poverty in his well-known book published in 1776: An Inquiry into the Nature and Causes of the Wealth of Nations (known as the Wealth of Nations) (Whittaker 2011). Although Adam Smith advocated an individual’s self-interest and free market economy and criticised restrictions on these concepts imposed by states (governments), he took a different angle and discussed sympathy, benevolence and justice in another book The Theory of Moral Sentiments, emphasising consideration of others’ interests (Hunt & Lautzenheiser 2011; Perelman 2010; Whittaker 2011). These apparently contradicting viewpoints have been acknowledged as ‘The Adam Smith Problem’ (Whittaker 2011, p.33). However, it could be argued that he presumably tried to acknowledge both sides and considered a trade-off between wealth and social justice. In fact, according to West (1992, p.2137): Adam Smith was not “an extreme dogmatic” defender of laissez-faire [free market] capitalism… [He] had several serious doubts about its moral conditioning of individuals under capitalism.
One might note that economic and social perspectives of sustainability are not seen as the central focus of the aforementioned studies, nevertheless, the environment could not be separated and it was, directly or indirectly, involved in these studies. Furthermore, Wells (2011) points to George Perkins Marsh—a North American environmentalist cited as the 26
prophet of conservation (Lowenthal 2009)—as one of the first to focus on the environmental aspect of sustainability. Marsh (1864, p.v) recognised ‘the general effects and prospective consequences of human action upon the earth’s surface’ and addressed some environmental challenges such as deforestation, waste, restoration and environmental degradation. Efforts for conservation were continued by John Muir who founded the Sierra Club in 1892 (one of the oldest conservation organisations), leading to increased public awareness of conservation and also more environmental concerns in the twentieth century (Moscardo 2013). It can be seen that Adam Smith’s notion of moral sentiment was transformed into social responsibility in the twentieth century. From the 1930s, many authors started to view businesses as ‘responsible’ for their activities and therefore the conceptual landscape of corporate social responsibility emerged in the literature (Carroll 1999). Carroll and Shabana (2010) note that sustainability and corporate social responsibility are similar concepts and Moscardo (2013, p.24) states that they are usually regarded ‘in the same way’. He quotes the World Business Council for Sustainable Development which defines corporate social responsibility as: … the commitment of business to contribute to sustainable economic development, working with employees, their families, the local community and society at large to improve their quality of life. (p.24)
According to Carroll (1979), the first definitive book on corporate social responsibility was published by Howard R. Bowen entitled Social responsibilities of the businessman (Bowen 1953) in which many of the central principles of sustainable business are addressed. Bowen (1953) argues that the decision-making processes of large companies around the world influence people’s lives and suggests that these processes should consider the objectives and values of society. Bowen (1953, p.48) posits that companies should consider the interests of ‘workers, suppliers, consumers, the immediate community and the general public’. Specifically, Bowen (1953) proposes that firms’ goals include economic progress and stability as well as the development of employees beyond wages. He suggests companies should also focus on improving working conditions, justice, freedom and the community. Since Bowen’s work was obviously ahead of its time and due to his
27
contributions to the field, he was cited as the ‘Father of Corporate Social Responsibility’ (Carroll 1999, p.270). Following Bowen’s seminal study, the concept of corporate social responsibility has continued to grow over the decades in academia and practice, and has become the subject of considerable debate. This can be traced in scholars’ arguments against or in favour of the concept (Carroll & Shabana 2010). For instance, Friedman (1970, p.211) posited that the social responsibility of a business ‘is to increase its profits’ and argued ‘few trends would so thoroughly undermine the very foundations of our free society as the acceptance by corporate officials of a social responsibility other than to make as much money for their shareholders as they possibly can’ (Friedman 1962, p.133). Similarly, Levitt (1958, p.42) affirmed that there is no way for social responsibility (sustainability) in business if it ‘does not make economic sense’. On the other hand, scholars who have argued in favour of social responsibility since the 1970s mainly link long-term economic viability to social responsibility and maintain that ‘it is in business’s long-term self-interest—enlightened self-interest—to be socially responsible’ (Carroll & Shabana 2010, p.88). Furthermore, it could be assumed that the trajectory of events, ideas and disasters, as noted earlier, from the 1960s and 1970s has driven sustainability to achieve widespread international recognition as a major policy challenge for business and governments (Moscardo 2013). Freeman’s (1984) development of stakeholder theory provides a significant contribution to the field through providing a theoretical ground wherein the new position of sustainability and sustainable business could be reinforced (Freeman 1984; Lee 2008; Moura-Leite & Padgett 2011). He defines stakeholders as ‘groups and individuals who can affect or are affected by the achievement of an organisation’s mission’ (Freeman 1984, p.52) and argues that stakeholders can force companies to incorporate social responsibility into decision-making processes (Freeman, Harrison & Wicks 2008). Similarly, Carroll and Shabana (2010, p.89) state that a ‘business should engage in CSR [corporate social responsibility] because the public strongly supports it’. Sustainability might make economic sense for a business mainly because of stakeholder pressures (Carroll & Shabana 2010).
28
Table 2.1 lists and briefly explains some major events and ideas relating to the rising trajectory of sustainability, drawing from a number of sources (S&P Dow Jones Indices, n.d.; Global Reporting Initiative 2013; International Organization for Standarization 2009; Moscardo 2013; Paul 2008; Social Accountability International n.d.; Sarkis, Zhu & Lai 2011; Wells 2011). Table 2.1 Some major events which raise awareness about sustainability
Year
Events
1962
Publication of Rachal Carson well-known book Silent spring: The book brings issues of environmental pollution to public attention, leading to the formation of the US Environmental Protection Agency (EPA) in 1970 and to the introduction of regulations impacting on industry (such as the banning of DDT production).
1972
United Nations Conference on the Human Environment in Stockholm, Sweden: It leads to the creation of the United Nations Environmental Program (UNEP) and calls for integrated and coordinated approaches to address environmental issues.
1987
Publication of the Brundtland Report (Our Common Future): Sustainable development is defined and acquires political and public momentum.
1992
United Nations Conference on the Environment and Development (the Earth Summit) in Rio de Janeiro, Brazil: Twenty-seven principles are articulated to formalise the Brundtland Report and countries are encouraged to develop a national plan for sustainable development.
1996
Appearance of ISO 14000 for environmental management: International Organization for Standardization (ISO) promotes the ISO 14000 family of international standards for environmental management for firms.
1997
1) United Nations conference on climate change in Kyoto: Many developed countries sign the Kyoto Protocol and agree to set specific targets for cutting their GHG emissions which leads the European Union to adopt an emission trading scheme in 2003 and also some other countries such as Australia to legalise a carbon tax in 2011. 2) Promotion of SA8000 for social accountability: Social Accountability International (SAI), founded by the Council on Economic Priorities (CEP), publishes SA8000 as a social accountability standard for firms, including an array of UN and International Labour Organization conventions.
29
1999
2002
2012
Launch of The Dow Jones Sustainability World Index: It aims to track the stock performance of the world’s leading companies in terms of economic, environmental and social criteria. This shows the importance of sustainability for financial sectors. Publication of the first version of the Global Reporting Initiative as a framework in line with the triple bottom line approach: The mission of the Global Reporting Initiative is to provide a comprehensive framework in order to promote economic, environmental and social aspects of sustainability in organisations worldwide. In its latest version, G4, four indicators in the economic category, twelve indicators in the environmental category, and thirty indicators in the social category are introduced. United Nations Conference on Sustainable Development (known as Rio 2012 or Rio+20) with participants from 192 countries: Rio+20 focuses on two themes: (1) how to build a green economy to achieve sustainable development and lift people out of poverty, including support for developing countries that will allow them to find a green path for development; and (2) how to improve international coordination for sustainable development by building an institutional framework.
Taking into account the issues raised by listed events in Table 2.1 and their affiliated publications, declarations and reports, it is evident that ‘sustainability principles represent a dramatic expansion of the earlier scope of corporate social responsibility’ (Wells 2013, p.7). The Global Reporting Initiative framework is a good example of this wide-ranging scope. The Global Reporting Initiative is a non-profit organisation, based in the Netherlands, which provides a framework for measuring sustainability performance indicators (Global Reporting Initiative, n.d.). This framework (the latest version is known as G4) is cited as one of the most popular sustainability reporting frameworks (Searcy 2006; Skouloudis, Evangelinos & Kourmousis 2009). Despite this fact and according to a number of studies, the Global Reporting Initiative has attracted strong criticism for its practical ineffectiveness, mainly because it includes too many indicators where the trade-offs are unclear, which in turn can lead to flawed decisionmaking (Aras & Crowther 2009; Fonseca, McAllister & Fitzpatrick 2013; Moneva, Archel & Correa 2006; Morhardt 2009). For example, water, energy and the interplay between them can be seen as social and economic issues; however, they are categorised under the environmental dimension in the Global Reporting Initiative guideline (Jägerskog et al. 30
2014). Its underlying factors are shown in Table 2.2, including four aspects in the economic category, twelve aspects in the environmental category, and thirty aspects in the social category. Table 2.2 Categories and aspects of the Global Reporting Initiative G4 sustainability framework (Global Reporting Initiative 2013) Category Aspects
Economic
Environmental
• • • •
• • • • • • • • • • • •
Category
Social Labour Practices and Decent Work
Subcategories Aspects
Economic Performance Market Presence Indirect Economic Impacts Procurement Practices
• Employment • Labour/Management Relations • Occupational Health and Safety • Training and Education • Diversity and Equal Opportunity • Equal Remuneration for Women and Men • Supplier Assessment for Labour Practices • Labour Practices Grievance Mechanisms
Materials Energy Water Biodiversity Emissions Effluents and Waste Products and Services Compliance Transport Overall Supplier Environmental Assessment Environmental Grievance Mechanisms
Human Rights
Society
• Investment • Nondiscrimination • Freedom of Association and Collective Bargaining • Child labour • Forced or Compulsory Labour • Security Practices • Indigenous Rights • Assessment • Supplier Human Rights Assessment • Human Rights Grievance Mechanisms
• Local Communities • Anti-corruption • Public Policy • Anticompetitive Behaviour • Compliance • Supplier Assessment for Impacts on Society • Grievance Mechanisms for Impacts on Society
Product Responsibility • Customer Health and Safety • Product and Service Labelling • Marketing Communications • Customer Privacy • Compliance
Companies need to meet some economic, social and environmental requirements in order to claim to be socially responsible or sustainable (Montiel 2008). However, there may be trade-offs between social, environmental and economic objectives according to Dahlsrud (2008). Keeping a balance between the three dimensions might be costly yet increasingly 31
essential for firms. Carroll and Shabana (2010) assert that this may require organisations to sacrifice some profits. Sustainability is linked with supply chain management. Firstly, it can be argued that sustainable development requires integrated and collaborative approaches and integration and collaboration have been conceptualised and highlighted in the supply chain management literature. Secondly, the importance and significance of supply chain management and its contribution towards sustainability are acknowledged in the corporate social responsibility literature and also in the contemporary sustainability reporting frameworks such as the Global Reporting Initiative, as shown in Table 2.2 (Moscardo 2013; Wells 2011). Therefore, the area of sustainable supply chain management has emerged to explore the linkage between supply chain management and sustainability (Carter & Easton 2011). This emerging area and its main aspects are discussed in the next section.
2.4
Sustainable supply chain management
Sustainable supply chain management has evolved as an extension of the traditional supply chain management, by adding environmental and social aspects (Wittstruck & Teuteberg 2012). Scholars have proposed various definitions of this extension. Ahi and Searcy (2013) identified and analysed some published definitions of sustainable supply chain management in the literature, as summarised and presented in Table 2.3. Table 2.3 Definitions of sustainable supply chain management Definition
Author(s)
Ahi and Searcy (2013, p.17)
‘…the creation of coordinated supply chains through the voluntary integration of economic, environmental, and social considerations with key inter-organizational business systems designed to efficiently and effectively manage the material, information, and capital flows associated with the procurement, production, and distribution of products or services in order to meet stakeholder requirements and improve the profitability, competitiveness, and resilience of the organization over the short- and long-term.’
Hassini, Surti and Searcy
‘…the management of supply chain operations, resources, information, and funds in order to maximize the supply chain profitability while at the same time
32
(2012, p.2)
minimizing the environmental impacts and maximizing the social well-being.’
Wittstruck and Teuteberg (2012, p.142)
‘…an extension to the traditional concept of Supply Chain Management by adding environmental and social/ethical aspects.’
Wolf (2011, p.223)
‘…the degree to which a manufacturer strategically collaborates with its supply chain partners and collaboratively manages intra- and inter-organization processes for sustainability.’
Pagell and Wu (2009, p.38)
‘…the specific managerial actions that are taken to make the supply chain more sustainable with an end goal of creating a truly sustainable chain.’
Font et al.(2008, p.260)
‘…adding sustainability to existing supply chain management processes, to consider environmental, social and economic impacts of business activities.’
Ciliberti Pontrandolfo and Scozzi (2008, p.1580)
‘…the management of supply chains where all the three dimensions of sustainability, namely the economic, environmental, and social ones, are taken into account.’
Seuring and Muller(2008, p.1700)
‘…the management of material, information and capital flows as well as cooperation among companies along the supply chain while taking goals from all three dimensions of sustainable development, i.e., economic, environmental and social, into account which are derived from customer and stakeholder requirements.’
Carter and Rogers(2008, p.368)
‘…the strategic, transparent integration and achievement of an organization’s social, environmental, and economic goals in the systemic coordination of key inter-organizational business processes for improving the long-term economic performance of the individual company and its supply chains.’
Jorgensen and Knudsen(2006, p.450)
‘…the means by which companies manage their social responsibilities across dislocated production processes spanning organizational and geographical boundaries.’
Among these definitions, two of them have been cited widely in the literature (Ahi & Searcy 2013). According to Ahi and Searcy (2013), the first definition is proposed by Carter and Rogers (2008) who define sustainable supply chain management based on the triple bottom line approach, and the second definition is proposed by Seuring and Müller (2008) who advocate that the social and environmental aspects of a supply chain need to be enhanced by supply chain members, while the overall economic feasibility is maintained. It appears that they view membership in a sustainable supply chain as a business opportunity for all involved partners and merit consideration of the economic, social and environmental 33
performance indicators when a focal company aims to design its supply chain (Seuring & Müller 2008). Bearing these two definitions in mind, this thesis adopts the definition by Hassini, Surti and Searcy (2012, p.70): …the management of supply chain operations, resources, information, and funds in order to maximize the supply chain profitability while at the same time minimizing the environmental impacts and maximizing the social well-being.
This definition implies the multi-objective nature of decisions in sustainable supply chain management, in line with one of the central themes of the recent United Nations Conference on Sustainable Development (known as Rio+20) that is the balanced consideration and integration of economic, social and environmental objectives in decisionmaking (United Nations 2012). This emphasises the fact that both researchers and practitioners should take a holistic integrated approach in decision-making to improve sustainability. Growing environmental concerns such as global warming, and social issues such as human rights and safety, have forced many companies to include a wider range of objectives rather than just economic-related goals (Gupta & Palsule-Desai 2011). In addition, the pressure from stakeholders, including customers, governments, non-governmental organisations, public authorities and trade unions have compelled companies to manage more sustainable supply chains (Andersen & Skjoett-Larsen 2009). Moreover, the introduction of various environmental legislation (e.g. carbon tax) as well as standards (e.g. ISO 14000 series or Social Accountability 8000) and reporting frameworks should force companies at the chain level to move beyond the single economic objective and incorporate a multi-objective approach in order to balance all three dimensions of sustainability. Balancing multiple objectives could often be conflicting due to trade-offs between the three dimensions of sustainability (Winter & Knemeyer 2013). Taking the supplier selection into account, for instance, suppliers may be based on different economic-, social- and environmental-related performance indicators. As a result, many focal companies are faced with the dilemma of choosing between suppliers. The aforementioned point highlights that sustainability issues could significantly impact decisions at the supply chain level. While some social or environmental initiatives may 34
provide financial benefits (e.g. managing energy usage efficiently), there are other factors that could undermine the financial performance directly or indirectly (Ross, Parker & Benavides 2012). Despite this fact, the implications of social and environmental aspects in decision-making are now more essential than before. That is why many senior managers state that sustainability issues will be critical for their businesses’ future success, as mentioned in the report published by the United Nations Global Compact and Business for Social Responsibility (2010). Focal companies seek to implement sustainability at the supply chain level as they increasingly form strategic alliances with their supply chain partners (Tan, Lyman & Wisner 2002). The area of sustainable supply chain management has been researched and published, particularly in the last decade, to provide insights for practitioners to enhance sustainability. This is described further in the following sections depicting the linkage between sustainability and supply chain management. 2.4.1 The economic aspect The economic aspect of supply chain management has been widely researched in the literature (Hutchins & Sutherland 2008; Seuring 2013). It can be seen that this aspect has attracted several researchers to address the long-term economic stability of a supply chain (Pullman, Maloni & Carter 2009). Some of the relatively unexplored areas have been examined by some scholars, such as the governance structure of multi-tier supply chains (Mena, Humphries & Choi 2013). In order to improve the economic aspect, supply chain management should also consider its links with the environment and society because it deals with a large number of activities from extracting raw materials to production, packaging, transporting and recycling of products, all of which can have a profound impact on both the environment and society (Pagell & Wu 2009; Wisner, Tan & Leong 2008). In other words, economic sustainability means achieving economic objectives whilst protecting the environment and safeguarding society (Yusuf et al. 2013). Some studies in the literature have not addressed the economic aspect directly, and only examined how environmental or social dimensions can be embedded into supply chain management (Hsu & Hu 2009; Hutchins & Sutherland 2008; Sonesson & Berlin 2003). 35
However, a number of researchers have investigated the linkage between the three dimensions and maintained that implementation of environmental and/or social initiatives could provide economic advantages such as: •
cost minimisation due to recycling, redesigning to re-use, remanufacturing and reducing packaging waste (Carter & Rogers 2008; Christmann 2000; Mollenkopf et al. 2005; Rosenau et al. 1996; Van Hoof & Lyon 2013; Zailani et al. 2012);
•
economic performance improvement due to environmental purchasing and sustainable packaging (Carter 2005; Carter & Jennings 2004; Zailani et al. 2012);
•
health and safety cost reduction as a result of safer warehousing, transportation and manufacturing (Carter & Rogers 2008; Carter & Stevens 2007; Zailani et al. 2012);
•
lower recruitment and labour turnover costs associated with better working conditions (Carter & Rogers 2008);
•
improving product quality and lead times resulting from the implementation of environmental management standards, such as ISO 14000 (Carter & Rogers 2008; Hanson, Melnyk & Calantone 2004; Montabon et al. 2000);
•
difficult to replicate competitive advantage—if companies proactively implement sustainability initiatives, they are not compelled by future regulations to change their supply chain structure, leading to competitive advantages for focal companies and their suppliers (Carter & Rogers 2008);
•
enhancing reputation through proactively engaging in sustainability initiatives (Ellen, Webb & Mohr 2006; Klassen & McLaughlin 1996; Roberts 2003);
•
improving focal firms’ marketing performance due to greater customer satisfaction, and increasing supplier’s capabilities to innovate and gain trust in suppliers as a result of engaging in sustainability initiatives (Ageron, Gunasekaran & Spalanzani 2012; Bai & Sarkis 2010a; Rao & Holt 2005); and,
•
improvement in key performance measures such as quality of service and products, market share, customer loyalty and profit (Yusuf et al. 2013).
Following the above studies, Golicic and Smith (2013) conducted a meta-analysis of over twenty years of research on environmental supply chain management. Underpinned by a theoretical lens (i.e. natural resource-based view), their research intended to determine 36
whether the overall effect of environmental practices at the supply chain level on focal firm performance is positive. Although the meta-analysis did not include the social aspect, their empirical results show that the link between environmental supply chain practices and accounting-, operational- and market-based forms of firm performance is ‘positive and significant’ (Golicic & Smith 2013, p.78). They concluded that these results would encourage firms to go beyond regulatory requirements and adopt sustainability practices with respect to associated positive financial outcomes (Golicic & Smith 2013). Similarly, based on a survey study, Wong (2013) suggests that the environmental management capabilities of firms at the supply chain level improve financial performance through reducing resource consumption, waste and disposal as well as capturing additional market share and profit. Conversely, Pullman, Maloni, and Carter (2009) found that the financial benefits of social and environmental sustainability initiatives may be difficult to recognise, calling for future research to help practitioners understand and clarify trade-offs between the three dimensions. Linton, Klassen and Jayaraman (2007) as well as Corbett and Klassen (2006) argued that sustainability initiatives would add an additional level of complexity to existing supply chains at both strategic (e.g. supply chain design) and operational levels which may need short-term investment. In fact, it could be drawn from some recent studies that the implementation of environmental or social initiatives might ‘increase costs, especially in the short term’ (Wu & Pagell 2011, p.577), which in turn highlights the importance of analysing trade-offs in decision-making when environmental and social aspects are to be considered along with the traditional economic aspect (Epstein & Yuthas 2012; Linton, Klassen & Jayaraman 2007; Pullman, Maloni & Carter 2009; Ross, Parker & Benavides 2012). In this way, decision-makers in focal companies gain a clear understanding of the cost-benefit analysis of environmental and social initiatives in their short-term and longterm strategic plans. The next section provides an overview of how the literature has dealt with the environmental and social aspects, and discusses some related findings. 2.4.2 The environmental aspect According to Carter and Easton (2011), the environmental aspect of supply chain management has been a prominent research area over the past two decades. Stemming from 37
both the environmental management and supply chain management literature, this area has been referred to as ‘green supply chain management’ and aims to integrate environmental issues with supply chain processes and functions such as product design, supplier selection, operations, transportation as well as the end-of-life management of used products (Srivastava 2007). Some of the important environmental concerns investigated in the supply chain management literature include GHG emissions (Paksoy, Bektaş & Özceylan 2011), resource depletion (Yusuf et al. 2013), waste generation (Tsai & Hung 2009), hazardous substances in products (Hsu & Hu 2009), energy consumption (Cholette & Venkat 2009) and water consumption (Brent 2005). Of these concerns stated, the issue of GHG emissions, CO2 in particular, has been highlighted in the literature as the most prominent because of the hazardous consequences on ecosystems and human health. Concerns about these hazards has led to the introduction of several regulations worldwide (Gupta & Palsule-Desai 2011; Paksoy, Bektaş & Özceylan 2011). According to the report published by the World Meteorological Organization (2013, p.3), the world experienced significant climate extremes during the 2000s, which ‘was the warmest decade on record since modern meteorological records began around the year 1850’. More national temperature records were reported broken than in any previous decade. Many factors impact upon the Earth’s climate which can be classified into natural and anthropogenic (human-induced) factors. The GHGs are the main cause of anthropogenic climate change of which CO2 is cited as the most significant (World Meterological Organization 2013). In their report ‘Climate Change 2007: The Physical Science Basis’, the Intergovernmental Panel on Climate Change (2007, p.10) stated that ‘most of the observed warming over the last 50 years is likely to have been due to the increase in greenhouse gas concentrations’. The possible effects of global warming are ice melting in the Arctic and the Antarctic, ocean warming and sea level rising, ozone layer depletion, temperature extremes, weather patterns changing and biodiversity issues, which pose serious threats to the economic and social well-being of society (Spoolman & Miller 2008). Consequently, the Kyoto Protocol came into existence in 2005 as the predominant international agreement to stabilise GHG emissions. Several countries then introduced carbon-related tax regulations to comply with 38
the Kyoto protocol through monitoring and controlling emitted GHGs such as CO2 (Gupta & Palsule-Desai 2011). Hence the level of CO2 emissions is one of the most frequently mentioned environmental performance indicators in the sustainable supply chain management literature (Benjaafar, Yanzhi & Daskin 2013; Elhedhli & Merrick 2012; Ramudhin, Chaabane & Paquet 2010; Seuring 2013). Apart from regulatory requirements, focal companies are driven to ‘go green’ mainly by two stakeholder groups: customers and non-governmental organisations. Firstly, there is empirical evidence that customers’ buying behaviour is increasingly influenced by the conditions under which goods are produced (Ageron, Gunasekaran & Spalanzani 2012; González-Benito & González-Benito 2006; Wolf 2011). In fact, more and more customers desire merchandise which has been produced in an environmentally friendly condition (Collins, Steg & Koning 2007) and are even willing to pay more for these products in an attempt to build a ‘green society’ (Ageron, Gunasekaran & Spalanzani 2012). Secondly, non-governmental organisations are important for a number of reasons. For instance, they promote environmental awareness among consumers and encourage them to improve their quality of life by changing their buying behaviour (Farahani, Asgari & Davarzani 2009), thereby exerting pressure on focal companies to change their ‘business as usual’ scenarios (Wu & Pagell 2011). Instead of following a reactive approach in green supply chain management, practitioners should adopt a value-seeking mindset and integrate environmental initiatives with their processes and functions at the supply chain level (Golicic & Smith 2013; Pagell & Wu 2009; Srivastava 2007). In this way, greening a supply chain could be seen as a potential source of competitive advantage (Christmann 2000; Markley & Davis 2007; Van Hoek 1999). To integrate environmental practices into supply chain management, two main streams can be distinguished in the literature, namely green operations and green design, according to a comprehensive study by Srivastava (2007). He extensively reviewed the green supply chain management literature from 1990 to 2007 and classified studies based on the context, as depicted in Figure 2.4. Although this classification does not cover some other studies such as those addressing the important issue of the green supplier selection (Cousins, Lamming & Bowen 2004; Handfield et al. 2002; Lu, Wu & Kuo 2007; Rao 39
2002, 2005; Rao & Holt 2005), it is used to briefly address supply chain managementrelated processes or functions in green or environmentally sustainable supply chain management.
Green Supply Chain Management
Network Design and Reverse Logistics
Manufacturing and Remanufacturing
Green Operations
Waste Management
Environmentally Concious Design Green Design Environmental Life Cycle Anaysis
Figure 2.4 The classification of the green supply chain management literature, adapted from Srivastava (2007)
According to Srivastava (2007), green operations seek to address issues such as remanufacturing, designing the supply chain to include reverse logistics, and waste management. It can be argued that these challenges are closely linked since effective remanufacturing may require well-designed reverse logistics and waste management systems. Jayaraman, Guide and Srivastava (1999) define remanufacturing as the transformation of used products into products which satisfy exactly the same standards as new products and argue that it is profitable and environmentally conscious. However, they affirm that a focal firm must simultaneously consider the traditional forward and also the
40
reverse flow of goods (i.e. a closed-loop system) for a remanufacturing process to function effectively. The flow of materials and products from suppliers to consumers is called ‘forward logistics/supply chain’ in the literature. Issues concerning the after-use process such as recovery, re-use, repair, disassembly, remanufacturing and recycling combined with all associated facilities (e.g. collection and inspection centres), transportation and operations lie in ‘reverse logistics/supply chain’. The integration of forward and reverse supply chain is addressed as a ‘closed-loop supply chain’ (Paksoy, Bektaş & Özceylan 2011). Since these issues may affect supply chain configuration and network design, Ilgin and Gupta (2010) note that most of the studies on reverse and closed-loop supply chains focus on supply chain configuration. There are also some studies about other problems in this context, such as uncertainty with respect to the quantity, timing, condition and selection criteria of used products and the end-of-life process, selection and evaluation of suppliers, performance measurement and marketing (Ilgin & Gupta 2010). Despite these efforts, Presley, Meade and Sarkis (2007) and Nikolaou, Evangelinos and Allan (2013) note that studies on reverse and closed-loop supply chains mainly consider economic and environmental aspects. The initial product design directly affects how a product can be recovered, re-used, repaired, disassembled, remanufactured, recycled or disposed of (Linton, Klassen & Jayaraman 2007). For example, if an electronic product is designed with a high level of the lead content, there will be complications for its recycling and disposal due to the toxicity of lead and associated restricted environmental regulations (Linton, Klassen & Jayaraman 2007). Green design is concerned with the integration of environmental considerations and the product’s design process which traditionally considers some objectives including cost, functionality and manufacturability (Ilgin & Gupta 2010). It aims to increase understanding of how the product design may impact the environment, which could improve a product’s environmental compatibility during its life cycle throughout its supply chain (Srivastava 2007). Two approaches have been mostly used to address green design, namely environmentally conscious design and environmental life cycle analysis/assessment (Farahani, Asgari & Davarzani 2009; Ilgin & Gupta 2010; Srivastava 2007). 41
According to Farahani, Asgari and Davarzani (2009), environmentally conscious design aims to improve a product’s ease of disassembly, re-use, recycle and recovery, in addition to its production process enhancement. It also involves minimising the use of scarce resources or hazardous materials in order to be in compliance with several legislations and regulations. This consideration will largely influence the relationship between a focal company and its upstream suppliers that supply components or raw materials. Several studies related to environmentally conscious design exist in the literature which cover design for recycling (Masanet & Horvath 2007), design for disassembly (Krikke, Bloemhof-Ruwaard & Van Wassenhove 2003), design for waste minimisation (He et al. 2004), design under legislation and regulation (Fleischmann et al. 2001), customer-oriented environmentally conscious design (Madu, Kuei & Madu 2002), and green material selection (Ilgin & Gupta 2010). The second approach, environmental life cycle assessment, uses an analytical tool to quantify and assess the environmental impacts of a product’s life cycle, i.e. from extraction and processing of raw materials, production, transportation and use, to the product’s afteruse treatment and waste management through recycling, remanufacturing and final disposal (Finnveden et al. 2009). Practitioners can optimise resource usage and, more importantly, identify those stages of the product’s life cycle where major improvements can be made to protect the environment. Guinée et al. (2010) present a useful overview of the historical evolution of this approach. They state that the first related scientific publication was released in 1993 and since then, according to Finnveden et al. (2009), the interest in environmental life cycle assessment has grown rapidly resulting in its broad application in practice. In line with these studies, the International Organization for Standardization (ISO) developed two standards: •
ISO 14040: Environmental management – Life cycle assessment: Principles and framework (International Organization for Standarization 2006a)
•
ISO 14044: Environmental management – Life cycle assessment: Requirements and guidelines (International Organization for Standarization 2006b).
These two standards categorise the life cycle assessment framework into four phases: 1) Goal and Scope Definition, 2) Life Cycle Inventory Analysis, 3) Life Cycle Impact 42
Assessment and 4) Interpretation (International Organization for Standarization 2006a). The first phase involves establishing system boundaries, functional units and study objectives. The second phase includes collecting data and estimating inputs (e.g. resources and energy) and outputs (pollutants released into the land, water and air) throughout a product’s life cycle. It should be noted that the term ‘product’ refers to both goods and services in these two standards (International Organization for Standarization 2006a). In the third phase, various potential impacts of inventory data (i.e. both inputs and outputs) on human health and the environment are evaluated and finally the results are interpreted as a set of recommendations for decision-makers. There is a large body of research which incorporates environmental life cycle analysis in supply chain management (Bojarski et al. 2009; Brent 2005; Guillén-Gosálbez & Grossmann 2010; Hagelaar & van der Vorst 2001; Hugo & Pistikopoulos 2005; Matos & Hall 2007; Puigjaner & Guillén-Gosálbez 2008; Ruiz-Femenia, Caballero & Jiménez 2011; Sonesson & Berlin 2003). For instance, Hugo and Pistikopoulos (2005) used life cycle assessment criteria in supply chain design and planning. Further, Bojarski et al. (2009) extended this study, analysed different configurations of a supply chain in the chemical industry expanded in Germany, Netherland, France, Spain and Portugal, and discussed the trade-off between the most profitable and the best environmentally-friendly supply chain configurations. The abovementioned studies have addressed the environmental and economic dimensions. However, according to Pagell and Wu (2009), a truly sustainable supply chain considers all three dimensions. Based on an empirical study, Matos and Hall (2007) advocated that having a broad integrated approach to examine interactions amongst environmental, economic and social dimensions is better than examining only one of them (e.g. the environmental aspect) in detail. They asserted that practitioners usually consider economic and environmental criteria in decision-making and do not recognise the importance of social factors or ‘do not know how to deal with social factors’ (Matos & Hall 2007, p.1090). It can be inferred that their findings and arguments highlight two existing gaps in the literature: how practitioners can deal with social criteria and, at the same time, how they can follow a broad integrated approach to design and manage sustainable supply 43
chains despite its inevitable limitations (Carter & Easton 2011; Pullman, Maloni & Carter 2009; Seuring 2013; Seuring & Müller 2008; Winter & Knemeyer 2013; Wu & Pagell 2011). Bearing this in mind, it should be acknowledged that there have been relatively few studies that address the social aspect. This is discussed in the following section. 2.4.3 The social aspect Sustainability dimensions, including the social aspect, require focal firms to consider issues beyond organisational boundaries (Brammer, Hoejmose & Millington 2011; Zailani et al. 2012). Klassen and Vereecke (2012, p.103) define social issues in supply chain management as ‘product- or process-related aspects of operations that affect human safety, welfare and community development’. The literature proposes various yet similar explanations for social issues in the supply chain. Social responsibility-related standards, codes of conduct, and reporting frameworks consider to some extent similar social criteria (Global Reporting Initiative 2013; Social Accountability International n.d.). For example, Pullman, Maloni and Carter (2009) described social sustainability practices as ensuring quality of life, encouraging diversity and community connectedness, and providing equitable opportunities and democratic processes along with an open and accountable governance structure. Empirical evidence has shown that many focal companies have struggled to implement social sustainability practices in their supply chains (Andersen & Skjoett-Larsen 2009; Klassen & Vereecke 2012). It could be argued that the social dimension is still not directly incorporated into supply chain performance measures of many organisations (Cetinkaya 2011). One reason lies in the fact that implementing social initiatives (e.g. codes of conduct) across global supply chains is a complicated undertaking because there is a large number of supply chain members for many multinational companies (Mamic 2005). The empirical study by Mamic (2005) highlighted this problem across sports footwear, apparel and retail sectors in which there may be more than 5000 suppliers, and suggested that focal companies should prioritise their suppliers according to their importance and the nature of the relationship with them. In addition, the research found that the influence of a focal company is the major determinant of adopting the social initiatives at the supplier level (Mamic 2005). 44
Ansett (2007) stated that only a few organisations have acknowledged the strategic advantage of being socially responsible at the supply chain level even though by doing so they may be rewarded with enhanced credibility and reputation, license to operate, risk mitigation and strategic innovation. It could be inferred from the available literature that the implementation of social sustainability (or corporate social responsibility) practices at the supply chain level would face two interlinked challenges: firstly, how focal companies and their executives can make a long-term commitment to social sustainability (Andersen & Skjoett-Larsen 2009); and secondly, how committed focal companies are able to develop decision-making processes and business models underlying social issues so as to design and manage sustainable supply chains (Wu & Pagell 2011). Notwithstanding these challenges, focal companies in today’s business environment are increasingly under pressure to minimise the number of incidents in terms of the social aspect at the supply chain level (Green 2013; Greenhouse 2013a; Manik & Yardley 2013) which could harm their reputation. It is now believed that reputation is ‘a valuable corporate asset, hard to build, yet easy to diminish’ (Roberts 2003, p.168). With regard to the aforementioned challenges, some scholars have tried to examine the linkage between social sustainability initiatives and financial performance outcomes in order to encourage focal firms to make a long-term commitment to social initiatives (Carter 2005; Pullman, Maloni & Carter 2009; Roberts 2003; Wu & Pagell 2011; Yusuf et al. 2013). While these efforts have addressed the first challenge (as noted earlier) and may encourage organisations to create ‘a business case for sustainability’ (Schaltegger, Bennett & Burritt 2006; Schaltegger, Lüdeke-Freund & Hansen 2011), few studies in recent years have taken one step further and examined how social concepts can be applied to supply chains in order to provide insights for practitioners. The study by Carter (2000) is among the first to practically analyse social issues in the global (i.e. international) supplier management. He advocated that setting up a mechanism for formally communicating codes of conduct and reporting violations of codes would dissuade supply chain members from engaging in unethical behaviours. His findings show that engaging in socially sustainable practices could create a win-win situation in supply chains, providing a secure business opportunity for all partners involved. Carter and 45
Jennings (2002) elaborated the concept of ‘logistics social responsibility’ and proposed a framework in which social issues such as safety, human rights, quality of life, ethics, diversity, community and philanthropy are considered. The above studies are limited to the purchasing function and supplier management, and they do not address other supply chain members such as distributors. This limitation exists in more recent publications as well—those who address only suppliers’ audit by means of available social accountability standards and initiatives (Andersen & Skjoett-Larsen 2009; Ciliberti, Pontrandolfo & Scozzi 2008; Courville 2003; Koplin, Seuring & Mesterharm 2007; Welford & Frost 2006). This trend of studies might be justifiable. Given the power dynamics of the relationships between a focal company and its upstream and downstream supply chain members, it could be argued that the upstream supply chain members are more likely to be governed by the focal company than the downstream members. For instance, a focal company might easily conduct an environmental audit of the upstream suppliers while this could turn out to be highly problematic in the downstream retailers (Abbasi & Nilsson 2012). Furthermore, it is evident that the risks associated with sustainability misconducts have been mostly created by upstream chain members rather than downstream networks in recent years (Manik & Yardley 2013). That is why researchers have followed the problemoriented approach to focus more on upstream supply networks; nevertheless, this highlights an associated gap in the literature, which is addressing the important issue of making downstream chain members more sustainable. This may create wholly sustainable supply chains for focal firms, if it is coupled with a successful extension of governance structure to upstream supply chain members. The extant literature also suggests setting social standards and principles, codes of conduct and auditing processes to help focal firms ensure socially acceptable practices amongst supply chain members, especially those with extended supply chains into developing countries (Awaysheh & Klassen 2010). Gopalakrishnan et al. (2012) stated that the publicity and importance of social standards such as the Social Accountability 8000 should be equal to that of environmental standards such as the ISO 14000 series. These standards set basic social responsibility requirements for supply chain members. For instance, the 46
Social Accountability 8000 explicitly examines nine areas including child labour, forced labour, health and safety, freedom of association and collective bargaining, discrimination, disciplinary practices, working hours, remuneration, and related management systems (Social Accountability International n.d.). By means of these standards, a focal company may administer social audits on its supply chain members in order to assess their social performance (Awaysheh & Klassen 2010). Based on the audit outcomes, a focal company can identify those areas or links in the chain which need improvement so as to enhance the overall chain performance (Klassen & Vachon 2003). The adoption of codes of conduct could help focal companies to dictate specific guidelines in their interaction with suppliers (Mamic 2005), which in turn can be used in the worst cases to terminate contracts with those suppliers that violate the codes of conduct (Awaysheh & Klassen 2010). According to Andersen and Skjoett-Larsen (2009, p.78), a code of conduct is ‘a document stating a number of social and environmental standards and principles that a firm’s suppliers are expected to fulfil’. Principles stated in codes of conduct are often derived from a focal firm’s values, local legislation, standards (e.g. Social Accountability 8000), sustainability reporting frameworks (e.g. the Global Reporting Initiative) or international conventions such as the International Labour Organization Declaration on Fundamental Principles and Rights at Work (Andersen & Skjoett-Larsen 2009). Although codes of conduct have been widely used in contracts between focal firms and their suppliers (Welford & Frost 2006), Andersen and Skjoett-Larsen (2009) observed that they have not been truly implemented in many cases across supply chains. It could be argued that this issue exists because the social sustainability has not been successfully embedded into business models or decision-making processes. There are only very few studies in the literature which attempt to address this problem such as the study by Hutchins and Sutherland (2008). They proposed a framework to examine how some social sustainability indicators (labour equity, health, safety and philanthropy) can be quantified and incorporated in a supply chain decision-making process. They provided a hypothetical example to demonstrate how the framework can be applied in practice. While this approach to test a framework seems to be justified in decision-making related research (Bai, Sarkis 47
& Wei 2010; Benjaafar, Yanzhi & Daskin 2013; Chaabane, Ramudhin & Paquet 2012; Nagurney & Yu 2012; Sarkis 2003), their study did not include environmental or economic indicators. The social, environmental and economic dimensions should be analysed through integrated approaches. Otherwise, it tends to be difficult for decision-makers to examine the linkages between the three sustainability dimensions and balance strategic priorities. This identifies a major gap in the literature (Brandenburg et al. 2014; Walker et al. 2014), and as Wu and Pagell (2011, p.577) state: ‘existing research has not addressed the business models and decision-making processes underlying sustainable supply chain management’. It can be seen that this gap has been frequently highlighted by several studies including recent comprehensive literature review papers (Brandenburg et al. 2014; Carter & Easton 2011; Klassen & Vereecke 2012; Pullman, Maloni & Carter 2009; Seuring 2013; Seuring & Müller 2008; Winter & Knemeyer 2013). This gap is also reinforced through a systematic literature review method presented in the next section. 2.4.4 Analysis of the literature: Some research gaps A literature review is a systematic and explicit approach to identify and evaluate the existing body of recorded documents (Fink 2010). For this study, an extensive literature review was conducted to categorise and analyse peer-reviewed papers in order to identify research gaps and significant findings in the area of sustainable supply chain management, as summarised in the previous sections. A systematic review provides a high level of transparency and clearly communicates how a review was conducted (Denyer & Neely 2004), hence it reduces researcher bias concerning the inclusion or exclusion of articles (Carter & Easton 2011). Following the studies by Soni and Kodali (2011) and Winter and Knemeyer (2013), the literature review method employed in this study encompassed a fivestep process which included a time horizon of review, database selection, article selection, article classification and analysis of articles. Figure 2.5 shows the schematic overview of the literature review method.
