<![CDATA[DMA Medika Pharmacy faces challenges in managing its drug inventory due to unstable fluctuations in sales, which tend to increase at the beginning of each month and after Eid al-Fitr. Consequently, inventory planning is still based on estimates without optimally utilizing historical data analysis. This study aims to develop a web-based drug inventory forecasting system and implement the Seasonal Autoregressive Integrated Moving Average (SARIMA) method to predict drug inventory requirements based on sales data. This study used a quantitative approach using the Data Analytics Research Method and the Cross-Industry Standard Process for Data Mining (CRISP-DM) development model. The data consisted of historical sales records for 61 types of drugs from January to December 2025, which were processed using the Python programming language. The results showed that the system successfully processed historical data, developed SARIMA models, and generated drug sales forecasts for the period from January to June 2026, presented in the form of tables and graphs, with model performance evaluated using Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE). Evaluation of the 61 types of drugs using MAE and MAPE resulted in an average MAE value of 80.0638 and an average MAPE value of 488.11%, with 24 types of drugs classified in the moderate category and 37 types classified in the poor category. For Actifed Hijau, the evaluation produced an MAE value of 2.2644 and a MAPE value of 60.81%, which was classified in the poor category. These results indicate that the implementation of SARIMA has not yet produced optimal forecasting performance because the historical data used covered only a 12-month period. Nevertheless, the developed system can serve as supporting information to assist the drug inventory planning and management process at DMA Medika Pharmacy.]]>
Copyrights © 2026
