Drug forecasting and supply model design using Artificial Neural Network (ANN) and Continuous Review (r, q) to minimize total supply cost

Abstract

The Mentawai Islands Regency Regional General Hospital faces a significant challenge with an 83% overstock of Medical Consumables, leading to increased inventory costs and potential damage and expiration of items. This exceeds the 1% pharmaceutical drug storage standards the Ministry of Health set. This study aims to optimize demand forecasting and minimize total inventory costs through a two-stage process. Firstly, demand forecasting is conducted using Artificial Neural Network (ANN), predicting a future demand of 10,036 units of Medical Consumables. Subsequently, the optimal order quantity and reorder points are calculated using the continuous review (r, Q) approach. The results reveal the optimal order quantities and reorder points for four types of Medical Consumables. This research introduces a novel approach by employing ANN for demand forecasting, then calculating optimal order quantities and reorder points using continuous review (r, Q). The cost components considered in the inventory cost calculation include purchasing cost, holding cost, shortage cost, order cost, outdating cost, and inspection cost. The designed forecasting models aim to enhance inventory management efficiency, optimize cost control, and improve patient services. The limitation of this research is that it only used five types of consumable medical materials to carry out this research due to limited data access. It is hoped that future research can use other types of drugs as well as a periodic review and forecasting approach using GA.