48
•Time horizon for article selection: 1994‒2013 Step 1
•Database selection: Emerald, Elsevier, Springer and Wiley Step 2
•Article selection: keywords, delimitations Step 3
•Article classification: sustainability dimension and methodology used Step 4
•Analysis: identifying research gaps and discussing significant findings Step 5
Figure 2.5 The literature review method, adapted from Soni and Kodali (2011) and Winter and Knemeyer (2013)
Step1 – time horizon for article selection: The year 1994 was chosen as the starting point of the analysis time horizon, because sustainable supply chain management started to gain popularity from that point (Seuring & Müller 2008; Winter & Knemeyer 2013). Therefore, the data collection period for the review spans between 1994 and 2013, covering a 19-year period.
Step2 – Database selection: Relevant papers were obtained using major publishers’ databases including Emerald, Elsevier, Springer and Wiley
49
Step3 – Article selection: A set of keywords were used to find relevant articles. This included sustainable/green/carbon/ emission/ environmental/social/ corporate social responsibility combined with supply chain/logistics/ value chain/ supplier and networks. The search process based on these keywords resulted in the first article database, encompassing 378 publications. Defining clear boundaries is particularly important in a literature review to delimit the research (Seuring & Müller 2008). To narrow the scope of the literature review, 183 out of 378 publications were eliminated, mostly owing to the criteria set below: •
only peer-reviewed articles were included in this study and other publications so, for example, guest-editorial introductions to special issues were eliminated;
•
this research focuses on the forward supply chain management so publications related to reverse and closed-loop supply chain/logistics were excluded. While the importance of reverse and closed-loop supply chains is acknowledged, this delimitation was set to meet the research time constraint. However, there are comprehensive review papers addressing these issues (Srivastava 2007);
•
those publications on the environmental or social issues that did not address supply chain management functions or practices were eliminated.
Based on the above criteria, 195 peer-reviewed papers created the final database for the literature review.
Step4 – Article classification: All 195 papers were classified according to research methods adopted and the sustainability dimensions incorporated. Four research methods were differentiated including theoretical papers, case studies, surveys and modelling papers. Each paper was also classified according to which sustainability dimension it addressed. This could be social (i.e. if only social issues were considered), environmental (i.e. if only environmental aspects were covered) or social+environmental (i.e. if both dimensions were addressed). The economic
50
dimension was not emphasised because it might be more insightful to see how the traditional economic dimension of supply chain management has been extended to cover social and/or environmental dimensions in order to form the sustainable supply chain management literature.
Step5 – Analysis: The final step attempted to identify research gaps through descriptive analysis of the literature so as to reinforce the discussion provided in the previous section, which in turn justifies the research questions for this study on sustainable supply chain management. The analyses are illustrated in Figures 2.6 and 2.7. Figure 2.6 shows that a large number of articles (i.e. 132 papers or 70%) focus only on the environmental aspect of supply chain management while 17 papers (less than 9%) address the social issues relating to managing supply chains. Forty-six papers have addressed both dimensions and are coded as social+environmental.
132
Environment
17
Social
46
Environment+Social 0
20
40
60
80
100
120
140
Figure 2.6 Sustainability dimensions addressed in the reviewed papers (out of 195 papers)
Figure 2.7 shows a gap analysis of the literature through presenting the distribution of reviewed papers based on their methods as well as social and/or environmental dimensions. There is a clear deficit in the literature overall on how social issues can be captured and 51
incorporated in decision-making processes related to supply chain management (see modelling papers category in the bar chart), where only two papers (i.e. 1% ) address the social dimension out of 195 reviewed papers: the study by Hutchins and Sutherland (2008) and Erol, Sencer and Sari (2011). This gap has been highlighted by various scholars in operations and supply chain management (Brandenburg et al. 2014; Carter & Easton 2011; Eskandarpour et al. 2015; Seuring 2013; Seuring & Müller 2008; Walker et al. 2014; Winter & Knemeyer 2013).
Modelling papers
Surveys Environmental Social Environmental+Social
Case studies
Theoretical papers
0%
20%
40%
60%
80%
100%
Figure 2.7 Distribution of reviewed papers based on their methods as well as social and/or environmental dimensions
2.5
Summary and conclusion
The literature review presented in this chapter has uncovered some major gaps in the interplay between sustainability and supply chain management. While sustainability is the consideration of social, environmental and economic dimensions concurrently (Hassini, Surti & Searcy 2012), the findings show that the literature has predominantly dealt with one or two dimensions in isolation. Furthermore, sustainability has been misinterpreted as 52
green or environmental practices by many scholars with the general oversight of the social dimension of sustainability. These findings are also supported by several researchers such as Carter and Easton (2011), Ashby, Leat and Hudson-Smith (2012), and by literature review studies conducted by Hassini, Surti and Searcy (2012), Seuring (2013), Brandenburg et al. (2014) and Walker et al. (2014). Seuring and Müller (2008) conducted a comprehensive review of the literature spanning thirteen years from 1994 to 2007 and identified that less than 17% of published articles have really addressed and discussed all three dimensions; thereby resulting in an appeal for future research to firstly analyse the social aspect and ultimately adopt an integrated approach to investigate the interconnections among these dimensions. There is a need for more research to address this gap (Seuring 2013; Winter & Knemeyer 2013), particularly in analytical studies where scholars struggle to simultaneously analyse a combination of performance measures (Winter & Knemeyer 2013). Therefore, this research aims to bridge this frequently mentioned gap by providing insights for practitioners on how to design sustainable supply chains. This direction is in line with the primary focus of the recent United Nations Conference on Sustainable Development: a need for ‘integration and a balanced consideration of social, economic and environmental goals and objectives in both public and private decision-making’ (United Nations 2012). The next chapter develops a framework to incorporate economic, environmental and social indicators into supply chain design, which is one of the key elements of supply chain management (Lambert 2008). This may assist decision-makers in various industries in designing sustainable supply chains (Varsei et al. 2014).
53
3
3.1
Addressing the gap: A framework for sustainable supply chain design
Introduction
For designing sustainable supply chains, all three dimensions of sustainability should be incorporated in the decision-making process and the associated analytical models. However, as discussed in Chapter 2, the literature is still scarce on research simultaneously incorporating economic, environmental and social sustainability dimensions into supply chain analytical models (Brandenburg et al. 2014; Seuring 2013), particularly into the models representing supply chain design despite its crucial role in supply chain management (Lambert 2008). Many researchers agree that the analytical modelling methodology and the techniques developed in the field of operations research and management science (Bertrand & Fransoo 2002; Kotzab et al. 2006; Wacker 1998) can provide decision-makers with strategies that may significantly reduce the total supply chain cost (Drezner & Hamacher 2002; Shapiro 2007; Simchi-Levi, Kaminsky & Simchi-Levi 2008). It can be argued that such approaches have the strength to include multiple performance measures relating to the three dimensions of sustainability in an attempt to adopt balanced decision-making, encompassing all three dimensions. This may ultimately help focal companies optimise sustainable supply chain design (Varsei et al. 2014). This chapter proposes a framework for the simultaneous incorporation of the economic, environmental and social performance measures in supply chain design models (see Varsei et al. 2014), and explains the associated research methodology. Next, a more detailed elaboration of supply chain design is provided in Section 3.2. Section 3.3 examines why the analytical methodology should be employed to overcome the supply chain design problem. Then the proposed framework is presented in Section 3.4. The section is divided into four sub-sections to elaborate the relevant literature on the interfaces between supply 54
chain design and the economic, environmental and social dimensions of sustainability. Based on the proposed framework, the next chapter (Chapter 4) develops a novel integrated model for sustainable supply chain design customised for an industrial sector, and Chapter 5 concerns a case study to illustrate the applicability of the proposed analytical model.
3.2
Supply chain design
Supply chain design (or supply chain network design) aims to find the best supply chain configuration in accordance with a focal firm’s competitive strategy so as to achieve its long-term goals. It typically involves long-term integrated decisions related to the number, location and capacity of facilities (e.g. production plants and distribution centres); the flow of raw materials, intermediate and finished products throughout a supply chain; and a set of suppliers to select (Chopra & Meindl 2010; Kaminsky, Simchi-Levi & Simchi-Levi 2004; Vidal & Goetschalckx 1997). These decisions are integrated because they impact upon each other (Farahani, Asgari & Davarzani 2009). For instance, the number of distribution centres in a market zone will affect transportation activities between distribution centres and retailers. An effective way to identify the optimal (i.e. the most suitable) configuration of a supply chain is to use supply chain design models (Beckman & Rosenfield 2008). These models are integrated because they cannot only configure the structure of a supply chain, but also determine product flow between facilities and locations. According to Simchi-Levi, Kaminsky and Simchi-Levi (2008), supply chain design plays a significant role in improving efficiency through determining the following integrated strategic decisions: •
sourcing strategies (i.e. supplier evaluation and selection)
•
the number, location and capacity (size) of facilities (e.g. production plants and distribution centres)
•
distribution strategies and allocation decisions (e.g. allocation of demand points to distribution centres).
For example, for a textile company aiming to expand its existing supply chain to enter a new market, the challenges might lie in selecting new cotton suppliers and determining the number, location and capacity of plants and distribution centres to meet the global 55
demands, and in allocating demand zones to plants and distribution centres. The textile company’s challenges are typical of the issues facing companies around the world, such as Amazon, Pepsi, ZARA and Procter & Gamble (Beckman & Rosenfield 2008). These issues can be examined and resolved by a detailed analysis of the design of supply chains. The next section provides further arguments about why companies should revisit their existing supply chains, or carefully examine their plans for developing and designing supply chains. 3.2.1 The need for the detailed analysis of supply chain design The strategic nature of supply chain design and its associated decisions necessitate careful examination. Supply chain design decisions involve large amounts of capital investment required for establishing facilities, the links between them and supplier selection (Current, Min & Schilling 1990; Watson et al. 2013). These decisions also affect a firm’s other strategic decisions such as inventory, transportation, distribution and pricing policies. While good supply chain design decisions could help firms and their supply chains to be efficient, poorly located facilities might threaten their efficiency and responsiveness. For instance, according to Chopra and Meindl (2013), Toyota’s first United States plant in Kentucky in 1988 allowed Toyota to be responsive to the United States’ market. Amazon realised that managing its supply chain with a single distribution centre in Seattle (in the United States) was unlikely to be cost effective in meeting customer demands throughout the United States and therefore the company redesigned the supply chain and established more distribution centres in other locations in order to minimise the total supply chain cost (Chopra & Meindl 2013). As markets evolve, demand patterns shift, and other economic, environmental and social norms and factors change, a need may arise to revisit and redesign supply chains through relocating, expanding, increasing, or decreasing facilities (Owen & Daskin 1998). Firms that seek cost leadership in the marketplace tend to relocate their production plants to regions with lower labour costs (Sameer & Jayavel 2003). A number of focal companies in the apparel, fast-moving consumer goods, and communications technologies industries, for example, have moved their facilities from developed countries to some countries in the Asia-Pacific region since the 1980s in hope of lowering costs (Chopra & Meindl 2013). 56
Companies may also gain a competitive advantage through fast responsiveness to market trends. Therefore, they redesign their supply chains in order to locate facilities (e.g. plants and distribution centres) closer to final customers even if this direction brings additional costs. For instance, the giant Spanish apparel producer, ZARA, has some manufacturing plants in Spain which enables ZARA to respond quickly to changing fashion trends in Europe despite the high production and labour costs in Europe (Chopra & Meindl 2013). In addition to the aforementioned drivers, some other factors such as energy prices and new environmental-related regulations such as carbon tax may force companies to revisit their supply chains. In 2008, Procter & Gamble, one of the largest companies in the fast-moving consumer goods industry, announced a plan to redesign its supply chain network, which was initially designed when the oil price was US$10 per barrel, only around one tenth of the oil price in 2008 at the time of the supply chain redesign (Chopra & Meindl 2013). Simchi-Levi (2010) argues that the dramatic increase in the price of oil over the past few years and its unprecedented volatility in recent years have significantly affected distribution strategies, network design decisions, offshoring policies and business strategies. To effectively cope with these changes, companies should analyse and revisit supply chain design decisions carefully and use supply chain design analytical models for this purpose (Beckman & Rosenfield 2008; Shapiro 2007). Some relevant models are discussed in the next section. 3.2.2 Facility location and supply chain design Supply chain design has received considerable attention in academia (Chopra & Meindl 2013; Daskin 1995; Drezner & Hamacher 2002; Farahani, Drezner & Asgari 2009; Wang, Lai & Shi 2011). Research in this area can be traced back to the pioneering study by Alfred Weber in 1909 on facility location models (Current, Min & Schilling 1990). Both terms (i.e. facility location and supply chain design, or network design) have often been used interchangeably in the literature (Melkote & Daskin 2001). In his book Theory of the location of industries, Alfred Weber analysed how to locate a plant so as to minimise the total weighted distance (as a cost function) between the plant and its several customer locations (Friedrich 1929). Since then, facility location has attracted researchers from 57
various disciplines including operations and supply chain management, industrial engineering, planning and regional science, telecommunication and energy (Current, Min & Schilling 1990; Melkote & Daskin 2001). Facilities denote places and buildings in the forms of production and manufacturing plants, distribution centres, retail outlets, airports, fire stations, bus stops, schools, hospitals, subway stations, electronic switching centres and computer terminals, among others (Ahuja, Magnanti & Orlin 1993; Drezner & Hamacher 2002). Facilities create networks in various industries and sectors, from electrical and national highway networks to supply chain networks (Ahuja, Magnanti & Orlin 1993). In general, a facility location model aims to find the best possible configuration in which entities (e.g. products, electricity, data or people) are moved between locations in such a way that a set of defined goals can be reached (Ahuja, Magnanti & Orlin 1993). Determining the best location for a new facility is a challenge (Chopra & Meindl 2013; Owen & Daskin 1998; ReVelle, Eiselt & Daskin 2008). To cope with this challenge, several mathematical models have been proposed in the literature including P-median, Pcentre, uncapacitated fixed charge, capacitated fixed charge and location-allocation models (Drezner & Hamacher 2002; Owen & Daskin 1998). These efforts are well studied by several scholars including Owen and Daskin (1998), ReVelle, Eiselt and Daskin (2008) and Melo, Nickel and Saldanha-da-Gama (2009) in their comprehensive reviews of facility location models and applications. Among the aforementioned modelling approaches, three models can be seen as the better representatives of supply chains: the uncapacitated fixed charge facility location model, the capacitated fixed charge facility location model and the location-allocation model (Chopra & Meindl 2013; Owen & Daskin 1998). These models aim to determine the number, location and capacity of facilities so as to minimise total establishment and transportation costs (Owen & Daskin 1998). The term ‘fixed charge’ denotes that a fixed cost or capital investment associated with establishing a facility is considered in a model (Owen & Daskin 1998). But, in the uncapacitated fixed charge facility location model, it is assumed that established facilities can serve demand without limits on capacities. For real-world problems, this assumption is not practical in many cases (Drezner & Hamacher 2002). 58
Accordingly, most studies published on supply chain design are drawn from the capacitated fixed charge facility location model (Chopra & Meindl 2013). According to Chopra and Meindl (2013), this model is a more realistic representation of real-world supply chains because a maximum capacity for each facility (e.g. a production plant) is considered. In addition, realistic network design models need to consider allocation decisions, which address how to best ship products between facilities (Owen & Daskin 1998). Accordingly, location-allocation models have been developed in the literature to consider product flow between facilities (Owen & Daskin 1998). Similar to many supply chain design models already developed in the literature (Chaabane, Ramudhin & Paquet 2012; Chopra & Meindl 2013; Pishvaee, Torabi & Razmi 2012), the proposed model in this thesis is based on a combination of the capacitated fixed charge facility location model and the locationallocation model,. Supply chain design or facility location models, similar to other analytical models, can be modified to encompass specific factors. Some scholars have modified network design models in order to consider exchange rates and tariffs in global supply chain design models (Bhutta et al. 2003; Goh, Lim & Meng 2007; Meixell & Gargeya 2005; Vidal & Goetschalckx 1997), while others have recently incorporated environmental issues, such as CO2 emissions, into models in the area of sustainable supply chain management (Chaabane, Ramudhin & Paquet 2012; Ferretti et al. 2007; Frota et al. 2008; Hugo & Pistikopoulos 2005; Mallidis, Dekker & Vlachos 2012). These modifications and considerations combined with other factors would make the models more challenging in terms of computation (Drezner & Hamacher 2002; Melo, Nickel & Saldanha-da-Gama 2009),which in turn would require analytical approaches to address the computational complexity. This is elaborated in the following section.
3.3
Analytical modelling for supply chain design: Research methodology
The ‘analytical mathematical research methodology’ (Wacker 1998, p.361), often referred to as analytical modelling or mathematical modelling, is a quantitative research methodology employed by several scholars mainly in the field of operations research and
59
management science in order to address various issues, including those related to supply chain management (Bertrand & Fransoo 2002; Kotzab et al. 2006; Wacker 1998). The analytical modelling is a ‘scientific approach to managerial decision-making’ which has been ‘adopted successfully’ to a wide variety of complex problems in business, industry and government (Render, Stair & Hanna 2012, p.2). The typical stages in the analytical modelling methodology include (1) defining a problem; (2) developing a mathematical model which can represent the problem as much as possible; (3) collecting real data (or preparing artificial data sets) in order to test the model and illustrate its applicability; (4) finding an appropriate solution approach to overcome the computational complexity of the model and to optimise it; and, (5) analysing the numerical results, discussing the managerial implications and disseminating the model for future improvement and development (Render, Stair & Hanna 2012). This approach reflects an iterative process. Few analytical modelling studies have collected real data to test models, particularly in the sustainable supply chain management field (Seuring 2013). Many scholars recommend conducting more case study research for ‘empirically informed analytical’ studies (Brandenburg et al. 2014; Choi & Guide 2012, p.507). In their recent review of modelling studies in the field, Brandenburg et al. (2014, p.310) conclude that ‘employing case study research is highly recommended not only for SSCM [sustainable supply chain management] models in particular but also for model-based research in general’. From the research methodology perspective, it can be argued that a combination of the ‘analytical mathematical research methodology’ (Wacker 1998, p.361) and case study research (Meredith 1998; Stuart et al. 2002) may result in an ‘empirically informed analytical study’ (Choi & Guide 2012, p.507). Chapter 5 elaborates this further. Section 3.3 examines why analytical mathematical research methodology should be used to approach the supply chain design problem, taking into account its high computational complexity (Farahani, Drezner & Asgari 2009; Owen & Daskin 1998; Watson et al. 2013).
60
3.3.2
Trade-offs and analytical approaches
The supply chain design problem often involves several trade-offs. One effective approach to the trade-offs is to use analytical models and optimisation techniques (Ahuja, Magnanti & Orlin 1993; Snyder 2011). The trade-offs may exist in every aspect of a typical supply chain design problem. For example, Simchi-Levi, Kaminsky and Simchi-Levi (2008) argue that increasing the number of distribution centres may result in: •
service level improvement
•
inventory and setup cost increase
•
outbound transportation cost decrease (e.g. from distribution centres to retailers and customers)
•
inbound transportation cost increase (e.g. from production plants to distribution centres).
This example clearly demonstrates the trade-off between inbound and outbound transportation costs, and the trade-off between service level and inventory cost. Focal firms can analyse these trade-offs (sometimes it may be a win-win scenario) by means of quantitative tools in order to maintain a balance between supply chain costs and other strategic objectives (Beckman & Rosenfield 2008). Therefore, it is not surprising to see that the most commonly applied methodology to approach supply chain design is analytical modelling (Ahmadi Javid & Azad 2010; Chaabane et al. 2008; Fahimnia 2011; Fahimnia, Farahani et al. 2013; Melkote & Daskin 2001; Mula et al. 2010; Pishvaee & Razmi 2012; Ramudhin, Chaabane & Paquet 2010; ReVelle, Eiselt & Daskin 2008). To overcome models and solve mathematical problems, different approaches can be used including optimisation and simulation techniques. Optimisation techniques have been broadly applied in studies on supply chain design and facility location, while the importance of simulation techniques has also been acknowledged in the literature (Daskin 1995; Drezner & Hamacher 2002; Shapiro 2007). The comprehensive studies by Fahimnia (2011) and Simchi-Levi, Kaminsky and SimchiLevi (2008) evaluate different techniques and affirm the value of optimisation techniques despite their limitation owing to the size and complexity of the typical supply chain 61
analytical models. Shapiro (2007) states that the use of optimisation techniques in strategic supply chain studies has considerably increased in recent years due to their ability to help managers gain a better understanding of the trade-offs between desired objectives. Unlike simulation techniques, this can be achieved through iteratively validating and analysing the strategic directions identified in an optimisation process, excluding non-desirable scenarios and identifying good solutions in a relatively short period of time (Fahimnia 2011; Sadjady & Davoudpour 2012). The next section elaborates the concepts of optimisation and model. 3.3.3
Optimisation and model
Optimisation refers to the process of finding the best possible solution for a mathematical model through systemically analysing a set of alternatives. According to Eykhoff (1974, p.1), a model is ‘a representation of the essential aspects of an existing system (or a system to be constructed) which presents knowledge of that system in usable form’. A model is a description of a system using mathematical concepts and language (Eykhoff 1974). In the supply chain management area, models use mathematical language to describe a supply chain network encompassing various facilities or nodes (e.g. suppliers, production plants, distribution centres and demand points) and product flow between facilities. Typically, a supply chain model may consist of a set of objectives (i.e. business goals) and a set of constraints (e.g. resource limitations or other business constraints), in addition to a set of input parameters and a set of decision variables (Shapiro 2007). Objectives and constraints are presented as mathematical functions. Models and optimisation techniques are ‘tools designed to help solve managerial, planning, and design problems in which the decision maker must allocate limited resources among various activities to optimise a measurable goal’ (Sarker & Newton 2008, p.31). In an optimisation problem, a desired business objective or a set of objectives are minimised or maximised, subject to a set of constraints imposed on the problem (Belegundu & Chandrupatla 1999). Models and optimisation problems may consist of one objective function or more than one objective function, which are known as single-objective or multi-objective respectively and the latter may be more complex to solve (Fahimnia 2011; Mavrotas 2009; Sarker & Newton 2008).
62
Supply chain design models are typically difficult to approach. They are known as NP-hard in terms of computational complexity (Owen & Daskin 1998); hence sophisticated optimisation techniques and solution approaches have been adopted in various studies for several supply chain design mathematical models (Current, Min & Schilling 1990; Drezner & Hamacher 2002; Fahimnia 2011; Simchi-Levi, Kaminsky & Simchi-Levi 2008). These optimisation techniques and solution approaches are well studied and reviewed by a number of scholars such as Drezner and Hamacher (2002), Farahani, Asgari and Davarzani (2009), Fahimnia, Farahani et al. (2013), and Melo, Nickel and Saldanha-da-Gama (2009). For over fifty years, researchers and practitioners have adopted mathematical modelling techniques to analyse often complex problems with large data sets, assisting decisionmakers in strategic, tactical and operational decisions (Bertrand & Fransoo 2002; Kotzab et al. 2006; Shapiro 2007). Several modelling techniques have been developed in the literature which include linear programming, nonlinear programming and mixed integer programming (Bazaraa, Sherali & Jarvis 2010; Shapiro 2007). First introduced in 1947 shortly after World War II (Dantzig 2002), linear programming has played a central role in many types of supply chain applications and has become one of the most applied techniques (Shapiro 2007). The term ‘linear’ denotes that all functions and formulations (i.e. objectives and constraints) are represented by linear relationships. The ability of linear programming to analyse complex and large problems in industrial as well as governmental planning areas has made it popular in the field of operations research and management science. As a result, ‘hundreds of books and an uncountable number of articles’ have been written on the subject (Bazaraa, Sherali & Jarvis 2010; Dantzig 2002, p.42). Nonlinear programming has been developed to address non-linearity which exists in many real-world applications. The term ‘nonlinear’ denotes that there is at least one function in a model (within its objective functions or constraints) which is formulated as non-linear; in other words, its mathematical relationship cannot be graphically represented as a straight line (Bazaraa, Sherali & Jarvis 2010). Despite the broad applications of linear and nonlinear models, they do not include integer decision variables and can only consider continuous decision variables. In reality, supply chain decisions involve both types of decision variables. That is why Simchi-Levi, 63
Kaminsky and Simchi-Levi (2008, p.93) argue that the importance of integer decision variables render linear and nonlinear programming ‘inappropriate’ and requires the use of mixed integer programming models in which both types could exist. The question then arises about the main difference between a continuous and an integer decision variable. In the context of supply chain modelling and optimisation, continuous decision variables are concerned with determining the optimal, or the best possible quantity of product flow in a supply chain network (Simchi-Levi, Kaminsky & Simchi-Levi 2008). The quantity of products transported from a production plant to a distribution centre is a good example of a continuous decision variable. While these variables are allowed to take any value (including real numbers), integer decision variables must take only integers, and according to Shapiro (2007), the most frequently applied integer variables are zero-one (0-1), or binary variables. They are often also called selection decision variables. These variables are constrained to take on values of 0 or 1 to address selection-related operational, tactical or strategic considerations in supply chain models. For instance, assume that there are two cotton suppliers in two locations and the objective is to decide which supplier to select. Each supplier gets a binary (0-1) variable; where ‘0’ indicates that the supplier is not chosen to be in the network and ‘1’ otherwise. Covering both types of decisions, mixed integer programming models are capable of providing a realistic representation of supply chain problems for decision-makers. According to Shapiro (2007), a supply chain network design analysis typically involves a variety of yes-no (i.e. selection) decisions including: •
Which of the established (i.e. existing) production plants or distribution centres should remain open or closed?
•
Which of the candidate (i.e. non-established) production plants or distribution centres should be established? At what capacity and size?
•
Which of the candidate distribution centres will serve a particular market zone or demand point (point of sale, retailer or customer)?
•
Which of the candidate suppliers will serve a particular production plant? (Shapiro 2007)
64
Due to the selection characteristic of strategic decisions inherent in supply chain design, most studies have employed mixed integer programming models (Ramos, Gomes & Barbosa-Póvoa 2014). Following these studies, the proposed model in this research for sustainable supply chain design is within the class of mixed integer programming models. The next section explains how supply chain design models can be developed and modified in order to capture economic, environmental and social objectives simultaneously.
3.4
Sustainable supply chain design
Sustainable supply chain design has emerged as an extension of the traditional supply chain design to incorporate economic, environmental and social sustainability (Chaabane, Ramudhin & Paquet 2012). While the globally acknowledged comprehension of sustainability refers to the consideration of all three dimensions (Elkington 1999), most studies have addressed economic and environmental aspects and the social component has received considerably less attention to date (Ashby, Leat & Hudson-Smith 2012; Boloori Arabani & Farahani 2012; Brandenburg et al. 2014; Eskandarpour et al. 2015; Pagell & Wu 2009; Seuring 2013; Walker et al. 2014). In fact, the social dimension has been ‘neglected’ in the literature (Brandenburg et al. 2014, p.308). According to the recent review of modelling approaches for sustainable supply chain management (Seuring 2013, p.1518), ‘the social dimension is almost completely missing’ in this research field, in addition to ‘the integration of the three dimensions’. This section shows how economic, environmental and social dimensions can be incorporated in an integrated sustainable supply chain design model. It begins with the economic dimension as follows. 3.4.1
Economic performance measures
The economic aspect of supply chain design has been well studied and published (Snyder 2011). Modelling approaches for supply chain design traditionally focus on economicrelated performance indicators such as cost or profit, customer satisfaction, responsiveness or service level (Chaabane et al. 2008; Simchi-Levi, Kaminsky & Simchi-Levi 2008). Vidal and Goetschalckx (1997) presented a review of strategic supply chain models, including supply chain design models, and identified their main characteristics. Most 65
reviewed articles employed mixed integer programming models with cost minimisation as the desired objective function in static single-period deterministic problems. The number of associated variables and constraints ranged between 60 and 19,841. Vidal and Goetschalckx (1997) stated that mixed integer programming had an undeniable role in supply chain design process and decision support tools (Vidal & Goetschalckx 1997). Beamon (1998) provided a review of the literature and categorised articles based on supply chain performance measures. The study considered both qualitative performance measures, for which there is no direct numerical measurement (such as customer satisfaction and supply chain flexibility), and quantitative performance measures, which can be described numerically (such as supply chain costs). In general, these performance measures define objective functions in optimisation models, which in turn are used to find the best value of the decision variables (Shapiro 2007). Beamon (1998) concluded that the reviewed studies mainly considered a single measure: cost minimisation. A similar conclusion was also drawn by Meixell and Gargeya (2005) who reviewed the relevant research articles in the field published from 1982 to 2005. Dullaert et al (2007) presented a taxonomy of modelling approaches for supply chain design published since 1999. They found that the most applied modelling approach is discrete mixed integer programming. This taxonomy provides a better understanding of the related factors that affect supply chain design problems. The most important factors are listed below. •
Objective functions: minimising costs, maximising profits, and multiple objective functions (e.g. cost and lead time minimisation).
•
Number of supply chain stages: determining the number of stages that comprise a supply chain, e.g. suppliers, plants, distribution centres and retailers (these stages also determine the number of echelons).
•
Nature of demand: deterministic and stochastic.
•
Number of time periods: single and multiple time periods.
•
Optimisation techniques: e.g. mixed integer programming.
66
•
Data source for testing the model: real case study and artificial illustration by numerical examples.
According to Dullaert et al. (2007), the majority of the reviewed studies aimed to minimise total supply chain costs in multi-stage supply chains, while only one study (the study by Santoso et al. (2005), maximised after-tax profit as the objective function. Their study is the only research which considers the stochastic nature of demand instead of assuming deterministic market demand. In addition, most reviewed studies applied mixed integer linear models (Bidhandi et al. 2009; Dhaenens-Flipo & Finke 2001; Jang et al. 2002; Sadjady & Davoudpour 2012; Santoso et al. 2005) and only one study formulated the problem as a mixed integer non-linear model (Park, Lee & Sung 2010). To evaluate the applications of models in practice, authors have mainly focused on artificial illustration using benchmark industry data or generating numerical examples randomly; very few studies have been grounded in real cases. This observation is evident in the supply chain modelling literature in general (Mula et al. 2010; Seuring 2013). In fact, few scholars have embraced the challenge of approaching industry and conducting pragmatic and empirically-grounded analytical modelling studies drawing on real case studies (Chen, Zhang & Delaurentis 2014; Choi & Guide 2012; Seuring 2013). These observations are similar to the findings of Dullaert et al. (2007), Meixell and Gargeya (2005), Beamon (1998) and Vidal and Goetschalckx (1997). Table 3.1 summarises relevant sustainable supply chain design studies in the literature based on desired objectives, modelling and solution approaches, data sources, the industry sectors and the locations from which data were gathered by the authors to evaluate and validate their models.
