Traffic state prediction and improvement based on qualitative dynamic probabilistic networks
Kesheng Tang · Yunnan Daxue xuebao. Shehui kexue ban · 2012
The traffic problem has become a major obstacle of cities' development.The traffic state of cities could be predicted and improved.An efficient traffic state prediction can improve the traffic state and reduce the traffic obstruction.Qualitative Dynamic Probabilistic Networks(QDPNs) is one of the most efficient models in the uncertain knowledge and dynamically reasoning field.A traffic state predictionand improvement method based on QDPNs has been presented in this paper.The method can systematically model the traffic state of city.By reasoning it,the model can help us find the crux of the problem so that we will be able to take targeted measures to address the traffic congestion.