Particle swarm optimization algorithm for design of an adaptive Kanban system based on optimization via simulation
Khouloud Elloumi, Ahmad Ammar, Mounir Benaissa · Journal of Industrial and Production Engineering · 2025
Manufacturing systems face challenges in adapting to fluctuating demand, making traditional fixed-parameter Kanban systems inefficient. This study addresses these limitations by proposing an adaptive Kanban-based pull control mechanism that dynamically adjusts the number of Kanbans based on inventory levels and backorders. To design an adaptive Kanban system, a metaheuristic optimization approach using particle swarm optimization is integrated with simulation to account for stochastic and time-varying customer demand. Previous studies have either overlooked demand fluctuations in Kanban systems or underutilized the advantages of particle swarm optimization. The results demonstrate that the proposed algorithm significantly reduces total production costs compared to existing metaheuristic methods. These findings underscore the importance of integrating simulation and optimization to enhance decision-making in manufacturing, providing a robust framework for improving production flow control in uncertain environments.