67
Table 3.1 Summary of the relevant sustainable supply chain design studies in the literature
Author(s), (Year)
Selected Objectives for Sustainable Supply Chain Network Design Economic
Environment
Social
Modelling Approach – Solution approach
Data Source
Industry sector – Location
Cost
Carbon dioxide equivalent (CO2e)
Not Indicated
Mixed Integer Programming –Optimisation
Artificial illustration
Aluminium– Not Indicated
Elhedhli and Merrick (2012)
Cost
Carbon dioxide
Not Indicated
Mixed Integer Programming –Optimisation
Artificial illustration
Not Indicated
Abdallah et al. (2012)
Cost
Carbon dioxide
Not Indicated
Mixed Integer Programming –Optimisation
Artificial illustration
Computer– the United States
Mallidis, Dekker and Vlachos (2012)
Cost
Carbon dioxide, fine dust
Not Indicated
Mixed Integer Programming –Optimisation
Real case study
Logistics– Europe
Pishvaee, Torabi and Razmi (2012)
Cost
Carbon dioxide equivalent (CO2e)
Not Indicated
Mixed Integer Programming –Optimisation
Real case study
Medical device–Iran
Wang, Lai and Shi (2011)
Cost
Carbon dioxide
Not Indicated
Mixed Integer Programming –Optimisation
Artificial illustration
Procurement– China
Yeh and Chuang (2011)
Cost, time and quality
Supplier’s green appraisal score
Not Indicated
Mixed Integer Programming –Optimisation
Artificial illustration
Electronic– Taiwan
Pinto-Varela, Barbosa-Póvoa and Novais (2011)
Profit
Eco-Indicator 99 (Emissions, Resource Depletion)
Not Indicated
Mixed Integer Programming –Optimisation
Artificial illustration
Pulp and paper– Portugal
Ramudhin, Chaabane and Paquet (2010)
Cost
Carbon dioxide
Not Indicated
Mixed Integer Programming –Optimisation
Artificial illustration
Steel–Not Indicated
Nagurney and Nagurney (2010)
Cost
Carbon dioxide
Not Indicated
Mixed Integer Programming –Optimisation
Artificial illustration
Not Indicated
GuillénGosálbez and Grossmann (2010)
Net present value
Eco-Indicator 99 (Emissions, Resource Depletion)
Not Indicated
Mixed Integer Programming –Optimisation
Artificial illustration
Chemical– Not Indicated
Chaabane, Ramudhin and Paquet (2012)
68
Distance
Carbon dioxide, Energy consumption
Not Indicated
Simulation based method
Artificial illustration
Wine–Not Indicated
Bojarski et al. (2009)
Net present value
IMPACT 2002+ (Climate change, Ecosystem quality, Human health and Resources)
Not Indicated
Mixed Integer Programming –Optimisation
Artificial illustration
Chemical– Not Indicated
Chaabane et al. (2008)
Cost
Carbon dioxide
Not Indicated
Mixed Integer Programming –Optimisation
Artificial illustration
Not Indicated
Neto et al. (2008)
Cost
Carbon dioxide, Waste
Not Indicated
Mixed Integer Programming –Optimisation
Artificial illustration
Pulp and paper– Europe
Ferretti et al. (2007)
Cost
Emissions
Not Indicated
Mixed Integer Programming –Optimisation
Artificial illustration
Aluminium– Not Indicated
Hugo and Pistikopoulos (2005)
Net present value
Eco-Indicator 99 (Emissions, Resource Depletion)
Not Indicated
Mixed Integer Programming –Optimisation
Artificial illustration
Chemical– Not Indicated
Cholette and Venkat (2009)
It is evident that most authors have focused on supply chain cost minimisation as the desired economic objective within supply chain design models (Daskin 1995; Drezner & Hamacher 2002; Melo, Nickel & Saldanha-da-Gama 2009; Shapiro 2007). Similarly, this research follows the same approach. The total supply chain cost in supply chain design models involves various fixed and variable costs. For example, fixed costs include a capital investment required for establishing and equipping facilities (e.g. production plants and distribution centres) at multiple geographical locations. Variable costs may comprise the costs of procurement, manufacturing, transporting raw materials between suppliers and production plants, and shipping the finished product from distribution centres to demand points (Chopra & Meindl 2013; Fahimnia 2011; Pishvaee & Razmi 2012; Ramudhin, Chaabane & Paquet 2010). These cost components can be adjusted by decision-makers depending on a problem context. For instance, Pishvaee, Torabi and Razmi (2012) did not consider the cost of shipment between suppliers and production plants.
69
The next section presents the interplay between the environmental dimension and supply chain design. 3.4.2
Environmental performance measures
The most predominant environmental indicator used to incorporate the environmental aspect in supply chain design is the amount of CO2 emissions, or, GHG emissions (i.e. carbon dioxide equivalent (CO2-e)) generated by a supply chain. This observation is evident in the third column of Table 3.1. However, there are obviously more environmental-related indicators. Some environmental concerns highlighted in the literature on supply chain management include GHG emissions (Paksoy, Bektaş & Özceylan 2011), waste generation (Tsai & Hung 2009), energy consumption (Cholette & Venkat 2009), water usage (Brent 2005), and the use of hazardous and toxic substances in products (Hsu & Hu 2009). With regard to these concerns, environmental performance measures could be targeted at air, water or solid waste (Sarkis 2003). The list of environmental indicators that can be used as performance measures is quite extensive (Global Reporting Initiative 2013). Given the extensive list of environmental performance measures, the issue of GHG emissions has been stressed in the literature as the most prominent because of its significant consequences on ecosystems and human health (Howard-Grenville et al. 2014), which in turn has led to the introduction of several emission-control regulatory policies worldwide (Gupta & Palsule-Desai 2011; Paksoy, Bektaş & Özceylan 2011). It should be noted that the factors contributing to the Earth’s climate change can be classified into natural and anthropogenic (i.e. human-induced) factors. GHGs are the main causes of anthropogenic climate change, of which CO2 is referred to as the most harmful (Gupta & Palsule-Desai 2011, p.235). As seen in Table 3.1, relatively few studies incorporate multiple environmental metrics in supply chain design analysis. Hugo and Pistikopoulos (2005) developed a multi-objective mixed integer model to minimise the environmental impact of a chemical supply chain through adopting the Eco-Indicator 99 as a quantitative life cycle assessment tool (Goedkoop & Spriensma 2001). As an impact assessment method in life cycle assessment, the Eco-Indicator 99 is a damage-oriented approach utilising a procedure for weighting 70
three categories: damage to human health, damage to ecosystem quality, and damage to resources (Goedkoop & Spriensma 2001). Guillén-Gosálbez and Grossmann (2010) applied the Eco-Indicator 99 to measure environmental performance of supply chains in order to design a sustainable supply chain in the chemical industry. Yeh and Chuang (2011) proposed a supplier’s green appraisal score as an environmental indicator, and supply chain cost as an economic indicator formulated in a multi-objective mixed integer model. They evaluated the model by numerical examples, similar to the majority of studies shown in Table 3.1 They illustrated a company in the electronic industry in Taiwan so as to provide decision-makers with some scenarios in which economic and environmental objectives can be addressed. The authors argued that their findings could help firms gain visibility on how to manage cost-efficient and green supply chains. Most studies incorporated carbon emissions in supply chain design models (see Table 3.1). Ferretti et al. (2007) proposed a multi-objective mixed integer model to minimise supply chain costs as well as emissions generated by facilities and transportation activities. Claiming to propose the first model to consider carbon pricing, Chaabane et al. (2008) presented a multi-objective mixed integer model to minimise the supply chain cost while minimising the cost of CO2 emissions. They considered CO2 generated by transportation activities and the authors employed emission factors associated with transportation modes. Their proposed model was solved by the IBM ILOG CPLEX Optimiser software, which is one of the most widely used mathematical programming solvers (Fahimnia, Sarkis et al. 2013; IBM 2012). Chaabane et al. (2008) examined trade-offs between two objectives and showed that CO2 emissions could be reduced by 10% with 2% increase in the total supply chain cost. A very similar approach was adopted by Ramudhin, Chaabane and Paquet (2010) in order to minimise supply chain costs and CO2 emissions. The study showed that the proposed optimisation model could help decision-makers examine trade-offs between the objectives, i.e. supply chain costs and CO2 emissions. Some recent studies have followed a more comprehensive approach and selected the amount of CO2-e emissions as an environmental indicator. Pishvaee, Torabi and Razmi (2012) and Chaabane et al. (2012) incorporated CO2-e emissions in their proposed models for supply chain design. According to Brander (2012, p.2), CO2-e ‘is a term for describing 71
different greenhouse gases in a common unit’. These gasses include CO2, methane (CH4), nitrous oxide (N2O), hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), sulfur hexafluoride (SF6), and nitrogen trifluoride (NF3). Given these seven types of GHGs, for ‘any quantity and type of greenhouse gas, CO2e signifies the amount of CO2 which would have the equivalent global warming impact’ (Brander 2012, p.2). Following the majority of the reviewed studies, the proposed framework in this thesis selects the amount of carbon emissions and in particular CO2-e emissions as the environmental performance measure. To incorporate CO2-e emissions into an optimisation model, two approaches can be distinguished in the literature. Firstly, and in the presence of environmental regulations such as carbon pricing/trading schemes in some countries, the amount of GHGs emitted within manufacturing, storage, and in-bound and out-bound transportation can be converted into the equivalent cost of emissions. Some authors such as Chaabane, Ramudhin and Paquet (2012), Paksoy (2010), Ramudhin, Chaabane and Paquet (2010) and Bojarski et al. (2009) have followed this approach and investigated the impact of carbon tax or carbon trading schemes on supply chain configuration. The second approach, which may be more pragmatic and adopted in this research, is to formulate an individual environmental objective representing supply chain emissions. Emissions can be measured by ‘emission factors’, which are mostly available in the guidelines developed by industrial sectors or in the Emission Factor Data Base (n.d.). The studies by Pan, Ballot and Fontane (2013), Pishvaee, Torabi and Razmi (2012), Elhedhli and Merrick (2012), Wang, Lai and Shi (2011), Pinto-Varela, Barbosa-Póvoa and Novais (2011) and Sundarakani et al. (2010) have adopted this approach. It can provide companies with a set of scenarios consisting of various costs and emission levels. This could ultimately allow companies to select an appropriate configuration based on their strategic goals. The second approach, as mentioned above, may also help decision-makers incorporate social performance measures in multi-objective models. In other words, a supply chain design model involves three different objectives: economic, environmental and social. Multi-objective optimisation techniques can be adopted in order to overcome the
72
complexity inherited in the multi-objective models (Branke et al. 2008; Ehrgott 2005; Mavrotas 2009; Ramos, Gomes & Barbosa-Póvoa 2014). Multi-objective optimisation, which is also known as multi-objective programming, multicriteria optimisation, multi-attribute optimisation and Pareto optimisation, is concerned with mathematical models consisting of more than one objective function to be simultaneously optimised (Branke et al. 2008; Ehrgott 2005; Mavrotas 2009; Ramos, Gomes & Barbosa-Póvoa 2014). It has been applied in many fields including logistics and supply chain management, engineering and economics where good solutions (i.e. optimal decisions) need to be taken in the presence of trade-offs between two or more conflicting objectives. In the sustainable supply chain management area, minimising supply chain cost while improving environmental and social performance indicators (e.g. minimising carbon emissions) are examples of multiple objectives, as elaborated in this chapter. In some problems, there might be more than three objectives, hence more trade-offs exist between several conflicting objectives. This thesis adopts multi-objective optimisation in Chapter 5, where various types of the associated techniques are discussed, grounded in the multi-objective optimisation literature (Branke et al. 2008; Ehrgott 2005; Mavrotas 2009; Ramos, Gomes & Barbosa-Póvoa 2014). The next section examines how supply chain design models can address the social dimension of sustainability. 3.4.3
Social performance measures
According to several review studies, practical modelling efforts incorporating the social dimension of sustainability along with environmental and economic objectives are virtually non-existent (Boloori Arabani & Farahani 2012; Brandenburg et al. 2014; Carter & Easton 2011; Carter & Rogers 2008; Seuring 2013; Seuring & Müller 2008). Based on a comprehensive literature review, Brandenburg et al. (2014, p.300) conclude that ‘SSCM [sustainable supply chain management] research tends to focus primarily on environmental issues while social facets are widely neglected in empirical and in analytical SSCM modelling research’. This is evident in the fourth column of Table 3.1 as well.
73
It has been argued that social performance measures cannot be easily embedded into supply chain analytical models since they are difficult to capture and measure (Chaabane, Ramudhin & Paquet 2010, 2012; Ramudhin, Chaabane & Paquet 2010). However, social performance measures can play a significant role in determining selection decision variables, which exist in supply chain design models as discussed earlier. Social indicators can affect the optimal configuration of a supply chain in terms of the social sustainability dimension. For example, with the recent disaster in Bangladesh as discussed earlier in Chapter 2 (Greenhouse 2013a; Manik & Yardley 2013), selecting socially responsible suppliers will help companies avoid the reputation loss caused by suppliers’ social misconducts. This consideration may change the optimal configuration of a focal company’s supply chain. This thesis introduces a method to assess, score and measure a supply chain’s social performance based on the analytic hierarchy process, in order to incorporate the social aspect in supply chain design models (Saaty 2008; Varsei, Soosay & Fahimnia 2013; Varsei et al. 2014). In brief, supply chain members are scored based on their performance in terms of a set of social categories (e.g. employment, human rights in workplaces, the wellbeing of communities) determined by a focal company or decision-maker, and the scores form normalised social impact coefficients associated with those businesses or facilities involved in a supply chain (i.e. supply chain members). A choice of social categories does not impact the proposed method and would be case-specific. Established and non-established supply chain members may be associated with different social categories. Established supply chain members refer to already existing facilities, while nonestablished members denote potential facilities that can be located/built in the future to serve the supply chain. The selection from established supply chain members (e.g. current suppliers) could be subject to more social categories because businesses are already established and people work at them; hence a decision-maker can potentially consider related social categories such as working conditions and human rights at workplaces. However, the selection from non-established supply chain members (e.g. a new production plant) entails the regional social impacts of facility locations, such as the employment opportunity for a region at which a plant will be located.
74
Social categories can be drawn from available guidelines and standards for social sustainability as well as socioeconomic considerations. As a comprehensive guideline, the Global Reporting Initiative proposes four primary categories (including labour practices and decent work conditions, human rights, society, and product responsibility), and thirty sub-categories, such as employment, investment, supplier assessment for impacts on society, and supplier human rights assessment (Global Reporting Initiative 2013). These categories are also congruent with the guidelines of Social Accountability 8000 standard (Social Accountability International n.d.), International Labour Organisation (n.d.) and social life cycle assessment tools (Dreyer, Hauschild & Schierbeck 2006, 2010; Hauschild, Dreyer & Jørgensen 2008). Established and non-established supply chain members can be assessed based on a set of these categories, or based on other social categories determined by a focal company or decision-maker. A score is assigned to each supply chain member based on its performance against the desirable performance in each social category. For scoring scales, the pairwise comparison scale can be used which has been employed by many scholars in the analytic hierarchy process literature. Saaty (1990) was the first to propose the analytic hierarchy process, and to introduce the pairwise comparison scale. The analytical hierarchy process is one of the multi-criteria decision making methods (Brandenburg et al. 2014). These methods have been widely used in the literature to help managers assess and find the best possible decision when there are several objectives and criteria (Govindan et al. 2015). Multi-criteria decision making methods include analytical hierarchy process, analytical network process and data envelopment process (Govindan et al. 2015). According to Govindan et al. (2015, p.71) who reviewed the applications of these methods in green supplier evaluation and selection, ‘the most widely used multi-criteria decision-making approach is analytical hierarchy process’. The analytic hierarchy process has been extensively utilised for analysing and comparing alternatives (e.g. supplier selection) in the management science and supply chain management literature (Handfield et al. 2002; Ho 2008; Lin & Juang 2008; Opasanon & Lertsanti 2013; Partovi 2006; Sarkis 1998; Wang & Gupta 2011). Brandenburg et al.
75
(2014) reviewed quantitative models for sustainable supply chain management and summarised studies on the analytic hierarchy process. A modified version of Saaty’s (1990) comparison scales is presented in Table 3.2. As elaborated by Saaty (1990), the unit difference between successive scales in Table 3.2 is based on the well-known psychological theory presented by Miller (1956). The total number of scales is determined taking into account the observation that most individuals cannot make a comparison among more than seven objects (plus/minus two) simultaneously (Saaty & Ozdemir 2003; Wang & Gupta 2011). Given the current supply chain design models, it can be argued that comparison between alternatives may not help decision-makers make a judgment about the actual social performance. Hence, this study considers the comparison between an alternative and what a focal company or decisionmaker desires (Varsei et al. 2014). A score is assigned to each supply chain member based on the degree of conformance, which is the variation between a focal company’s desirable social performance and the actual social impact of a supply chain member influenced by its actions or location (i.e. spatial and regional characteristics). A larger score means better social performance. Table 3.2 The proposed pairwise comparison scale Score
Explanation (variation between a focal firm’s desirable social performance and the actual social impact of a supply chain member influenced by its location or actions)
9
No variation
7
Weak variations
5
Essential or strong variations
3
Demonstrated variations
1
Absolute variations
2, 4, 6, 8
Intermediate scores
Given the comparison scale presented in Table 3.2, every supply chain member receives a score (on a scale of 1 to 9) reflecting its performance against the focal company’s desired social performance, for example in terms of impact on community and working condition. The smaller the score, the greater would be the deviation from the desirable standards (i.e. the worse the social performance). Once scores are assigned to each supply chain member, 76
the normalised summation of the scores for each supply chain member divided by the number of social categories determines the supply chain member’s social impact coefficient, which is a number between 0 and 1. Obviously, members with higher social impact coefficients are the better members in terms of social sustainability. These coefficients can be embedded into an objective function which represents the social aspect of supply chain design. Therefore, decision-makers could analyse and compare resulting alternative supply chain configurations (i.e. scenarios) with respect to associated social impacts. By defining social and environmental objectives as described above, supply chain design models can be characterised by three different objective functions. In other words, models become multi-objective. One effective way to overcome these models is to use multiobjective optimisation, in which multiple scenarios (i.e. solutions) are developed, each of them with a specific total supply chain cost, emission and social impact (Branke et al. 2008; Ehrgott 2005; Mavrotas 2009; Ramos, Gomes & Barbosa-Póvoa 2014). Determining balanced solutions would be case-specific depending on the economic, environmental and social goals of a focal company or decision-maker. Multi-objective optimisation and scenario-based approaches have been adopted in the literature providing insights into sustainability trade-offs (or win-win situations) in complex supply chain models (Bai & Sarkis 2010b; Bojarski et al. 2009; Chaabane, Ramudhin & Paquet 2012; Frota et al. 2008; Guillén-Gosálbez & Grossmann 2010; Nagurney & Yu 2012; Pinto-Varela, Barbosa-Póvoa & Novais 2011; Ramos, Gomes & Barbosa-Póvoa 2014; Seuring 2013). 3.4.4
The proposed framework
Based on the abovementioned analysis and discussion, the proposed framework for sustainable supply chain design is formed. Figure 3.1 illustrates how economic, environmental and social sustainability indicators can all be incorporated into supply chain design models, including the proposed model in this thesis. As depicted in Figure 3.1, the framework and the associated models have the following characteristics: •
Economic, environmental and social sustainability indicators can directly impact on supply chain design models through selection decision variables. 77
•
Economic and environmental sustainability indicators can also directly impact on supply chain design models through continuous decision variables.
•
Supply chain cost is selected in this thesis as an economic sustainability indicator.
•
Supply chain social impact is selected as a social sustainability indicator, which considers a set of the Global Reporting Initiative’s four social categories (i.e. labour practices and decent work conditions, human rights, society, and product responsibility) and the underlying aspects such as employment.
•
Supply chain emission (CO2-e) is selected as an environmental sustainability indicator.
•
The proposed sustainable supply chain design model in this thesis (presented in Chapter 4) is based on two classical facility location models: the capacitated fixed charge facility location model and the location-allocation model. These two models involve typical mathematical formulations in the forms of objective functions and constraints, which are employed in Chapter 4 to develop the mixed integer programming model.
•
The model is multi-objective, and a multi-objective optimisation technique can be employed as a solution approach so as to determine a set of integrated strategic decisions.
It is worth noting that while this research does not intend to generalise the proposed framework to other supply chain analytical models (e.g. at the tactical and operational levels), the framework may have the potential to be modified and used for developing such models so as to encompass multiple social, environmental and economic performance measures. Moreover, while the proposed model involves supply chain cost, emission and social impact, supply chain design models can be adjusted in order to include more sustainability performance indicators (e.g. water usage).
78
Economic sustainability indicators Integrated decisions: •supply chain cost •Selecting suppliers •Plants’ locations, numbers, and capacities
Social sustainability indicators •supply chain social impact (a set of the Global Reporting Initiative’s social categories and aspects)
Sustainable supply chain design •Multi-objective mixed integer programming model
Environmental sustainability indicators
Multi-objective optimisation
•Distribution centres’ locations, numbers and capacities •Product flows (between suppliers, plants, distributions centres, and demand points) •Allocation decisions
•supply chain emission (carbon dioxide equivalent emissions) Impact on selection decision variables Impact on continuous decision variables
Figure 3.1 The proposed framework for designing sustainable supply chains
3.5
Summary and conclusion
The analytical modelling methodology has been utilised by several scholars to approach supply chain design and this study adopts this methodology. A supply chain design model typically aims to minimise total supply chain costs (including facility and transportation costs) while meeting given annual demands. This approach has been followed in most modelling studies in the supply chain management literature where there are trade-offs, for example, between the service level (i.e. serving customer demands appropriately) and the total supply chain cost (e.g. fixed cost of facilities and variable cost of production and transportation). However, the nature of trade-offs may change over time. Although most companies in various industries have designed their supply chains only in the pursuit of profits (Chaabane, Ramudhin & Paquet 2012), they are increasingly forced by various stakeholders to consider more performance indicators pertaining to the environmental and social dimensions (Brandenburg et al. 2014; Burritt & Schaltegger 2014; Sarkis, Zhu & Lai 79
2011; Soosay, Fearne & Dent 2012; Soosay, Fearne & Varsei 2014; Varsei et al. 2014; Wu & Pagell 2011). With the aim of addressing the economic, environmental and social dimensions simultaneously, this chapter has developed a unique framework for designing sustainable supply chains (Varsei et al. 2014). Grounded in the facility location and supply chain design literature (Drezner & Hamacher 2002; Owen & Daskin 1998), the proposed framework has extended prevalent supply chain design models to encompass all three dimensions simultaneously (Varsei et al. 2014). The capacitated fixed charge facility location model and the location-allocating model have laid the foundations of the proposed framework, along with the analytic hierarchy process method which could help decisionmakers incorporate a set of social performance indicators into the analytical models for sustainable supply chain design. The following chapter presents how the framework is used to develop a novel mathematical model (multi-objective mixed integer programming) tailor-made for an industrial sector.
80
4 4.1
A model for sustainable supply chain design
Introduction
Based on the framework elaborated in Chapter 3, this chapter aims to develop a unique mathematical model so as to incorporate the economic, environmental and social sustainability indicators into supply chain design in the context of the wine industry. The model could assist focal wine companies in designing sustainable supply chains, and it may shed light on the application of such models in other industries. Model development is the critical stage of the analytical modelling research, as explained in Chapter 3 (Render, Stair & Hanna 2012) . Mathematical models ‘should be solvable, realistic, and easy to understand and modify, and required input data should be obtainable’ (Render, Stair & Hanna 2012, p.4). Model development strongly influences the subsequent stages (i.e. collecting data, solving the model and analysing the results), hence researchers should adopt an iterative approach and consider the subsequent stages during the model development stage (Render, Stair & Hanna 2012). Considering this issue, this chapter presents the objective functions and constraints of the model in the form of mathematical formulations, along with the assumptions, indices, parameters (i.e. inputs to the model) and decision variables (i.e. outputs) of the model. To illustrate the applicability of the model, the next chapter (Chapter 5) concerns the case study of a major Australian wine company, for which a customised model is presented and solved by a multi-objective optimisation method (Mavrotas 2009). Hence Chapter 4 and 5 collectively aim to present an ‘empirically-informed analytical’ model (Choi & Guide 2012, p.507) based on a year-long case study and iterative data collection process conducted in 2014 in Australia.
81
This chapter begins with an elaboration on why the wine supply chain was selected for this research. Then an overview of the wine supply chain is presented in Section 4.3. The analytical model for sustainable wine supply chain design is proposed in Section 4.4, followed by the summary and conclusion of the chapter in Section 4.5.
4.2
Why wine?
The wine industry is a fertile ground for research on sustainable supply chain design and management from the economic, environmental and social perspectives (Cholette & Venkat 2009; Christ & Burritt 2013; Soosay, Fearne & Dent 2012; Soosay, Fearne & Varsei 2014; Szolnoki 2013). Over the past two decades, more than 25 billion litres of wine per annum have been produced and distributed globally (Winetitles 2014). Supply chain decisions related to the large-scale production and distribution of wine can have critical environmental and social implications, notably for ‘climate change’ (Holland & Smit 2010, p.125) and ‘employment’ (Anderson et al. 2009, p.5; Land and Water Australia 2008, p.3), and influence the profitability of the companies and actors involved in the wine supply chain (Michalewicz, Michalewicz & Spitty 2011; Moccia 2013). Despite this significance, research on the links between the wine industry, sustainability and supply chain management is still in its infancy (Garcia et al. 2012; Szolnoki 2013). Some researchers have recently addressed the economic, environmental and/or social aspects of the wine supply chain. Pullman, Maloni and Dillard (2010) interviewed and surveyed more than fifty wine producers in the United States to investigate typical sustainability practices in the wine supply chain. They observed that the current focus is primarily on economic survival and then environmental sustainability, with little direct recognition for the social aspect such as supporting the region in which wine producers operate or improving employees’ working conditions (Pullman, Maloni & Dillard 2010). Soosay, Fearne and Dent (2012) conducted value chain and life cycle analyses in order to highlight the importance of taking a holistic view when considering the sustainability of a wine product. Using a real case study of a wine supply chain, they analysed trade-offs between environmental impacts (carbon emissions) and consumer perceptions of value (Soosay, Fearne & Dent 2012). Garcia et al. (2012) proposed a framework for measuring wine logistics performance, in terms of quality, timeliness, cost and productivity based on a 82
benchmarking study in Argentina. Christ and Burritt (2013) and Christ (2014) have raised the importance of water management in the wine supply chain. ‘Water is a vital input for winegrape cultivation and the production of wine, and wine organisations are often exposed directly and through the supply chain to water management issues’ (Christ 2014, p.380). While addressing some important aspects of the wine supply chain, none of the abovementioned studies has used analytical models or examined how to design a wine supply chain, which is likely the subject of the operations research and management science field (i.e. modelling and optimisation). Relatively few scholars in the area of operations research and management science have developed analytical models to optimise the wine operations and supply chain. Moccia (2013) has recently reviewed and summarised these modelling studies. Most of them have primarily addressed the operational or tactical aspects of the wine supply chain management. In addition, the existing publications have incorporated the economic and/or environmental-related indicators (mainly cost and carbon emission) into the models (Arnaout & Maatouk 2010; Berruto, Tortia & Gay 2006; Bohle, Maturana & Vera 2010; Cholette 2007; Cholette & Venkat 2009; Correa & Queyranne 2012; Dunstall, Higgins & Sciberras 2008; Ferrer et al. 2008; Moccia 2013; Vázquez-Rowe et al. 2012). In other words, the social dimension is missing. Cholette and Venkat (2009) utilised a web-based tool, CargoScope, to calculate the energy and carbon emissions associated with several potential distribution networks available to some selected wineries in the United States. They affirm that modelling and optimisation approaches ‘will help in quantifying the costs and benefits of different supply chain options and will support management decisions’(Cholette & Venkat 2009, p.1412). They also assert that ‘no single supply chain configuration is ideal for all wineries’ (Cholette & Venkat 2009, p.1412), hence each wine company should develop to some extent a unique decision support model customised for its own business characteristics in order to design the most efficient supply chain network. Focusing more on the operational side, the study by Bohle, Maturana and Vera (2010) proposed a mathematical model to optimise wine grape harvesting. Similarly, Arnaout and Maatouk (2010) addressed the vineyard
83
harvesting problem and developed a model to minimise the operational costs and optimise scheduling. Despite some existing literature considering modelling/optimisation in the wine industry context, it appears that research concerning how the wine supply chain design can be modelled and optimised is non-existent (Moccia 2013). This research is among the first to investigate this strategic aspect of wine supply chain management. How to design wine supply chains is an important strategic decision because the wine industry has economic, environmental and social implications for several countries and regions around the world. This thesis has been conducted in Australia which is among the top ten producers and top five exporters of wine (Winetitles 2014). According to Osmond and Anderson (1998, p.20), ‘the wine industry has contributed very substantially to growth of the Australian economy’. The industry also has profound implications for communities and the society as a whole in terms of ‘regional contributions’ and ‘employment’ (Anderson et al. 2009, p.5). In 2011, for example, the indirect economic value of the Australian wine industry exceeded AUD 51 billion and involved around 2400 wine companies, with around 6700 grape growers (i.e. suppliers) and over 84,000 people across various regions (Australian Grape and Wine Authority 2012). Some national reports published by the Australian government indicate that providing employment is the most important social contribution of the wine industry (LWA 2008). It is evident that the wine industry’s contributions to the Australian economy and society are significant. Indeed, economic and social contributions are significant to the wine industry of several other countries, including the major producers such as France, Italy, Spain, the United States, China, Chile, Argentina, South Africa, Germany and Russia (Australian Grape and Wine Authority 2012). Furthermore, the wine industry is inextricably linked to ‘a large number of environmental concerns’ (Christ & Burritt 2013, p.232). While the industry impacts the physical environment within which it operates, its viability is affected by environmental conditions (Christ 2014; Christ & Burritt 2013; Holland & Smit 2010; Moccia 2013). According to
84
Christ and Burritt (2013), these concerns include water use, GHG emissions, solid waste, chemical use, land use issues, and the impact on ecosystems. Among these concerns, the issue of GHG emissions and in particular CO2 emissions has received considerable attention (Cholette & Venkat 2009; Point, Tyedmers & Naugler 2012). On the one hand, given the post-production logistics in the wine supply chain, the distribution of wine is carbon intensive mainly due to the reliance on bulky and heavy forms of wine packaging (Cholette & Venkat 2009; Christ & Burritt 2013). On the other hand, the industry is highly vulnerable to the impacts of climate change (Anderson et al. 2009; Holland & Smit 2010). Christ and Burritt (2013, p.232) assert that the extant literature consistently demonstrates ‘the potentially devastating consequences of global warming for a number of the world’s wine regions’. Moccia (2013, p.54) argues ‘climate change is forecasted to severely hamper existing vineyards due to extreme heat waves’. This may necessitate optimising wine supply chain design, in addition to implementing other initiatives at both the organisation and industry levels, so as to minimise the amount of emissions. Strategic decisions related to supply chain design affect CO2 emissions in the wine supply chain. Wine supply chain design considerably influences carbon intensive transportation activities during post-production logistics (Cholette & Venkat 2009). The CO2 emissions of wine distribution and post-production logistics account for around 50% of the total emissions of the wine production and consumption life cycle (Cholette & Venkat 2009; Christ & Burritt 2013). The study by Cholette and Venkat (2009, p.1402) shows that ‘wine itself comprises just half the weight and under 40% of the volume of a case of twelve 750 mL glass bottles’. It should be noted that the 0.75 litre glass bottle is still the most widely utilised container for bottling and packaging in the wine industry (Colman & Päster 2009; Ghidossi et al. 2012). The above discussion has highlighted the importance and significance of the wine supply chain and examined why it deserves further investigation. Before presenting the model for sustainable wine supply chain design, it is important to clarify the wine supply chain and elaborate on several members who play various roles in the network.
85
4.3
The wine supply chain: An overview
4.3.1 From wine to the wine industry Wine is one of the world’s oldest commodities; however, systemic approaches to winerelated business practices have only been adopted recently (Aylward 2004). ‘It has been referred to as an “industry” only within the past 20 years’, according to Aylward (2004, p.425). The industry has attracted considerable interest particularly with the rapid emergence and growth of wine companies in the so-called New World (Aylward 2004). The term New World primarily refers to the vineyards (i.e. where grapevines are grown) and wineries (i.e. where grapes are transformed to wine) of Australia, New Zealand, the United States, South Africa, Chile, and Argentina (Banks & Overton 2010). In contrast, the Old World denotes the traditional wine-growing areas of the Middle East as well as Western and Sothern Europe (Banks & Overton 2010). The importance of the wine industry worldwide is evident in its business and production volume (Garcia et al. 2012). Top countries ranked by total grape production, wine production and wine exports are presented in Table 4.1, 4.2 and 4.3 respectively (Winetitles 2013). Table 4.1 Top countries ranked by total grape production in 2009 (Winetitles 2013)
Country 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Grape production million tonnes
% of world
8,039 7,603 6,643 6,098 5,760 4,265 2,742 2,256 2,182 1,878 1,749 1,684 1,457 1,235
11.8 11.2 9.7 8.9 8.4 6.3 4.0 3.3 3.2 2.8 2.6 2.5 2.1 1.8
China Italy USA France Spain Turkey Chile Iran Argentina India South Africa Australia Brazil Germany
86
World total
88,178
Table 4.2 Top countries ranked by total wine production in 2009 (Winetitles 2013)
Country 1 2 3 4 5 6 7 8 9 10 11 12 13
Wine production million litres
% of world
4,731 4,627 3,609 2,196 1,280 1,213 1,178 1,010 999 929 713 670 587
17.5 17.1 13.3 8.1 4.7 4.5 4.3 3.7 3.7 3.4 2.6 2.5 2.2
Italy France Spain USA China Argentina Australia Chile South Africa Germany Russia Romania Portugal World total
27,106
Table 4.3 Top countries ranked by total wine exports in 2009 (Winetitles 2013)
Country 1 2 3 4 5 6 7 8 9 10 11
Italy Spain France Australia Chile USA South Africa Germany Argentina Portugal New Zealand World total
Wine exports million litres
% of world
1,952 1,461 1,227 772 693 398 396 356 283 231 113
22.3 16.7 14.0 8.8 7.9 4.5 4.5 4.1 3.2 2.6 1.3
8,760
87
As seen in Table 4.1 and 4.2, the worldwide grape production was 88.178 billion tonnes and the total wine production was 27.106 billion in 2009. Around 33% of the total wine production was distributed internationally according to Table 4.3. Many businesses are involved in the wine supply chain in several countries around the world. Due to competition at the national and global levels, many wine producers have been ‘under enormous pressure’ to improve their performance during the last few years (Garcia et al. 2012; Lin 2013; Michalewicz, Michalewicz & Spitty 2011, p.74; Soosay, Fearne & Dent 2012). To cope with the challenge, ‘wine companies around the world are realising the importance of supply chains and the impact of their performance on the business’ (Garcia et al. 2012, p.284), hence the increasing significance of supply chain management. That is why several recent wine industry reports have highlighted the importance of the supply chain view of the industry (Australian Grape and Wine Authority 2012; Lin 2013). Improving the efficiency and effectiveness of wine supply chains becomes ‘a critical factor to remain competitive in a marketplace that is more and more global, and where competition is tougher and tougher’ (Garcia et al. 2012, p.284). One effective approach is to use analytical models. Some wine companies have applied analytical models to ‘optimise supply chain activities’ (Michalewicz, Michalewicz & Spitty 2011, p.74). These models can help wine companies minimise transportation and other logistics costs, improve resource allocation, improve the accuracy of demand forecast, create optimal harvest schedules, and optimise other business processes at the operational, tactical and strategic decision levels (Bohle, Maturana & Vera 2010; Michalewicz, Michalewicz & Spitty 2011; Moccia 2013). Before presenting the proposed model for the wine supply chain design, the next section elaborates a typical supply chain network in the wine industry. 4.3.2 A typical wine supply chain Recent studies have identified and illustrated to some extent a similar supply chain for wine products (Cholette & Venkat 2009; Christ 2014; Considine & Frankish 2014; Garcia et al. 2012; Petti et al. 2006; Soosay, Fearne & Dent 2012; Soosay, Fearne & Varsei 2014). A typical wine supply chain is shown in Figure 4.1, adapted from the studies by Garcia et al. (2012) and Soosay, Fearne and Varsei (2014). It consists of grape growers and other raw 88
materials providers (i.e. suppliers), wine producers at winery facilities where grapes are transformed into bulk wine, bottlers and packers at bottling plants where bottles or other containers are filled with bulk wine and packed, distributors at distribution centres, and retailers at demand points (i.e. customer zones) close to final consumers. These supply chain members are elaborated below. It should be noted that companies may use various transportation modes to transport grapes, glass bottles and other raw materials, bulk wine and finished products (e.g. wine bottles) from one location to another throughout the supply chain (Garcia et al. 2012). Transportation modes used by various companies include road, rail, water, air or a combination of them (Dekker, Bloemhof & Mallidis 2012).
Suppliers of grapes
Wine producers
Suppliers of yeast, glass bottles, closures, labels, etc.
