A novel short-term traffic forecast model based on travel distance estimation and ARIMA

Short-term traffic forecast is vital for intelligent transportation systems. In this paper, a short-term traffic forecast model is proposed for the freeway network. This model proposed travel distance estimation (TDE) algorithm to predict the vehicles' positions in the near future. Moreover, historical traffic data are also considered by utilizing Autoregressive Integrated Moving Average (ARIMA) model. As such, both the spatial-temporal relation in freeway network and the historical traffic data analysis are integrated in the proposed model. The forecast results are more reliable compared to conventional methods. Experiments with real traffic scenarios validated the efficiency and accuracy of the proposed model.