This paper addresses the characteristics of urban rail transit fires and proposes a predictive model combining evacuation simulation and random forests for rapid forecasting of evacuation targets. Firstly, the characteristics of urban rail transit fires and key factors affecting evacuation are analyzed. Secondly, a three-dimensional model for crowd evacuation simulation is constructed. Finally, a predictive model based on random forests is developed, with an analysis of the importance of predictive variables. The random forest approach can effectively deal with complex nonlinear relationships and large amounts of data, and improve the prediction accuracy and robustness of the model. The results indicate that this method can rapidly predict emergency evacuation scenarios in urban rail transit fires, achieving an accuracy of 95%.


    Access

    Download


    Export, share and cite



    Title :

    Analysis of Fire Characteristics and Emergency Evacuation Influencing Factors in Urban Railway Transportation


    Additional title:

    Advances in Engineering res



    Conference:

    International Symposium on Traffic Transportation and Civil Architecture ; 2024 ; Tianjin, China June 21, 2024 - June 23, 2024



    Publication date :

    2024-09-24


    Size :

    8 pages





    Type of media :

    Article/Chapter (Book)


    Type of material :

    Electronic Resource


    Language :

    English





    EMERGENCY EVACUATION DEVICE AND RAILWAY VEHICLE

    CHEN DONGDONG / ZHAO JIANGTAO / DUAN HONGLIANG et al. | European Patent Office | 2025

    Free access

    Emergency Evacuation Simulation of Urban Railway Station Based on EXODUS Software

    Liu, Zhigang / Huang, Yuanchun / Pan, Haize | ASCE | 2011


    Simulating Passenger Evacuation in Railway Station under Fire Emergency using Safe Zone Approach

    Salarian, Amir Hossein / Mashhadizadeh, Aida / Bagheri, Morteza | Transportation Research Record | 2020



    Transportation Guide for All-Hazards Emergency Evacuation

    D. Matherly / B. Wolshon / J. Renne et al. | NTIS | 2013