Transportation modes
Retailers at demand points
Distributors
Bottlers / Packers
Figure 4.1 A typical wine supply chain (photos: courtesy of Google.com)
89
Suppliers (grape growers and suppliers of other raw materials): In a typical wine supply chain, as illustrated in Figure 4.1, grape growers produce and harvest grapes at vineyards, and then deliver grapes to wine producers which operate at wineries. A vineyard is a piece of land in which grapevines are grown, and it is often close to a winery. Grape growers are among the most important suppliers within the supply chain since the quality of grapes directly influences the quality of wine (i.e. the finished product of the supply chain) (Considine & Frankish 2014). In addition, there are other raw materials suppliers (except grape growers) that provide wine producers and bottlers with all the supplies needed for winemaking and bottling such as yeast, glass bottles and other container forms, corks, closures and labels (Garcia et al. 2012). Raw materials such as glass bottles, corks, closures and labels are sometimes referred to as ‘dry goods’ in the wine industry (Lin 2013). Wine producers: Wine producers receive grapes from grape growers, transform them into different types of wine, and send bulk wines to bottlers to produce the finished products of the supply chain. The term ‘bulk wine’ denotes wine in large containers before final packaging (International Organisation of Vine and Wine 2014). According to Grainger and Tattersall (2005), the main operational activities for wine production include destemming and crushing, must preparation, fermentation and temperature control, maceration, racking, pressing, blending and maturation. Bottlers/Packers: Bottlers/Packers are primarily responsible for the filling, packing and shipment of finished products. They receive bulk wines from wine producers, and fill up the different types of containers (such as glass bottles, tetra packs and bag-in-box) with various types of wine. They then pack finished products into cartons and pallets or other units used for distribution (Garcia et al. 2012). It should be noted that bottling plants can be located near to or far from wineries. For example, some Australian wine companies have bottling facilities in the UK and transport bulk wine from Australia to the UK so as to meet demand in the UK and other European countries (Lin 2013). Some others are ‘all-in-onelocation’ which means that their winery, bottling facility and distribution centre are geographically established at one location from which they transport wine products to several demand points. 90
Distributors: Distributors are responsible for all operations related to the transportation of finished products from plants to retailers or final consumers (Martínez-Salazar et al. 2014). Distributors receive cartons and pallets from bottling plants, store them, and then dispatch them to final demand points (i.e. retailers). Distributors may re-pack and re-label wine bottles so as to meet the specific requirements of retailers or other customers (Garcia et al. 2012). Retailers: They receive wine products from distributors and sell them to final consumers. Retailers include supermarkets, liquor stores, hotels, restaurants and clubs (Garcia et al. 2012). It should be noted that the actual number of supply chain members and the sequence of operational activities are case-specific in practice. For example, in Figure 4.1, it is assumed that a distributor acts as a supply chain member between a bottling plant and a set of retailers in order to facilitate product movement; however, some wine supply chains may consist of multi-tier distribution networks including distributors, importers, freight forwarders as well as wholesalers (Garcia et al. 2012). In contrast, some other wine companies may directly ship finished products from a bottling plant to retailers. Acknowledging the complexity of wine supply chain networks, this section has presented a typical wine supply chain in line with recent studies in the literature (Cholette & Venkat 2009; Christ 2014; Considine & Frankish 2014; Garcia et al. 2012; Petti et al. 2006; Soosay, Fearne & Dent 2012). Based on this typical wine supply chain network, the next section concerns the model development for sustainable wine supply chain design.
4.4
Model development
This section formulates the problem (i.e. sustainable wine supply chain design) as a multiobjective mixed-integer linear programming model. As discussed in Chapter 3, the proposed multi-objective model is based on two classical facility location models: capacitated fixed charge facility location and location-allocation (Drezner & Hamacher 2002; Melkote & Daskin 2001). However, these two models are extended in this research in order to represent sustainable wine supply chain design encompassing economic, environmental and social objectives. The proposed model aims to help a decision-maker 91
find a solution (i.e. a supply chain configuration) that would minimise supply chain cost, reduce emitted GHGs and enhance the social impact of the supply chain. Section 4.4.1 gives the formal statement of the problem. Section 4.4.2 introduces assumptions, indices, parameters and decision variables for the purpose of mathematical modelling. Section 4.4.3 and 4.4.4 develop formulations of objective functions and constraints respectively. 4.4.1 Problem statement Based on the typical wine supply chain illustrated in Figure 4.1, Figure 4.2 shows the network under investigation, decision variables and the flows of materials and products between the supply chain members. As seen in this figure, suppliers (s) supply raw materials (k) to wineries (p) and bottling plants (b), where bulk wines (j) and bottled wines (i) are produced respectively. As mentioned earlier, bulk wines can be packed in various packaging forms including glass bottles and bag-in-box (Colman & Päster 2009). Nonetheless, the term ‘bottled wine’ is used in this research as the representative because a glass bottle is still the most widely used packaging form in the industry (Colman & Päster 2009). Bottled wines (i) are shipped from bottling plants to distribution centres (w) and then to retailers at demand points (m). Wineries, bottling plants and distribution centres can be established with various capacities (i.e. sizes) denoted as c, a, and u respectively. Considering different sizes (e.g. small-, medium-, and large-sized plants) allows the decision-maker to fix the capacity bottleneck issue whenever the available capacity of a facility in the network (i.e. a winery, bottling plant or distribution centre) is not sufficient to fulfil demand. Each capacity has its own variable and fixed costs, hence the impact of economies of scale is addressed in the model. The shipment of materials or products may occur through different transportation modes (r) including road, rail, sea, and intermodal transport (e.g. rail-sea) with various associated costs and emissions. Intermodal transport refers to the use of a single transport load unit (e.g. a container) over at least two transport modes (Bauer, Bektas & Crainic 2009; Dekker, Bloemhof & Mallidis 2012). With appropriate coordination and planning (Dekker, Bloemhof & Mallidis 2012), intermodal transport could offer ‘an advanced platform for more efficient, reliable, flexible, and sustainable freight transportation’ (SteadieSeifi et al. 2014, p.1). Considering the case of rail-road intermodal transport, Dekker, Bloemhof and 92
Mallidis (2012, p.673) argue that this intermodal transport can ‘save thousands of truck kilometres in congestion sensitive areas and thus reduce the environmental impact’ and increase efficiency. In addition, given the fact that both trains and container ships emit less GHG emissions (per tonne kilometre) than trucks, the use of intermodal transport may minimise the total emissions generated in a supply chain (Bauer, Bektas & Crainic 2009; Bouchery & Fransoo 2015). Incorporation of intermodal transport into the proposed model may be of particular importance in countries like Australia, where a combination of various transportation modes potentially can be utilised by companies. Despite this fact, ‘road transport is the main mode of transport for the majority of commodities produced and/or consumed in Australia’ according to the Bureau of Infrastructure, Transport and Regional Economics (2014, p.4). Therefore, intermodal transport has received considerable attention recently in Australia due to the aforementioned advantages (Parsons & van Duyn 2014), and some major logistics companies (e.g. the Toll Group company) provide this freight service in Australia claiming it as a ‘cost effective freight transport solution’ (Toll Group n.d.). Considering the above context, the ultimate goal of the proposed multi-objective model is to find a set of optimised or balanced solutions in terms of the economic, environmental and social objectives through determining several interconnected decisions related to: •
the number and location of suppliers;
•
the number, location and capacity (i.e. size) of wineries, bottling plants and distribution centres;
•
quantities of raw materials, bulk wines and bottled wines transported throughout the supply chain;
•
allocation of demand to each facility in the supply chain; and,
•
utilised transportation modes.
93
Supplier sites
Wineries
Bottling plants
Distribution centres (DC)
Demand points
Supplier 1
Winery 1
Bottling 1
DC 1
Demand point 1
Supplier 2
Winery 2
Bottling 2
DC 2
Demand point 2
Supplier S
Winery P
Bottling B
DC W
Demand point M
sb Y skbr
Figure 4.2 Decision variables in a typical wine supply chain (X: binary variables, Y: flow variables)
94
4.4.2 Assumptions, indices, parameters and decision variables This section presents the assumptions, indices, parameters and decision variables considered and utilised for the purpose of mathematical modelling. Assumptions: The development of the model is based on the following assumptions. It should be noted that most of the assumptions are to some extent typical and used in several analytical modelling studies in the literature (Chopra & Meindl 2013; Fahimnia, Sarkis et al. 2013; Farahani, Drezner & Asgari 2009; Owen & Daskin 1998; Pishvaee, Torabi & Razmi 2012). •
The model is deterministic. The aggregate demand of a product type at a demand point is assumed to be known and must be satisfied annually. The aggregate demand of a product type is the total demand for the product ordered by several retailers or other outlets at a demand point or a customer zone (Simchi-Levi, Kaminsky & Simchi-Levi 2008).
•
The model is discrete. There is a known finite set of suppliers, wineries, bottling plants and distribution centres.
•
Limitations on supply and production capacities at supplier sites, wineries, bottling plants and distribution centres are known.
•
It is assumed that the different types of raw materials, bulk wines and bottled wines are known.
•
Transportation costs are known.
•
Raw materials can be supplied from more than one supplier, but the shipments of raw materials between suppliers are not allowed.
•
Bulk wines can be supplied from more than one winery, but the shipments of bulk wines between wineries are not allowed.
•
Bottled wines can be supplied from more than one bottling plant, but the shipments of bottled wines between bottling plants are not allowed. 95
•
Bottled wines can be supplied from more than one distribution centre, but the shipments of bottled wines between distribution centres are not allowed.
•
Direct shipments from suppliers to bottling plants (e.g. empty glass bottles) are allowed.
•
Environmental and social-related parameters are known.
•
Emission factors are used for calculating emissions. These factors are in CO2-e and the model considers emissions within transportation activities. As mentioned earlier, carbon emissions generated by transportation activities account for the considerable portion of total emissions in the wine supply chain (Cholette & Venkat 2009).
Indices: The following indices are used in the model. i:
index of bottled wine, i = 1, …, I
j:
index of bulk wine, j = 1, …, J
k:
index of raw material, k = 1, …, K
s:
index of supplier, s = 1, …, S
p:
index of winery location, p = 1, …, P
c:
index of winery capacity, c = 1, ..., C
b:
index of bottling plant location, b = 1, …, B
a:
index of bottling plant capacity, a = 1, ..., A
w:
index of distribution centre location, w = 1, …, W
u:
index of distribution centre capacity, u = 1, …, U
m:
index of demand point, m = 1, …, M
r:
index of transportation mode, r = 1, …, R
Parameters: Model parameters can be classified as economic, environmental and social. Parameters are ‘inputs’ to the model.
Economic-related parameters include:
96
l sks
capacity of supplier s to supply raw material k
v sks
unit purchasing cost of raw material k from supplier s
β kj
amount of raw material k needed to produce one unit bulk wine j
sp t skpr
unit transportation cost of raw material k from supplier s to winery p through transportation mode r
sb t skbr
unit transportation cost of raw material k from supplier s to bottling plant b through transportation mode r
f pcp
annual fixed-cost of building and equipping winery p with capacity c
πp
the maximum number of wineries to be utilised
p v pcj
unit production cost of bulk wine j in winery p with capacity c
l pcp
maximum production volume of winery p with capacity c
δ ki
amount of raw material k needed to produce one unit bottled wine i
θ ji
amount of bulk wine j needed to produce one unit bottled wine i
pb t pjbr
unit transportation cost of bulk wine j from winery p to bottling plant b through transportation mode r
f bab
annual fixed-cost of building and equipping bottling plant b with capacity a
b v bai
unit production cost of bottled wine i in bottling plant b with capacity a
l bab
maximum production volume of bottling plant b with capacity c
πb
the maximum number of bottling plants to be utilised
bw t biwr
unit transportation cost of bottled wine i from bottling plant b to distribution centre w through transportation mode r
f wuw
annual fixed-cost of building and equipping distribution centre w with capacity u
97
v ww
unit holding cost in distribution centre w
w lwu
maximum storage space in distribution centre w with capacity u
πw
the maximum number of distribution centres to be utilised
wm twimr
unit transportation cost of bottled wine i from distribution centre w to demand point m through transportation mode r
µ
maximum budget for annual fixed-cost of facilities used in the budgetary constraint of the model
D im
annual demand of bottled wine i at demand point m
Environmental-related parameters include the following:
sp e skpr
emitted CO2-e in the transportation of one unit raw material k from supplier s to winery p through transportation mode r (g CO2-e/unit)
sb e skbr
emitted CO2-e in the transportation of a unit raw material k from supplier s to bottling plant b through transportation mode r (g CO2-e/unit)
pb e pjbr
emitted CO2-e in the transportation of a unit bulk wine j from winery p to bottling plant b through transportation mode r (g CO2-e/unit)
bw ebiwr
emitted CO2-e in the transportation of a unit bottled wine i from bottling plant b to distribution centre w through transportation mode r (g CO2-e/unit)
wm ewimr
emitted CO2-e in the transportation of a unit bottled wine i from distribution centre w to demand point m through transportation mode r (g CO2-e /unit)
Social-related parameters are as follows:
98
m ss
social coefficient associated with the selection of supplier s
m pp
social coefficient associated with the selection of winery p
m bb
social coefficient associated with the selection of bottling plant b
mww
social coefficient associated with the selection of distribution centre w
Decision variables: Decision variables are ‘outputs’ of the model, and include:
X ss X
p pc
binary variable takes the value of 1 if supplier s is selected; 0 otherwise binary variable takes the value of 1 if winery p with capacity c and is established; 0 otherwise
X bab
binary variable takes the value of 1 if bottling plant b with capacity a is established; 0 otherwise
w X wu
binary variable takes the value of 1 if distribution centre w with capacity u is established; 0 otherwise
sp Y skpr
quantity of raw material k transported from supplier s to winery p through transportation mode r
sb Y skbr
quantity of raw material k transported from supplier s to bottling plant b through transportation mode r
Y pcjp
quantity of bulk wine j produced in winery p with capacity c
pb Y pjbr
quantity of bulk wine j shipped from winery p to bottling plant b through transportation mode r
Y baib
quantity of bottled wine i produced in bottling plant b with capacity a
bw Y biwr
quantity of bottled wine i shipped from bottling plant b to distribution centre w through transportation mode r
99
quantity of bottled wine i shipped from distribution centre w to demand point m through
wm Y wimr
transportation mode r
After describing the assumptions and defining indices, parameters, and decision variables, the next two sections present how they are used in formulating the objective functions and constraints of the model. 4.4.3 Formulation of objective functions The proposed model is multi-objective mixed integer programming, in which all functions are linear and decision variables are either binary (i.e. 0 or 1) or continuous. The model includes three objectives (i.e. economic, environmental and social). Economic objective: The economic objective identifies aforementioned decision variables in such a way that minimise the total supply chain fixed and variable cost within the defined scope. These cost components encompass the cost of purchasing raw materials from suppliers, the fixed-cost of establishing wineries, bottling plants and distribution centres, the variable cost of producing bulk wines and bottled wines and storage, and the total cost of transportation activities between suppliers, wineries, bottling plants, distribution centres and demand points. The formulations of these cost components are presented in three equations below. Equation (1) formulates the cost of purchasing raw materials (e.g. grapes) from suppliers (first term) and shipping them to wineries (second term) and to bottling plants (third term).
∑∑∑∑∑v
Z1
s
k
p
∑∑∑∑ t s
k
b
b
r
s sk
sp sp sp sb (Y skpr ) + ∑∑∑∑ t skpr Y skpr +Y skbr + s
k
p
r
sb skbr
sb Y skbr
(1)
r
The second equation presents the annual fixed cost of building and equipping wineries (first term), bottling plants (second term) and distribution centres (third term).
100
Z2 =
∑∑ f p
p pc
X
p pc
c
w + ∑∑ f bab X bab + ∑∑ f wuw X wu b
a
w
(2)
u
Equation (3) calculates the annual cost of producing bulk wines in wineries (first term) and shipping them to bottling plants (second term), producing bottled wines in bottling plants (third term) and shipping them to distribution centres (fourth term), and, storing bottled wines in distribution centres and shipping them to demand points (Term 5). p Z 3 = ∑∑∑v pcj Y pcip p
c
i
+ ∑∑∑∑ t b
i
w
+
bw biwr
Y
r
bw biwr
∑∑∑∑ t p
+
j
b
pb pjbr
pb Y pjbr
i
m
∑∑∑v b
r
∑∑∑∑ (v w
+
w w
+t
wm wimr
a
i
)Y
wm wimr
b bai
Y baib (3)
r
The sum of the above three linear cost components comprising the supply chain fixed and variable costs yields the total cost function of the model denoted as GF1 and presented in Equation (4). The total cost function (i.e. GF1 ) represents the economic performance of the supply chain. The economic objective of the proposed model is to minimise GF1 .
GF1 = Z 1 + Z 2 + Z 3
(4)
Environmental objective: As discussed earlier, modelling the environmental impact of supply chain design is restricted to capturing emitted GHGs (in CO2-e) within the transportation activities between suppliers, wineries, bottling plants, distribution centres and demand points. The amount of CO2-e emissions across the supply chain can be calculated by incorporating emission factors into the following four equations. It should be noted that the emission factors are known and available in the last version (Version 1.2) of the International Wine Carbon Calculator Protocol (FIVS 2008). Released in July 2008, the International Wine Carbon Calculator Protocol Version 1.2 is an emission accounting protocol developed through a partnership between four national wine industry bodies in four countries: the Winemakers’ Federation of Australia, the Wine Institute of 101
California (the United States), the New Zealand Winegrowers and South Africa’s Integrated Production of Wine program. Given the increased attention to climate change and GHG emissions in the wine industry, the protocol was aimed at providing wine industry specific, easy-to-use and free guidelines that would help wine companies calculate the GHG emissions of their facilities in order to develop strategies for reducing emissions and ensure the long-term viability and health of their businesses (FIVS 2008). Equation (5) calculates CO2-e emissions generated by the transportation of raw materials from suppliers to wineries (first term) and to bottling plants (second term), of bulk wines from wineries to bottling plants (third term), of bottled wines from bottling plants to distribution centres (fourth term), and of bottled wines from distribution centres to demand points (fifth term). Denoted as GF2 , Equation (5) represents the emission function of the model. The environmental objective of the proposed model is to minimise this function.
sp sp GF2 = ∑∑∑∑ e skpr Y skpr s
+
k
p
∑∑∑∑ e b
i
w
r
+
∑∑∑∑ e s
bw biwr
Y
r
bw biwr
+
k
b
sb pb pb Y skbr + ∑∑∑∑ e pjbr Y pjbr
sb skbr
r
∑∑∑∑ e
p
wm wimr
w
i
m
Y
j
b
r
wm wimr
(5)
r
Social objective: The social objective is to maximise the total social impact associated with the selection of suppliers, wineries, bottling plants and distribution centres in the network design model. To measure the total social impact, the method proposed by Varsei et al. (2014) and elaborated in Chapter 3 can be utilised, which is rooted in the analytical hierarchy process (Saaty 1990, 2008). A supply chain member is scored based on its performance in terms of a set of social categories, which are determined by a focal company (i.e. decision-maker). The scores, drawn from the proposed pairwise comparison scale (see Table 3.2 in Chapter 3), form the normalised social impact coefficient of the supply chain member. The coefficients are used to define the mathematical social objective function in Equation (6). For the purpose of defining the social objective in this research, two social categories are utilised for the selection of wineries, bottling plants and distribution centres: 102
unemployment and regional gross domestic product (regional GDP). ‘Employment’ and ‘regional GDP’ are among the key contributions of the wine industry, according to the comprehensive study by Anderson et al. (2009, pp.5,7). Furthermore, as one of the Australian government reports concludes, ‘the wine industry contributes socially by providing employment’ (LWA 2008, p.3). It can be argued, solely from the social sustainability perspective, that a wine company will improve the social impact of its supply chain network by selecting the supply chain members which are located in regions with high unemployment rates (i.e. impact on employment) and low regional GDP rates (i.e. investment in regions). For the purpose of this study, data on the unemployment and regional GDP rates can be found in the Australian Bureau of Statistics (2014) and the SGS Economics & Planning (2013), respectively. The selected categories are also among the social categories proposed by the Global Reporting Initiative (2013). For the selection of suppliers (e.g. grape growers), four primary categories proposed by the Global Reporting Initiative can be used: labour practices and decent work conditions, human rights, society, and product responsibility (Global Reporting Initiative 2013). A score is assigned to each supplier based on the degree of conformance, which is the variation between a focal company’s desirable social impact and the social impact of the supplier. The social objective denoted as GF3 is specified as follows in Equation (6).
GF3 = ∑ m ss X ss + ∑∑ m pp X s
p
p pc
c
w + ∑∑ m bb X bab + ∑∑ mww X wu b
a
w
(6)
u
The first term of Equation (6) calculates the social impact of selecting suppliers, and other terms denote the social impacts associated with the selection of wineries, bottling plants and distribution centres respectively. Given the above economic, environmental and social objectives, the proposed model could help decision-makers measure different performance indicators and find the most suitable set of decision variables aligned with their business strategies and sustainability goals in an 103
optimisation process. Similar to most (if not all) analytical models and optimisation problems, the proposed model is subject to a set of constraints which are presented in the next section. 4.4.4 Formulation of constraints The proposed sustainable wine supply chain design model is subject to the following linear constraints which can be imposed by a decision-maker. It should be mentioned that most of the following constraints are typical in supply chain design or strategic facility location models (Chaabane, Ramudhin & Paquet 2012; Chopra & Meindl 2013; Melkote & Daskin 2001; Owen & Daskin 1998). Constraints on supplier sites: Constraint set (7) ensures that given the selection of supplier s, the total amount of raw material k shipped from the supplier to wineries and bottling plants cannot exceed its capacity for supplying the raw material.
∑∑∑ (Y p
b
sp skpr
sb +Y skbr ) ≤ l sks X ss
∀s , k
(7)
r
Constraints on wineries: There are six sets of constraints related to wineries. Constraint set (8) enforces that there is enough amount of raw material k to produce bulk wine j at each winery, according to the bulk wine j bill of material (i.e. the combination of raw materials needed to produce bulk wine j).
∑∑ β Y kj
c
p pcj
j
sp = ∑∑Y skpr s
∀p , k
(8)
r
Constraint set (9) enforces the flow balance at wineries.
∑Y c
p pcj
pb = ∑∑Y pjbr ∀p , j b
(9)
r
104
Constraint set (10) is a capacity constraint expressing that the total production volume in a winery cannot exceed its capacity.
∑Y
p pcj
≤ l pcp X
p pc
∀p , c
(10)
j
Constraint set (11) allows winery p to be characterised with maximum one capacity when established.
∑X
p pc
≤ 1 ∀p
(11)
c
Constraint set (12) prohibits establishment of a winery when it does not produce any bulk wine.
X
p pc
≤ ∑Y pcjp
∀p , c
(12)
j
Using constraint set (13), a decision-maker can bound the maximum number of wineries to be utilised by π p .
∑∑ X p
p pc
≤π p
(13)
c
This constraint may also enable a decision-maker to analyse the impact of the number of open facilities on objective functions, decision variables, and eventually on the design of a supply chain (Melkote & Daskin 2001; Owen & Daskin 1998). Constraints on bottling plants: There are seven sets of constraints related to bottling plants. Constraint set (14) ensures that there is enough amount of bulk wine j for producing bottled wine i at each bottling plant, according to bottled wine i bill of material.
105
∑∑θ Y ji
a
pb = ∑∑Y pjbr
b bai
i
p
∀b , j
(14)
r
Similarly, constraint set (15) ensures that there is enough amount of raw material k for producing bottled wine i at each bottling plant, according to bottled wine i bill of material.
∑∑ δ a
sb Y baib = ∑∑Y skbr ∀b , i
(15)
ki
i
s
r
Constraint set (16) enforces the flow balance at bottling plants.
∑Y
b bai
a
bw = ∑∑Y biwr ∀b , i w
(16)
r
Constraint set (17) states that the total production volume of bottling plant b cannot exceed its capacity.
∑Y
≤ l bab X bab
b bai
∀b , a
(17)
i
Constraint set (18) allows bottling plant b to be characterised with maximum one capacity when established.
∑X
b ba
≤ 1 ∀b
(18)
a
Constraint set (19) prohibits establishment of a bottling plant when it does not manufacture any bottled wine.
X bab ≤ ∑Y baib
∀b , a
(19)
i
106
Constraint set (20) bounds the maximum number of utilised bottling plants by π b .
∑∑ X b
b ba
≤π b
(20)
a
Constraints on distribution centres: Four constraints on distribution centres are defined as follows. Constraint set (21) enforces the flow balance at distribution centres representing that the total amount transported from a distribution centre cannot exceed the quantity of products received.
∑∑Y m
wm wimr
r
bw = ∑∑Y biwr b
∀w , i
(21)
r
Constraint set (22) specifies that the total amount transported through a distribution centre cannot exceed its capacity (i.e. storage space).
∑∑∑Y i
m
wm wimr
w w ≤ lwu X wu
∀w ,u
(22)
r
Constraint set (23) states that distribution centre w is characterised by one capacity (i.e. size) if it is established.
∑X
w wu
≤ 1 ∀w
(23)
u
Constraint set (24) limits the maximum number of utilised distribution centres to π w .
∑∑ X w
w wu
≤π w
(24)
u
Budgetary constraint: In addition to the bounds on the number of opened wineries, bottling plants, and distribution centres, constraint (25) imposes a budgetary limitation such that the total sum over facilities’ annual fixed costs of building and equipping must not 107
exceed a given amount of µ determined by a decision-maker (i.e. focal company). These constraints also restrict establishing as many wineries, bottling plants and distribution centres as possible to yield the maximum possible social impact.
∑∑ f p
p pc
X
p pc
c
w + ∑∑ f bab X bab + ∑∑ f wuw X wu ≤µ b
a
w
(25)
u
Demand satisfaction constraint: Constraint set (26) ensures that the amount of bottled wine i transported to each demand point must cover the demand per annum.
∑∑Y w
wm wimr
= D im
∀m , i
(26)
r
Constraints on selection decision variables: Constraint sets (27)–(30) enforce that the selection decision variables are binary (i.e. 0 or 1). For instance, given the constraint set (29), a decision-maker ensures that each bottling is either closed, if it gets the value of 0, or opened otherwise.
X ss ∈ {0,1}
∀s
(27)
∈ {0,1}
∀p , c
(28)
X bab ∈ {0,1}
∀b , a
(29)
w X wu ∈ {0,1}
∀w ,u
(30)
X
p pc
Constraints on flow decision variables: Constraint sets (31)–(37) ensure that the flow decision variables are non-negative. sp Y skpr ≥0
∀s , k , p , r
(31)
sb Y skbr ≥0
∀s , k , b , r
(32)
108
Y pcjp ≥ 0
∀p , c , j
(33)
pb Y pjbr ≥0
∀p , j , b , r
(34)
Y baib ≥ 0
∀b , a, i
(35)
bw Y biwr ≥0
∀b , i ,w , r
(36)
wm ≥0 Y wimr
4.5
∀w , i , m , r
(37)
Summary and conclusion
The chapter has developed a unique model for sustainable supply chain design in the wine industry based on the framework proposed in Chapter 3. It has presented an overview of the wine industry and explained why the wine supply chain is under investigation in this thesis, followed by an illustration of a typical wine supply chain so as to outline its various participants. Finally, the chapter has defined the indices, parameters and decision variables used in the three linear objective functions (economic, environmental and social) and 37 linear constraints of the proposed model. As the specific model for supply chain design in the wine industry is virtually non-existent (Moccia 2013), the proposed model is among the first to address this important strategic aspect of wine supply chain management. The economic objective is to minimise the total supply chain cost including the cost of purchasing raw materials from suppliers, the variable cost of producing bulk wines and bottled wines and storage, the fixed-cost of establishing wineries, bottling plants and distribution centres, and the total cost of transportation activities between suppliers, wineries, bottling plants, distribution centres and demand points. The environmental objective is to minimise the CO2-e emissions generated by the transportation activities between suppliers, wineries, bottling plants, distribution centres and demand points. The social objective is aimed at maximising the total social impact associated with the selection of suppliers, wineries, bottling plants and distribution centres in the network design model. 109
Following the method proposed by Varsei et al. (2014), and rooted in the analytical hierarchy process (Saaty 2008), the supply chain members can be evaluated and scored based on a set of social categories determined by a focal company (i.e. decision-maker). In this research, it is considered that unemployment and regional GDP could be used as the social categories for the selection of wineries, bottling plants and distribution centres, while labour practices and decent work conditions, human rights, society, and product responsibility could be utilised for supplier (e.g. grape grower) selection. The multi-objective mixed integer model can help decision-makers analyse trade-offs between multiple objectives and find a set of optimised/balanced solutions, which in turn may assist in configuring sustainable wine supply chains. While this research does not intend to generalise the proposed model to other industries, the model may have the potential to be modified and used for the application of such models in other industries (e.g. in the soft drink bottling industry). Chapters 3 and 4 have developed an innovative framework and model in order to answer the central research question of this thesis. The next chapter concerns the case study of a major Australian wine company to test the model and illustrate its applicability. Chapter 5 also presents the associated numerical results and discusses some realistic scenarios and their implications for economic, environmental and social sustainability.
110
5
Case study for model application: Description, results and discussion
5.1
Introduction
This chapter aims to illustrate the applicability of the proposed model by conducting a case study. Relatively few analytical studies in the sustainable supply chain management literature have adopted case study research and used real data (Brandenburg et al. 2014; Seuring 2013; Seuring & Müller 2008). However, this research has conducted case study research and collected real data. Using the model proposed in Chapter 4, this chapter examines supply chain design of a major wine company in Australia in terms of a set of sustainability performance indicators. The proposed model in this thesis is multi-objective encompassing various economic, environmental and social objectives. Sophisticated solution methods have been developed in the modelling and optimisation literature for the multi-objective optimisation problems (Branke et al. 2008; Ehrgott 2005; Figueira, Greco & Ehrgott 2005). Among the solution methods, the ‘augmented ε-constraint method’ was employed to solve the model because of its ability to efficiently overcome the multi-objective optimisation problems (Mavrotas 2009; Mavrotas & Florios 2013; Zhang & Reimann 2014). The IBM ILOG CPLEX Optimiser software (often referred to as CPLEX) was utilised for the model deployment (i.e. computer programming and coding). It is a popular optimisation software package for the deployment of various mathematical models (Simchi-Levi, Kaminsky & Simchi-Levi 2008). This chapter is organised as follows. Section 5.2 elaborates case study research and its application for the analytical modelling studies, including the sustainable supply chain design and management models. Section 5.3 introduces the case and its characteristics (i.e. 111
case setting). It consists of two sub-sections in order to (i) describe the data collection process and the problem statement, and (ii) develop a customised multi-objective mixed integer model based on the model presented in Chapter 4. Section 5.4 presents the solution method for the model, reports its results (i.e. a set of trade-off solutions and associated supply chain design scenarios) and discusses their implications. Finally, the chapter’s summary and conclusion are outlined in Section 5.5.
5.2
Case study research for analytical modelling studies
In analytical modelling studies, one effective approach to the development and testing of mathematical models is to conduct case study research and collect real data and information about a phenomenon and its various associated parameters (Brandenburg et al. 2014; Seuring 2008, 2013). Such an approach can help researchers conduct an ‘empirically informed analytical study’ which has been recommended by several scholars over the recent years (Brandenburg et al. 2014; Choi & Guide 2012, p.507). Brandenburg et al. (2014, p.310) assert that ‘employing case study research is highly recommended not only for SSCM [sustainable supply chain management] models in particular but also for modelbased research in general’. From the research methodology perspective, an empirically informed analytical study (Choi & Guide 2012) may require a combination of the ‘analytical mathematical research methodology’ (Kotzab et al. 2006; Wacker 1998, p.361) as discussed earlier, and case study research (Meredith 1998; Stuart et al. 2002) as elaborated below. Case study research concerns an in-depth examination of a contemporary phenomenon (e.g. the supply chain design problem of a wine company) where the investigator has little control over the contextual conditions and investigates the phenomenon in its natural setting (McCutcheon & Meredith 1993; Yin 2014). Case research is explorative in nature (Quak & de Koster 2007) and typically uses multiple methods and tools for data collection from a number of data sources for triangulation (Meredith 1998). Quantitative and qualitative approaches can be used for data collection and multiple data sources can be utilised (Meredith 1998). In the field of operations management, these data sources include operational and financial data, interviews, business plans, plant layouts, maps, documents and websites, questionnaires, as well as the direct observations of the different aspects 112
relating to a phenomenon (Meredith 1998; Seuring 2008; Voss, Tsikriktsis & Frohlich 2002). Using multiple data sources in case study research can assist in investigating a phenomenon from several angles and getting a more accurate picture of it (McCutcheon & Meredith 1993). ‘The goal is to understand as fully as possible the phenomenon being studied’ (Meredith 1998, p.443), hence the validity and reliability of the research could be increased (Voss, Tsikriktsis & Frohlich 2002). Despite some traditional criticisms of case study research, such as the generalisability limitation (Seuring 2008; Stuart et al. 2002), conducting case research can considerably assist in bridging ‘the gap between operations management theory and practice’ (Flynn et al. 1990, p.251). However, relatively few analytical modelling studies have adopted case study research (Brandenburg et al. 2014; Seuring 2013; Seuring & Müller 2008) while there have been several calls for case research particularly in supply chain modelling studies over the recent years (Brandenburg et al. 2014; Chen, Zhang & Delaurentis 2014; Seuring 2013). It can be argued that building a relationship with a company (or multiple companies) to the point where you can conduct an in-depth investigation of its supply chain for an analytical modelling study may often be a very complex, intensive and timeconsuming process (McCutcheon & Meredith 1993; Watson et al. 2013). The main issue is that the process could add another complexity dimension to the analytical problems which already deal with the mathematical modelling and/or optimisation complexities. In a supply chain design study, typically the first challenge is collecting data from the different parts of a focal company. For example, the study may require collecting demand data from the sales group, production costs from the operations/production team, transportation modes and rates as well as information on suppliers from the logistics/supply chain group, and fixed plant costs (i.e. capital investment) from the finance department of a company (Watson et al. 2013). This data collection process is usually intensive and timeconsuming (Watson et al. 2013), nonetheless, collecting these necessary data sets provides ‘a validated starting point for the model’ (i.e. whether the mathematical model is formulated correctly) and also ‘a baseline’ to which researchers can compare a set of feasible scenarios (i.e. model output) and alternative supply chain configurations (Watson et al. 2013, p.20). Furthermore, incorporating all three sustainability dimensions into the 113
study and conducting sustainable supply chain design research requires additional effort to collect environmental and social-related performance indicators from the focal company and other organisations in its supply chain. The second challenge is estimating data for several new potential locations and product flow paths which may require collecting data from other organisations (Watson et al. 2013). For example, given the supply chain design study of a wine company, the fixed and variable costs of new bottling plants (i.e. new locations) as well as the transporting modes and rates in the various paths between new locations must be researched and calculated. Although obtaining data in a supply chain design study can be difficult and timeconsuming, this is often ‘the whole point of a study’ (Watson et al. 2013, p.21). The purpose of a supply chain design study is not only to evaluate the existing configuration of a supply chain, but to also compare it with new alternatives and business models so as to find the most suitable network for product movement in terms of the set of objectives determined by a decision-maker. Taking into account the abovementioned challenges, to date relatively few studies have employed the case study approach, collected real data and conducted an in-depth investigation of single or multiple cases in the literature on sustainable supply chain modelling and optimisation (Brandenburg et al. 2014; Seuring 2013; Seuring & Müller 2008; Walker et al. 2014). Seuring (2013) has recently reviewed analytical modelling approaches for sustainable supply chain management and found that only two papers out of thirty-six reviewed studies have used real data; ‘On most occasions, the illustration is made-up’ (Seuring 2013, p.1518) and scholars have utilised artificial numerical examples in order to test the models and illustrate their applicability. Because of this issue, some researchers such as Kotzab et al. (2006, p.434) argue that the majority of the analytical modelling studies in the literature have lost their ‘empirical foundations’. However, while this shortcoming is to some extent justifiable due to the complexities of the mathematical model development and optimisation, it can be argued that the analytical models which are grounded in case study research (i.e. empirically informed analytical studies) could provide more real insights into prevalent business
114
problems. In other words, conducting case study research could address the issue and add more empirical foundations to many analytical modelling studies. In addition, when following a structured case research methodology for analytical modelling studies, the validity and reliability of research are enhanced (Seuring 2008). Stuart et al. (2002) have proposed a five-stage process for case study research: (1) defining the research question, (2) selecting the case, (3) gathering data, (4) analysing data, and (5) disseminating the results. Seuring (2008, p.130) argues that the actual process is iterative and ‘might have to repeat several stages’. After collecting and analysing data, researchers should return to the previous stages, review them and take necessary actions (Seuring 2008). This research has followed the aforementioned five stages (Stuart et al. 2002) and collected real data and information in order to illustrate the applicability of the proposed model in Chapter 4, by examining the supply chain design of a major Australian wine company. This is elaborated in detail in the next sections.
