Optimizing (s, S) inventory policy with lost sales, uncertain demand and lead time using simulation and heuristics: a case study of food products

Effective inventory management is essential for minimizing operational costs and maintaining high service levels in supply chains, particularly under conditions of uncertainty. This study addresses inventory optimization using the periodic review (s,S) policy, focusing on systems characterized by stochastic demand, variable lead times, and limited storage capacity. A simulation-optimization framework was developed tominimize total annual costs: comprising holding, shortage, and over-storage costs, ensuring high service levels. The methodology utilizes historical demand data from three product categories (slow-, medium-, and fast-moving) to model demand variability and employs random number generation to simulate both demand and lead time. Two optimization techniques, Grid Search and Golden Section Search (GSS), were applied to identify the optimal reorder point (s) and order-up-to level (S). The results demonstrate the effectiveness of these techniques in reducing costs and handling complex inventory scenarios involving lost sales and storage constraints. The study contributes a robust decision-making tool for inventory control, offering practical value for industries facing fluctuating demand and constrained resources, such as healthcare and retail. The findings highlight that simulation-based approaches, integrated with heuristic optimization, significantly enhance the adaptability of the (s, S) policy in real-world applications.