<![CDATA[Drug stock imbalances, including shortages or surpluses, remain a major challenge in pharmacy management at community health centers. This study aims to implement the Triple Exponential Smoothing (TES) method to forecast drug requirements by considering seasonal patterns and monthly consumption trends. The dataset used in this study consists of drug expenditure reports from May 2024 to May 2025, including Paracetamol 500 mg, Antacid DOEN Tablets, and Acetylcysteine 200 mg as sample drugs. The forecasting process was conducted using the RapidMiner software add the Holt-Winters Triple Exponential Smoothing model with smoothing parameters = 0.3, = 0.1, and = 0.3, as well as a seasonal length of four periods. The results indicate that the model achieved Mean Absolute Percentage Error (MAPE) values ranging from 15,37% to 17,97% and Mean Absolute Error (MAE) values between 159.922 to 495.938. These findings demonstrate a good level of forecasting accuracy, indicate that the TES method is suitable as a decision support tool for drug inventory planning in community health centers. Furthermore, the forecasting results provide a more measurable estimation of future drug demand compared to the average consumption approach commonly used in the preparation of Drug Requirement Plan (Rencana Kebutuhan Obat – RKO).]]>
Copyrights © 2026
