A robust design of a circular supply chain network based on the resilience and responsiveness dimensions: A data-driven model

In recent years, the importance of the Supply Chain Network Design Problem (SCNDP) has drastically increased in the business environments. Although in the traditional perspective usually supply chains were configured only based on the financial aspect, nowadays owing to environmental concerns, unpredictable disruptions, and customer expectations, several other aspects have been incorporated into the mentioned problem. In this regard, this study focused on the SCNDP by considering crucial aspects namely responsiveness, Circular Economy (CE), and resilience. This study suggests a novel Data-Driven Approach (DDA) to investigate the mentioned problem. In the first stage of the developed method, the weights of the potential locations for establishing repairing and recycling centers are determined using the combinations of the Fuzzy Best-Worst Method (FBWM) and Light Gradient Boosting Machine (LightGBM) Then, in the second stage, a closed-loop supply chain with the mentioned dimensions (i.e., resilience, CE, and responsiveness) is configured by proposing a Multi-objective Mathematical Model (MOM). It should be noted that a DDA by combining the Robust Fuzzy-Stochastic (RFS) programming and the Dynamic Regression is developed to deal with mixed uncertainty. Moreover, this study focuses on a real-world case study in the healthcare industry for demonstrating the application of the developed DDA. Besides, this research develops an efficient solution method to solve the proposed MOM. The obtained results confirm the application and efficiency of the proposed DDA.