<![CDATA[This study aims to forecast the sales of four consignment partner products using the Seasonal Autoregressive Integrated Moving Average (SARIMA) method based on monthly historical sales data from January 2023 to August 2025. The consignment system faces a high risk of product returns due to demand uncertainty, making accurate and reliable forecasting methods essential. The novelty of this research lies in the application of automatic parameter optimization using Grid Search to reduce subjectivity in selecting SARIMA models. The results indicate that the optimized SARIMA models provide good predictive performance and satisfy residual diagnostic tests. The KSP product shows the highest accuracy with a MAPE value of 7.69%, followed by KSO at 17.28%, while MO and MSM yield MAPE values of 25.54% and 27.55%, respectively, which are still acceptable for short-term operational planning. These findings confirm that a Grid Search–based SARIMA approach can serve as a reliable basis for decision-making in inventory control and in mitigating the risks of overstock and stockout in consignment schemes.]]>
Copyrights © 2026
