Shop Floor Scheduling Method by Integrating Deep Reinforcement Learning and Attention Mechanism
The flexible job shop scheduling problem is a typical scheduling problem in the manufacturing industry.In a complex and changeable production environment,traditional scheduling methods are difficult to ensure both production efficiency and quality.In this paper,an end-to-end deep reinforcement learning method for dynamic flexible job shop scheduling problem is proposed.In order to improve solution quality,an Heterogeneous Graph Attention Network(HGAN)is constructed as a decision module,allowing reinforcement learning agents to be searched in its solution space.The network model is trained by the Proximal Policy Optimization(PPO)algorithm,and the trained model can be used to generate the optimal scheduling strategy.Experimental results on random instances and benchmark instances show that the proposed method is superior to the scheduling rules and meta-heuristics in terms of solution quality and running time.
万方数据
维普
