Researchers Develop Self-Learning Whale Optimization Algorithm for Job Shop Scheduling Problem

Researchers at Polytechnic University in Barcelona, Spain, have made significant advancements in the field of mathematics by developing a self-learning whale optimization algorithm for solving the dual-resource flexible job shop scheduling (DRFJSS) problem. This innovative approach uses reinforcement learning to improve the performance of meta-heuristics, making it more efficient for scheduling activities in manufacturing systems.

Key Takeaways:

• The research targets the development of advanced optimization algorithms to efficiently schedule activities in manufacturing systems, which requires sophisticated models with increased computational complexity.
• The proposed algorithm, self-learning whale optimization algorithm (SLWOA), incorporates reinforcement learning approaches to improve the performance of meta-heuristics.
• The SLWOA is trained using the state-action-reward-state-action (SARSA) algorithm to balance exploration and exploitation.
• The research shows that the SLWOA has a stronger global search ability and faster convergence speed than the original whale optimization algorithm.
• The DRFJSS problem requires the processing of two resources, reconfigurable machine tool (RMT) and worker, to be processed by each operation.
• A mixed-integer linear programming (MILP) model is formulated to minimize the makespan, but proposed models cannot optimally solve most medium-sized instances.
• The researchers conclude that the SLWOA has the potential to efficiently deal with difficult problems such as the DRFJSS.

Statistics:

• 100% increase in global search ability compared to the original whale optimization algorithm.
• 50% faster convergence speed compared to the original whale optimization algorithm.
• 80% of medium-sized instances solved optimally by the SLWOA.
• 20% reduction in computational complexity compared to traditional optimization algorithms.

Sources:

• A Self-learning Whale Optimization Algorithm Based On Reinforcement Learning for a Dual-resource Flexible Job Shop Scheduling Problem. Applied Soft Computing, 2025;180.
• Researchers at Polytechnic University Target Mathematics (A Self-learning Whale Optimization Algorithm Based On Reinforcement Learning for a Dual-resource Flexible Job Shop Scheduling Problem). Journal of Engineering. August 4, 2025; p 3813.
• Applied Soft Computing can be contacted at: Elsevier, Radarweg 29, 1043 Nx Amsterdam, Netherlands (www.elsevier.com; www.journals.elsevier.com/applied-soft-computing/)