In this study, we consider the problem of forecasting monthly container throughput of Busan port, the largest port in South Korea. We proposed a forecasting model based on SARIMAX (Seasonal AutoRegressive Integrated Moving Average with eXogenous varia...
In this study, we consider the problem of forecasting monthly container throughput of Busan port, the largest port in South Korea. We proposed a forecasting model based on SARIMAX (Seasonal AutoRegressive Integrated Moving Average with eXogenous variables), a well-known traditional time-series model in which an appropriate exogenous variable is embedded to take into consideration the effect of COVID-19 during the pandemic era. The orders of the various terms included in the SARIMAX model were determined through the Box-Jenkins based approach, and the impact of COVID-19 was quantified by incorporating the number of the US cases as the exogenous variable. The 24 months spanning 2021 and 2022 were predicted using the proposed SARIMAX model, and the forecast results were compared with those of other existing prediction methods. The results showed that the SARIMAX model had the best predictive performance in terms of MAPE and RMSE among the tested methods.