5.3
Case setting
This section presents the case study setting. Sub-section 5.3.1 explains the case selection and the problem statement, and describes the data collection process. Sub-section 5.3.2 presents the modified version of the model elaborated in Chapter 4, in order to represent the case and its specific characteristics. It should be stated that the proposed model in Chapter 4 can be seen as a generic model and may need modification for actual implementation in practice. 5.3.1 Sampling, problem statement and data collection Purposeful sampling was conducted in order to find one information-rich wine company willing to take part in this research (Patton 2002). Purposeful sampling has been widely employed by researchers for the selection of information-rich cases related to the phenomenon of interest (Patton 2002). According to Patton (2002), who has provided a comprehensive explanation and discussion of purposeful sampling:
115
The logic and power of purposeful sampling lie in selecting information-rich cases for study in depth. Information-rich cases are those from which one can learn a great deal about issues of central importance to the purpose of the inquiry, thus the term purposeful sampling. Studying information-rich cases yields insights and in-depth understanding rather than empirical generalizations. (Patton 2002, p. 230)
An in-depth preliminary study was conducted to obtain an overview of the wine industry, wine companies, their supply chains and distribution activities in Australia. The researcher initially approached the South Australian Wine Industry Association, the Winemakers’ Federation of Australia and the Australian Wine Research Institute who then responded by inviting the researcher to present his research. The presentation at the Winemakers’ Federation of Australia was attended by some major Australian wine companies’ executives including plant managers as well as supply chain and distribution managers. Following the purposeful sampling and after careful consideration, some major Australian wine companies were contacted and invited to participate in this research. Given the hectic harvest season (i.e. around February in Australia), only one of them accepted to participate in this study and provide the information and data related to the mathematical model. There are around 2000 wine companies in Australia according to the published industry reports (Australian Grape and Wine Authority 2012; Lin 2013; Winetitles 2013). However, a few major companies are among the key players in the industry (Lin 2013). In fact, ‘the top 20 companies accounted for 78%’ of industry output in 2013 (Winetitles 2013, p.15). It could be argued that the major wine companies might be better information-rich cases. The majority of published industry reports often tend to focus on the major companies, their supply chains and their production and distribution practices (Lin 2013; Winetitles 2013), more published information could already be found related to these firms and their various supply chain-related practices. The case company is designated as ABC in this research in order to maintain its anonymity. The obtained data sets are not published to maintain the confidentiality as mentioned in the data collection protocol used for this research. Currently, ABC is based in New South Wales (NSW) and produces different types of red and white wines in a winery plant designated here as CLW and bottles them at the same location in a bottling plant denoted as CLB. Bottled wines are then transported by road (in most routes) and rail to Australia’s states so as to serve demand. These states include NSW, 116
Victoria (VIC), Tasmania (TAS), Queensland (QLD), South Australia (SA), Northern Territory (NT), and Western Australia (WA). The company commenced its wine business several decades ago as a small-sized winery, but the growing demand for its products transformed ABC into a major wine company in Australia. However, according to ABC’s supply chain manager, the company has expanded the winery and bottling facilities but has not changed its supply chain design and facility location following an ‘all-in-one-location strategy’. This means that, despite a change in demand, its current large-sized winery and bottling facility have been geographically in one location from which ABC transports finished products to several demand points. This issue has also motivated the research because firms should carefully examine and optimise their supply chain design once every few years or when there is a significant change in demand characteristics, according to Simchi-Levi (2010) and Chopra and Meindl (2013). Therefore, the scope of this case study is the winery-bottling plant-demand point network and the study aims to examine the possible business models in which bottling plant/s can be located in the location/s which are different from the existing one. This could help analyse how far the existing network is from some alternative optimised networks in terms of the economic, environmental and social objectives. This scope excludes suppliers (e.g. grape growers) in order to delimit the study. Realistically, it is assumed that ABC’s large-sized winery facility is fixed in the current location for the purpose of the mathematical model deployment. In other words, the optimisation process will not change the location of the winery during the model deployment. Wineries should be adjacent to their major grape suppliers (Christ 2014), therefore it may not seem feasible to change the winery location in the analysis. Five capital cities in Australia were determined as possible locations for new bottling plants (Sydney, Melbourne, Brisbane, Adelaide, and Perth) because these locations collectively accounted for over 95% of the company’s demand. Rail-sea intermodal and sea transportation modes were incorporated into the model where applicable in addition to the road and rail. Figure 5.1 illustrates the case company’s current supply chain within the scope as well as some potential network scenarios. In this figure, the solid and dash lines represent the 117
current and alternative supply chain design, respectively. Since shipment from CLB to TAS is operated through VIC and given the relatively low demand in TAS, demand in TAS was aggregated with demand in VIC in the analysis. Following the same logic, demand in SA and NT were aggregated for the purpose of the analysis.
118
Wineries
bulk wine
Bottling plants
bottled wine
Demand points
QLD
NSW
VIC, TAS
SA, NT
WA
Figure 5.1 ABC’s supply chain from winery to demand points
119
Given the generic model presented in Chapter 4 and its characteristics and assumptions, below specific indices, parameters and decision variables were used to mathematically formulate the sustainable supply chain design problem of ABC. Their definitions are as follows.
Indices:
i:
index of bottled wine, i = 1, …, I
j:
index of bulk wine, j = 1, …, J
p:
index of winery location, p = 1, …, P
b:
index of bottling plant location, b = 1, …, B
a:
index of bottling plant capacity, a = 1, ..., A
m:
index of demand point, m = 1, …, M
r:
index of transportation mode, r = 1, …, R
Parameters:
Economic-related parameters include:
l pp
maximum production volume of the winery p
θ ji
amount of bulk wine j needed to produce one unit bottled wine i
pb t pjbr
unit transportation cost of bulk wine j from winery p to bottling plant b through transportation mode r
f bab
annual fixed-cost of building and equipping bottling plant b with capacity a
b v bai
unit production cost of bottled wine i in bottling plant b with capacity a
120
l bab
maximum production volume of bottling plant b with capacity c
µ
maximum budget for annual fixed-cost of bottling facilities used in the budgetary constraint of the model
bm t bimr
unit transportation cost of bottled wine i from bottling plant b to demand point m through transportation mode r
D im
annual demand of bottled wine i at demand point m
Environmental-related parameters include:
pb e pjbr
emitted CO2-e in the transportation of a unit bulk wine j from winery p to bottling plant b through transportation mode r (g CO2-e/litre)
bm ebimr
emitted CO2-e in the transportation of one unit bottled wine i from bottling plant b to demand point m through transportation mode r (g CO2e/bottle)
Social-related parameters include:
m bb
social coefficient associated with the selection of bottling plant b
Decision variables:
X bab
binary variable takes the value of 1 if bottling plant b with capacity a is established; 0 otherwise
pb Y pjbr
quantity of bulk wine j shipped from winery p to bottling plant b through transportation mode r
Y baib
quantity of bottled wine i produced in bottling plant b with capacity a
121
bm Y bimr
quantity of bottled wine i produced at bottling plant b and shipped to demand point m through transportation mode r
It should be noted that since ABC operates the distribution activities at its bottling plant (i.e. in one geographical location), some specific parameters and decision variable were defined in order to modify the generic model presented in Chapter 4 in accordance with bm
bm bm ABC’s specific supply chain network. These include t bimr , ebimr and Y bimr .
To mathematically model the supply chain design of ABC, we consider that ABC has one fixed and established winery (p) that could supply bulk wines (j) to six candidate bottling plants (b) in six locations, where bottled wines (i) are made. Three different capacity options (a) are considered for each bottling plant in the model and this can help the decision-maker incorporate the economies of scale into the mathematical model. Bottled wines are transported to five demand points (m). Four transportation modes (r) exist including road, rail, sea, and intermodal rail-sea modes. The mathematical model aims to optimise the economic, environmental and social objectives (which will be defined in the next section) through examining and determining the interconnected decisions below: •
numbers, locations, and capacities of the bottling plants,
•
quantities of the transported bulk wines and bottled wines,
•
demand allocation to the bottling plants,
•
transportation modes to be utilised.
To determine the parameters of the model for the model deployment, information and data sets were gathered in two main steps: first, an iterative process of data collection and semistructured interviews with ABC; second, semi-structured interviews with a few other companies to collect other required data for the study (e.g. transportation cost in a new route and/or new mode in the alternative supply chain networks). These steps were carried out throughout 2014, and are elaborated below. In the first step of data collection, an iterative process of data collection and semistructured interviews with the supply chain manager of ABC was carried out. Throughout these interactions, an in-depth understanding of a large-sized wine company’s operations 122
and supply chain was gained. Given the abovementioned parameters of the model, collected information and data included: •
ABC’s supply chain design and the locations of the supply chain members
•
the capital and operational costs (i.e. fixed and variable costs per unit product as defined in the abovementioned parameters of the model)
•
bulk and bottled wine transportation costs per unit
•
transportation modes and routes utilised by the company to meet demand
•
the aggregate demand for several product types (red and white wines) in each customer zone.
One may argue that the demand or product aggregation in a supply chain design study may be a simplification. However, based on multiple supply chain design analyses, SimchiLevi, Kaminsky and Simchi-Levi (2008) have found that the impact of demand aggregation and product aggregation on the analysis accuracy are less than 0.05 and 0.03%, respectively. Following their findings and discussion, different types of wine products and also adjacent demand points (i.e. customer zones) were aggregated for the purpose of this study. Since ABC provided some of the required data sets for the model deployment, in the second step of data collection, semi-structured interviews with a few other companies were carried out to collect other required data for the study. These included iterative semistructured interviews with the managing director and operations manager of a smaller-sized wine company who provided the detailed fixed and variable costs of smaller-sized bottling plants. This information assisted with analysing the alternative supply chain networks and with incorporating the economies of scale into the model. In particular, it helped determine these parameters: f bab which is the annual fixed-cost of building and equipping bottling b plant b with capacity a), and v bai which is the unit production cost of bottled wine i in
bottling plant b with capacity a. ABC provided the transportation costs between the existing location of ABC and its customers. Since other transportation routes (e.g. from a potential bottling plant in
123
Adelaide to demand point in QLD) as well as modes (e.g. sea) were not utilised by the company, they were collected from three independent logistics companies which use sea, rail and road modes of transportation. All shipments were assumed to be in the full container load, which is a typical assumption in many supply chain design studies due to the strategic nature of the study (Watson et al. 2013). CO2-e emissions were calculated based on the emission factors provided in the International Wine Carbon Calculator Protocol Version 1.2 (FIVS 2008). These factors are shown in Table 5.1. Table 5.1 Emission factor in gram CO2-e per tonne kilometre, adapted from the Wine Carbon Calculator Protocol Version 1.2 (FIVS 2008)
Transportation Emission mode factor
Type
Road
115.0
Rail
26.4
heavy or articulated truck average locomotive
Sea
14.0
container ship
As elaborated in Chapter 4, two social categories associated with the location of bottling plants were considered in this study in order to incorporate the social impact of ABC’s supply chain network into the model: unemployment and regional gross domestic product (GDP). These two categories were used for the purpose of this study to determine social coefficients (i.e. m bb ) representing the social impact of locating a facility, and to illustrate the applicability of the model. It was assumed that they are equally weighted in the optimisation model of this study. As seen in Table 2.2 in Chapter 2, These categories are also mentioned within the social aspects proposed by the Global Reporting Initiative: employment as an underlying factor of Labour Practices and Decent Work, and local community as an underlying factor of Society (Global Reporting Initiative 2013). To calculate the social coefficients, Varsei et al. (2014) proposed the simplified version of the analytic hierarchy process method (Saaty 1990; Saaty & Ozdemir 2003) and its associated 1 to 9 pairwise comparison scale (see Table 3.2 in Chapter 3). The method was 124
adopted in order to score each bottling plant location in terms of two selected social categories, and to determine the associated social coefficients. A score was assigned to each bottling plant location based on the degree of conformance, which is the variation between ABC’s desirable social impact and the actual social impact of a candidate bottling plant influenced by its regional characteristics. A larger score means better social performance. Five states in Australia were selected where candidate bottling plants can be located near their capital cities. Table 5.2 presents the unemployment rate and regional GDP in these states based on the statistics published by the Australian Bureau of Statistics (2014) and the SGS Economics & Planning (2013) respectively. Table 5.2 shows the associated scores as well. The current bottling plant (i.e. CLB) is located in NSW. The scores in Table 5.2 were determined by the researcher (using scores presented in Table 3.2 in Chapter 3) for the purpose of this study based on the unemployment rate and the regional GDP of these states. Table 5.2 Scores for the selected social categories Candidate bottling plant location NSW VIC QLD SA WA
Unemployment Rate
Score
Regional GDP (% of total GDP)
Score
5.8 6.7 6.6 6.8 5
7 8 8 8 6
30.9 21.9 19.1 6.20 16.6
5 6 6 8 7
It can be seen that SA has priority over other locations due to its considerably lower regional GDP and slightly higher unemployment rate in comparison. Based on these scores, Table 5.3 presents the normalised social coefficients adjusted by the bottling plant capacity options. For example, the NSW associated scores for two selected categories are 7 and 5, and given the 1 to 9 scale, the normalised coefficient is approximately equal to 0.667 for a �7�9+5�9� 2
large-sized plant �
≅ 0.667�. Three capacity options were incorporated into the
model and it was realistically assumed that the larger plants may have proportionally greater social effects on locations and regions where plants are located.
125
Table 5.3 Normalised social coefficients Candidate bottling plant location NSW VIC QLD SA WA
Normalised social coefficients adjusted to capacity options Large-sized Medium-sized Small-sized plant plant plant 0.667 0.148 0.067 0.778 0.173 0.078 0.778 0.173 0.078 0.889 0.198 0.089 0.722 0.160 0.072
Using these coefficients (i.e. m bb ), the social dimension was incorporated into the proposed supply chain design model. As elaborated in Chapter 3, the social coefficients could influence the selection of integer (binary) variables in supply chain design optimisation. 5.3.2 Customised sustainable supply chain design model for the problem The customised model has the objectives as outlined below. Economic objective: The economic objective aims to minimise the total cost of building and equipping the bottling plants as well as distributing bulk and bottled wines between the winery, bottling plants and demand points. Eq. (38) formulates the annual fixed cost of building and equipping the bottling plants in the multiple locations.
Z 4 = ∑∑ f bab X bab b
(38)
a
Eq. (39) calculates the cost of transporting bulk wines from the winery to the bottling plants (first term), producing bottled wines in the bottling plants (second term), and transportation to the demand points (third term). pb pb Z 5 = ∑∑∑∑ t pjbr Y pjbr p
j
b
r
+
∑∑∑v b
a
b bai
i
126
Y baib
+
∑∑∑∑ t b
i
m
r
bm bimr
bm Y bimr
(39)
The summation of above linear cost components represents the total cost function of the model, which is denoted as CF1 and presented in Eq. (40). The economic objective of the proposed model is to minimise this function.
CF= Z4 + Z5 1
(40)
Environmental objective: Eq. (41) calculates the CO2-e emissions generated in the transportation of bulk wines from the winery to the bottling plants (first term), and in the transportation of bottled wines from the bottling plants to the demand points (second term). Denoted as CF2 , the formula represents the emission function. The environmental objective of the model is to minimise this function.
= CF2
∑∑∑∑ e p
j
b
r
pb bm bm Y pjbr + ∑∑∑∑ ebimr Y bimr
pb pjbr
b
i
m
(41)
r
Social objective: CF3 in Eq. (42) aims to measure the total social impact of the bottling plant locations. The social objective of the model is to maximise this function in order to enhance the social sustainability of the supply chain design.
CF3 = ∑∑ m bb X bab b
(42)
a
Following the constraints presented in Chapter 4, the customised model is subject to linear constraints as follows: Constraints on wineries: Constraint set (43) is a capacity constraint expressing that the total production volume in the winery cannot exceed its capacity.
∑∑∑Y j
b
pb pjbr
≤ l pp
∀p
(43)
r
Constraints on bottling plants: Constraint set (44) ensures that there is enough amount of bulk wine j for producing bottled wine i at each bottling plant.
127
∑∑θ Y ji
a
b bai
i
pb = ∑∑Y pjbr p
∀b , j
(44)
r
Constraint set (45) enforces the flow balance at the bottling plants.
∑Y
b bai
a
bm = ∑∑Y bimr ∀b , i m
(45)
r
Constraint set (46) states that the total quantity of bottled wines produced in a bottling plant cannot exceed its capacity.
∑Y
≤ l bab X bab
b bai
∀b , a
(46)
i
Constraint set (47) states that bottling plant b is characterised by one capacity (i.e. size) if it is established.
∑X
b ba
≤ 1 ∀b
(47)
a
Constraint set (48) prohibits establishment of a bottling plant when it does not produce any bottled wine.
X bab ≤ ∑Y baib
∀b , a
(48)
i
Constraint (49) enables a focal company to impose a budgetary limitation such that the total investment over the annual fixed costs of building and equipping bottling plants must not exceed a given amount of µ . This constraint also restricts establishing as many bottling plants as possible to yield the maximum possible social impact.
∑∑ f b
b ba
X bab ≤ µ
(49)
a
128
Demand satisfaction constraints: The constraint set (50) ensures that the amount of bottled wine i transported to each demand point must cover the demand.
∑∑Y b
bm bimr
= D im
∀i , m
(50)
r
Constraints on selection decision variables: Constraint set (51) enforces that the selection decision variables are binary.
X bab ∈ {0,1} ∀b , a
(51)
Constraints on flow decision variables: Constraint sets (52)-(54) ensure that the flow decision variables are non-negative.
pb Y pjbr ≥0
∀p , j , b , r
(52)
Y baib ≥ 0
∀b , a, i
(53)
bm Y bimr ≥0
∀b , i , m , r
(54)
The next section is concerned with the solution method for the above multi-objective mathematical model, followed by results and discussion.
5.4
Solution method, results and discussion
5.4.1 Solution method The sustainable supply chain design model presented in this research involves the economic, environmental and social objectives, therefore, a multi-objective optimisation method should be employed as the appropriate solution approach (Branke et al. 2008; Ehrgott 2005; Figueira, Greco & Ehrgott 2005). In general, many real-world problems 129
consist of multiple objectives which often conflict with each other (Zhang & Reimann 2014). In most cases, ‘there is no single optimal solution that simultaneously optimises all the objective functions’, hence ‘decision-makers are looking for the “most preferred” solution’ instead of the optimal solution (Mavrotas 2009, p.455). The most preferred solution is one of the so-called Pareto optimal solutions (Mavrotas 2009), which are often referred to as non-dominated, efficient, non-inferior or trade-off solutions in the modelling and optimisation literature (Zhou et al. 2011). The Pareto optimal solutions are the solutions that cannot be improved (either minimised or maximised) in one objective function without deterioration of their performance in at least another objective function (Mavrotas 2009). To solve a multi-objective model, a decisionmaker should find the Pareto optimal solutions first, and then select the most preferred solution based on objectives, priorities or preferences. Since there may be no single optimal solution, every Pareto optimal solution may have its own shortcomings in terms of the objective functions (i.e. trade-offs). To solve the multi-objective optimisation models and find the associated Pareto optimal solutions, researchers have developed some specific methods. According to Ehrgott (2005), Mavrotas (2009) and Zhang and Reimann (2014), the two most widely used methods are the weighted sum and ε-constraint. In the weighted sum method, a decision-maker weights each objective and then considers the summation of the weighted objectives as the new single-objective. In other words, a multi-objective model is mathematically formulated as a single-objective model (Ehrgott 2005). The weighted sum method may not work correctly for many multi-objective models, for example when objective functions correlate with each other (Branke et al. 2008; Ehrgott 2005; Figueira, Greco & Ehrgott 2005). Branke et al. (2008, p.12) state that ‘it is also possible that a small change in the weights may cause big differences in objective values’. Given these shortcomings, the ε-constraint method was introduced in 1971 and since then many studies have adopted this approach to multi-objective optimisation problems (Mavrotas 2009; Zhang & Reimann 2014). In the ε-constraint method, a decision-maker selects one of the objective functions and optimises this function while the others are incorporated as the additional constraints on a mathematical model (Branke et al. 2008). 130
Many scholars have successfully used the ε-constraint method to solve various mathematical models including linear programming and mixed-integer programming (Ehrgott 2005; Mavrotas 2009). Despite its advantages over the weighting method, the main drawback of the ε-constraint method is ‘production of weak efficient solutions’ (Zhang & Reimann 2014, p.16). Using the ε-constraint method, it is not guaranteed that all obtained solutions are Pareto optimal. In order to overcome this drawback, the method was improved by Mavrotas (2009) who introduced ‘the augmented ε-constraint method’ (Ramos, Gomes & Barbosa-Póvoa 2014; Zhang & Reimann 2014). Unlike the original version, the augmented ε-constraint method can generate only Pareto optimal solutions, hence the efficiency of the solution method is enhanced (Mavrotas 2009; Zhang & Reimann 2014). The augmented ε-constraint method was adopted in this research to obtain the Pareto optimal solutions. In the optimisation process, the economic objective was optimised while the environmental and social objectives were treated as the constraints of the mathematical model. Using this method to obtain the Pareto optimal solutions, a decision-maker can assess the trade-offs among the three objectives and select an appropriate solution (i.e. the most preferred). The augmented ε-constraint method consists of the following three steps, according to Ramos, Gomes and Barbosa-Póvoa (2014) and Mavrotas (2009). In the first step, the method generated the ‘pay-off table’. It is a table comprising the ranges of the Pareto optimal solutions (Mavrotas 2009). To generate this table, three rounds of the optimisation process were followed in order to generate three Pareto optimal solutions representing the ranges of the objective values. In the optimisation algorithm, it was realistically assumed that the economic objective gets the first priority, followed by the environmental and social objectives respectively. It can be argued that this prioritisation is more pragmatic and also in congruence with the definition of sustainable supply chain management as it aims to ‘maximize the supply chain profitability while at the same time minimizing the environmental impacts and maximizing the social well-being’ (Hassini, Surti & Searcy 2012, p.70). In the first round, the method optimised the economic objective function
131
* solely. Assume that CF1 = z1 was the result (i.e. the minimum supply chain cost was equal *
to z1 ). Then it optimised the environmental objective function, however, this time it * * considered CF1 = z1 as an additional constraint to the model. When CF2 = z2 was obtained
(i.e. the minimum supply chain emission), it optimised the social objective function by * * adding both CF1 = z1 and CF2 = z2 as the constraints. The method proceeded in the similar
way in the subsequent second and third rounds with different sequences. In the second round it started with the environmental objective, and in the third round it began with the social objective respectively. In the second step, the method determined the minimum and the maximum values of the environmental and social objective functions within the set of Pareto optimal solutions obtained in the previous step. This gave the differences between the maximum and minimum values of the environmental and social objective. Using these differences, the number and value of constraints for environmental and social objective functions were obtained. In the last step, the method solved the model iteratively to generate the Pareto optimal solutions by incorporating the determined environmental and social constraints into the model. Adopting this method, the next section discusses the results including the pay-off table and Pareto optimal solutions. 5.4.2 Results and discussion This section presents the Pareto optimal solutions for the problem, which are associated with some alternative supply chain design scenarios, and discusses the implications. The section aims to compare these scenarios with the supply chain design currently employed by ABC, in terms of the economic, environmental and social objectives. The IBM ILOG CPLEX Optimiser software was used for the computer programming and coding of the augmented ε-constraint optimisation method. As mentioned earlier, the IBM ILOG CPLEX Optimiser software is a popular optimisation software package for the computer programming and coding of various mathematical models including the mixed integer models. 132
Table 5.4 presents the pay-off table generated by the augmented ε-constraint method. The pay-off table shows the supply chain cost, emission and social impact in three scenarios (named as scenarios S2–S4) in which one objective function was given priority over two others following the augmented ε-constraint method (Mavrotas 2009; Zhang & Reimann 2014). Scenario S1 represents ABC’s existing supply chain design, as illustrated in Figure 5.2, in which ABC transports bottled wines from CLB to demand points through rail and road modes of transportation. Table 5.4 also presents the location and capacity of bottling plant/s in each scenario. Table 5.4 The pay-off table generated by the augmented ε-constraint method and existing scenario at ABC
Optimisation priority
Scenarios
minimum S2 cost function minimum S3 CO2-e function maximum S4 social function current S1 scenario
Supply chain cost (AUD per year)
Supply chain emission (tonnes of CO2-e per year)
Supply chain social impact
Bottling plant location (size)
37,529,611
2,215
0.778
Melbourne (largesized)
38,664,053
1,780
0.845
47,287,876
7,151
0.967
44,603,553
4,936
0.667
Melbourne (largesized) and Sydney (small-sized) Adelaide (large-sized) and Brisbane (smallsized) CLB (large-sized)
As seen in Table 5.4, ABC is required to implement scenario S2 and relocate the bottling plant to Melbourne if it aims to minimise the total supply chain cost (i.e. the economic objective). According to this scenario, ABC will transport bulk wines from CLW to Melbourne by rail, bottle in a new large-sized facility, and then transport bottled wines from Melbourne to demand points: NSW by rail, QLD by sea, SA by rail and WA by sea. This solution will provide ABC with a massive cost saving, equal to AUD 7.1 million per annum which accounts for 15.9% of the current scenario (i.e. scenario S1). In addition, the solution will improve the social objective by 14.3%. More interestingly, it will reduce 2721 tonnes of CO2-e emissions per annum which is equal to 55.1% of the current level.
133
Figure 5.2 An illustration of ABC’s current supply chain network (solid lines) and the alternative scenario S5 (dashed lines)
Scenario S3 was obtained when the CO2-e emissions was minimised (i.e. the environmental objective). Under this scenario, the company will establish two bottling plants: a largesized bottling plant in Melbourne and a small-sized plant in Sydney. The company will transport bulk wines to these facilities by rail. The Melbourne plant will serve demand in VIC as well as in SA and WA. The Sydney plant will serve demand in NSW and QLD. All shipments from bottling plants to demand points will be through sea transportation mode since its emission factor is 8.2 times and 1.9 times smaller than the road and rail modes respectively (see Table 5.1). Given this scenario, ABC will reduce supply chain emission and cost by 63.9% and 13.3% respectively, and improve its social impact about 26.7%. The main reason behind this significant reduction in CO2-e emission and cost in both scenario S2 and S3 is that the current bottling plant is located far from ABC’s major 134
demand points, which are Australia’s eastern capital cities. Therefore the company utilises more bottled wine transportation than bulk wine transportation throughout its supply chain network. According to Cholette and Venkat (2009), the bottled wine transportation accounts for the major share of carbon equivalent emissions in the wine supply chain because the bottled wine packages are bulky and heavy. According to their analysis, ‘wine itself comprises just half the weight and under 40% of the volume of a case of twelve 750 ml. glass bottles’ (Cholette & Venkat 2009, p.1402). Scenario S4 was obtained by maximising the supply chain social impact. Although this solution will improve the social performance of supply chain design significantly by 44.9%, the economic and environmental objectives will deteriorate by 6% and 44.8% respectively. In this scenario, ABC will open a large-sized plant in Adelaide to serve demand in all states except QLD, for which a small-sized bottling plant will be located in Brisbane. However, the distance between Adelaide and Melbourne (i.e. the nearest major demand point) is around 725 kilometres and the company will need to pay a high transportation cost associated with the product movement in this route if it relocates from CLB to Adelaide. The argument is valid for the Adelaide–Sydney route as well, where the distance is around 1410 kilometres. Incorporating the pay-off table into the optimisation process (i.e. the augmented εconstraint method), the Pareto optimal solutions were obtained. Table 5.5 contains 54 combinations of the environmental and social constraints which were considered to minimise the supply chain cost. As seen in Table 5.5, some combinations resulted in infeasible solutions and they are marked as ‘infeasible’ in Table 5.5. The term infeasible means that the optimisation process did not find any solution that could satisfy the associated environmental and social constraints as well as the model constraints, therefore, a decision-maker should select the most appropriate solution from the feasible solutions.
135
Table 5.5 The Pareto optimal solutions (cost in million AUD) Emission constraints, in tonnes of CO2-e
Social impact constraints
0.778
0.816
0.854
0.891
0.929
0.967
7151
6480
5808
5137
4466
3794
3122
2452
1780
$37.53
$37.53
$37.53
$37.53
$37.53
$37.53
$37.53
$37.53
$38.66
2215
2215
2215
2215
2215
2215
2215
2215
1780
0.778
0.778
0.778
0.778
0.778
0.778
0.778
0.778
0.845
$38.04
$38.04
$38.04
$38.04
$38.04
$38.04
$38.04
$38.04
2054
2054
2054
2054
2054
2054
2054
2054
0.856
0.856
0.856
0.856
0.856
0.856
0.856
0.856
$38.04
$38.04
$38.04
$38.04
$38.04
$38.04
$38.04
2054
2054
2054
2054
2054
2054
2054
0.856
0.856
0.856
0.856
0.856
0.856
0.856
0.856
$47.18
$49.05
$52.97
6994
6480
5808
infeasible
infeasible
infeasible
infeasible
infeasible
infeasible
0.956
0.956
0.956
$47.18
$49.05
$52.97
6994
6480
5808
infeasible
infeasible
infeasible
infeasible
infeasible
infeasible
$47.29
$49.47 infeasible
infeasible
infeasible
infeasible
infeasible
infeasible
infeasible
S6 0.956 S70.956 S8 0.956 7151 0.967
S4
6480 0.967
S9
S2
S5
S3
$38.66 1780 0.845
$38.04 2054
infeasible
Eight different scenarios were obtained (scenarios S2–S9) including three solutions discussed earlier (scenarios S2–S4). As seen in Table 5.5, trade-offs exist between the economic, environmental and social objectives when we consider any pair of the selected scenarios. In scenario S5, the company will establish a large-sized facility in Melbourne. Unlike scenario S3, a small-sized bottling plant is located in Brisbane in this scenario which could result in 1.3 % and 1.6% improvements in the social and economic objectives, respectively. However, the solution will be associated with 274 tonnes more CO2-e emissions in comparison. Similar to scenario S4, scenarios S6–S9 will improve the social objective due to the establishment of a large-sized plant in Adelaide. But, these scenarios will not be efficient in terms of the economic and environmental objectives and will considerably increase the supply chain cost and emission since the major demand points are located in VIC, NSW and QLD. Compared to ABC’s current business model, the results suggest that scenarios S3 and S5 will improve all three objectives. Therefore, the company should redesign its supply chain based on one of these scenarios. 136
While the rail-sea mode of transportation was incorporated into the model, none of the obtained solutions used this mode in the transportation routes. Given the results, it can be argued that the cost, emission and social impact advantages that the company will gain from relocating its bottling plant will be more preferable than keeping the existing bottling location open and utilising the rail-sea mode of transportation for the product movement. That is why the optimisation process did not select a solution in which the rail-sea mode of transportation was incorporated. The model was also run with a new setting in order to analyse the impacts of the prevalent freight transport practices on the objective functions. The supply chain practitioners in Australia often argue that the rail or sea modes should only be used in the long west–east routes, i.e. between WA (in west) and SA, VIC, NSW, QLD (in east). In one of the conducted interviews, a manger stated that ‘road dominates domestic movements less than 2,000 kilometres’ in Australia. This observation is reflected in the 2014 Australian Logistics Council report as well (Australian Logistics Council 2014). Although Australia’s major capital cities are situated on the coasts, the coastal shipping only accounts for about 17% of the domestic freight share. ‘Road transport is the main mode of transport for the majority of commodities produced and/or consumed in Australia’ according to the Bureau of Infrastructure, Transport and Regional Economics (2014, p.4). Given this fact, it was assumed in the new setting that ABC would transport bottled wines by road from the bottling plants to all demand points, except WA to which sea mode can be utilised (i.e. the distance is more than 2,000 kilometres). Except for this change, all other parameters of the model stayed the same. Table 5.6 presents the pay-off table associated with the new setting as elaborated above. Similar to scenario S2, scenario S10 was obtained when minimising the economic objective. Compared to scenario S2, scenario S10 will increase the supply chain cost and emission by about 3% and 69% respectively. This significant increase in the supply chain emissions is due to the difference between the rail and road emission factors (see Table 5.1). In contrast, scenario S10 is better than scenario S2 in terms of the social objective, since two bottling plants will be located given this solution. Minimising the environmental objective yielded scenario S11, and scenario S12 was obtained when maximising the social 137
objective. Given scenarios S10 and S11 in Table 5.6, the results suggest that ABC could improve all three objectives by redesigning its supply chain network even if it follows the prevalent road-dominant logistics practice in Australia.
Table 5.6 The pay-off table based on the prevalent freight transport practice in Australia
Optimisation priority
Scenarios
S10
S11
S12 S1
minimum cost function minimum CO2-e function maximum social function current scenario
Supply chain cost (AUD per year)
Supply chain emission (tonnes of CO2-e per year)
Supply chain social impact
38,661,673
3,739
0.856
38,968,456
3,332
0.845
56,314,991
30,384
0.967
44,603,553
4,936
0.667
Bottling plant location (size) Melbourne (largesized) and Brisbane (small-sized) Melbourne (largesized) and Sydney (small-sized) Adelaide (largesized) and Brisbane (small-sized) CLB (large-sized)
According to the results presented in this section, the company should reconfigure its current supply chain and the associated business model if it aims to design a more sustainable network. A number of solutions were presented which can be used as the basis for this decision-making. Taking into account one of the solutions, for example scenario S3, the company may significantly reduce around 3000 tonnes of carbon emissions per year. In addition, given this scenario, the results suggest that the case company will benefit from a new supply chain design in terms of the economic sustainability objective (see Table 5.4). Therefore, it can be argued that the business model driven by the existing supply chain design may not be efficient in terms of the economic, environmental and social objectives. ABC has grown from a being small-sized winery to a major Australian wine business due to a significant increase in demand; however, the company still uses almost the same supply chain design and post-production distribution it had when it was a small-sized 138
business. In other words, the current supply chain design may not be the most efficient and purposeful network that the company needs now and for the future. This reflects the Choi, Dooley and Rungtusanatham (2001, p.351) observation that ‘many supply networks emerge rather than result from purposeful design’. However, it appears that non-purposeful design might have negative economic, environmental and social implications. Many policymakers and scholars around the world have increasingly noticed the aforementioned issue and called for ‘reshaping value chains’ in order to address sustainability-related global problems such as climate change (Howard-Grenville et al. 2014, p.618). The design of supply chains strongly influences the amount of transportation for product movement, and ‘the transportation of products, components, raw materials, and people is a major consumer of energy, accounting for about one-quarter of total energy consumption in a developed country’ (Howard-Grenville et al. 2014, p.618). Using analytical modelling analyses and optimisation techniques, companies should examine their supply chain design once every few years (Chopra & Meindl 2013) or, similar to ABC’s case, when there is a significant change in demand (Simchi-Levi 2010), and optimise the design in terms of multiple sustainability performance indicators. In such analyses, the results would be case-specific; hence may not be generalisable (Varsei et al. 2014). In addition, it is not guaranteed to obtain a solution (i.e. a supply chain design scenario) in which every single objective is optimised given the possible trade-offs between various sustainability performance indicators (Varsei, Christ & Burritt 2015). However, the analyses could highlight the trade-offs and help decision-makers explicitly investigate the impacts of their strategic supply chain design decisions in terms of various sustainability indicators, and make informed decisions. Equipped with informed decisions in the case of trade-offs, companies may prefer to make a sacrifice in one objective in order to improve another performance indicator in line with specific strategic business goals. This may assist with creating the business cases for sustainability (Schaltegger, LüdekeFreund & Hansen 2011; Varsei et al. 2014).
139
5.5
Summary and conclusion
This chapter has presented a case study to illustrate the applicability of the model. Given the generic model introduced in Chapter 4, this chapter has made a customised model for the sustainable supply chain design of a wine company in Australia. The economic, environmental and social dimensions of sustainability have been incorporated in the mixed integer multi-objective model. Specifically, supply chain cost and CO2-e emission have been used as the measures of the economic and environmental objectives respectively. To address the social dimension, the study has used the method proposed by Varsei et al. (2014) to calculate unemployment and regional GDP as the selected social impact coefficients and incorporate them into the model. To solve the model, the augmented ε-constraint method was adopted which is an efficient solution method for multi-objective optimisation problems (Mavrotas 2009; Ramos, Gomes & Barbosa-Póvoa 2014; Zhang & Reimann 2014). The IBM ILOG CPLEX Optimiser software was utilised for computer programming and coding. The pay-off tables and several Pareto optimal solutions were presented, and the associated trade-offs were discussed. Despite the trade-offs, the results of this specific case study suggest that the redesign of the supply chain could improve some sustainability performance measures.
