Multiagent System Approach for Dynamic Lot-Sizing in Supply Chains

Supply chains are distributed systems and hence multiagent based systems are considered the best approach to manage supply chains. In multiagent systems for supply chain management, each agent is assigned to model a facility and the interaction protocol to connect the agents is defined. The intelligent agents interact with each other to dynamically access, transfer, and evaluate the information of the supply chain. We introduce a specific multiagent system for dynamic lot-sizing in supply chains. The dynamic lotsizing problem over a planning horizon makes a plan that minimizes cost while satisfying forecasted external demands. The common drawbacks of the majority of dynamic lot-sizing algorithms are that they offer centralized solutions with a monolithic view of the problem. On the other hand, supply chains are struggling with expensive inefficiencies due to the lack of information sharing. Therefore, there is a need to design coordination mechanisms that are capable of motivating information sharing. Supply chain entities will be willing to share information when they are guaranteed to get the right benefits in return for information sharing. Such trustworthy mechanisms may not produce the global optimum of monolithic models, but they would be more reasonable, considering the reality of supply chains.