Long-term Traffic Flow Forecasting Based on an Artificial Neural Network

Abstract

There is no doubt that a good knowledge of traffic demand has a direct impact on improving traffic management. Road traffic is strongly correlated with many factors such as day of week, time of day, season and holidays which make it suitable for prediction. In this paper, we develop a neural network model for hourly traffic prediction that makes full use of these temporal characteristics. The proposed algorithm is tested on a real-world case, and the experiment results is presented to evaluate its accuracy.