International Journal of Environmental Science and Technology | 2021
Designing a multi-objective green supply chain network for an automotive company using an improved meta-heuristic algorithm
Abstract
This paper aims to present a multi-product, five-echelon green supply chain network for an automotive company. The proposed green supply chain network consists of various echelons including suppliers, storages, plants, distribution centers, and customers. The first objective of the problem in this study is to minimize transportation and construction costs, and the second objective is to minimize the carbon dioxide emissions during the transportation process across all echelons of the supply chain. The proposed model considers construction of new centers by creating a balance between transportation and construction costs. The model is first validated on a small-scale instance problem, and then, it is solved using the weighted method. Further, the Pareto optimal solution of the model is obtained. Due to the NP-hardness of the problems, the studied case is solved with two meta-heuristic methods of NSGA II and SPEA II on large-scale instance problems and the Taguchi method is utilized to set the parameters of these two meta-heuristic algorithms. Moreover, the priority-based method is used to encode the bi-objective model. The minimum capacity constraint is also considered in designing the meta-heuristic algorithm and a new repairing algorithm is proposed in order to prevent the model from producing infeasible solutions. Several sample problems are created to evaluate the performance of the proposed meta-heuristic algorithms. Finally, the results of the proposed algorithms are compared to each other using some performance measures. The results indicate the superiority of the SPEA II over NSGA II in terms of all comparison metrics.