<![CDATA[Managing the number of railway passengers in Indonesia presents a significant challenge for PT Kereta Api Indonesia, particularly in relation to transport capacity planning, scheduling, and resource optimization. Forecasting therefore plays a crucial role in supporting effective decision-making. This study aimed to forecast railway passenger volumes using the Holt–Winters Triple Exponential Smoothing method and the Seasonal Autoregressive Integrated Moving Average (SARIMA) model, and to compare their forecasting performance. This applied research utilized secondary monthly data published by Statistics Indonesia (BPS), covering the period from January 2022 to December 2024, with forecasts generated for January to June 2025. Model performance was evaluated using the Mean Absolute Percentage Error (MAPE) criterion. The results indicated that the SARIMA model outperformed the Holt–Winters variants and other SARIMA specifications, achieving the lowest MAPE value of approximately 3%. Based on this evaluation, the SARIMA model was identified as the most accurate model for forecasting Indonesian railway passenger volumes. The findings suggest that SARIMA-based models provide a reliable approach for supporting railway passenger demand forecasting in Indonesia.]]>
Copyrights © 2026
