Improve urban passenger transport management by rationally forecasting traffic congestion probability

Abstract

A Bayesian network (BN) approach is proposed in this study to analyse the overall traffic congestion probability of an urban road network in consideration of the influence of applying various transport policies. The continually expanding urbanised region of Beijing has been chosen as the study area because of its rapid expansion and motorisation, which lead to the severe traffic congestion occurring nearly every day. It is demonstrated that the proposed BN approach is able to rationally predict the probability of the overall traffic congestion that will take place given a certain transport policy. It is also proven that increasing the number of buses providing convenient passenger transport service in the urbanised region of Beijing will most effectively reduce the probability of the traffic congestion in this area, especially when the newly constructed roads in the same region are put into use.