<![CDATA[Accurate demand forecasting is essential for salt-processing enterprises because fluctuating demand can affect production planning, inventory control, and distribution efficiency. This study examines the demand for table salt products at CV Restu Ibu by applying the Autoregressive Integrated Moving Average with Exogenous Variables (ARIMAX) model using product price as an external predictor. Monthly sales and price data for 1 kg, 2 kg, and 4 kg packaging sizes from January 2018 to December 2025 were analyzed through the Box–Jenkins procedure, including stationarity testing, model identification, parameter estimation, diagnostic checking, forecast evaluation, and projection for 2026–2028. The results show that all demand series became stationary after first-order differencing. The best models were ARIMAX (2,1,1) for the 1 kg product, ARIMAX (1,1,1) for the 2 kg product, and ARIMAX (0,1,1) for the 4 kg product. Diagnostic testing using the Ljung–Box test confirmed that all selected models satisfied the white-noise residual assumption. The forecasting accuracy was excellent, with Mean Absolute Percentage Error values of 2.363%, 2.824%, and 1.559%, respectively. The forecast results indicate a declining demand trend for the 1 kg product, an increasing trend for the 2 kg product, and a fluctuating pattern for the 4 kg product. These findings demonstrate that incorporating price information improves the representation of demand dynamics and supports product-specific forecasting for price-sensitive salt products. The study also confirms that packaging size produces heterogeneous demand behavior, making differentiated forecasting more appropriate than a single aggregate demand model in practice.]]>
