Online Optimization of Reconfiguration Planning for SRMS Based on DQN

Reconfigurable machine tools (RMTs) as key equipment of smart reconfigurable manufacturing systems (SRMS) can promote its flexibility when demand changes. The fundamental problem lies in dynamically reconfiguring the RMTs in SRMS efficiently and accurately by considering the flexibility of production precedence and operation sequences simultaneously. Therefore, a dynamic reconfiguration planning method for SRMS based on deep reinforcement learning is proposed in this chapter. The reconfiguration processes of SRMS are modeled by considering reconfiguration cost, moving cost, and processing cost. Deep Q-network (DQN) is adopted to find the optimal reconfiguration scheme with the highest return. A case study is presented to demonstrate the effectiveness and efficiency of the proposed method.