Two hybrid meta-heuristic algorithms for a dual-channel closed-loop supply chain network design problem in the tire industry under uncertainty

Abstract

There is so much interest in online purchasing within supply chain networks nowadays. After expanding the internet access and services, customers’ behavior has changed. Today, a customer’s shopping manner usually begins with the internet search. With this approach, we face some new trends in this field, such as online-to-offline (O2O) commerce that aims to balance online and offline sales. Regarding the supply chain management, the O2O commerce can help the managers to conduct both online and offline businesses. The tire industry is one of the applications of the O2O approach, which also directly affects the supply chain network design (SCND). Therefore, this work for the first time proposes a dual-channel, multi-product, multi-period, multi-echelon closed-loop SCND under uncertainty for the tire industry. To tackle the uncertain parameters of the problem (e.g., prices and demand), a fuzzy approach, so-called the Jimenez’s method, is applied. Another main innovation of this work is two new hybrid meta-heuristic algorithms with new procedures. Two recent nature-inspired algorithms (i.e., red deer algorithm (RDA) and whale optimization algorithm (WOA)) are hybridized with the genetic algorithm (GA) and simulated annealing (SA) to strengthen the diversification and intensification phases, respectively. The numerical experiments demonstrate that the hybrid version of WOA and SA returns high-quality solutions and requires an acceptable amount of computational time. The conducted sensitivity analyses underline the importance of tire remanufacturing. Furthermore, setting the appropriate prices in different channels for the available tire types is critical for sustainable tire supply chain management.