Minimization of Total Setup Time on a Bottleneck Machine Using Particle Swarm Optimization

In today's competitive global market, efficiency and cost-effectiveness in production processes are vital for businesses. The optimization of setup times resulting from scheduling and product changeovers is a strategic factor that directly impacts production line efficiency, capacity utilization, and resource sustainability. This study aims to minimize total setup times by optimizing product sequencing on a bottleneck machine within a tire manufacturing line. Due to its combinatorial nature and large solution space, the problem presents a complex optimization challenge. To address this, Particle Swarm Optimization (PSO) is applied and compared with several metaheuristic algorithms—Tabu Search (TS), Ant Colony Optimization (ACO), and the Lin-Kernighan Heuristic (LKH). These algorithms were evaluated using real production data and compared with the existing sequencing method used by the supply chain department. Extensive analysis based on 10-day production plans revealed significant improvements by all metaheuristic methods over the current approach. PSO achieved a 12.83% improvement, TS 10.71%, LKH 12.71%, and ACO demonstrated the best performance with a 16.64% improvement. The statistical significance of the results was confirmed using a T-test, showing that metaheuristic solutions significantly outperform traditional sequencing methods. This study highlights the value of setup time optimization in complex manufacturing systems and offers a practical model for both academic research and industrial decision-making. The detailed comparison of algorithm performance also provides a strong foundation for future studies in this field.