Integration of the BBN-NK-Boltzmann model of tunnel fire network scenarios with coupled forward and reverse rendition analysis

Guan ding Yang,Jie Liu,Wan qing Wang,Hao wen Zhou,Xiao dong Wang,Feng Lu,Li ting Wan,Liang yun Teng,Huyun Zhao
DOI: https://doi.org/10.1016/j.ress.2023.109546
IF: 7.247
2023-12-01
Reliability Engineering & System Safety
Abstract:One hundred tunnel fire cases worldwide were analyzed, and the key risk factors and maximum risk coupling forms were identified at the system level to improve tunnel safety. The BBN-NK model was introduced to positively analyze tunnel fires. Among the combined model, the BBN model of tunnel fire accident risk coupling was constructed to visualize the risk factor interactions, and the NK model was used for calculation to quantify the coupling degrees of various types of risk factors. Actual statistical data were used for inverse radar data comparison, and nonlinear curve fitting and residual analysis were performed to complete the correction and validation of the integrated BBN-NK model. The forward and reverse analysis deductions form a closed-loop process of the BBN-NK-Boltzmann model. The results show that controlling the coupling of four risk factors, namely human-machine-environment-management coupling, can effectively avoid tunnel fire accidents. Moreover, the typical adverse scenario evolution path and seven key nodes were found to be the key risks that must be considered during prevention and control. With the increase of the coupling factors, the frequency of tunnel fire accidents exhibit an increasing trend, and also conformed to the Boltzmann inversion curve equation.
engineering, industrial,operations research & management science
What problem does this paper attempt to address?