Simulation Optimization Approach for Dynamic and Stochastic Closed Loop Supply Chain Network

In this paper, four simulation optimization (SO) models are developed by combining simulation and genetic algorithm. In proposed models, optimal values of inventory control parameters and the number of facilities to be opened are determined simultaneously for periodic review and continuous review systems, respectively. Furthermore, single product and multi-components of closed-loop supply chain (CLSC) network are created considering two different objective functions of review systems to gain a sustainable competitive advantage for companies. We seek to offer valuable insights for creating robust and user-friendly CLSC network where the forward network includes suppliers, plants, retailers, and customers, and reverse network includes collection centers, disassembly centers, refurbishing centers, and disposal center. The results of this study demonstrated that four SO models have a significant potential to satisfy the customer’s needs since average service level of the models is at least 81.8%. The total supply chain cost can be decreased at least 3% and at most 22% on average with proposed continuous review model whose objective is the minimization of differences between the total overordering cost and the total underordering cost (C-D). Furthermore, the total lost sales cost can be improved at least 15% and at most 89% on average with C-D model.