140
6 6.1
Conclusion
Introduction
Companies are increasingly forced by various stakeholders and institutional norms to consider the performance indicators pertaining to the environmental and social sustainability aspects, in addition to the traditional economic dimension (Carter & Easton 2011; Sarkis, Zhu & Lai 2011; Schaltegger & Burritt 2014; Soosay, Fearne & Varsei 2014; Varsei et al. 2014; Winter & Knemeyer 2013). Accordingly, many scholars have called for an integrated holistic approach that covers all three sustainability dimensions (Brandenburg et al. 2014; Etzion 2007; Seuring 2013; Seuring & Müller 2008; Tang, Cao & Schvaneveldt 2008; Walker et al. 2014; Winter & Knemeyer 2013). With the aim of addressing this frequent call, this study contributes to the sustainable supply chain management literature by offering a multidimensional framework for sustainable supply chain design that incorporates some key performance indicators with the economic, environmental and social dimensions (Varsei & Soosay 2013; Varsei, Soosay & Fahimnia 2013; Varsei et al. 2014). Given that the practical modelling efforts incorporating all three sustainability dimensions are ‘almost completely missing’ (Seuring 2013, p.1518) and ‘neglected’ in the literature (Brandenburg et al. 2014, p.308), the framework could help researchers as well as focal companies in various industries investigate potential trade-offs or interplay between multiple sustainability indicators, which in turn may assist in designing sustainable supply chains and ultimately in creating business cases for sustainability (Schaltegger & Wagner 2006). In doing so, this research offers significant theoretical contribution and implications for sustainable supply chain management (Schaltegger & Burritt 2014; Varsei & Soosay 2013; Varsei et al. 2014).
141
This study has been conducted in Australia as one of the top ten wine producers in the world (Winetitles 2014), and research in the links between the wine industry, sustainability and supply chain management is still in its infancy despite its significance (Christ & Burritt 2013; Garcia et al. 2012; Soosay, Fearne & Dent 2012; Szolnoki 2013). Using the proposed framework in the context of the wine industry, this research is among the first to develop a mathematical model for optimising sustainable wine supply chain design. It conducts a year-long case study in Australia in order to illustrate the applicability of the model. To solve the model, it proposes a multi-objective optimisation method which could overcome the complexity of such a model and find the Pareto optimal (i.e. trade-off) solutions (Mavrotas 2009; Zhang & Reimann 2014). The results from the case study helps to explicitly highlight the trade-offs or interconnections required when optimising sustainable supply chain design. Decision-makers in the wine industry in Australia and other wine producing countries as well as researchers can employ the framework, model and solution method to examine and optimise the wine supply chains in terms of a set of the economic, environmental and social objectives. The model and the proposed solution approach could also shed light on the application of the sustainable supply chain design models in other industries. The conclusion of this study is presented in this chapter. After outlining a summary of the study and discussion in Section 6.2, Section 6.3 discusses the contributions of the study and the associated managerial implications. The limitations of the study are acknowledged in Section 6.4, along with presenting several directions for future research. Section 6.5 makes concluding remarks.
6.2
A summary and discussion
This study was aimed at addressing a highlighted gap in the sustainable supply chain management literature (Brandenburg et al. 2014; Seuring 2013) and the study’s central research question: How can the economic, environmental and social dimensions of sustainability be incorporated simultaneously into the supply chain design optimisation model in order to help decision-makers develop a more sustainable supply chain?
142
The sustainable supply chain management literature has evolved to meet a need for the inclusion of various sustainability aspects in the traditional supply chain management area (Carter & Easton 2011; Soosay, Fearne & Varsei 2014; Winter & Knemeyer 2013). Many scholars from multiple academic communities (e.g. operations and supply chain management, industrial engineering, operations research and management science, systems analysis, computer science and optimisation, marketing, accounting, strategy) have contributed to this emerging area (Burritt & Schaltegger 2014; Guillén-Gosálbez & Grossmann 2010; Linton, Klassen & Jayaraman 2007; Markman & Krause 2014; Sarkis, Zhu & Lai 2011; Senge 2010; Soosay, Fearne & Dent 2012; Tang & Zhou 2012; Wu & Pagell 2011). Nevertheless, as discussed in Chapter 2, the literature is still scarce in embedding the economic, environmental and social dimensions simultaneously in supply chain management modelling studies, including the research on supply chain network design (Brandenburg et al. 2014; Chaabane, Ramudhin & Paquet 2012; Seuring 2013; Seuring & Müller 2008). Supply chain design or network design is ‘a strategic decision level problem that provides an optimal platform for the effective and efficient supply chain management’ (Tiwari et al. 2010, p.95). It typically involves several strategic decisions about the number, location and capacity of facilities (e.g. production plants, distribution centres), decisions about supplier selection, and decisions about the optimal flow of material/finished products from suppliers to the demand points (Chopra & Meindl 2013; Simchi-Levi, Kaminsky & Simchi-Levi 2008). Rooted in the pioneering book by Alfred Weber, Theory of the location of industries (Friedrich 1929), it has received considerable attention in academia and its application has proven to be highly important for a broad range of private and public companies (Drezner & Hamacher 2002; Farahani, Drezner & Asgari 2009; Melkote & Daskin 2001; Owen & Daskin 1998). Several companies in various industries, such as the beverage industry, emphasise such decisions to manage product movements from sources to demand points efficiently and/or responsively (Chopra & Meindl 2013; Shapiro 2007; Soosay, Fearne & Dent 2012). The efficient design of a supply chain in terms of even single economic performance measure (i.e. the supply chain cost) is often a challenge to scholars particularly in the fields 143
of operations management, operations research and management sciences (Drezner & Hamacher 2002; Farahani, Drezner & Asgari 2009; Klibi, Martel & Guitouni 2010; ReVelle, Eiselt & Daskin 2008). Yet, considering multiple environmental and/or social sustainability indicators may add more dimensions of complexity in terms of the modelling and optimisation analyses (Brandenburg et al. 2014; Tang & Zhou 2012). Nonetheless, the growing prevalence of the interrelated sustainability challenges and associated global problems such as climate change may necessitate considering multiple indicators and optimising supply chains accordingly (Howard-Grenville et al. 2014). This is in line with the well-known triple bottom line approach which aims to encourage companies to assess performance using the traditional goal of increasing profit as well as considering the associated social and environmental impacts (Elkington 1999). Following such an approach and in order to answer the study’s central research question, a multidimensional framework for sustainable supply chain design (i.e. the first contribution of this study) was proposed in Chapter 3. The main characteristic of the framework is that sustainability indicators—such as the social aspects which have been argued as difficult-toincorporate into the supply chain design optimisation model (Chaabane, Ramudhin & Paquet 2010, 2012)—could influence the selection decision variables (i.e. binary variables) of the supply chain design optimisation model, hence they can be incorporated in the models as the coefficients of the selection decision variables (Varsei et al. 2014). To quantify and determine the coefficients, a modified version of Saaty’s (1990) comparison scales was proposed (Varsei et al. 2014) which has been developed in the analytic hierarchy process literature (Barker & Zabinsky 2011; Govindan et al. 2015; Ho 2008; Saaty 2008; Sarkis 1998, 2003). The selection decision variables play a crucial role in supply chain design models particularly in the associated strategic location decisions. These variables in turn impact other integrated supply chain design decisions including the flow decision variables representing product movement throughout the supply chain (ReVelle, Eiselt & Daskin 2008; Simchi-Levi, Kaminsky & Simchi-Levi 2008). To delimit the study, supply chain cost, emission (CO2-e) and social impact were considered in the framework as the economic, environmental and social sustainability indicators respectively. Supply chain social impact may comprise any set of the Global Reporting Initiative’s four social categories (i.e. labour practices and decent work conditions, human 144
rights, society, and product responsibility) and the underlying aspects such as employment (Global Reporting Initiative 2013). The selection of social categories would be contextspecific. Chapter 4 used the framework to develop a novel integrated multi-objective model for optimising sustainable supply chain design in the context of the wine industry (i.e. the second contribution of this study). Despite the importance of the wine industry and its supply chain in several countries at the national and global levels (Garcia et al. 2012), it appears that the literature on wine supply chain design modelling and optimisation even with a single economic performance indicator (e.g. supply chain cost) is non-existent (Moccia 2013; Szolnoki 2013). The chapter presented the economic, environmental and social objective functions and constraints of the model in the form of mathematical formulations, along with the assumptions, indices, parameters and decision variables. Drawing from the literature on supply chain network design/facility location, two classical facility location models formed the basis of the proposed model, namely capacitated fixed charge facility location and location-allocation (Drezner & Hamacher 2002; ReVelle, Eiselt & Daskin 2008). To illustrate the applicability of the proposed model, Chapter 5 adopted case study research (McCutcheon & Meredith 1993; Stuart et al. 2002) in order to investigate the sustainable supply chain design of a major Australian wine company. Given the scope, the chapter defined and measured the supply chain cost, emission (CO2-e) and social impact (based on unemployment and regional GDP) for the economic, environmental and social objectives respectively. It should be noted that, in comparison, few mathematical modelling studies in the operations and supply chain management literature have conducted case study research and employed real data to develop and/or test a mathematical model (Brandenburg et al. 2014; Seuring 2013). Drawing from the literature on multi-objective optimisation (Branke et al. 2008; Ehrgott 2005; Figueira, Greco & Ehrgott 2005), the study used the augmented ε-constraint method (Mavrotas 2009; Zhang & Reimann 2014) to solve the model in the IBM ILOG CPLEX Optimiser software, which is a popular optimisation software package (Fahimnia, Sarkis et al. 2013; IBM 2012). A number of the resulted Pareto optimal (i.e. trade-off) solutions 145
were presented and the associated supply chain scenarios were discussed in terms of the economic, environmental and social objectives. By means of the case study, the chapter showed how multiple sustainability objectives can be examined and optimised in one integrated model which in turn highlights the trade-offs between the selected sustainability objective functions (i.e. the third contribution of this study). In addition, the specific case study found that the company’s current supply chain design is not efficient in terms of the defined objectives, hence the company may benefit by redesigning the supply chain. Supply chain design plays a key role in supply chain management (Lambert 2008) and has a significant impact on firms’ financial performance (Chaabane, Ramudhin & Paquet 2012). Given the strategic importance of the supply chain design decisions, companies in various industries should examine their supply networks once every few years (e.g. every five years), and optimise or redesign them accordingly (Chopra & Meindl 2013; SimchiLevi, Kaminsky & Simchi-Levi 2008). In line with the extant literature (Brandenburg et al. 2014), this research suggests that companies should go beyond a single-objective approach and optimise their supply networks in terms of multiple sustainability indicators (i.e. economic, environmental and social dimensions) by employing appropriate decision-making frameworks, models, and solution methods such as those proposed in this research. Using multiple objectives to reshape supply networks might provide firms with opportunities (e.g. new business models, innovative operations and distribution strategies) in which multidimensional sustainability performance measures would be improved (i.e. win-win-win). In this case, multi-objective supply chain design would help sustainable supply chain management meet its ultimate goals: … the management of supply chain operations, resources, information, and funds in order to maximize the supply chain profitability while at the same time minimizing the environmental impacts and maximizing the social well-being. (Hassini, Surti & Searcy 2012, p.70)
However, obtaining win-win-win solutions may not be possible. It is true to state that companies might find complex trade-offs between multiple competing objectives when 146
conducting the supply chain design analysis. Nonetheless, such analysis would be still beneficial because at least it could help decision-makers make informed and balanced decisions. In addition, sometimes it may be preferable for an organisation to make a small sacrifice in profit in order to achieve a meaningful improvement in an environmental or social performance indicator (Varsei, Christ & Burritt 2015). This in turn would help the organisation to be in line with a long-term sustainability strategy or a risk-mitigation strategy, given changing institutional norms as well as growing pressures from stakeholders (e.g. customers) for more sustainable practices (Carter & Easton 2011; Howard-Grenville et al. 2014; Soosay, Fearne & Dent 2012; Walker et al. 2014). In particular, optimising multi-objective supply chain design could assist wine companies in the development of innovative business models in order to address sustainability challenges such as water use and GHG emissions (Varsei, Christ & Burritt 2015). The global wine industry currently faces several areas of sustainability concern influencing wine companies’ profitability and even long-term viability (Christ & Burritt 2013; Soosay, Fearne & Dent 2012). That is why wine businesses have increasingly paid more attention to improvement initiatives such as better land use, waste management or less chemical use (Christ & Burritt 2013). In addition to these initiatives, the optimisation of wine supply chain design may address sustainability concerns but at the same time might also open up new strategic avenues for improvement of wine supply chain management which could unlock new streams for value creation. For example, according to the recent analysis by Varsei, Christ and Burritt (2015), redesigning a wine supply chain by changing the location of bottling plants can minimise supply chain cost and GHG emissions. The next section discusses the contributions of the study as well as the associated implications for practice.
147
6.3
Contributions and managerial implications
This research contributes to the literature by addressing the interface between supply chain management and sustainability. The research extends the literature by explicitly examining sustainable supply chain design (Varsei & Soosay 2013; Varsei et al. 2014), and, in particular, by making three key contributions. First, a generic multidimensional framework is proposed for sustainable supply chain design (Chapter 3). Second, an integrated multiobjective mathematical model is developed for optimising sustainable supply chain design in the wine industry (Chapter 4). Third, a real-world case study conducted in Australia shows how multiple sustainability objectives can be examined in a single analytical model. It highlights trade-offs between the objectives which should be considered when optimising supply chain design (Chapter 5). Moreover, a number of useful managerial implications could be derived from (a) the proposed framework for sustainable supply chain design, (b) the associated multi-objective mathematical model for the wine supply chain design, and (c) the specific application in the real-world case study using the multi-objective optimisation analysis. Three contributions and the associated managerial implications of the study are discussed as follows. 6.3.1 A novel framework Proposing a multidimensional framework for sustainable supply chain design (Varsei & Soosay 2013; Varsei, Soosay & Fahimnia 2013; Varsei et al. 2014), the study is among the first to address one of the highlighted gaps in the sustainable supply chain management literature (Brandenburg et al. 2014; Seuring 2013; Walker et al. 2014), and to incorporate some of the key aspects of all three dimensions of sustainability in a single overarching framework. The framework unpacks how various sustainability performance measures can be incorporated in the supply chain design analyses. In doing so, the framework offers new insight into how the current supply chain design (i.e. network design) models could be modified in order to simultaneously consider some performance indicators with all three sustainability dimensions. In addition, the framework might have the potential to be modified and used for developing other supply chain analytical models (e.g. tactical and
148
operational supply chain planning models) to include social, environmental and economic performance indicators. The framework can help researchers as well as practitioners examine supply chains (both already established networks or to-be-established supply chains) in various industries and analyse the alternative supply chain scenarios and their associated implications for a set of sustainability indicators (Varsei, Christ & Burritt 2015; Varsei et al. 2014). This may help address the pressing calls for ‘reshaping value chains’ from the management scholars and other stakeholders including the public, customers, and policy-makers around the world (Howard-Grenville et al. 2014, p.618). These calls aim to address some global sustainability challenges such as climate change, to which supply chains and the associated activities such as transportation contribute significantly (Howard-Grenville et al. 2014). As highlighted in one of the recent editorials of the Academy of Management Journal, climate change is ‘one of the greatest challenges we confront in the 21st century’ and ‘responses to it will fundamentally reshape many of the phenomena, interactions, and relationships that are of central concern to management scholars’ (Howard-Grenville et al. 2014, p.615). The framework is practical by providing a generic decision-making tool for sustainable supply chain design (Varsei et al. 2014). Given the specific real-world situations, the framework could help managers structure their supply chain design decision, and prioritise and determine the sustainability indicators. This may assist with focusing on the supply chain design optimisation opportunities that could deliver greater improvements in terms of the selected set of sustainability indicators, echoing a focal company’s sustainability goals, long-term strategic plans, as well as the specific characteristics of the industry within which the business operates. 6.3.2 A novel model Research on the modelling and optimisation of the wine supply chain design is virtually non-existent (Garcia et al. 2012; Moccia 2013; Szolnoki 2013). This study contributes to the literature not only by developing a wine supply chain design mathematical model, but also by incorporating the environmental and social sustainability indicators to the model (in addition to the economic measures such as the supply chain cost). Using the proposed 149
generic framework and building on the extant literature on facility location, sustainable supply chain design (Chaabane, Ramudhin & Paquet 2012; Simchi-Levi, Kaminsky & Simchi-Levi 2008), and wine supply chain management (Cholette & Venkat 2009; Christ & Burritt 2013; Moccia 2013; Soosay, Fearne & Dent 2012; Szolnoki 2013), the study proposes a novel multi-objective mathematical model for optimising sustainable supply chain design in the wine industry. Researchers can use the proposed mixed-integer programming model to develop and formulate the mathematical models with similar characteristics (e.g. similar objective functions and constraints) customised for other wine companies, or for other industries such as the beverage industry. For this reason, and to facilitate ‘putting sustainability into supply chain management’ (Beske & Seuring 2014, p.322), this research aimed to clearly elaborate all mathematical formulas in detail and in a less-complex way of introducing and interpreting mathematical formulas. Hence, it may be easier to understand by scholars from multiple academic communities as well as practitioners. Many operations and supply chain practitioners have used mathematical models to analyse complex business problems and make ‘informed’ decisions at the strategic, tactical and operational levels (Bertrand & Fransoo 2002; Kotzab et al. 2006; Shapiro 2007). Likewise, practitioners in the wine industry can employ the proposed model in optimising wine supply chain design in terms of a set of the economic, environmental and social objectives, and make the associated supply chain design decisions. The decisions include (a) number, location and capacities of facilities, (b) the selection of suppliers, and (c) the optimal flow of material or finished products from suppliers to the demand points (Simchi-Levi, Kaminsky & Simchi-Levi 2008). In addition, the model can be modified in order to be used by decision-makers in similar industries (e.g. in the soft drink bottling industry). 6.3.3 Optimising multidimensional sustainability objectives To illustrate the applicability of the proposed multi-objective model, this research is among few modelling studies in the sustainable supply chain management literature to conduct case study research and obtain real data (Brandenburg et al. 2014; Seuring 2013). The research involved conducting a year-long case study in Australia to investigate the 150
sustainable supply chain design of a major Australian wine company. Drawing from the literature on multi-objective optimisation (Branke et al. 2008; Ehrgott 2005; Figueira, Greco & Ehrgott 2005), it employs the augmented ε-constraint method (Mavrotas 2009; Zhang & Reimann 2014) in order to overcome the complexity of such a model, solve it and find the trade-off solutions. The case study examines how the supply chain configuration of a wine company is changed if it is optimised in terms of the economic, environmental and social indicators, suggests some alternative supply chain design scenarios, and discusses their implications. Given the multi-objective characteristic of the sustainable wine supply chain design model, it may not be guaranteed to obtain a solution (i.e. a supply chain design scenario) in which every single objective is optimised taking into account the complex interplay between various sustainability indicators. The case study highlights the trade-offs and the implications of the alternative wine supply chain design scenarios for various sustainability indicators. In addition, the case study may shed light on the application of such multiobjective optimisation models in other industries, which in turn may assist researchers and decision-makers in conducting similar sustainable supply chain design studies. These studies ultimately may assist in ‘reshaping value chains’ (Howard-Grenville et al. 2014, p.615), and creating business cases for sustainability (Schaltegger, Lüdeke-Freund & Hansen 2011).
6.4
Limitations and directions for future research
As with any study, research limitations exist. It is acknowledged that this study has limitations, but these limitations can provide several directions for future research. The first limitation is that the proposed framework and model for sustainable supply chain design only incorporates some specific sustainability indicators: supply chain cost, emission and social impact (based on unemployment and regional GDP). In line with the triple bottom line approach (Elkington 1999), the sustainability measures were collected drawing from the literature on sustainable supply chain management as well as existing guidelines for sustainability such as the Global Reporting Initiative (n.d.). It can be argued that a comprehensive framework and the associated model would ultimately require a wider consideration of economic, environmental and social performance measures. However, 151
given the complexity of supply chain design, its modelling and optimisation, as well as the scope of a potential research project, it may be impractical to take into account every single aspect of sustainability when developing a framework that aims to adopt an integrated approach (Hubbard 2009; Matos & Hall 2007; Varsei et al. 2014). Pagell and Wu (2009) discuss that truly sustainable supply chain networks may not exist, and related studies have their limitations to some extent. In addition, based on an empirical study, Matos and Hall (2007) advocate that having a broad integrated approach to examine interactions amongst a few environmental, economic and social dimensions might be better than applying deep, but disconnected expertise in each dimension (e.g. those sustainable supply chain management studies that only consider several environmental measures). This limitation, however, offers new avenues for future research. Future studies can extend this work and consider more or different sets of sustainability performance indicators for examining and optimising sustainable supply chain design (Varsei, Christ & Burritt 2015). For example, future research may consider different sets of the social indicators customised for specific cases. More social categories can be drawn from the comprehensive guidelines and standards such as the Global Reporting Initiative and Social Accountability 8000. Given the environmental dimension, it is true to state that the environmental sustainability challenges are not limited to GHG emissions. There are some other interlinked sustainability indicators such as water usage (Christ 2014; Christ & Burritt 2013) that can be incorporated and examined, hence many opportunities exist for future research in the development of such models. For instance, Varsei, Christ and Burritt (2015) have recently developed a mathematical model and examined the trade-offs between supply chain cost, water usage and carbon emissions when optimising global wine supply chain design. With regard to the multiple criteria decision-making tool used in the framework (i.e. a simplified version of the analytic hierarchy process), the proposed framework, although easily implementable due to its relatively straightforward and flexible characteristics (Varsei et al. 2014), may be further improved by other multiple criteria decision-making tools (i.e. other than the analytic hierarchy process). For example, the integration of other tools such as TOPSIS (i.e. the Technique for Order of Preference by Similarity to Ideal Solution) may possibly be pursued (Seuring 2013), as the integration of more sophisticated 152
hybrid approaches such as Fuzzy TOPSIS may assist in formulating uncertainty (Kannan, Jabbour & Jabbour 2014). In addition, a comparative analysis with other tools would be beneficial to identify further the strengths and weaknesses of the framework and eventual improvement (Varsei et al. 2014). Despite the important benefits of multiple criteria decision-making tools, including the analytic hierarchy process, we acknowledge the traditional criticism of such tools including the judgement bias as well as the subjective assignment of the comparison scales throughout the decision-making process (Bhutta & Huq 2002; Govindan et al. 2015; Ho, Xu & Dey 2010; Kannan, Jabbour & Jabbour 2014). As Govindan et al. (2015, p.70) conclude: AHP [the analytic hierarchy process] is not without its critics, however, so ample consideration should be given to its limitations. The growth and application of AHP may derive more from a convenience and simplification perspective rather than from a strong theoretical mathematical perspective.
Future research should also examine the interface between the economic, social and environmental sustainability and other quantitative supply chain models. As discussed earlier, supply chain design models address the strategic level decisions (Chopra & Meindl 2013). The framework has the potential to be modified and used for the development of other analytical tools such as the tactical and operational supply chain planning models so as to consider multidimensional sustainability indicators. Incorporating the social, environmental and economic performance measures into such models may provide new insights into the trade-offs between the indicators at the tactical and operations decision levels. The work by Ramos, Gomes and Barbosa-Póvoa (2014) is an example of such efforts. For planning a sustainable reverse logistics system, the authors consider the truck drivers’ working hours as the social measure, the distance between distribution facilities as the economic measure, and the associated transportation emission as the environmental measure (Ramos, Gomes & Barbosa-Póvoa 2014). Another set of limitations lies in the proposed mathematical model, and particularly in the assumptions of the model. As with any analytical research that develops a mathematical model, the proposed multi-objective mathematical model has limitations. Eykhoff (1974) defines a model as a representation of a real-world system using mathematical concepts and language. A ‘representation’ of a real-world context will likely require some 153
assumptions to be made, hence will automatically entail limitations (Michalewicz 2012). Given the inherited complexity of a supply chain system, researchers in the field consider several assumptions when proposing analytical models (Chaabane, Ramudhin & Paquet 2012; Pishvaee, Torabi & Razmi 2012; Ramos, Gomes & Barbosa-Póvoa 2014; Render, Stair & Hanna 2012). Likewise, some assumptions were made in this study for model development in Chapter 4 and 5. Although we acknowledge some of them as simplification, it should be noted that most of them are to some extent typical and considered by other similar supply chain design analytical modelling studies in the literature (Chopra & Meindl 2013; Farahani, Drezner & Asgari 2009; Owen & Daskin 1998; Pishvaee, Torabi & Razmi 2012). Future studies, however, may not consider one or some of the assumptions, and develop more complex mathematical models for sustainable wine supply chain design. For example, one of the assumptions is that the model is deterministic. While this should be a typical assumption in supply chain design models given the strategic nature of such studies (Chopra & Meindl 2013), future research can develop stochastic models for sustainable supply chain design in the wine industry or other industries in order to capture the uncertainty and parameter variability in a set of parameters (e.g. change in demand) and/or the supply chain risks such as supply disruption (Jabbarzadeh, Fahimnia & Seuring 2014; Santoso et al. 2005). In addition, the presented model is single-period to realistically address the case problem. When a real problem requires a dynamic approach, future research can extended the model to develop a multi-period model which can be solved using dynamic programming (Owen & Daskin 1998). The case study research methodology has the generalisability limitation (Seuring 2008; Stuart et al. 2002). It is acknowledged that the results of the case study may not be generalisable, and in fact the results of any analytical model deployment and application may vary depending on the case setting. However, the case study aimed to illustrate the applicability of the model and provided insights for sustainable wine supply chain design as discussed earlier. The model and the proposed efficient multi-objective optimisation solution method could be used in other similar case studies (e.g. in terms of problem size) with associated modifications, hence the ‘transferability’ of the case study (Cassell & 154
Symon 2011, p.635). Conducting a single or multiple case study research (McCutcheon & Meredith 1993; Stuart et al. 2002) will provide more insights and also enhance the validity of the model. It is also acknowledged that conducting sensitivity analyses considering multiple values for the parameters of the model (e.g. considering a range of transportation rates in a transportation mode) will add more insights into to the analysis and discussion of the case study. While this delimitation would provide another avenue for future research, it should be mentioned that the case study was aimed at representing the specific current real setting with associated exact parameters and focusing more on analysing its implications for the set of sustainability indicators. Moreover, while the generic model presented in Chapter 4 includes multiple wine supply chain tiers (i.e. supplier, winery, bottling plant, distribution centre and demand point), the specific model of the case company’s supply chain design (as elaborated in Chapter 5) excludes suppliers to delimit the study and the data collection process. Future research can embed suppliers to the optimisation analysis of the model during the deployment in the IBM ILOG CPLEX Optimiser software, or other optimisation software packages. This may provide additional insights into the multi-objective optimisation of sustainable wine supply chain design, particularly the interplay between selected sustainability indicators and the winery location decisions since considering suppliers may influence the decisions and subsequently the entire supply chain network. In addition, the case study used a relatively small-sized data set representing the parameters of the model. When optimising the larger-sized supply chain design problems which may involve more interlinked decision variables, parameters and/or constraints in which the computational times may become a concern (e.g. wine supply chain design in Europe which may include a greater number of suppliers, wineries, bottling plants, distributers and demand points), future research could use different solutions methods such as heuristic methods to overcome the complexity of the problem in terms of the size of the search space resulting in unmanageable solution time, conflicting objectives and constraints (Michalewicz 2012). They also can employ a range of available meta-heuristic algorithms such as the evolutionary algorithms (e.g. genetic algorithm) to solve the problems and find 155
the best possible solutions (i.e. perhaps near-optimum) since these optimisation methods have been used by several authors in the literature particularly to overcome the complexity of operational and tactical supply chain models (Altiparmak et al. 2009; Dehghanian & Mansour 2009; Yeh & Chuang 2011). However, when adopting the heuristic or metaheuristic methods for such large-sized complex problems, scholars should address the robustness of solutions and should acknowledge the associated limitations (Ehrgott 2005; Michalewicz 2012) which may be justifiable given the complexity inherited in analytical models. It is true to state that the development of more theoretical mathematical models and/or solution methods for sustainable supply chain design is a research avenue. But, given the emerging nature of the field, it can be argued that the focus should be given more to conducting case-based analytical research often named as empirically informed analytical study (Brandenburg et al. 2014; Choi & Guide 2012) in various companies, industries, and countries around the world. Effective modelling and solution approaches could then be pursued and adopted drawing from the literature on modelling and optimisation (i.e. overlapped operations research, management science and computer science disciplines) (Michalewicz 2012). Given the fact that relatively few modelling studies in the supply chain management literature have adopted case study research and used real data (Brandenburg et al. 2014; Seuring 2013), the latter ‘could broaden the scientific field of SSCM with regards to the focused industries’, hence is ‘highly recommended’ (Brandenburg et al. 2014, p.310). This pragmatic approach, as followed recently by Varsei, Christ and Burritt (2015), might also provide a broader impact across the entire business discipline as well as corporate thinking for more expeditious implementation of sustainability.
6.5
Concluding remarks
Companies should be equipped with new managerial decision-making tools in order to incorporate multiple sustainability performance indicators into decision-making in various areas including supply chain design and management (Varsei et al. 2014). The importance of developing these decision-making frameworks and tools has been highlighted by several scholars (Brandenburg et al. 2014; Chaabane, Ramudhin & Paquet 2012; Seuring 2013) 156
and by the 2012 United Nations Conference on Sustainable Development, which made its central theme on ‘integration and a balanced consideration of social, economic and environmental goals and objectives in both public and private decision-making’ (United Nations 2012). To address this need as well as the complexity associated with the sustainable supply chain design and management studies, scholars should make effort through conducting ‘needed’ multidisciplinary research (Markman & Krause 2014, p.1; Varsei, Christ & Burritt 2015). Winter and Knemeyer (2013, p.33) assert that researchers should ‘consider the applicability of a more multidisciplinary approach towards research activities in this stream’. This study may be an example of such effort. It sets the stage for addressing some highlighted gaps in the sustainable supply chain management literature. It is believed that this area still requires significant examination as these managerial decision-making tools may guide business sustainable progress (Varsei et al. 2014), hence ample opportunities exist for future research.
157
References
Abbasi, M & Nilsson, F 2012, ‘Themes and challenges in making supply chains environmentally sustainable’, Supply Chain Management: An International Journal, vol. 17, no. 5, pp. 517-530. Abdallah, T, Farhat, A, Diabat, A & Kennedy, S 2012, ‘Green supply chains with carbon trading and environmental sourcing: formulation and life cycle assessment’, Applied Mathematical Modelling, vol. 36, no. 9, pp. 4271-4285. Accenture 2012, Long-term growth, short-term differentiation and profits from sustainable products and services: a global survey of business executives, viewed 3 April 2012, . Ageron, B, Gunasekaran, A & Spalanzani, A 2012, ‘Sustainable supply management: an empirical study’, International Journal of Production Economics, vol. 140, no. 1, pp. 168-182. Ahi, P & Searcy, C 2013, ‘A comparative literature analysis of definitions for green and sustainable supply chain management’, Journal of Cleaner Production, vol. 52, no. 1, pp. 329-341. Ahmadi Javid, A & Azad, N 2010, ‘Incorporating location, routing and inventory decisions in supply chain network design’, Transportation Research Part E: Logistics and Transportation Review, vol. 46, no. 5, pp. 582-597. Ahuja, R, Magnanti, T & Orlin, J 1993, Network flows: theory, algorithms, and applications, Prentice Hall, Englewood Cliffs, New Jersey. Altiparmak, F, Gen, M, Lin, L & Karaoglan, I 2009, ‘A steady-state genetic algorithm for multi-product supply chain network design’, Computers and Industrial Engineering, vol. 56, no. 2, pp. 521-537.
158
Andersen, M & Skjoett-Larsen, T 2009, ‘Corporate social responsibility in global supply chains’, Supply Chain Management: An International Journal, vol. 14, no. 2, pp. 75-86. Anderson, K, Nelgen, S, Valenzuela, E & Wittwer, G 2009, The economic contributions and characteristics of grapes and wine in Australia’s wine regions, The University of Adelaide, Australia, viewed 23 January 2014, . Ansett, S 2007, ‘Mind the gap: a journey to sustainable supply chains’, Employee Responsibilities and Rights Journal, vol. 19, no. 4, pp. 295-303. Aras, G & Crowther, D 2009, ‘Corporate sustainability reporting: a study in disingenuity?’, Journal of Business Ethics, vol. 87, no. 1, pp. 279-288. Arnaout, J-PM & Maatouk, M 2010, ‘Optimization of quality and operational costs through improved scheduling of harvest operations’, International Transactions in Operational Research, vol. 17, no. 5, pp. 595-605. Ashby, A, Leat, M & Hudson-Smith, M 2012, ‘Making connections: a review of supply chain management and sustainability literature’, Supply Chain Management: An International Journal, vol. 17, no. 5, pp. 497-516. Australian Bureau of Statistics 2014, Labour force, no. 6202.0., Canberra, viewed 14 June 2014, . Australian Grape and Wine Authority 2012, The grape and wine research and development corporation strategic plan: 2012-17, The Australian Grape and Wine Authority (AGWA), Australia, viewed 18 November 2013, . Australian Logistics Council 2014, The economic significance of the Australian logistics industry, Australian Logistics Council, viewed 5 July 2014, . Awaysheh, A & Klassen, RD 2010, ‘The impact of supply chain structure on the use of supplier socially responsible practices’, International Journal of Operations & Production Management, vol. 30, no. 12, pp. 1246-1268.
