<![CDATA[Accurately predicting ship loading and unloading volumes at ports is essential for optimizing operations and strategic planning. This study compares three time series forecasting models: SARIMA, LSTM, and Prophet, to predict ship loading and unloading volumes at Tanjung Priok, Indonesia's busiest port. The dataset covers the period from January 2017 to August 2025 in monthly increments. Hyperparameter tuning was performed on each model to determine the optimal hyperparameter values. Model performance was evaluated on the testing dataset using the Mean Absolute Percentage Error (MAPE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and accuracy metrics. Based on these four metrics, SARIMA had the best performance with an accuracy rate of 87.97%, followed by the LSTM model at 85.38%, and the Prophet model at 69.40%. SARIMA can capture seasonal patterns and trends in loading and unloading data, making it useful for decision-making related to ship scheduling, crane allocation, and dock management. This study emphasizes the importance of selecting a model based on the characteristics of the data and demonstrates that traditional statistical models, such as SARIMA, are competitive with deep learning models for time series with strong seasonality.]]>
Copyrights © 2025
