Enhancing production inventory management for imperfect items using fuzzy optimization strategies and Differential Evolution (DE) algorithms

Abstract

In the complex landscape of production inventory management for manufacturing industries, this research brings to light the significant impact of imperfect items—often an overlooked aspect that can result in elevated costs and operational inefficiencies. Addressing this oversight, we introduce a tailored mathematical model that harmoniously integrates demand forecasting, production planning, quality checks, and inventory control measures, specifically when imperfect items are in play. Notably, our model leverages the power of hybrid intelligent algorithms, fusing fuzzy systems with both genetic and differential evolution (DE) algorithms. Through sensitivity analysis, we discern the intricate effects of varying parameters on system performance. To cater to different operational exigencies, we introduce two distinct decision-making approaches. Our results highlight a promising reduction in anticipated yearly expenses, bolstering the reliability of events and ensuring costs are kept within predefined financial thresholds. Furthermore, we unveil a pioneering hybrid DE algorithm that distinctly outperforms the foundational DE in optimization problem-solving. A side-by-side comparison of our introduced methods elucidates invaluable insights into inventory cash flow management. Practitioners stand to gain immensely from our findings, which pave the way for cost-effective strategies, superior quality control, and an uptick in overall operational efficiency in the face of imperfect items. This seminal work not only offers actionable strategies but also beckons future research opportunities, thereby pushing the frontiers of manufacturing systems.

Keywords

• Production inventory
• Low quality and imperfect items
• Inventory control
• Differential Evolution
• Fuzzy Simulation