159
Aylward, D 2004, ‘Innovation–export linkages within different cluster models: a case study from the Australian wine industry’, Prometheus, vol. 22, no. 4, pp. 423-437. Bai, C & Sarkis, J 2010a, ‘Integrating sustainability into supplier selection with grey system and rough set methodologies’, International Journal of Production Economics, vol. 124, no. 1, pp. 252-264. Bai, C & Sarkis, J 2010b, ‘Green supplier development: analytical evaluation using rough set theory’, Journal of Cleaner Production, vol. 18, no. 12, pp. 1200-1210. Bai, C, Sarkis, J & Wei, X 2010, ‘Addressing key sustainable supply chain management issues using rough set methodology’, Management Research Review, vol. 33, no. 12, pp. 1113-1127. Ballou, RH 2001, ‘Unresolved issues in supply chain network design’, Information Systems Frontiers, vol. 3, no. 4, pp. 417-426. Banerjee, SB 2003, ‘Who sustains whose development? Sustainable development and the reinvention of nature’, Organization Studies, vol. 24, no. 1, pp. 143-180. Banerjee, SB 2008, ‘Corporate social responsibility: the good, the bad and the ugly’, Critical Sociology, vol. 34, no. 1, pp. 51-79. Banerjee, SB 2011, ‘Embedding sustainability across the organization: a critical perspective’, Academy of Management Learning & Education, vol. 10, no. 4, pp. 719-731. Banks, G & Overton, J 2010, ‘Old world, new world, third world? Reconceptualising the worlds of wine’, Journal of Wine Research, vol. 21, no. 1, pp. 57-75. Barker, TJ & Zabinsky, ZB 2011, ‘A multicriteria decision making model for reverse logistics using analytical hierarchy process’, Omega, vol. 39, no. 5, pp. 558-573. Bauer, J, Bektas, T & Crainic, TG 2009, ‘Minimizing greenhouse gas emissions in intermodal freight transport: an application to rail service design’, Journal of Operational Research Society, vol. 61, no. 3, pp. 530-542. Baumol, WJ, Litan, RE & Schramm, CJ 2007, Good capitalism, bad capitalism, and the economics of growth and prosperity, Yale University Press, New Haven, CT. Bazaraa, MS, Sherali, HD & Jarvis, JJ 2010, Linear programming and network flows, 4th edn, Wiley, New York. 160
Beamon, B 1998, ‘Supply chain design and analysis: models and methods’, International Journal of Production Economics, vol. 55, no. 3, pp. 281-294. Beckman, S & Rosenfield, D 2008, Operations strategy: competing in the 21st century, McGraw-Hill/Irwin, Boston. Belegundu, AD & Chandrupatla, TR 1999, Optimization concepts and applications in engineering, Prentice Hall, Upper Saddle River, New Jersey. Benjaafar, S, Yanzhi, L & Daskin, M 2013, ‘Carbon footprint and the management of supply chains: insights from simple models’, Automation Science and Engineering, IEEE Transactions, vol. 10, no. 1, pp. 99-116. Berruto, R, Tortia, C & Gay, P 2006, ‘Wine bottling scheduling optimization’, Transactions of the American Society of Agricultural and Biological Engineers, vol. 46, no. 1, pp. 291-295. Bertrand, J & Fransoo, J 2002, ‘Operations management research methodologies using quantitative modeling’, International Journal of Operations & Production Management, vol. 22, no. 2, pp. 241-264. Beske, P & Seuring, S 2014, ‘Putting sustainability into supply chain management’, Supply Chain Management: An International Journal, vol. 19, no. 3, pp. 322-331. Bhutta, KS & Huq, F 2002, ‘Supplier selection problem: a comparison of the total cost of ownership and analytic hierarchy process approaches’, Supply Chain Management: An International Journal, vol. 7, no. 3, pp. 126-135. Bhutta, KS, Huq, F, Frazier, G & Mohamed, Z 2003, ‘An integrated location, production, distribution and investment model for a multinational corporation’, International Journal of Production Economics, vol. 86, no. 3, pp. 201-216. Bidhandi, HM, Mohd. Yusuff, R, Megat Ahmad, MMH & Abu Bakar, MR 2009, ‘Development of a new approach for deterministic supply chain network design’, European Journal of Operational Research, vol. 198, no. 1, pp. 121-128. Bohle, C, Maturana, S & Vera, J 2010, ‘A robust optimization approach to wine grape harvesting scheduling’, European Journal of Operational Research, vol. 200, no. 1, pp. 245-252. Bojarski, AD, Laínez, JM, Espuña, A & Puigjaner, L 2009, ‘Incorporating environmental impacts and regulations in a holistic supply chains modeling: an LCA approach’, Computers and Chemical Engineering, vol. 33, no. 10, pp. 1747-1759. 161
Boloori Arabani, A & Farahani, RZ 2012, ‘Facility location dynamics: An overview of classifications and applications’, Computers and Industrial Engineering, vol. 62, no. 1, pp. 408-420. Bouchery, Y & Fransoo, J 2015, ‘Cost, carbon emissions and modal shift in intermodal network design decisions’, International Journal of Production Economics, vol. 184, pp. 388-399. Bowen, HR 1953, Social responsibilities of the businessman, Harper & Brothers, New York. Brammer, S, Hoejmose, S & Millington, A 2011, Managing sustainable global supply chains: a systematic review of the body of knowledge, Network for Business Sustainability, viewed 11 May 2012, . Brandenburg, M, Govindan, K, Sarkis, J & Seuring, S 2014, ‘Quantitative models for sustainable supply chain management: developments and directions’, European Journal of Operational Research, vol. 233, no. 2, pp. 299-312. Brander, M 2012, Greenhouse gases, CO2, CO2e, and carbon: what do all these terms mean?, Ecometrica, viewed 19 May 2012, . Branke, J, Deb, K, Miettinen, K & Słowiński, R 2008, Multiobjective optimization: interactive and evolutionary approaches, Springer, Berlin, Heidelberg, Brent, AC 2005, ‘Integrating LCIA and LCM: evaluating environmental performances for supply chain management in South Africa’, Management of Environmental Quality, vol. 16, no. 2, pp. 130-142. Brundtland, GH 1987, Our common future: the world commission on environment and development, Oxford University Press, Oxford. Bureau of Infrastructure, Transport and Regional Economics 2014, Freight line 1: Australian freight transport overview, Bureau of Infrastructure, Transport and Regional Economics, Canberra, Australia, viewed 20 October 2014, . Burgess, K, Singh, PJ, Koroglu, R 2006, ‘Supply chain management: a structured literature review and implications for future research’, International Journal of Operations & Production Management, vol. 26, no. 7, pp.703-729. 162
Burritt, R & Schaltegger, S 2014, ‘Accounting towards sustainability in production and supply chains’, The British Accounting Review, vol. 46, no. 4, pp. 327-343. Carroll, AB 1979, ‘A three-dimensional conceptual model of corporate performance’, The Academy of Management Review, vol. 4, no. 4, pp. 497-505. Carroll, AB 1999, ‘Corporate social responsibility: evolution of definitional construct’, Business and Society, vol. 38, no. 3, pp. 268-295. Carroll, AB & Shabana, KM 2010, ‘The business case for corporate social responsibility: a review of concepts, research and practice’, International Journal of Management Reviews, vol. 12, no. 1, pp. 85-105. Carson, RT, Mitchell, RC, Hanemann, M, Kopp, RJ, Presser, S & Ruud, PA 2003, ‘Contingent valuation and lost passive use: damages from the Exxon Valdez oil spill’, Environmental and Resource Economics, vol. 25, no. 3, pp. 257-286. Carter, CR 2000, ‘Precursors of unethical behavior in global supplier management’, Journal of Supply Chain Management, vol. 36, no. 1, pp. 45-56. Carter, CR & Jennings, MM 2002, ‘Logistics social responsibility: an integrative framework’, Journal of Business Logistics, vol. 23, no. 1, pp. 145-180. Carter, CR & Jennings, MM 2004, ‘The role of purchasing in corporate social responsibility: a structural equation analysis’, Journal of Business Logistics, vol. 25, no. 1, pp. 145-186. Carter, CR 2005, ‘Purchasing social responsibility and firm performance: the key mediating roles of organizational learning and supplier performance’, International Journal of Physical Distribution & Logistics Management, vol. 35, no. 3, pp. 177194. Carter, CR & Stevens, CK 2007, ‘Electronic reverse auction configuration and its impact on buyer price and supplier perceptions of opportunism: a laboratory experiment’, Journal of Operations Management, vol. 25, no. 5, pp. 1035-1054. Carter, CR & Rogers, D 2008, ‘A framework of sustainable supply chain management: moving toward new theory’, International Journal of Physical Distribution & Logistics Management, vol. 38, no. 5, pp. 360-387.
163
Carter, CR & Easton, P 2011, ‘Sustainable supply chain management: evolution and future direction’, International Journal of Physical Distribution & Logistics Management, vol. 41, no. 1, pp. 46-62. Cassell, C & Symon, G 2011, ‘Assessing “good” qualitative research in the work psychology field: a narrative analysis’, Journal of Occupational & Organizational Psychology, vol. 84, no. 4, pp. 633-650. Cetinkaya, B 2011, Sustainable supply chain management: practical ideas for moving towards best practice, Springer, Heidelberg. Chaabane, A, Ramudhin, A, Paquet, M & Benkaddour, MA 2008, ‘An integrated logistics model for environmental conscious supply chain network design’, Proceedings of the 2008 Americas Conference on Information Systems, Association for Information Systems, pp. 175-183. Chaabane, A, Ramudhin, A & Paquet, M 2010, ‘Optimization and evaluation of sustainable supply chains’, Proceedings of the 8th International Conference of Modeling and Simulation, Institute of Electrical and Electronics Engineers (IEEE), pp. 1-10. Chaabane, A, Ramudhin, A & Paquet, M 2012, ‘Design of sustainable supply chains under the emission trading scheme’, International Journal of Production Economics, vol. 135, no. 1, pp. 37-49. Chen, C, Zhang, J & Delaurentis, T 2014, ‘Quality control in food supply chain management: an analytical model and case study of the adulterated milk incident in China’, International Journal of Production Economics, vol. 152, pp. 188-199. Chen, IJ & Paulraj, A 2004, ‘Towards a theory of supply chain management: the constructs and measurements’, Journal of Operations Management, vol. 22, no. 2, pp. 119150. Choi, TY, Dooley, KJ & Rungtusanatham, M 2001, ‘Supply networks and complex adaptive systems: control versus emergence’, Journal of Operations Management, vol. 19, no. 3, pp. 351-366. Choi, TY & Guide, VDR 2012, ‘Updates from the editors’, Journal of Operations Management, vol. 30, pp. 507-508. Cholette, S 2007, ‘A novel problem for a vintage technique: using mixed-integer programming to match wineries and distributors’, Interfaces, vol. 37, no. 3, pp. 231-239.
164
Cholette, S & Venkat, K 2009, ‘The energy and carbon intensity of wine distribution: a study of logistical options for delivering wine to consumers’, Journal of Cleaner Production, vol. 17, no. 16, pp. 1401-1413. Chopra, S & Meindl, P 2010, Supply chain management: strategy, planning, and operation, 4th edn, Prentice Hall, Upper Saddle River, New Jersey. Chopra, S & Meindl, P 2013, Supply chain management: strategy, planning, and operation, 5th edn, Pearson, Upper Saddle River, New Jersey. Christ, KL & Burritt, RL 2013, ‘Critical environmental concerns in wine production: an integrative review’, Journal of Cleaner Production, vol. 53, pp. 232-242. Christ, KL 2014, ‘Water management accounting and the wine supply chain: empirical evidence from Australia’, The British Accounting Review, vol. 46, no. 4, pp. 379396. Christmann, P 2000, ‘Effects of “best practices” of environmental management on cost advantage: the role of complementary assets’, Academy of Management Journal, vol. 43, no. 4, pp. 663-680. Christopher, M 1998, Logistics and supply chain management, 2nd edn, Financial Times Press, New York. Christopher, M 2005, Logistics and supply chain management, 3rd edn, Financial Times Prentice Hall, New York. Ciliberti, F, Pontrandolfo, P & Scozzi, B 2008, ‘Investigating corporate social responsibility in supply chains: a SME perspective’, Journal of Cleaner Production, vol. 16, no. 15, pp. 1579-1588. Collins, CM, Steg, L & Koning, MAS 2007, ‘Customers’ values, beliefs on sustainable corporate performance, and buying behavior’, Psychology & Marketing, vol. 24, no. 6, pp. 555-577. Colman, T & Päster, P 2009, ‘Red, white, and “green”: the cost of greenhouse gas emissions in the global wine trade’, Journal of Wine Research, vol. 20, no. 1, pp. 15-26. Considine, JA & Frankish, E 2014, A complete guide to quality in small-scale wine making, Academic Press, San Diego.
165
Cooper, MC, Lambert, DM & Pagh, JD 1997, ‘Supply chain management: more than a new name for logistics’, The International Journal of Logistics Management, vol. 8, no. 1, pp. 1-14. Corbett, CJ & Klassen, RD 2006, ‘Extending the horizons: environmental excellence as key to improving operations’, Manufacturing and Service Operations Management, vol. 8, no. 1, pp. 5-22. Correa, JR & Queyranne, M 2012, ‘Efficiency of equilibria in restricted uniform machine scheduling with total weighted completion time as social cost’, Naval Research Logistics, vol. 59, no. 5, pp. 384-395. Courville, S 2003, ‘Use of indicators to compare supply chains in the coffee industry’, Greener Management International, no. 43, pp. 93-105. Cousins, P, Lamming, R & Bowen, F 2004, ‘The role of risk in environment-related supplier initiatives’, International Journal of Operations & Production Management, vol. 24, no. 6, pp. 554-565. Current, J, Min, H & Schilling, D 1990, ‘Multiobjective analysis of facility location decisions’, European Journal of Operational Research, vol. 49, no. 3, pp. 295-307. Dahlsrud, A 2008, ‘How corporate social responsibility is defined: an analysis of 37 definitions’, Corporate Social Responsibility & Environmental Management, vol. 15, no. 1, pp. 1-13. Dantzig, GB 2002, ‘Linear programming’, Operations Research, vol. 50, no. 1, pp. 42-47. Daskin, M 1995, Network and discrete location: models, algorithms and applications, Wiley, New York. Dehghanian, F & Mansour, S 2009, ‘Designing sustainable recovery network of end-of-life products using genetic algorithm’, Resources, Conservation and Recycling, vol. 53, no. 10, pp. 559-570. Dekker, R, Bloemhof, J & Mallidis, I 2012, ‘Operations research for green logistics: an overview of aspects, issues, contributions and challenges’, European Journal of Operational Research, vol. 219, no. 3, pp. 671-679. Denyer, D & Neely, A 2004, ‘Introduction to special issue: Innovation and productivity performance in the UK’, International Journal of Management Reviews, vol. 5-6, nos. 3-4, pp. 131-135.
166
Dhaenens-Flipo, C & Finke, G 2001, ‘An integrated model for an industrial production– distribution problem’, IIE Transactions, vol. 33, no. 9, pp. 705-715. Dreyer, L, Hauschild, M & Schierbeck, J 2006, ‘A framework for social life cycle impact assessment’, International Journal of Life Cycle Assessment, vol. 11, no. 2, pp. 8897. Dreyer, L, Hauschild, M & Schierbeck, J 2010, ‘Characterisation of social impacts in LCAPart 1: development of indicators for labour rights’, International Journal of Life Cycle Assessment, vol. 15, no. 3, pp. 247-259. Drezner, Z & Hamacher, HW 2002, Facility location: applications and theory, Springer, New York. Du Pisani, JA 2006, ‘Sustainable development: historical roots of the concept’, Environmental Sciences, vol. 3, no. 2, pp. 83-96. Dullaert, W, Bräysy, O, Goetschalckx, M & Raa, B 2007, ‘Supply chain (re)design: support for managerial and policy decisions’, European Journal of Transport and Infrastructure Research, vol. 7, no. 2, pp. 73-91. Dunstall, S, Higgins, A & Sciberras, P 2008, Innovation opportunities in bulk wine transport, Technical report, The Commonwealth Scientific and Industrial Research Organisation (CSIRO) and Sinclair Knight Merz (SKM), Australia. Emission Factor Data Base n.d., Emission Factor Data Base, The Intergovernmental Panel on Climate Change, viewed 7 June 2012, . Ehrgott, M 2005, Multicriteria optimization, 2nd edn, Springer, Berlin. Elhedhli, S & Merrick, R 2012, ‘Green supply chain network design to reduce carbon emissions’, Transportation Research Part D: Transport and Environment, vol. 17, no. 5, pp. 370-379. Elkington, J 1999, Cannibals with forks: the triple bottom line of 21st century business, Capstone, Oxford. Ellen, P, Webb, D & Mohr, L 2006, ‘Building corporate associations: consumer attributions for corporate socially responsible programs’, Journal of the Academy of Marketing Science, vol. 34, no. 2, pp. 147-157.
167
Elliott, SR 2005, ‘Sustainability: an economic perspective’, Resources, Conservation and Recycling, vol. 44, no. 3, pp. 263-277. Epstein, M & Yuthas, K 2012, ‘Analyzing sustainability impacts’, Strategic Finance, vol. 93, no. 7, pp. 27-33. Erol, I, Sencer, S & Sari, R 2011, ‘A new fuzzy multi-criteria framework for measuring sustainability performance of a supply chain’, Ecological Economics, vol. 70, no. 6, pp. 1088-1100. Eskandarpour, M, Dejax, P, Miemczyk, J & Péton, O 2015, ‘Sustainable supply chain network design: an optimization-oriented review’, Omega, vol. 54, no. 1, pp. 11-32. Etzion, D 2007, ‘Research on organizations and the natural environment, 1992-present: a review’, Journal of Management, vol. 33, no. 4, pp. 637-664. Eykhoff, P 1974, System identification: parameter and state estimation, John Wiley & Sons, New York. Fahimnia, B 2011, Planning and optimisation of complex supply chains: an integrated approach for solving aggregate production-distribution planning problems, LAP Lambert Academic Publishing, Saarbrucken. Fahimnia, B, Farahani, RZ, Marian, R & Luong, L 2013, ‘A review and critique on integrated production–distribution planning models and techniques’, Journal of Manufacturing Systems, vol. 32, no. 1, pp. 1-19. Fahimnia, B, Reisi, M, Paksoy, T & Özceylan, E 2013, ‘The implications of carbon pricing in Australia: an industrial logistics planning case study’, Transportation Research Part D: Transport and Environment, vol. 18, pp. 78-85. Fahimnia, B, Sarkis, J, Dehghanian, F, Banihashemi, N & Rahman, S 2013, ‘The impact of carbon pricing on a closed-loop supply chain: an Australian case study’, Journal of Cleaner Production, vol. 59, pp. 210-225. Farahani, R, Drezner, Z & Asgari, N 2009, ‘Single facility location and relocation problem with time dependent weights and discrete planning horizon’, Annals of Operations Research, vol. 167, no. 1, pp. 353-368. Farahani, RZ, Asgari, N & Davarzani, H 2009, Supply chain and logistics in national, international and governmental environment, Springer, London.
168
Ferrer, J-C, Mac Cawley, A, Maturana, S, Toloza, S & Vera, J 2008, ‘An optimization approach for scheduling wine grape harvest operations’, International Journal of Production Economics, vol. 112, no. 2, pp. 985-999. Ferretti, I, Zanoni, S, Zavanella, L & Diana, A 2007, ‘Greening the aluminium supply chain’, International Journal of Production Economics, vol. 108, nos. 1-2, pp. 236245. Figueira, J, Greco, S & Ehrgott, M 2005, Multiple criteria decision analysis: state of the art surveys, Springer, New York. Fine, CH 2000, ‘Clockspeed-based strategies for supply chain design’, Production and Operations Management, vol. 9, no. 3, pp. 213-221. Fink, A 2010, Conducting research literature reviews: from the internet to paper, 3rd edn, SAGE Publications, USA. Finnveden, G, Hauschild, MZ, Ekvall, T, Guinée, J, Heijungs, R, Hellweg, S, Koehler, A, Pennington, D & Suh, S 2009, ‘Recent developments in life cycle assessment’, Journal of Environmental Management, vol. 91, no. 1, pp. 1-21. Fisher, ML 1997, ‘What is the right supply chain for your product?’, Harvard Business Review, vol. 75, pp. 105-117. FIVS 2008, International wine carbon calculator protocol version 1.2, FIVS, viewed 19 August 2014, . Fleischmann, M, Beullens, P, Bloemhof-Ruwaard, JM & Van Wassenhove, LN 2001, ‘The impact of product recovery on logistics network design’, Production and Operations Management, vol. 10, no. 2, pp. 156-173. Flynn, BB, Sakakibara, S, Schroeder, RG, Bates, KA & Flynn, EJ 1990, ‘Empirical research methods in operations management’, Journal of Operations Management, vol. 9, no. 2, pp. 250-284. Fonseca, A, McAllister, ML & Fitzpatrick, P 2013, ‘Sustainability reporting among mining corporations: a constructive critique of the GRI approach’, Journal of Cleaner Production, vol. 84, pp. 70-83.
169
Font, X, Tapper, R, Schwartz, K & Kornilaki, M 2008, ‘Sustainable supply chain management in tourism’, Business Strategy and the Environment, vol. 17, no. 4, pp. 260-271. Freeman, E 1984, Strategic management: a stakeholder approach, Pittman, Marshfield, Boston. Freeman, E, Harrison, JS & Wicks, AC 2008, Managing for stakeholders: survival, reputation, and success, Yale University Press, New Haven, USA. Friedman, M 1962, Capitalism and freedom, University of Chicago Press, Chicago. Friedman, M 1970, ‘The social responsibility of a business is to increase its profits’, The New York Times Magazine, 13 September, viewed 28 June 2013, . Friedrich, CJ 1929, Alfred Weber's Theory of the location of industries, University of Chicago Press, Chicago, Illinois, viewed 3 June 2012, . Frost, S & Burnett, M 2007, ‘Case study: the Apple iPod in China’, Corporate Social Responsibility and Environmental Management, vol. 14, no. 2, pp. 103-113. Frota, JQ, Bloemhof-Ruwaard, JM, van Nunen, JAEE & van Heck, E 2008, ‘Designing and evaluating sustainable logistics networks’, International Journal of Production Economics, vol. 111, no. 2, pp. 195-208. Garcia, FA, Marchetta, MG, Camargo, M, Morel, L & Forradellas, RQ 2012, ‘A framework for measuring logistics performance in the wine industry’, International Journal of Production Economics, vol. 135, no. 1, pp. 284-298. Ghidossi, R, Poupot, C, Thibon, C, Pons, A, Darriet, P, Riquier, L, De Revel, G & Mietton Peuchot, M 2012, ‘The influence of packaging on wine conservation’, Food Control, vol. 23, no. 2, pp. 302-311. Gibson, BJ, Mentzer, JT & Cook, RL 2005, ‘Supply chain management: the pursuit of a consensus definition’, Journal of Business Logistics, vol. 26, no. 2, pp. 17-25. Giddings, B, Hopwood, B & O’Brien, G 2002, ‘Environment, economy and society: fitting them together into sustainable development’, Sustainable Development, vol. 10, no. 4, pp. 187-196.
170
Gini, A 2005, Case studies in business ethics, 5th edn, Pearson Prentice Hall, Upper Saddle River, New Jersey. Global Reporting Initiative n.d., About GRI, Global Reporting Initiative (GRI), viewed 21 June 2013, . Global Reporting Initiative 2013, G4 Sustainability Reporting Guidelines, Global Reporting Initiative (GRI), viewed 12 December 2013, . Goedkoop, M & Spriensma, R 2001, The Eco-indicator 99: a damage oriented method for life cycle impact assessment, PRé Consultants, viewed 14 March 2013, . Goh, M, Lim, JYS & Meng, F 2007, ‘A stochastic model for risk management in global supply chain networks’, European Journal of Operational Research, vol. 182, no. 1, pp. 164-173. Golicic, SL & Smith, CD 2013, ‘A meta-analysis of environmentally sustainable supply chain management practices and firm performance’, Journal of Supply Chain Management, vol. 49, no. 2, pp. 78-95. González-Benito, J & González-Benito, Ó 2006, ‘The role of stakeholder pressure and managerial values in the implementation of environmental logistics practices’, International Journal of Production Research, vol. 44, no. 7, pp. 1353-1373. Gopalakrishnan, K, Yusuf, YY, Musa, A, Abubakar, T & Ambursa, HM 2012, ‘Sustainable supply chain management: a case study of British Aerospace (BAE) Systems’, International Journal of Production Economics, vol. 140, no. 1, pp. 193-203. Govindan, K, Rajendran, S, Sarkis, J & Murugesan, P 2015, ‘Multi criteria decision making approaches for green supplier evaluation and selection: a literature review’, Journal of Cleaner Production, vol. 98, pp. 66-83. Grainger, K & Tattersall, H 2005, Wine production: vine to bottle, Blackwell Publishing, Williston, USA. Green, J 2013, ‘Bangladesh 1,000 dead recall Triangle-to-Bhopal tragedies’, Bloomberg, 10 May, viewed 16 June 2013, . 171
Greenhouse, S 2013a, ‘Retailers are pressed on safety at factories’, The New York Times, 10 May, viewed 16 June 2013, . Greenhouse, S 2013b, ‘As firms line up on factories, Wal-Mart plans solo effort’, The New York Times, 14 May, viewed 8 June 2013, . Guillén-Gosálbez, G & Grossmann, I 2010, ‘A global optimization strategy for the environmentally conscious design of chemical supply chains under uncertainty in the damage assessment model’, Computers & Chemical Engineering, vol. 34, no. 1, pp. 42-58. Guinée, JB, Heijungs, R, Huppes, G, Zamagni, A, Masoni, P, Buonamici, R, Ekvall, T & Rydberg, T 2010, ‘Life cycle assessment: past, present, and future’, Environmental Science & Technology, vol. 45, no. 1, pp. 90-96. Gupta, S & Palsule-Desai, OD 2011, ‘Sustainable supply chain management: review and research opportunities’, IIMB Management Review, vol. 23, no. 4, pp. 234-245. Hagelaar, G & van der Vorst, J 2001, ‘Environmental supply chain management: using life cycle assessment to structure supply chains’, The International Food and Agribusiness Management Review, vol. 4, no. 4, pp. 399-412. Handfield, R & Nichols, E 1998, Introduction to supply chain management, Prentice-Hall, New Jersey. Handfield, R, Walton, SV, Sroufe, R & Melnyk, SA 2002, ‘Applying environmental criteria to supplier assessment: a study in the application of the analytical hierarchy process’, European Journal of Operational Research, vol. 141, no. 1, pp. 70-87. Hanson, JD, Melnyk, SA & Calantone, RJ 2004, ‘Core values and environmental management: a strong inference approach’, Greener Management International, vol. 46, no. 1, pp. 29-40. Hassini, E, Surti, C & Searcy, C 2012, ‘A literature review and a case study of sustainable supply chains with a focus on metrics’, International Journal of Production Economics, vol. 140, no. 1, pp. 69-82.
172
Hauschild, MZ, Dreyer, LC & Jørgensen, A 2008, ‘Assessing social impacts in a life cycle perspective: lessons learned’, CIRP Annals - Manufacturing Technology, vol. 57, no. 1, pp. 21-24. He, C, Gao, Y, Yang, S & Edwards, D 2004, ‘Optimization of the process for recovering caprolactam from wastewater in a pulsed-sieve-plate column using green design methodologies’, Journal of Loss Prevention in the Process Industries, vol. 17, no. 3, pp. 195-204. Ho, W 2008, ‘Integrated analytic hierarchy process and its applications: a literature review’, European Journal of Operational Research, vol. 186, no. 1, pp. 211-228. Ho, W, Xu, X & Dey, PK 2010, ‘Multi-criteria decision making approaches for supplier evaluation and selection: a literature review’, European Journal of Operational Research, vol. 202, no. 1, pp. 16-24. Holland, T & Smit, B 2010, ‘Climate change and the wine industry: current research themes and new directions’, Journal of Wine Research, vol. 21, nos. 2-3, pp. 125136. Howard-Grenville, J, Buckle, SJ, Hoskins, BJ & George, G 2014, ‘Climate change and management’, Academy of Management Journal, vol. 57, no. 3, pp. 615-623. Hsu, C & Hu, AH 2009, ‘Applying hazardous substance management to supplier selection using analytic network process’, Journal of Cleaner Production, vol. 17, no. 2, pp. 255-264. Hubbard, G 2009, ‘Measuring organizational performance: beyond the triple bottom line’, Business Strategy and the Environment, vol. 18, no. 3, pp. 177-191. Hugo, A & Pistikopoulos, EN 2005, ‘Environmentally conscious long-range planning and design of supply chain networks’, Journal of Cleaner Production, vol. 13, no. 15, pp. 1471-1491. Hult, GTM, Ketchen Jr, DJ, Cavusgil, ST & Calantone, RJ 2006, ‘Knowledge as a strategic resource in supply chains’, Journal of Operations Management, vol. 24, no. 5, pp. 458-475. Hunt, EK & Lautzenheiser, M 2011, History of economic thought, 3rd edn, M.E. Sharpe, Armonk, New York.
173
Hutchins, MJ & Sutherland, JW 2008, ‘An exploration of measures of social sustainability and their application to supply chain decisions’, Journal of Cleaner Production, vol. 16, no. 15, pp. 1688-1698. IBM 2012, CPLEX Optimizer, viewed 1 May 2012, . Ilgin, MA & Gupta, SM 2010, ‘Environmentally conscious manufacturing and product recovery (ECMPRO): a review of the state of the art’, Journal of Environmental Management, vol. 91, no. 3, pp. 563-591. Intergovernmental Panel on Climate Change, 2007, ‘Summary for Policymakers’, in S Solomon, D Qin, M Manning, Z Chen, M Marquis, KB Averyt, M Tignor and HL Miller (eds), Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. International Labour Organization n.d., Introduction to International Labour Standards, viewed 18 June 2012, International Organization for Standarization 2006a, ISO 14040 international standard, in Environmental Management – Life Cycle Assessment – Principles and Framework, International Organisation for Standardization, Geneva, Switzerland, viewed 10 September 2012, . International Organization for Standarization 2006b, ISO 14044 International Standard, in Environmental Management – Life Cycle Assessment – Requirements and Guidelines, International Organisation for Standardisation, Geneva, Switzerland, viewed 10 September 2012, . International Organization for Standarization 2009, Environmental management: the ISO 14000 family of International Standards, International Organization for Standardization, viewed 7 November 2012, . International Organisation of Vine and Wine 2014, Modify bulk wine definition in the harmonized system, The International Organisation of Vine and Wine, viewed 12 September 2014, . 174
Irwin, E & Clark, J 2006, The local costs and benefits of Walmart, The Ohio State University, viewed 21 December 2012, . Jabbarzadeh, A, Fahimnia, B & Seuring, S 2014, ‘Dynamic supply chain network design for the supply of blood in disasters: a robust model with real world application’, Transportation Research Part E: Logistics and Transportation Review, vol. 70, no. 1, pp. 225-244. Jägerskog, A, Clausen, TJ, Holmgren, T & Lexén, K 2014, Energy and water: the vital link for a sustainable future, Report No. 33. Stockholm International Water Institute (SIWI), Stockholm, viewed 3 October 2014, . Jang, Y-J, Jang, S-Y, Chang, B-M & Park, J 2002, ‘A combined model of network design and production/distribution planning for a supply network’, Computers and Industrial Engineering, vol. 43, nos. 1-2, pp. 263-281. Jayaraman, V, Guide, VDR & Srivastava, R 1999, ‘A closed-loop logistics model for remanufacturing’, Journal of the Operational Research Society, vol. 50, no. 5, pp. 497-508. Jorgensen, AL & Knudsen, JS 2006, ‘Sustainable competitiveness in global value chains: how do small Danish firms behave?’, Corporate Governance, vol. 6, no. 4, pp. 449462. Kaminsky, P, Simchi-Levi, D & Simchi-Levi, E 2004, Managing the supply chain: the definitive guide for the business professional, McGraw-Hill, New York. Kannan, D, Jabbour, ABLdS & Jabbour, CJC 2014, ‘Selecting green suppliers based on GSCM practices: using Fuzzy TOPSIS applied to a Brazilian electronics company’, European Journal of Operational Research, vol. 233, no. 2, pp. 432-447. Ketchen, DJ & Hult, GTM 2007, ‘Bridging organization theory and supply chain management: the case of best value supply chains’, Journal of Operations Management, vol. 25, no. 2, pp. 573-580. Khiewnavawongsa, S 2011, ‘Barriers to green supply chain implementation in the electronics industry’, PhD thesis, Purdue University, USA.
175
Kim, YH, Sting, FJ & Loch, CH 2014, ‘Top-down, bottom-up, or both? Toward an integrative perspective on operations strategy formation’, Journal of Operations Management, vol. 32, nos. 7-8, pp. 462-474. Klassen, RD & McLaughlin, CP 1996, ‘The impact of environmental management on firm performance’, Management Science, vol. 42, no. 8, pp. 1199-1214. Klassen, RD & Vachon, S 2003, ‘Collaboration and evaluation in the supply chain: the impact on plant-level environmental investment’, Production & Operations Management, vol. 12, no. 3, pp. 336-352. Klassen, RD & Vereecke, A 2012, ‘Social issues in supply chains: capabilities link responsibility, risk (opportunity), and performance’, International Journal of Production Economics, vol. 140, no. 1, pp. 103-115. Kleindorfer, PR, Singhal, K & Van Wassenhove, LN 2005, ‘Sustainable operations management’, Production and Operations Management, vol. 14, no. 4, pp. 482492. Klibi, W, Martel, A & Guitouni, A 2010, ‘The design of robust value-creating supply chain networks: a critical review’, European Journal of Operational Research, vol. 203, no. 2, pp. 283-293. Koplin, J, Seuring, S & Mesterharm, M 2007, ‘Incorporating sustainability into supply management in the automotive industry – the case of the Volkswagen AG’, Journal of Cleaner Production, vol. 15, nos. 11-12, pp. 1053-1062. Kotzab, H, Seuring, S, Müller, M & Reiner, G 2006, Research methodologies in supply chain management, Springer, New York Krikke, H, Bloemhof-Ruwaard, J & Van Wassenhove, LN 2003, ‘Concurrent product and closed-loop supply chain design with an application to refrigerators’, International Journal of Production Research, vol. 41, no. 16, pp. 3689-3719. La Londe, BJ 1997, ‘Supply chain management: myth or reality?’, Supply Chain Management Review, vol. 1, pp. 6-7. Lambert, DM, Cooper, MC & Pagh, JD 1998, ‘Supply chain management: implementation issues and research opportunities’, The International Journal of Logistics Management, vol. 9, no. 2, pp. 1-20. Lambert, DM, Stock, JR & Ellram, LM 1998, Fundamentals of logistics management, Irwin/McGraw-Hill, Boston. 176
Lambert, DM & Cooper, MC 2000, ‘Issues in supply chain management’, Industrial Marketing Management, vol. 29, no. 1, pp. 65-83. Lambert, DM 2008, Supply chain management: processes, partnerships, performance, 3rd edn, Supply Chain Management Institute, Sarasota, Florida. Land and Water Australia 2008, Signposts for Australian agriculture: the Australian wine industry, The National Land & Water Resources Audit (LWA), report No. PN21914, Australia, viewed 9 March 2014, . Larson, PD, Poist, RF & Halldórsson, Á 2007, ‘Perspectives on logistics vs. SCM: a survey of SCM professionals’, Journal of Business Logistics, vol. 28, no. 1, pp. 1-24. Lee, HL 2004, ‘The triple-A supply chain’, Harvard Business Review, vol. 82, no. 10, pp. 1-12. Lee, M-DP 2008, ‘A review of the theories of corporate social responsibility: its evolutionary path and the road ahead’, International Journal of Management Reviews, vol. 10, no. 1, pp. 53-73. Levitt, T 1958, ‘The dangers of social responsibility’, Harvard Business Review, vol. 36, no. 5, pp. 41-50. Lin, R 2013, Wine Production in Australia, Industry research report no: C1214, IBISWorld, Australia, viewed 24 October 2013, . Lin, SS & Juang, YS 2008, ‘Selecting green suppliers with analytic hierarchy process for biotechnology industry’, Operations and Supply Chain Management, vol. 1, no. 2, pp. 115-129. Linton, JD, Klassen, R & Jayaraman, V 2007, ‘Sustainable supply chains: an introduction’, Journal of Operations Management, vol. 25, no. 6, pp. 1075-1082. Lowenthal, D 2009, George Perkins Marsh: prophet of conservation, University of Washington Press, Seattle. Lu, LYY, Wu, CH & Kuo, TC 2007, ‘Environmental principles applicable to green supplier evaluation by using multi-objective decision analysis’, International Journal of Production Research, vol. 45, nos. 18-19, pp. 4317-4331.
177
Lumley, S & Armstrong, P 2004, ‘Some of the nineteenth century origins of the sustainability concept’, Environment, Development and Sustainability, vol. 6, no. 3, pp. 367-378. Madu, CN, Kuei, C & Madu, IE 2002, ‘A hierarchic metric approach for integration of green issues in manufacturing: a paper recycling application’, Journal of Environmental Management, vol. 64, no. 3, pp. 261-272. Mallidis, I, Dekker, R & Vlachos, D 2012, ‘The impact of greening on supply chain design and cost: a case for a developing region’, Journal of Transport Geography, vol. 22, no. 1, pp. 118-128. Mamic, I 2005, ‘Managing global supply chain: the sports footwear, apparel and retail sectors’, Journal of Business Ethics, vol. 59, no. 1, pp. 81-100. Manik, J & Yardley, J 2013, ‘Building collapse in Bangladesh leaves scores dead’, The New York Times, 24 April, viewed 15 June 2013, . Markley, MJ & Davis, L 2007, ‘Exploring future competitive advantage through sustainable supply chains’, International Journal of Physical Distribution & Logistics Management, vol. 37, no. 9, pp. 763-774. Markman, G & Krause, D 2014, ‘Special topic forum on theory building surrounding sustainable supply chain management’, Journal of Supply Chain Management, vol. 50, no. 2, pp. 1-2. Marsh, G 1864, Man and nature, Sampson Low, Son and Marston, London. Martínez-Salazar, IA, Molina, J, Ángel-Bello, F, Gómez, T & Caballero, R 2014, ‘Solving a bi-objective transportation location routing problem by metaheuristic algorithms’, European Journal of Operational Research, vol. 234, no. 1, pp. 25-36. Masanet, E & Horvath, A 2007, ‘Assessing the benefits of design for recycling for plastics in electronics: a case study of computer enclosures’, Materials & Design, vol. 28, no. 6, pp. 1801-1811. Matos, S & Hall, J 2007, ‘Integrating sustainable development in the supply chain: the case of life cycle assessment in oil and gas and agricultural biotechnology’, Journal of Operations Management, vol. 25, no. 6, pp. 1083-1102.
