Optimizing the strategic and operational levels of demand-driven MRP using a hybrid GA-PSO algorithm

Designing an efficient production planning system will reduce production costs, increase customer satisfaction, and ensure sustainability in the global competitive market. This paper focuses on multi-objective optimization in production planning through the utilization of the Demand-Driven material requirement planning (DDMRP) approach. We employ a simulation-optimization method that uses a novel hybrid genetic algorithm and particle swarm optimization, to simultaneously optimize conflicting inventory cost and stockout objectives. To the best of our knowledge, there are rarely studies in the literature that optimized and reviewed the suitable value of lead time and variability factors in this approach. For the first time, this study considers the integration of the strategic level (strategic buffer positioning phase) and operational level (planning phase) of the DDMRP approach due to the interaction of these two issues. This study focuses on the subjective to prove the performance of the DDMRP approach by eliminating its effect. The results demonstrate that an integrated review of both the strategic and operational levels in determining the lead time and variability factors results in lower costs. The average inventory cost, stockout, and total cost are reduced by 21%, 68%, and 78% respectively, compared with the results of the state-of-the-art methods.