Drug Stock Prediction at Dandy Primary Clinic Using the Single Moving Average Method

Abstract

This study addresses the issue of ineffective drug stock management at Klinik Pratama Dandy, often resulting in shortages and surpluses due to manual recording and suboptimal planning. This study assesses the accuracy of the Single Moving Average (SMA) method for forecasting drug stock demand and its utility in pharmaceutical logistics planning. Utilizing 24 months of historical data (January 2023–December 2024), partitioned into 80% for training and 20% for testing, prediction accuracy was evaluated with the Mean Absolute Percentage Error (MAPE). The findings reveal that the SMA method performs with 'High Accuracy' (MAPE < 10%). A key insight is that the optimal forecasting period (n) is specific to each drug, as demonstrated with Paracetamol (n=5, MAPE 1.19%) and Hydrocortisone Ointment (n=2, MAPE 6.62%). Consequently, a web-based system was created to automate this forecasting process, serving as a practical instrument for more efficient and objective drug inventory management.