Elaf Al Hashmi - Department of Computer Science Amirkabir University of Technology Tehran, Iran
Noor Al olawi - Department of Computer Science Amirkabir University of Technology Tehran, Iran
Mehdi ghatee - Department of Computer Science Amirkabir University of Technology Tehran, Iran
hamidreza eftekhari - Department of Computer Science Amirkabir University of Technology Tehran, Iran

Traffic congestion might be considered as a severe problem that faced the high population cities around the world. Transportation engineering tends to investigate its behaver based on information gathered different devices. In this paper basing on the GPS data, we tried to predict the congestion in the streets. For this aim neural network and regression models have been studied. The results show that NARX neural network, MLP preceptor neural network provides an accurate prediction. As the case of this work were areal historical GPS data gathered from the roads of Beijing – the Chinese city- during 2008 was considered. The proposed approach achieved a high level of prediction accuracy. The traffic congestion has been predicted with 90-94% accuracy

Traffic congestion prediction, Neural networks models, GPS data