Optimizing Inventory Management with Seasonal Demand Forecasting in a Fuzzy Environment

This study explores an inventory management model in today's business landscape, where organizations increasingly rely on Machine Learning for demand-driven stock control.The proposed model accounts for imperfect and deteriorating products within a fuzzy environment, allowing for shortages and partial backlogging.Degradation rates and faulty percentages are classified as fuzzy variables since they are unpredictable and impacted by undefined conditions.The goal is to calculate the appropriate replenishment cycle and ordering quantity while reducing the optimal overall cost, including carbon pollution costs, within a constrained planning horizon.The defuzzification technique uses the sign distance approximation technique.Leveraging Machine Learning, the study utilizes a seasonal demand forecasting methodology.A numerical illustration supports the mathematical approach by demonstrating its capacity to estimate demand for deteriorating products.This facilitates optimized inventory management aligned with forecasted demand.A comparative examination emphasizes the positive aspects of AI learning-based forecasting systems over determined demand circumstances.Sensitivity analysis provides insights into the impact of various parameters on optimal solutions, contributing valuable managerial perspectives.