178
Mavrotas, G 2009, ‘Effective implementation of the ε-constraint method in multi-objective mathematical programming problems’, Applied Mathematics and Computation, vol. 213, no. 2, pp. 455-465. Mavrotas, G & Florios, K 2013, ‘An improved version of the augmented ε-constraint method (AUGMECON2) for finding the exact pareto set in multi-objective integer programming problems’, Applied Mathematics and Computation, vol. 219, no. 18, pp. 9652-9669. McCarter, MW & Northcraft, GB 2007, ‘Happy together? Insights and implications of viewing managed supply chains as a social dilemma’, Journal of Operations Management, vol. 25, no. 2, pp. 498-511. McCutcheon, DM & Meredith, JR 1993, ‘Conducting case study research in operations management’, Journal of Operations Management, vol. 11, no. 3, pp. 239-256. Meadows, DH, Randers, J & Meadows, DL 2005, Limits to growth: the 30-year update, 3rd edn, Earthscan, London. Meixell, MJ & Gargeya, VB 2005, ‘Global supply chain design: a literature review and critique’, Transportation Research Part E: Logistics and Transportation Review, vol. 41, no. 6, pp. 531-550. Melkote, S & Daskin, MS 2001, ‘Capacitated facility location/network design problems’, European Journal of Operational Research, vol. 129, no. 3, pp. 481-495. Melo, MT, Nickel, S & Saldanha-da-Gama, F 2009, ‘Facility location and supply chain management: a review’, European Journal of Operational Research, vol. 196, no. 2, pp. 401-412. Mena, C, Humphries, A & Choi, TY 2013, ‘Toward a theory of multi-tier supply chain management’, Journal of Supply Chain Management, vol. 49, no. 2, pp. 58-77. Mentzer, JT, DeWitt, W, Keebler, JS & Min, S 2001, ‘Defining supply chain management’, Journal of Business Logistics, vol. 22, no. 2, pp. 1-26. Mentzer, JT, Stank, TP & Esper, TL 2008, ‘Supply chain management and its relationship to logistics, marketing, production, and operations management’, Journal of Business Logistics, vol. 29, no. 1, pp. 31-46. Meredith, J 1998, ‘Building operations management theory through case and field research’, Journal of Operations Management, vol. 16, no. 4, pp. 441-454.
179
Michalewicz, M, Michalewicz, Z & Spitty, R 2011, ‘Optimising the wine supply chain’, Proceedings of 14th Australian Wine Industry Technical Conference, The Australian Wine Research Institute, pp. 1-5. Michalewicz, Z 2012, ‘Quo vadis, evolutionary computation? On a growing gap between theory and practice’, in J Liu, C Alippi & B Bouchon-Meunier (eds), Advances in Computational Intelligence, Springer, Berlin, pp. 98-121. Miller, GA 1956, ‘The magical number seven, plus or minus two: some limits on our capacity for processing information’, Psychological Review, vol. 63, no. 2, pp. 8197. Moccia, L 2013, ‘Operational research in the wine supply chain’, Information Systems and Operational Research, vol. 51, no. 2, pp. 53-63. Mollenkopf, D, Closs, D, Twede, D, Lee, S & Burgess, G 2005, ‘Assessing the viability of reusable packaging: a relative cost approach’, Journal of Business Logistics, vol. 26, no. 1, pp. 169-197. Moneva, JM, Archel, P & Correa, C 2006, ‘GRI and the camouflaging of corporate unsustainability’, Accounting Forum, vol. 30, no. 2, pp. 121-137. Montabon, F, Melnyk, SA, Sroufe, R & Calantone, RJ 2000, ‘ISO 14000: assessing its perceived impact on corporate performance’, Journal of Supply Chain Management, vol. 36, no. 2, pp. 4-16. Montiel, I 2008, ‘Corporate social responsibility and corporate sustainability: separate pasts, common futures’, Organization & Environment, vol. 21, no. 3, p. 245-269. Morhardt, JE 2009, ‘General disregard for details of GRI human rights reporting by large corporations’, Global Business Review, vol. 10, no. 2, pp. 141-158. Moscardo, G 2013, Sustainability in Australian business: principles and practice, John Wiley and Sons Australia, Milton, QLD. Moura-Leite, RC & Padgett, RC 2011, ‘Historical background of corporate social responsibility’, Social Responsibility Journal, vol. 7, no. 4, pp. 528-539. Mula, J, Peidro, D, Díaz-Madroñero, M & Vicens, E 2010, ‘Mathematical programming models for supply chain production and transport planning’, European Journal of Operational Research, vol. 204, no. 3, pp. 377-390.
180
Nagurney, A & Nagurney, LS 2010, ‘Sustainable supply chain network design: a multicriteria perspective’, International Journal of Sustainable Engineering, vol. 3, no. 3, pp. 189-197. Nagurney, A & Yu, M 2012, ‘Sustainable fashion supply chain management under oligopolistic competition and brand differentiation’, International Journal of Production Economics, vol. 135, no. 2, pp. 532-540. Neto, JQF, Bloemhof-Ruwaard, JM, van Nunen, JAEE & van Heck, E 2008, ‘Designing and evaluating sustainable logistics networks’, International Journal of Production Economics, vol. 111, no. 2, pp. 195-208. Nikolaou, IE, Evangelinos, KI & Allan, S 2013, ‘A reverse logistics social responsibility evaluation framework based on the triple bottom line approach’, Journal of Cleaner Production, vol. 56, pp. 173-184. Oliver, RK & Webber, MD 1982, ‘Supply-chain management: logistics catches up with strategy’, in M Christopher (ed.), Logistics: the strategic issues, Chapman & Hall, London, pp. 63-75. Omar, A, Davis-Sramek, B, Myers, MB & Mentzer, JT 2012, ‘A global analysis of orientation, coordination, and flexibility in supply chains’, Journal of Business Logistics, vol. 33, no. 2, pp. 128-144. Opasanon, S & Lertsanti, P 2013, ‘Impact analysis of logistics facility relocation using the analytic hierarchy process (AHP)’, International Transactions in Operational Research, vol. 20, no. 3, pp. 325-339. Osmond, R & Anderson, K 1998, Trends and cycles in the Australian wine industry, 1850 to 2000, Centre for International Economic Studies, University of Adelaide, Australia, viewed 23 November 2013, . Owen, SH & Daskin, MS 1998, ‘Strategic facility location: a review’, European Journal of Operational Research, vol. 111, no. 3, pp. 423-447. Pagell, M & Wu, Z 2009, ‘Building a more complete theory of sustainable supply chain management using case studies of 10 exemplars’, Journal of Supply Chain Management, vol. 45, no. 2, pp. 37-56. Paksoy, T 2010, ‘Optimizing a supply chain network with emission trading factor’, Scientific Research and Essays, vol. 5, no. 17, pp. 2535-2546. 181
Paksoy, T, Bektaş, T & Özceylan, E 2011, ‘Operational and environmental performance measures in a multi-product closed-loop supply chain’, Transportation Research Part E: Logistics and Transportation Review, vol. 47, no. 4, pp. 532-546. Pan, S, Ballot, E & Fontane, F 2013, ‘The reduction of greenhouse gas emissions from freight transport by pooling supply chains’, International Journal of Production Economics, vol. 143, no. 1, pp. 86-94. Park, S, Lee, T-E & Sung, CS 2010, ‘A three-level supply chain network design model with risk-pooling and lead times’, Transportation Research Part E: Logistics and Transportation Review, vol. 46, no. 5, pp. 563-581. Parsons, H & van Duyn, P 2014, ‘Coastal shipping, a necessity for an island nation’, Supply Chain Advisory Network, 28 August, viewed 25 October 2014, . Partovi, FY 2006, ‘An analytic model for locating facilities strategically’, Omega, vol. 34, no. 1, pp. 41-55. Patton, MQ 2002, Qualitative research and evaluation methods, 3rd edn, Sage Publications, Thousand Oaks, California. Paul, BD 2008, ‘A history of the concept of sustainable development: literature review’, Annals of University of Oradea Economic Science Series, vol. 17, no. 2, pp. 576580. Perelman, M 2010, ‘Adam Smith: class, labor, and the industrial revolution’, Journal of Economic Behavior & Organization, vol. 76, no. 3, pp. 481-496. Petti, L, Raggi, A, De Camillis, C, Matteucci, P, Sàra, B & Pagliuca, G 2006, ‘Life cycle approach in an organic wine-making firm: an Italian case-study’, Proceedings of the 5th Australian Conference on Life Cycle Assessment, Australian Life Cycle Assessment Society, pp. 1-11. Pinto-Varela, T, Barbosa-Póvoa, APFD & Novais, AQ 2011, ‘Bi-objective optimization approach to the design and planning of supply chains: economic versus environmental performances’, Computers and Chemical Engineering, vol. 35, no. 8, pp. 1454-1468. Pishvaee, MS & Razmi, J 2012, ‘Environmental supply chain network design using multiobjective fuzzy mathematical programming’, Applied Mathematical Modelling, vol. 36, no. 8, pp. 3433-3446. 182
Pishvaee, MS, Torabi, SA & Razmi, J 2012, ‘Credibility-based fuzzy mathematical programming model for green logistics design under uncertainty’, Computers and Industrial Engineering, vol. 62, no. 2, pp. 624-632. Point, E, Tyedmers, P & Naugler, C 2012, ‘Life cycle environmental impacts of wine production and consumption in Nova Scotia, Canada’, Journal of Cleaner Production, vol. 27, pp. 11-20. Porter, ME 1985, Competitive advantage: creating and sustaining superior performance, Free Press, New York. Presley, A, Meade, L & Sarkis, J 2007, ‘A strategic sustainability justification methodology for organizational decisions: a reverse logistics illustration’, International Journal of Production Research, vol. 45, nos. 18-19, pp. 4595-4620. Puigjaner, L & Guillén-Gosálbez, G 2008, ‘Towards an integrated framework for supply chain management in the batch chemical process industry’, Computers & Chemical Engineering, vol. 32, nos. 4-5, pp. 650-670. Pullman, ME, Maloni, MJ & Dillard, J 2010, ‘Sustainability practices in food supply chains: how is wine different?’, Journal of Wine Research, vol. 21, no. 1, pp. 35-56. Pullman, ME, Maloni, MJ & Carter, CR 2009, ‘Food for thought: social versus environmental sustainability practices and performance outcomes’, Journal of Supply Chain Management, vol. 45, no. 4, pp. 38-54. Quak, HJ & de Koster, MBM 2007, ‘Exploring retailers’ sensitivity to local sustainability policies’, Journal of Operations Management, vol. 25, no. 6, pp. 1103-1122. Ramos, TRP, Gomes, MI & Barbosa-Póvoa, AP 2014, ‘Planning a sustainable reverse logistics system: balancing costs with environmental and social concerns’, Omega, vol. 48, pp. 60-74. Ramudhin, A, Chaabane, A & Paquet, M 2010, ‘Carbon market sensitive sustainable supply chain network design’, International Journal of Management Science and Engineering Management, vol. 5, no. 1, pp. 30-38. Rao, P 2002, ‘Greening the supply chain: a new initiative in South East Asia’, International Journal of Operations and Production Management, vol. 22, nos. 5-6, pp. 632-655. Rao, P 2005, ‘The greening of suppliers in the South East Asian context’, Journal of Cleaner Production, vol. 13, no. 9, pp. 935-945. 183
Rao, P & Holt, D 2005, ‘Do green supply chains lead to competitiveness and economic performance?’, International Journal of Operations & Production Management, vol. 25, no. 9, pp. 898-916. Render, B, Stair, R & Hanna, M 2012, Quantitative analysis for management, 11th edn, Pearson Prentice Hall, Upper Saddle River, New Jersey. ReVelle, CS, Eiselt, HA & Daskin, MS 2008, ‘A bibliography for some fundamental problem categories in discrete location science’, European Journal of Operational Research, vol. 184, no. 3, pp. 817-848. Roberts, S 2003, ‘Supply chain specific? Understanding the patchy success of ethical sourcing initiatives’, Journal of Business Ethics, vol. 44, nos. 2-3, pp. 159-170. Rosenau, WV, Twede, D, Mazzeo, MA & Singh, SP 1996, ‘Returnable/reusable logistical packaging: a capital budgeting investment decision framework’, Journal of Business Logistics, vol. 17, no. 2, pp. 139-165. Ross, A, Parker, H & Benavides, M 2012, ‘Sustainability and supply chain infrastructure developement’, Management Decision, vol. 50, no. 10, pp. 1891-1910. Ross, DF 1998, Competing through supply chain management: creating market-winning strategies through supply chain partnerships, Chapman & Hall, New York. Ruiz-Femenia, R, Caballero, JA & Jiménez, L 2011, ‘Minimization of the life cycle impact of chemical supply chain networks under demand uncertainty’, in EN Pistikopoulos, MC Georgiadis & AC Kokossis (eds), Computer aided chemical engineering, vol. 29, Elsevier, Philadelphia, pp. 1195-1199. S&P Dow Jones Indices (n.d). Dow Jones Sustainability Index, viewed 17 June 2013, . Saaty, TL 1990, ‘How to make a decision: the analytic hierarchy process’, European Journal of Operational Research, vol. 48, no. 1, pp. 9-26. Saaty, TL & Ozdemir, MS 2003, ‘Why the magic number seven plus or minus two’, Mathematical and Computer Modelling, vol. 38, nos. 3-4, pp. 233-244. Saaty, TL 2008, ‘Decision making with the analytic hierarchy process’, International Journal of Services Sciences, vol. 1, no. 1, pp. 83-98.
184
Sadjady, H & Davoudpour, H 2012, ‘Two-echelon, multi-commodity supply chain network design with mode selection, lead-times and inventory costs’, Computers and Operations Research, vol. 39, no. 7, pp. 1345-1354. Sadler, I 2007, Logistics and supply chain integration, SAGE Publications, London. Sahin, F & Robinson, EP 2002, ‘Flow coordination and information sharing in supply chains: review, implications, and directions for future research’, Decision Sciences, vol. 33, no. 4, pp. 505-536. Sameer, P & Jayavel, S 2003, ‘Factors influencing global supply chain efficiency: implications for information systems’, Supply Chain Management: An International Journal, vol. 8, no. 3, pp. 241-250. Santoso, T, Ahmed, S, Goetschalckx, M & Shapiro, A 2005, ‘A stochastic programming approach for supply chain network design under uncertainty’, European Journal of Operational Research, vol. 167, no. 1, pp. 96-115. Sarker, RA & Newton, CS 2008, Optimization modelling: a practical approach, CRC Press, Boca Raton, Florida. Sarkis, J 1998, ‘Evaluating environmentally conscious business practices’, European Journal of Operational Research, vol. 107, no. 1, pp. 159-174. Sarkis, J 2003, ‘A strategic decision framework for green supply chain management’, Journal of Cleaner Production, vol. 11, no. 4, pp. 397-409. Sarkis, J, Zhu, Q & Lai, K 2011, ‘An organizational theoretic review of green supply chain management literature’, International Journal of Production Economics, vol. 130, no. 1, pp. 1-15. Schaltegger, S, Bennett, M & Burritt, R 2006, Sustainability accounting and reporting, Springer, Dordrecht, Netherlands. Schaltegger, S & Wagner, M 2006, Managing the business case for sustainability: the integration of social, environmental and economic performance, Greenleaf Publishing, Sheffield. Schaltegger, S & Burritt, RL 2010, ‘Sustainability accounting for companies: catchphrase or decision support for business leaders?’, Journal of World Business, vol. 45, no. 4, pp. 375-384.
185
Schaltegger, S, Lüdeke-Freund, F & Hansen, EG 2011, Business cases for sustainability and the role of business model innovation: developing a conceptual framework, Centre for Sustainability Management (CSM), Leuphana Universität Lüneburg, Germany, viewed 3 June 2013, . Schaltegger, S & Burritt, R 2014, ‘Measuring and managing sustainability performance of supply chains’, Supply Chain Management: An International Journal, vol. 19, no. 3, pp. 232-241. Searcy, C 2006, ‘Designing sustainable development indicators at an electric utility’, Ph.D. thesis, University of Alberta, Canada. Senge, P 2010, ‘The sustainable supply chain’, Harvard Business Review, pp. 70-74. Seuring, S 2008, ‘Assessing the rigor of case study research in supply chain management’, Supply Chain Management: An International Journal, vol. 13, no. 2, pp. 128-137. Seuring, S & Müller, M 2008, ‘From a literature review to a conceptual framework for sustainable supply chain management’, Journal of Cleaner Production, vol. 16, no. 15, pp. 1699-1710. Seuring, S, Sarkis, J, Müller, M & Rao, P 2008, ‘Sustainability and supply chain management: an introduction to the special issue’, Journal of Cleaner Production, vol. 16, no. 15, pp. 1545-1551. Seuring, S 2013, ‘A review of modeling approaches for sustainable supply chain management’, Decision Support Systems, vol. 54, no. 4, pp. 1513-1520. SGS Economics & Planning 2013, GDP growth: how are Australia’s major cities performing?, viewed 17 September 2014, . Shapiro, JF 2007, Modeling the supply chain, 2nd edn, Thomson Brooks/Cole, Australia. Shen, ZJM 2007, ‘Integrated supply chain design models: a survey and future research directions’, Journal of Industrial and Management Optimization (AIMS), vol. 3, no. 1, pp. 1-27.
186
Shen, ZJM & Qi, L 2007, ‘Incorporating inventory and routing costs in strategic location models’, European Journal of Operational Research, vol. 179 no. 2, pp. 372-389. Simchi-Levi, D, Simchi-Levi, E & Kaminsky, P 2004, Managing the supply chain: the definitive guide for the business professional, McGraw-Hill, New York. Simchi-Levi, D, Kaminsky, P & Simchi-Levi, E 2008, Designing and managing the supply chain: concepts, strategies, and case studies, 3rd edn, McGraw-Hill/Irwin, Boston, Mass. Simchi-Levi, D 2010, Operations rules: delivering customer value through flexible operations, MIT Press, Cambridge. Skouloudis, A, Evangelinos, K & Kourmousis, F 2009, ‘Development of an evaluation methodology for triple bottom line reports using international standards on reporting’, Environmental Management, vol. 44, no. 2, pp. 298-311. Slack, N, Chambers, S, Johnston, R & Betts, A 2009, Operations and process management: principles and practice for strategic impact, 2nd edn, Prentice Hall Financial Times, New York. Snyder, LV 2011, Fundamentals of supply chain theory, Wiley, Hoboken, New Jersey. Social Accountability International n.d., SA8000 standard and documents, Social Accountability International, viewed 9 November 2012, . Sonesson, U & Berlin, J 2003, ‘Environmental impact of future milk supply chains in Sweden: a scenario study’, Journal of Cleaner Production, vol. 11, no. 3, pp. 253266. Soni, G & Kodali, R 2011, ‘A critical analysis of supply chain management content in empirical research’, Business Process Management Journal, vol. 17, no. 2, pp. 238266. Soosay, C, Fearne, A & Dent, B 2012, ‘Sustainable value chain analysis: a case study of Oxford Landing from “vine to dine”’, Supply Chain Management: An International Journal, vol. 17, no. 1, pp. 68-77. Soosay, C, Fearne, A & Varsei, M 2014, ‘Extending sustainable practices beyond organizations to supply chains’, in S Sandhu, S McKenzie & H Harris (eds), Linking Local and Global Sustainability, vol. 4, Springer, Netherlands, pp. 71-90.
187
Spoolman, S & Miller, GT 2008, Environmental science: problems, concepts, and solutions, 12th edn, Thomson Brooks/Cole, Belmont, CA. Srivastava, SK 2007, ‘Green supply chain management: a state of the art literature review’, International Journal of Management Reviews, vol. 9, no. 1, pp. 53-80. Stadtler, H 2005, ‘Supply chain management and advanced planning: basics, overview and challenges’, European Journal of Operational Research, vol. 163, no. 3, pp. 575588. SteadieSeifi, M, Dellaert, NP, Nuijten, W, Van Woensel, T & Raoufi, R 2014, ‘Multimodal freight transportation planning: a literature review’, European Journal of Operational Research, vol. 233, no. 1, pp. 1-15. Stewart, E 2005, ‘Achieving sustainable development through corporate sustainability: an assessment of environmental and social performance in the Caribbean tourism industry’, Ph.D. thesis, Stanford University, USA. Stock, JR, Boyer, SL (2009), ‘Developing a consensus definition of supply chain management: a qualitative study’, International Journal of Physical Distribution & Logistics Management, vol. 39, no.8, pp.690-711. Stuart, I, McCutcheon, D, Handfield, R, McLachlin, R & Samson, D 2002, ‘Effective case research in operations management: a process perspective’, Journal of Operations Management, vol. 20, no. 5, pp. 419-433. Sundarakani, B, de Souza, R, Goh, M, Wagner, SM & Manikandan, S 2010, ‘Modeling carbon footprints across the supply chain’, International Journal of Production Economics, vol. 128, no. 1, pp. 43-50. Szolnoki, G 2013, ‘A cross-national comparison of sustainability in the wine industry’, Journal of Cleaner Production, vol. 53, pp. 243 - 251. Tan, KC, Lyman, SB & Wisner, JD 2002, ‘Supply chain management: a strategic perspective’, International Journal of Operations & Production Management, vol. 22, no. 6, pp. 614-631. Tang, CS & Zhou, S 2012, ‘Research advances in environmentally and socially sustainable operations’, European Journal of Operational Research, vol. 223, no. 3, pp. 585594.
188
Tang, O, Cao, D & Schvaneveldt, SJ 2008, ‘Institutional perspectives on supply chain management’, International Journal of Production Economics, vol. 115, no. 2, p. 261. Tiwari, MK, Raghavendra, N, Agrawal, S & Goyal, SK 2010, ‘A hybrid Taguchi–Immune approach to optimize an integrated supply chain design problem with multiple shipping’, European Journal of Operational Research, vol. 203, no. 1, pp. 95-106. Toll Group n.d., Intermodal, Toll Group, viewed 1 January 2015, . Tsai, WH & Hung, SJ 2009, ‘A fuzzy goal programming approach for green supply chain optimisation under activity-based costing and performance evaluation with a valuechain structure’, International Journal of Production Research, vol. 47, no. 18, pp. 4991-5017. United Nations 2012, Green economy in the context of sustainable development and poverty eradication, United Nations Sustainable Development, viewed 3 June 2013, . United Nations Global Compact Office and Business for Social Responsibility 2010, Supply chain sustainability: a practical guide for continuous improvement, viewed 23 July 2012, . Van Hoek, RI 1999, ‘From reversed logistics to green supply chains’, Supply Chain Management: An International Journal, vol. 4, no. 3, pp. 129-135. Van Hoof, B & Lyon, TP 2013, ‘Cleaner production in small firms taking part in Mexico’s sustainable supplier program’, Journal of Cleaner Production, vol. 41, pp. 270-282. Varsei, M & Soosay, C 2013, ‘A conceptual framework for the design and management of sustainable supply chains’, Proceedings of the 27th Australian and New Zealand Academy of Management, Australian and New Zealand Academy of Management, pp. 1-19. Varsei, M, Soosay, C & Fahimnia, B 2013, ‘A multi-dimensional assessment framework for sustainable supply chains’, presented at the 73rd Annual Meeting of the Academy of Management, Orlando, Florida, USA, 9-13 August. Varsei, M, Soosay, C, Fahimnia, B & Sarkis, J 2014, ‘Framing sustainability performance of supply chains with multidimensional indicators’, Supply Chain Management: An International Journal, vol. 19, no. 3, pp. 242-257. 189
Varsei, M, Christ, K & Burritt, R 2015, ‘Bottling location and the global wine supply chain: dollar, water and carbon trade-offs’, in J Humphreys (ed.), Proceedings of the 75th Annual Meeting of the Academy of Management, Academy of Management, pp. 1-6. Vázquez-Rowe, I, Villanueva-Rey, P, Iribarren, D, Teresa Moreira, M & Feijoo, G 2012, ‘Joint life cycle assessment and data envelopment analysis of grape production for vinification in the Rías Baixas appellation (NW Spain)’, Journal of Cleaner Production, vol. 27, pp. 92-102. Vidal, CJ & Goetschalckx, M 1997, ‘Strategic production-distribution models: a critical review with emphasis on global supply chain models’, European Journal of Operational Research, vol. 98, no. 1, pp. 1-18. Voss, C, Tsikriktsis, N & Frohlich, M 2002, ‘Case research in operations management’, International Journal of Operations & Production Management, vol. 22, no. 2, pp. 195-219. Wacker, JG 1998, ‘A definition of theory: research guidelines for different theory-building research methods in operations management’, Journal of Operations Management, vol. 16, no. 4, pp. 361-385. Walker, H, Seuring, S, Sarkis, J & Klassen, R 2014, ‘Sustainable operations management: recent trends and future directions’, International Journal of Operations & Production Management, vol. 34, no. 5, pp. 2-12. Walmart 2008, Walmart takes action to end forced child labor in Uzbekistan, Walmart, viewed 11 July 2012, . Wang, F & Gupta, SM 2011, Green supply chain management: product life cycle approach, McGraw Hill, New York. Wang, F, Lai, X & Shi, N 2011, ‘A multi-objective optimization for green supply chain network design’, Decision Support Systems, vol. 51, no. 2, pp. 262-269. Watson, M, Lewis, S, Cacioppi, P & Jayaraman, J 2013, Supply chain network design: applying optimization and analytics to the global supply chain, Pearson, Upper Saddle River, New Jersey.
190
Welford, R & Frost, S 2006, ‘Corporate social responsibility in Asian supply chains’, Corporate Social Responsibility and Environmental Management, vol. 13, no. 3, pp. 166-176. Wells, G 2011, Sustainability in Australian business: fundamental principles and practice, John Wiley and Sons Australia, Milton, QLD. Wells, G 2013, Sustainable business: theory and practice of business under sustainability principles, Edward Elgar, UK. West, EG 1992, ‘Capitalism as a moral system: Adam Smith’s critique of the free market economy’, Journal of Economic Literature, vol. 30, no. 4, pp. 2137-2138. Whittaker, J 2011, ‘The evolution of environmentally responsible investment: an Adam Smith perspective’, Ecological Economics, vol. 71, pp. 33-41. Winetitles 2013, The Australian and New Zealand wine industry directory, 31st Annual Edition, Winetitles, Australia. Winetitles 2014, World comparisons, Winetitles Media, viewed 21 August 2014, . Winter, M & Knemeyer, AM 2013, ‘Exploring the integration of sustainability and supply chain management: current state and opportunities for future inquiry’, International Journal of Physical Distribution & Logistics Management, vol. 43, no. 1, pp. 18-38. Wisner, JD, Tan, KC & Leong, GK 2008, Principles of supply chain management: a balanced approach, 2nd edn, South-Western, Cengage Learning, Mason. Wittstruck, D & Teuteberg, F 2012, ‘Understanding the success factors of sustainable supply chain management: empirical evidence from the electrics and electronics industry’, Corporate Social Responsibility and Environmental Management, vol. 19, no. 3, pp. 141-158. World Meterological Organization 2013, The global climate 2001-2010: a decade of climate extremes summary report, viewed 5 July 2013, . Wolf, J 2008, The nature of supply chain management research: insights from a content analysis of international supply chain management literature from 1990 to 2006, Gabler-Verlag, Wiesbaden.
191
Wolf, J 2011, ‘Sustainable supply chain management integration: a qualitative analysis of the German manufacturing industry’, Journal of Business Ethics, vol. 102, no. 2, pp. 221-235. Wolfe, CJ 2012, ‘Dimensions of purchasing social responsibility in sustainable supply chain organizations’, Ph.D. thesis, Northcentral University, USA. Wong, CWY 2013, ‘Leveraging environmental information integration to enable environmental management capability and performance’, Journal of Supply Chain Management, vol. 49, no. 2, pp. 114-136. World Trade Organization 2012, International trade statistics, World Trade Organization, viewed 4 April 2012, . Wu, Z & Pagell, M 2011, ‘Balancing priorities: decision-making in sustainable supply chain management’, Journal of Operations Management, vol. 29, no. 6, pp. 577590. Yeh, W-C & Chuang, M-C 2011, ‘Using multi-objective genetic algorithm for partner selection in green supply chain problems’, Expert Systems with Applications, vol. 38, no. 4, pp. 4244-4253. Yin, RK 2014, Case study research: design and methods, 5th edn, SAGE Publications, Thousand Oaks, California. Yusuf, YY, Gunasekaran, A, Musa, A, El-Berishy, NM, Abubakar, T & Ambursa, HM 2013, ‘The UK oil and gas supply chains: an empirical analysis of adoption of sustainable measures and performance outcomes’, International Journal of Production Economics, vol. 146, no. 2, pp. 501-514. Zailani, S, Jeyaraman, K, Vengadasan, G & Premkumar, R 2012, ‘Sustainable supply chain management (SSCM) in Malaysia: a survey’, International Journal of Production Economics, vol. 140, no. 1, pp. 330-340. Zhang, W & Reimann, M 2014, ‘A simple augmented ∊-constraint method for multiobjective mathematical integer programming problems’, European Journal of Operational Research, vol. 234, no. 1, pp. 15-24. Zhou, A, Qu, B-Y, Li, H, Zhao, S-Z, Suganthan, PN & Zhang, Q 2011, ‘Multiobjective evolutionary algorithms: a survey of the state of the art’, Swarm and Evolutionary Computation, vol. 1, no. 1, pp. 32-49.
192
Appendix 1
The ethics approval letter Mon 06-Jan-14 9:41 AM Dear Applicant Re: Ethics protocol “HREC Application-Mohsen Varsei” (Application ID: 0000032203) Thank you for submitting your ethics protocol for consideration. Your protocol has been considered by the E1Committee Review Group. I am pleased to advise that your protocol has been granted ethics approval and meets the requirements of the National Statement on Ethical Conduct in Human Research. Please note that the E1 Committee Review Group’s decision will be reported to the next meeting of the Human Research Ethics Committee for endorsement. Please regard this email as formal notification of approval. Ethics approval is always made on the basis of a number of conditions detailed athttp://www.unisa.edu.au/res/forms/docs/humanresearchethics_conditions.doc; it is important that you are familiar with, and abide by, these conditions. It is also essential that you conduct all research according to UniSA guidelines, which can be found at http://www.unisa.edu.au/res/ethics/default.asp Please note, if your project is a clinical trial you are required to register it in a publicly accessible trials registry prior to enrolment of the first participant (e.g. Australian New Zealand Clinical Trials Registry http://www.anzctr.org.au/) as a condition of ethics approval. Best wishes for your research. Executive Officer UniSA’s Human Research Ethics Committee, CRICOS provider number 00121B 193
Appendix 2
The interview protocol Project Title: Sustainable supply chain management: An integrated model for optimising supply chain network design
Project Overview Many organisations are realising the growing importance of sustainability in today’s business environment and the need to incorporate various sustainability performance indicators into supply chain management. While this requires multi-dimensional frameworks for designing and managing supply chains, the existing literature lacks integrated multidimensional frameworks in which all three dimensions of sustainability (i.e. environmental, social and economic dimensions) can be incorporated and analysed. To bridge this gap, this project proposes a multi-objective and multi-dimensional framework for optimising the design of supply chains. In brief, by implementing this project, a focal company will measure, assess and examine: •
Whether its current supply chain structure is designed optimally given the organisational goals and objectives
•
If not: what is the most cost-efficient supply chain design for the company?
•
One step forward: what would the most sustainable supply chain network design for the company look like? (i.e. how can your supply chain be designed to function at the lowest possible cost while enhancing environmental and social impacts?) 194
Project Scope: The project focuses on strategic decisions related to supply chain network design. Supply chain network design is mainly concerned with adopting optimisation and modelling techniques to examine and determine strategic decisions including: • the optimal number, location and capacity of facilities, including plants and distribution centres • distribution strategies (e.g. the optimal allocation of retailers or customers to distribution centres). These decisions are among focal companies’ long-term decisions, and play significant roles in improving efficiency and profitability. Such strategic decisions influence tactical and operational decisions of a firm with a direct impact on the supply chain’s competitiveness. A leading company needs to often revisit, analyse and redesign its supply chain to ensure the network is operating optimally given the organisational goals. The literature agrees that several successful companies in various industries have revisited and redesigned their supply chain networks successfully. Examples include Procter & Gamble, Amazon and Toyota, among many others. In this project, the researcher aims to use the analytical modelling and optimisation approach to examine multiple economic, environmental and social objectives of your firm’s supply chain network. The focus is placed on analysis of the supply chain configuration/structure in terms of a set of multi-dimensional sustainability performance indicators. Participation Procedure The researcher at the University of South Australia Business School will assess and analyse your company’s supply chain network design. The researcher would like to seek your company’s voluntary participation through allowing him to conduct some interviews with the operations and supply chain or senior managers.
195
Each face to face semi-structured interview will last around 45 minutes and the interviews will be audio-recorded. All participation will be voluntary. You are free to withdraw at any time without repercussions and you may request that your contributions be removed from the project. You are free to indicate that all information provided is to be treated as ‘commercial-in-confidence’. The researcher would also like to assure you that your nonparticipation in this research will not result in any discrimination, reduction in the level of care, effect on employment or any other penalty. All records containing personal information will remain confidential and no information which could lead to identification of any individual will be released unless required by law. Which data will be collected? The following data are required in order to assess, analyse and optimise your company’s supply chain network design. -
Economic dimension: What is the number and location of market zones (i.e. demand points)? What are the number, location and capacity of the existing plants, distribution centres and suppliers? If your company plans to expand its business: what are the number, location and capacity of potential plants, distribution centres and suppliers? What is the annual demand of products in market zones (i.e. demand points)? What are the annual fixed costs of current plants and distribution centres? If your company plans to expand its business: what are the fixed costs of locating new potential plants and distribution centres? What are the distances, in kilometres, between existing supply chain members? What are the variable costs of raw materials? What are the variable costs of production? What are the associated transportation costs between the supply chain members? 196
-
Environmental dimension:
The following items will be estimated and calculated using emission factors in your industry. CO2 equivalent emission per unit product in plants and distribution centres CO2 equivalent emission per unit product transported from suppliers to plants CO2 equivalent emission per unit product transported from plants to distribution centres CO2 equivalent emission per unit product transported from distribution centres to retailers (i.e. demand points) The abovementioned data will be incorporated into a multi-objective mathematical optimisation model as inputs and parameters. Developed by the researcher, this mathematical model has been presented in reputable academic conferences.
Confidentiality / Anonymity If you agree to participate in this study, the data will be incorporated into a mathematical model. The researcher is committed to the principles that guide reputable research. The research will be undertaken with care and respect for the respondents’ welfare and the quality and validity of the research report/result. All confidential information will be treated with integrity and not be made public. Data will be coded to ensure the confidentiality of information. The back-up data stored on compact disks will be stored in a locked cupboard. Data will be stored for seven years in accordance with the University of South Australia Codes of Conduct policy. You are free to indicate that all information provided is to be treated as ‘commercial-in-confidence’. If it becomes necessary or possible to cite your company’s name in any publication resulting from the research, the event of this happening will seek written permission beforehand. Outcome / Publication of Results (if applicable) The research outcomes of the study will be in the forms of research reports, conference papers and scholarly journal articles. An electronic copy of the research results will be distributed to you. You are free to indicate that all information provided is to be treated as 197
‘commercial-in-confidence’. If it becomes necessary or possible to cite your company’s name in any publication resulting from the research, the event of this happening will seek written permission from all participants beforehand. Consent The researcher seeks your cooperation to participate in this study. By signing the attached consent form, you are agreeing to provide informed consent. Right to Withdraw Throughout the course of the proposed research program, you are free to withdraw at any time for whatever reason without repercussions and you may request that your contributions be removed from the project. Feedback A summary report will be prepared and disseminated to interested parties involved in the study. Data Storage Details The researcher will ensure the confidentiality of information by coding data. The back-up data on compact disks will be stored in a locked cupboard at the office of Mohsen Varsei, BH4-21, City West Campus, North Terrace, the University of South Australia. Data will be stored for seven years in accordance with the University of South Australia Code of Conduct policy. Concerns / Complaints Should you have any concerns or complaints about the nature and/or conduct of this research project, please contact the executive officer of the ethics committee: Ms Vicki Allen Ethics and Safety Officer Research and Innovation Services University of South Australia Mawson Lakes Campus Mawson Lakes Boulevard 198
Mawson Lakes SA 5095 Tel: +61 8 8302 3118 Email:
[email protected] Ms Allen will be available to discuss any ethical concerns about the project or answer questions about the rights of participants any issues you raise will be treated in confidence and investigated fully, and you will be informed of the outcome. This project has been approved by the University of South Australia’s Human Research Ethics Committee. THANK YOU
FOR
YOUR
199
PARTICIPATION
