Forecasting Demand and Managing Surgical Supply Inventory

A successful supply chain is essential to the functioning of many different industries, including the healthcare sector. Healthcare management of supply chains require demand forecasting and inventory control to guarantee the best possible patient outcomes, keep costs under control, and reduce waste. Numerous advanced methods for inventory control and demand forecasting have been made possible by technological and data analytics advancements. To lower costs and improve patient care, this study intends to take advantage of these developments to precisely forecast demand as well as control the surgical supply inventory. A long-short-term memory (LSTM) model is created to forecast the demand for frequently used surgical supplies to accomplish this goal. Furthermore, the number of scheduled surgeries impacts the demand for specific surgical supplies. A literature-based LSTM model is used to predict medical case volumes and supplies for specific procedures. The adopted model now includes new features to account for COVID-19-related variations in surgical case volumes in 2020. The study uses Mixed Integer Programming (MIP) to create a dynamic replenishment model for multiple items. Forecasting can frequently be inaccurate, and demand is rarely predetermined in the real world. To address these issues, we developed a Two-Stage Stochastic Programming (TSSP) model.