Optimization modelling for a sustainable closed-loop supply chain network using IoT: multiobjective metaheuristic algorithms

This paper investigates the optimization of a Closed-Loop Supply Chain Management to minimize total system costs and CO2 emissions. In addition, a mathematical model of mixed integer linear programming was designed to study the Closed-Loop Supply Chain Network. An Internet of Things (IoT) based platform was also developed to increase the planned network's speed, accuracy, and security. In this way, customers register their orders online and deposit the money to buy the product. After consuming the product, they can eventually sell it on the same platform when the waste is sent for recycling. In this context, the best location for the plant's construction and the amount of transportation in the network are selected for this problem. By generalizing this problem to a larger dimension, the number of decision variables of the model increases, and such a problem becomes difficult to solve. Therefore, metaheuristic algorithms such as Non-Dominated Genetic Sorting Algorithm II, Multiple Objective Particle Swarm Optimization, Multiobjective Grasshopper Optimization Algorithm, Multiobjective Salp Swarm Algorithm and Multiobjective Ant Lion Optimizer have been used to solve this Bi-Objective Optimization Problem. This paper proposes an optimization model for designing a sustainable SCLSCN that incorporates IoT technology. Moreover, several metaheuristic algorithms are used to solve the optimization problem. A real case study on plastic recycling was conducted in Mazandaran province to investigate the applicability of the model. The Taguchi method is used to adjust the parameters to obtain reliable results for these algorithms. In addition, we compared the responses of each algorithm to select the better solution and evaluate the effectiveness of each algorithm in the face of such a problem.