Dynamic Scheduling of Crane by Embedding Deep Reinforcement Learning into a Digital Twin Framework

This study proposes a digital twin (DT) application framework that integrates deep reinforcement learning (DRL) algorithms for the dynamic scheduling of crane transportation in workshops. DT is used to construct connection between workshop service system, logical simulation environment, 3D visualization model and physical workshop, DRL support core decision scheduling. First, problem constructed as Markov process (MDP), corresponding double Q-network (DDQN) designed interact with logic environment complete offline training algorithm. Second, trained DDQN embedded into framework, then connected system realize online based on real-time states workshop. Finally, case studies under job arrival equipment failure scenarios are presented demonstrate effectiveness proposed framework. The numerical analysis shows method superior traditional method, it also suitable large-scale problems.