Optimizing Production Manufacturing Using Reinforcement Learning

May 1998

Proceedings of the Eleventh International Florida Artificial Intelligence Research Society Conference (FLAIRS 1998)

Volume

Proceedings of the Eleventh International Florida Artificial Intelligence Research Society Conference (FLAIRS 1998)

All Papers

Many industrial processes involve making parts with an assembly of machines, where each machine carries out an operation on a part, and the finished product requires a whole series of operations. A well-studied example of such a factory structure is the transfer line, which involves a sequence of machines. Optimizing transfer lines has been a subject of much study in the industrial engineering and operations research fields. A desirable goal of a lean manufacturing system is to maximize demand, while keeping inventory levels of unfinished product as low as possible. This problem is intractable since the number of states is usually very large, and the underlying models are stochastic. In this paper we present an artificial intelligence approach to optimization based on a simulation-based dynamic programming method called reinforcement learning. We describe a reinforcement learning algorithm called SMART, and compare its performance on optimizing manufacturing systems with that of standard heuristics used in industry.