Research on Risk Assessment of Truck Rear-end Accident Based on Bayesian Network Model

  • Jinsong Dong, Yanhui Fan, Hao Zhang, Hongwei Zhang, Wei Zhao

Abstract

In order to minimize the occurrence of truck rear-end accidents and reduce the severity of accidents. Based on the in-depth investigation data of truck traffic accidents, this paper studies and determines risk factors set of truck rear-end accidents from four aspects: driver, vehicle, road and environment. Based on Bayesian theory, Bayesian network model was constructed by using Netica software. The Bayesian network structure was established by using data fusion method, and Expectation-Maximization algorithm that could process missing data was adopted to train the parameter in Netica. The model verification has been shown that the model was effective and can be used for the risk analysis and assessment of truck rear-end accidents. Bayesian network model in this paper has been used to quantitatively analyze the sensitivity of each risk factor and the significant risk factors of different risk levels in truck rear-end accidents have been put forward. The research results in this paper have been of important reference value for the formulation of prevention measures for truck rear-end accidents.

How to Cite
Jinsong Dong, Yanhui Fan, Hao Zhang, Hongwei Zhang, Wei Zhao. (1). Research on Risk Assessment of Truck Rear-end Accident Based on Bayesian Network Model. Forest Chemicals Review, 844-858. Retrieved from http://forestchemicalsreview.com/index.php/JFCR/article/view/970
Section
Articles