Residual Scheduling: A New Reinforcement Learning Approach to Solving Job Shop Scheduling Problem
This addresses scheduling inefficiencies in manufacturing industries with a novel method that shows strong performance gains.
The paper tackles the NP-hard job shop scheduling problem (JSP) and its flexible variant (FJSP) by proposing a new reinforcement learning approach called residual scheduling, which removes irrelevant machines and jobs to focus on remaining ones, achieving state-of-the-art results on most benchmarks and zero gap in 49 out of 50 large instances.
Job-shop scheduling problem (JSP) is a mathematical optimization problem widely used in industries like manufacturing, and flexible JSP (FJSP) is also a common variant. Since they are NP-hard, it is intractable to find the optimal solution for all cases within reasonable times. Thus, it becomes important to develop efficient heuristics to solve JSP/FJSP. A kind of method of solving scheduling problems is construction heuristics, which constructs scheduling solutions via heuristics. Recently, many methods for construction heuristics leverage deep reinforcement learning (DRL) with graph neural networks (GNN). In this paper, we propose a new approach, named residual scheduling, to solving JSP/FJSP. In this new approach, we remove irrelevant machines and jobs such as those finished, such that the states include the remaining (or relevant) machines and jobs only. Our experiments show that our approach reaches state-of-the-art (SOTA) among all known construction heuristics on most well-known open JSP and FJSP benchmarks. In addition, we also observe that even though our model is trained for scheduling problems of smaller sizes, our method still performs well for scheduling problems of large sizes. Interestingly in our experiments, our approach even reaches zero gap for 49 among 50 JSP instances whose job numbers are more than 150 on 20 machines.