Solving a Job Shop Scheduling Problem Using Q-Learning Algorithm

Job Shop Scheduling Problem (JSSP) is among the combinatorial optimization and Non-Deterministic Polynomial-time (NP) problems. Researchers have contributed to this area using several methods. Among them, we cite machine learning algorithms, more precisely Reinforcement Learning (RL). This algorithm suitable for discussed problem as agents can learn decisions optimize them according environment’s state. Once process efficient, RL be used in real-time cope with changes. paper deals JSSP algorithm, specifically a Q-learning algorithm. We propose new representation of state environment based on loads agent evaluated twice two different The actions selected by are dispatching rules. Our QL approach such compared results obtained literature.