Optimizing Production Manufacturing Using Reinforcement Learning

Manyindustrial processes involve makingparts with an assemblyof machines, where each machinecarries out an operation on a part, and the finished product requires a wholeseries of operations. Awell-studied exampleof sucha factory structure is the transfer line, which involves a sequence of machines. Optimizing transfer lines has beena subject of muchstudy in the industrial engineeringand operations research fields. A desirable goal of a lean manufacturing system is to maximizedemand,while keeping inventory levels of unfinished product as lowas possible. This problemis intractable since the numberof states is usually very large, and the underlying modelsare stochastic. In this paper wepresent an artificial intelligence approach to optimization based on a simulation-based dynamic programming method called reinforcement learning. Wedescribe a reinforcement learning algorithm called SMART,and compareits performanceon optimizing manufacturing systems with that of standardheuristics used in ...

• Type: Conference
• Year: 1998
• Where: FLAIRS
• Authors: Sridhar Mahadevan, Georgios Theocharous
• Keywords: Artificial Intelligence, FLAIRS 1998, Reinforcement Learning, Sucha Factory Structure, Transfer Lines