Scaling Adaptive Agent-Based Reactive Job-Shop Scheduling to Large-Scale Problems

Most approaches to tackle job-shop scheduling problems assume complete task knowledge and search for a centralized solution. In this work, we adopt an alternative view on scheduling problems where each resource is equipped with an adaptive agent that, independent of other agents, makes job dispatching decisions based on its local view on the plant and employs reinforcement learning to improve its dispatching strategy. We delineate which extensions are necessary to render this learning approach applicable to job-shop scheduling problems of current standards of difficulty and present results of an adequate empirical evaluation