Hybrid Adaptive Metaheuristic Scheduling (HAMS) for Efficient Real-Time Job Shop Scheduling

Efficient scheduling in dynamic job shop environments is critical to improving manufacturing productivity and responsiveness. This paper presents Hybrid Adaptive Metaheuristic Scheduling (HAMS)—a novel framework designed to address the limitations of traditional metaheuristic approaches such as premature convergence, fixed parameter settings, and limited adaptability. HAMS integrates three key components: Adaptive Metaheuristic Switching (AMS) using reinforcement learning to dynamically select optimal algorithms; Dynamic Parameter Tuning (DPT) to fine-tune algorithm parameters in real time; and a Hybrid Neuro-Heuristic Decision Model (HNHDM) powered by neural networks to predict job completion times and optimize resource allocation. The framework is evaluated using standard job shop scheduling benchmarks (FT06, FT10, FT20) and compared against Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Simulated Annealing (SA). Results show that HAMS achieves a 12.5% reduction in makespan, an 18.7% improvement in computational efficiency, and a 15.3% increase in resource utilization. These findings demonstrate the effectiveness and scalability of HAMS for real- time industrial scheduling applications.