Affiliations: Department of Industrial Engineering, Amirkabir University of Technology, Tehran, Iran | Collaborative Intelligent Systems Laboratory, European Centre for Soft Computing, Mieres, Asturias, Spain | Intelligent Signal Processing Group, Department of Electronics, University of Barcelona, Barcelona, Spain | Artificial Olfaction Group, Institute for Bioengineering of Catalonia, Barcelona, Spain
Abstract: In this paper, we present a new Multi-Agent System for reduction of the bullwhip effect in fuzzy supply chains. First, we show that a supply chain that uses an optimal ordering policy without data sharing among echelons still suffers from the bullwhip effect. Then, we propose the multi-agent solution to manage and reduce the bullwhip effect. The proposed multi-agent system includes four different types of agents in which each agent has its own list of actions. The proposed Multi-agent System applies a new Tabu Search algorithm for fuzzy rule generation, and a new data filtering algorithm for extraction of the bullwhip-free data from supply chain data warehouse. We validate the multi-agent system under different conditions and discuss how the system responds to different factors. The results show that the proposed multi-agent system reduces the bullwhip effect significantly in a rational time.
