Forecasting Methods for the Number of Exporting Containers to Indonesia in Covid-19 Situation: A Case Study of Freight Forwarder Company

  • กิตติเทพ ไชยเทพ Rangsit University
Keywords: Container; Forecasting; Time Series; Demand; Forecasting Model


The purpose of this study is to research container demand forecasting. Due to the Covid-19 situation, as a result, there are measures to reduce the number of employees in various places. These are the reasons that the operating time of the container at the terminal is longer than before. On the other hand, the volume of export demand began to increase, but the space and volume of containers on board were limited. As a result, Ocean freight or freight rates will increase by more than 300% in 2021. Forecasting sales or customer demand is an important business plan. The researcher recognizes the importance of forecasting and therefore studies appropriate forecasting techniques. In the case study where the researcher collected the data, Freight Forwarder is one of the service providers for booking containers for importers and exporters. Forecasting sales or customer demand is an important business plan. This research aims to compare various forecasting techniques that provide the best choice for the collected data by minimizing mean square error. In order to forecast the demand for the container, we collect the record of the demand for the container from Jan 2020 – Jun 2021 and use nine forecasting methods. The methods applied to compare are Simple Moving Average, Weighted Moving Average, Exponential Smoothing, Stationary with Additive and Multiplicative Seasonal Effect, Double Moving Average, Double Exponential Smoothing, and Holt-Winter’s Method for Additive and Multiplicative Seasonal Effect. The reason for selecting these methods is that there are many data types, such as random data, trend data, seasonal data, and trends with seasonal data. The results find that Exponential Smoothing is the best forecasting method that can minimize mean square error, which is 3,855.97. Therefore, an entrepreneur can apply such recommended methods to forecast the future demand for the container.