<![CDATA[PT SMS is a company engaged in the distribution of PVC ceiling panels, where inventory management plays a crucial role in maintaining operational continuity. The primary challenge faced by the company is unpredictable demand fluctuations, which frequently result in either overstocking or stock shortages (stockouts). This study aims to develop an inventory management optimization system capable of accurately forecasting future demand. The proposed methodology integrates two main techniques. First, the Z-Score method is employed for outlier detection to identify and handle extreme or abnormal sales data, thereby preventing bias in forecasting results. Second, the Weighted Moving Average (WMA) method is applied to forecast product demand based on historical sales data, assigning greater weight to more recent observations to better capture current market trends. The results indicate that the implementation of the Z-Score method effectively removes anomalies from the dataset prior to the forecasting process. Forecast accuracy was evaluated using the Mean Absolute Percentage Error (MAPE), demonstrating that the WMA method produces more accurate forecasts than conventional forecasting approaches. The proposed system enables PT SMS to improve inventory procurement planning, minimize inventory holding costs, and enhance the efficiency of meeting customer demand. ]]>
Copyrights © 2026
