Forecasting the Cargo Throughput for Haiphong Port in Vietnam

Port throughput forecasting is fundamental in port optimization. A reliable prediction model is essential for the terminal operators to make decisions on planning and renovation of building structure and other port facilities. By monitoring the changes in seasonal patterns and business cycles in months or quarters, the predicted values help port managers in decision making and planning in the context of small and unexpected changes. In this paper, the authors reviewed a various of commonly used forecasting methods applied for the time-series data in the short-term. By applying a set of monthly data of Haiphong port from January 2003 to February 2019 to these models and evaluating forecast accuracy by root mean squared error (RMSE), we found that the Winters exponential smoothing method appears to be the best model for forecasting total cargo throughput with trend and seasonal variations. The empirical results could be used as a reliable scientific source for the port managers and the departments to make short-term plans for upgrading facilities and setting up effective loading and unloading plans, and then contribute to avoiding congestion and reducing unnecessary waste.