Long-Term Vessel Arrival Forecasting at Port with Long Short-Term Memory: A Case Study

Predicting the arrival time of vessels is critical for a variety of applications such as port scheduling, cargo loading and unloading, and navigational safety. Accurate time arrival predictions can aid in increasing the efficiency and effectiveness of these operations, as well as avoiding delays and other disruptions. This research suggests employing a Long Short-Term Memory (LSTM) model, to forecast vessel arrival times. The outcomes are contrasted with the Autoregressive Integrated Moving Average (ARIMA) model, serving as a reference, and highlights the limitations of each model. To validate the models, historical data from the Mohammedia port terminal’s vessel records was used.