Risk analysis for hazardous chemical vehicle-bridge transportation system: A dynamic Bayesian network model incorporating vehicle dynamics

This study aims to analyze the risk of transporting hazardous chemicals on sea-crossing bridges using a dynamic Bayesian network (DBN) model that incorporates vehicle dynamics. Firstly, the cause-consequence relationship analysis is constructed using the bow-tie (BT) model, which is then translated into a Bayesian network (BN) by mapping algorithms. Based on the dynamic model, the occurrence probabilities of rollover and sideslip under different wind speeds are calculated as conditional probabilities. Secondly, a DBN model that satisfies the Markov assumption and time invariance is established to realize short-term risk prediction. Finally, the proposed model is applied to a sea-crossing bridge in Zhejiang, and other node parameters are obtained by combining the monitoring data of the vehicle-bridge transportation system (VBTS) monitoring platform and expert experience. The results indicate that vehicle failure has the highest impact on VBTS, and unsafe driver behavior and road alignment are the most vulnerable root causes, which should receive more attention. Additionally, wind sensitivity to VBTS is significant and cannot be ignored. The proposed method can effectively address the risks and challenges posed by hazardous chemical transportation on sea-crossing bridges and provides valuable insights with practical application to enhance transportation safety.

Automated summary

This study aims to analyze the risk of transporting hazardous chemicals on sea-crossing bridges using a dynamic Bayesian network (DBN) model. The study found that vehicle failure has the highest impact on the transportation system and should be given more attention. Additionally, wind sensitivity to the sea-crossing bridge transportation system is significant and cannot be ignored.