A Multi-agent Architecture for Coordination of Supply Chains with Local Information Sharing

Abstract

In supply chain management, improving the efficiency of the overall supply chain is important. It has been found that sharing information between supply chain members can increase the efficiency of the supply chain by varying degrees in different situations. There are, however, many practical difficulties in sharing information with all the companies in a complex supply chain. This paper presents a multi-agent system architecture for supply chain coordination where agents discover the structure of supply chains and share information on inventory levels only by local collaborations. Agents use the gathered information to estimate demands propagated from multiple markets through a complex network of companies with different lead times and market shares. The performance of the architecture is investigated with simulation experiments and the result is compared with those of alternative strategies.