Short Term Prediction of Hourly Traffic Volume Using Neural Network in Interurban Freeway

Abstract

Traffic congestion in metropolitan area such as Great Tunis, has become more and more serious. Over the past decades, research in this area has grown and become an ever-increasing problem. Many academic research and public authorities' efforts have been made to alleviate this issue. In this paper, we investigate the application of neural network time series model to predict hourly traffic volumes for a Tunisian national highway (N8). A total 1-year of traffic volume data of 14 stations in both directions, where used in this analysis. The study applies Artificial Neural Network (ANN) for short term forecasting of traffic flow using past traffic data. The model incorporates traffic volume as input variable. The simulation experimental results show that the model is with good stability and the mean square is used as evaluation criteria.