ARIMA Models for Bus Travel Time Prediction

In this paper, the time series model, Autoregressive Integrated Moving Average (ARIMA) is used to predict bus travel time. Travel time prediction is the main aspect that bus operator need to provide timetable for the operation and user’s information. The estimated parameters, appropriate model, and measures of model performance are presented. The objective of study is to find an appropriate time series model for predicting of bus travel time by considering the minimum of mean absolute relative error (MARE) and mean absolute percentage error (MAPE).
The data was collected from the case of bus service at the divided multi lane highway in Ipoh-Lumut corridor, Perak, Malaysia. Separately, those directions are Ipoh to Lumut and Lumut to Ipoh direction. The results show that the predicted travel times by using the moving average MA(2) and MA(1) model fits with the observed values for both travel directions, respectively. These fitted models have the minimum MARE and MAPE values among other tentative models. The conclusion, MA(2) and MA(1) models are able to be appropriately applied in the case, and those models can be used for bus travel time prediction in the timetable design.