DRL Approach for Flexible Job Shop Scheduling with Random Arrivals

A new paper proposes an event-based deep reinforcement learning (DRL) method to solve the Flexible Job Shop Scheduling Problem (FJSP) with random job arrivals. The approach uses the Proximal Policy Optimization algorithm and lightweight Multi-Layer Perceptrons to train an agent that minimizes total completion time. The state representation is directly accessible from the environment, and the agent selects from established dispatching rules. Simulations demonstrate that the DRL approach outperforms individual dispatching rules across datasets with varying heterogeneity and job arrival rates. The paper is available on arXiv.