Sustainable supply chain network design problem: Using the integrated BWM, TOPSIS, possibilistic programming, and ε-constrained methods
Introduction
Supply chain network design (SCND) is the elaboration of a complex structure of interaction between organizations for managing the flow of material, information, and money. In the early 90s, the prime concern of SCND was to maximize the overall profit and/or minimizing the costs by optimizing the material flow across the SC (Farahani et al., 2014, Pishvaee and Torabi, 2010). Due to diversifying consumer decisions towards greener products, rising regulatory pressures from the governments, and Non-Governmental Organizations, companies are now focusing more on the design of their SC to make them sustainable (de Oliveira et al., 2018, Wu and Pagell, 2011). The elucidation of sustainability was given in the Brundtland report as “meeting the needs of the present generations without compromising the needs of future generations” (Keeble, 1988). The sustainability concept often addresses the three major dimensions: social, economic, and environmental. Recently, sustainable supply chain network design (SSCND) has received notable consideration from scholars. Guo, Hu, Allaoui, and Boulaksil (2019) observed SSCND as a challenging task. Despite its importance, very limited studies have been conducted, especially addressing three dimensions of sustainability. In archived literature, only 17 percent of SCND models addressed sustainability issues (Govindan, Fattahi, & Keyvanshokooh, 2017). Hence, there is enormous scope for developing quantitative models for SSCND (Gunasekaran & Spalanzani, 2012).
SCND is a long-term strategic decision. Due to its complexity, it cannot be easily changed once it is formed (Tsao, Thanh, Lu, & Yu, 2018). The SCND is indeed an important part in implementing sustainability, as conventional SC activities account for approximately 50–70% of carbon emissions (Ahmed & Sarkar, 2018). Waltho, Elhedhli, and Gzara (2019) discussed different greenhouse gas reduction plans, such as carbon offset, carbon cap, cap-and-trade, and carbon tax, to lower the emission across the SC. However, limited studies were conducted addressing uncertainty and social sustainability. Social sustainability is defined as “developing a sustainable environment which upgrades living standards, social wellbeing by taking care of the need of the people where they live and work (McKenzie, 2004). In business organizations, social sustainability can be measured through labor working conditions, health and safety, human rights, fair labor wages, employment generation, work-life balance, equality of genders, empowerment, and more. Quantifying social sustainability is a difficult job, which needs to be addressed in the SCND (Das and Shaw, 2017, Tsao et al., 2018).
In addition to the sustainability dimension, a manager has to make decisions in a dynamic environment. A dynamic environment inflicts a massive level of uncertainty in decision making (Pishvaee & Torabi, 2010). Managing uncertainty is a real challenge for business. Currently, SSCND models have been formulated, taking deliberation of a deterministic environment, and limited models have been reported in a stochastic environment. To overcome these gaps, this study proposes a two-stage multi-objective possibilistic integer linear programming (MPILP) model, considering the three sustainability dimensions. As a theoretical contribution, we introduce a new sustainability dimension, supplier’s green image (SGI), in the design phase of SCND. In the first stage, SGI has been quantified by applying BWM and TOPSIS methods. The ratings have been subsequently integrated into the mathematical model.
In the second stage, a fuzzy possibilistic technique is proposed to handle uncertain scenarios. Further, this study applies the ε-constraint method for getting trade-off relationships among the three conflicting objectives. This work aims to reply to the following research questions:
- –
Which facilities should be opened to minimize the emission and cost of the supply chain?
- –
What would be the amount of material flow across the different echelons of the supply chain in an uncertain situation?
- –
What are the impacts of changing the overall carbon cap on the movements of goods across the chain?
- –
What are the effects of uncertainty over the model?
- –
What are the computational performances of the model in different scenarios?
The paper is organized as follows. The appropriate literature is presented in Section 2. Section 3 comprises the model development. The proposed methodology is presented in Section 4. Computational experiments are stated in Section 5. Lastly, conclusion, the managerial implication of the study and future scope are drawn in Section 6.
Section snippets
Literature review
SCND has grown as the most crucial research topic in SC management. As of now, a substantial amount of studies has already been carried out on the topic. A well-optimized SC network can save a considerable amount of resources and provide various benefits to the organization (Eskandarpour, Dejax, Miemczyk, & Péton, 2015). Nowadays, organizations are forced to develop their SC network sustainable due to increasing regulatory pressure towards environmental pollution (Shaw, Irfan, Shankar, & Yadav,
Model development
In this section, a two-stage multi-objective possibilistic integer linear programming based SCND model is proposed considering multi-product, multi-mode, and multi-echelons. The pictorial view of SCN is shown in Fig. 2. The proposed SCN is of forward type, and no reverse material flow is considered. In the literature, various studies have been reported on reverse logistics, which generally reincorporates the recycled material in the forward supply chain. For more details on reverse logistics,
The proposed solution technique
The research has been carried out in two stages, as shown in Fig. 4. In the first stage, different GI factors have been cataloged through literature review and taking views from academic professionals. This study uses the BWM method to calculate the weights. The BWM method is adopted due to its ability to handle the inconsistency efficiently and requires few evaluations from the decision-maker (Rezaei, 2015, Rezaei, 2016). In the management problem, the inconsistency of decision-makers is a
Implementation and computation
This section describes the practicality of the proposed SSCND model through a numerical example. The testing of the model was carried out considering 3 suppliers, 3 plants, 4 distribution centers, 5 customers, 2 transportation modes, and 2 products. The data for solving the model was generated randomly using Excel-based functions. In this study, the variable cost, variable emission, transportation cost, transportation emissions, capacities of the plants, distribution centers, suppliers, and
Conclusion, managerial importance and future scope
The study proposes a multi-echelon, multi-product SCND model considering carbon footprint, social sustainability, and different transportation modes under uncertain environment. The model has been motivated by the triple bottom line, where economic, environmental, and social aspects are dealt simultaneously. The current study has been led in two phases. In the primary phase, the supplier green image has been quantified by applying BWM and TOPSIS method. The BWM and TOPSIS have been infrequently
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
References (72)
- et al.
An integrated chance-constrained stochastic model for a mobile phone closed-loop supply chain network with supplier selection
Journal of Cleaner Production
(2019) - et al.
Impact of carbon emissions in a sustainable supply chain management for a second generation biofuel
Journal of Cleaner Production
(2018) - et al.
Large-scale reverse supply chain network design: An accelerated Benders decomposition algorithm
Computers & Industrial Engineering
(2018) - et al.
A new model for designing sustainable supply chain networks and its application to a global manufacturer
Journal of Cleaner Production
(2017) - et al.
A multi-objective stochastic programming approach for supply chain design considering risk
International Journal of Production Economics
(2008) Stochastic versus possibilistic programming
Fuzzy Sets and Systems
(1990)- et al.
Understanding approaches to complexity and uncertainty in closed-loop supply chain management: Past findings and future directions
Journal of Cleaner Production
(2018) - et al.
A systematic literature review on green supply chain management: Research implications and future perspectives
Journal of Cleaner Production
(2018) - et al.
A green supplier evaluation system based on a new multi-criteria sorting method: VIKORSORT
Expert Systems with Applications
(2018) - et al.
Performance evaluation of green suppliers using entropy-TOPSIS-F
Journal of Cleaner Production
(2019)