Decision Making in Agent-Based Manufacturing With A Reinforcement Learning Approach

Development of production systems, presents new ways of manufacturing. Agent-based systems are of emerging interest in the specification and implementation of manufacturing systems. In this paper, we have simulated a manufacturing system to introduce the role of intelligent decision making in manufacturing. The main idea is to develop independent agents which would create a formation for producing a desired product .Agents do not have enough information to know the highest quality that a combination of manufacturers can achieve in production process. By applying reinforcement learning, the agents "learn" which combination of manufacturers is best for a specific task. The simulation is a basic one with focus on agent's decision making; therefore different manufacturing phenomena such as scheduling are not its concern. Key-Words: Agent-Based Manufacturing, Agent Decision Making, Reinforcement Learning, Manufacturing Simulation