Univariate throughput forecasting models on container terminal equipment planning

Abstract

Planning of Container Terminal equipment has always been uncertain due to seasonal and fluctuating throughput demand, along with factors of delay in operation, breakdown and maintenance. Many timeseries models have been developed to forecast the unforeseen future of container throughput to project the needed amount of port equipments for optimum operation. Conventionally, a "ratio" method developed by port consultants at early port design stage is adopted for equipment planning, giving no consideration to the dynamic growth of the port in terms of improved layout and technological advancement in equipments. This study seeks first to enhance the empirical approach of the equipment planning at the end of planning time horizon by including assumed coefficient of port capacity parameters. The second is to compare the size of equipment purchase by receiving different terminal's future throughput demand from two univariate forecasting models at planning time horizon. The empirical method of equipment planning will be tested against the conventional yard equipment per quay crane ratio after deriving the throughput demand from forecasting models of Holt-Winter's exponential smoothing and seasonal ARIMA (autoregression integrated moving average) model. Results in the form of graphs and tables indicate similar forecasting pattern by two models and equipment estimation proofs to avail more redundancy for optimum operation. Suggestions for better estimation of equipments are also made for future models.