Development of a Data-Driven Decision-Making Model for Evaluating Agile and Resilient Suppliers in the Context of the Circular Economy (Case Study: Automotive Industry)

This study focuses on developing a data-driven model for evaluating and selecting suppliers in the automotive industry, emphasizing agility, resilience, and circular economy principles. Through an extensive literature review, 14 evaluation criteria were identified, which were subsequently weighted using expert survey data and the Fuzzy Best-Worst Method (FBWM). The analysis revealed that resilience is the most crucial dimension in supplier evaluation, with excess inventory and backup suppliers ranking as the most significant criteria. Building on these insights, a hybrid Weighted Support Vector Machine (WSVM) algorithm was employed to assess supplier performance, integrating feature weights derived from the FBWM. The developed model demonstrated over 90% accuracy in predicting supplier performance, showcasing its robustness and applicability. The findings highlight the importance of resilience in managing supply chain disruptions and underline the critical role of data-driven methodologies in enhancing decision-making processes. Additionally, this research offers practical recommendations for adopting data-driven models in supply chain management, particularly in contexts requiring agility and sustainability. By bridging theoretical frameworks with practical applications, the study provides valuable insights for industry practitioners aiming to optimize supplier selection and improve supply chain efficiency. The proposed approach paves the way for more resilient, sustainable, and adaptive supply chains in the automotive sector and beyond.