A Novel DQN-Based Hybrid Algorithm for Integrated Scheduling and Machine Maintenance in Dynamic Flexible Job Shops

This paper focuses on the dynamic flexible job shop scheduling problem with constrained maintenance resources (DFJSP-CMR), a pressing challenge in modern manufacturing systems. As traditional rigid scheduling models fall short in meeting the demands of today's dynamic production environments, there is a growing need for intelligent approaches that can seamlessly integrate production scheduling and maintenance planning under resource limitations. To tackle this challenge, we propose a novel hybrid algorithm aimed at minimising makespan while addressing machine deterioration, unexpected failures and constrained maintenance resources. The core of our approach is a deep Q-network with maintenance insertion algorithm (DQN-MI) specifically designed for efficient maintenance scheduling. The algorithm features a 5×3 action space, constructed as compound rules, along with a reward structure that balances machine utilisation efficiency with effective maintenance operations. Extensive computational experiments conducted on diverse problem instances demonstrate that DQN-MI delivers superior performance, further validating the effectiveness and versatility of the proposed method in addressing complex scheduling challenges while maintaining the stability and reliability of manufacturing systems. This research contributes to the advancement of intelligent manufacturing by presenting a robust and practical solution for the integrated management of production scheduling and maintenance planning.