Support vector regression (SVR) model for forecasting number of passengers on domestic flights at Sultan Hasanudin airport Makassar

Abstract

Sultan Hasanudin Airport is one of the largest airports in Indonesia, located in Makassar City. Its strategic location is the entrance of eastern Indonesia because it is a transit airport to other eastern regions of Indonesia. The number of airplane passengers at Sultan Hasanudin Airport has increased and decreased each time depending on certain moments. The increase in the number of passengers is closely related to the moments of religious holidays or year-end holidays. Whereas the decrease in the number of passengers was greatly influenced by the policy of rising plane ticket prices some time ago. Estimated number of passengers every month is needed in planning and making appropriate decisions from the government relating to fluctuations in the number of domestic flight passengers at Sultan Hasanudin Airport. Therefore, accurate forecasting techniques are needed to predict the number of passengers in the future. Because the data pattern of domestic flight passengers at Sultan Hasanudin Airport is not stationary, the ARIMA model can be used. However, the data on the number of passengers has a nonlinear data pattern, so we need a method that can overcome these problems. In this study the SVR model is used to overcome nonlinear patterns in the data. Compared to the ARIMA model, SVR has the advantage because it does not require stationary data assumptions as in ARIMA. The results of forecasting data on the number of domestic flight passengers at Sultan Hasanudin Airport using SVR show better accuracy or accuracy compared to the ARIMA model because it has a smaller MAPE value.