A fuzzy cognitive map based on Nash bargaining game for supplier selection problem: a case study on auto parts industry

Abstract

Supplier Selection (SS) is a critical issue due to intense competition in the current market and the need to provide customer necessities with acceptable quality. On the other hand, SS depends on various criteria that make it a Multi-Criteria Decision-Making problem. Hence, a novel framework has been proposed in the current study to evaluate and rank suppliers. The proposed framework by aggregating the Process Control Score (PCS) and Process Evaluation Score (PES) evaluate and rank suppliers. For calculating PCS, a new structure and logic of the Fuzzy Cognitive Map based on the Nash Bargaining Game (BG-FCM) has been proposed to solve FCM’s shortcoming in distinguishing between the important concepts in the real world. Moreover, for generating solutions with high separability and helping decision-makers to have a precise analysis of the system, a modified learning algorithm based on the Particle Swarm Optimization (PSO) and S-shaped transfer function (PSO-STF) has been utilized for training BG-FCM. For calculating PES, experimental mathematical equations in the inspected case have been utilized for important criteria of quality, delivery time, and price of the shipment. The proposed framework has been applied in an auto parts industry for validation. The results show that BG-FCM can successfully highlight the most important concepts and assign their original value. Also, PSO-STF in the comparison between other conventional FCMs’ learning algorithms has better performance in generating solutions with high separability. It can be concluded that BN-FCM with more progressive intelligence can analyze the complex systems and help decision-makers to have a vivid insight into the system.