This study proposes and validates an integrated model to reduce stockouts of medical surgical supplies in a Peruvian trading company, addressing a critical challenge in healthcare supply chain management where inventory failures directly affect service continuity and operational efficiency. A root cause analysis identified four key operational issues: reactive purchasing based on past sales, manual recording errors, lack of prioritization of near expiry products, and absence of expiration date verification during reception, which collectively lead to inventory inaccuracies and supply disruptions. To address these gaps, the proposed model integrates four complementary tools, Machine Learning for demand forecasting, Material Requirements Planning (MRP) for replenishment scheduling, Standard Work (SW) for process standardization, and First Expired, First Out (FEFO) for efficient inventory rotation, structured as a sequential and interdependent system that enables data driven decision making and improved operational control. Validation was carried out through a hybrid approach combining a field pilot for the Machine Learning and Standard Work components with a discrete event simulation in Arena for the MRP and FEFO components, allowing evaluation under both real and controlled conditions. The results demonstrate a 40.6% reduction in the Stockout Index (from 0.0577 to 0.0343), a 28% decrease in the Inventory Discrepancy Index (from 25% to 18%), a forecasting accuracy of 79%, and a 54.6% reduction in obsolescence (from 0.0793 to 0.036), evidencing a significant improvement in inventory performance.

Keywords

• Medical supply chain, inventory management, stockout, demand forecasting, FEFO.