A Roadside Safety Assessment Method Based on Bayesian Network
Heng Zheng, Liu Chang-cheng, Y U Liu · Journal of Highway and Transportation Research and Development · 2008
Roadside safety assessment is necessary for reducing the traffic accidents. An index system for roadside safety assessment was presented according to the studies of home and abroad. In addition, a method for roadside safety measurement using Bayesian network £¨BN£©, which has excellent abilities of expressing and reasoning knowledge under uncertain environment, was proposed to construct the safety assessment model. The assessment model is able to deal with complicated logic relationship as well as the different opinions of experts. The output of the safety assessment model can be used to predict and diagnose different safety conditions. The model was validated by application to safety assessment of a roadside, and the results show that the method is effective and reasonable.