Research on railway emergency rescue time prediction based on IFBN-GERTS network model

Due to the great uncertainty of railway emergencies, it is difficult for the management departments to manage and control the rescue process and time. Therefore, this paper establishes an emergency rescue graphical evaluation review technique stochastic (GERTS) network to predict the rescue time based on the logical relationship between the rescue processes. Considering the uncertainty of on-site information at the early stage of an emergency, Bayesian networks (BN) are combined with intuitionistic fuzzy set (IFS) to construct intuitionistic fuzzy Bayesian networks (IFBN) for emergency rescue decision inference. The direct fuzzy numbers of the probabilities of the IFBN output nodes are introduced into GERTS as branching probabilities of the network nodes, increasing its ability to express uncertainty. Finally, the network logic is modeled in MATLAB and Monte Carlo ideas are adopted for the selection of branching activities according to the branching probabilities, and the network simulation results are statistically analyzed to obtain the expected time for the whole rescue process. By predicting the emergency rescue time, it can provide a basis for the adjustment of the travel plan and assist emergency managers in making scientific decisions.