Research on Building Fire Risk Assessment Models and Evacuation Path Optimization Based on Big Data Analysis

This study aims to reduce the probability of congestion and optimize the utilization of emergency exits through scientific, professional, and refined research methods, ensuring the safe and efficient evacuation of personnel.A building fire risk assessment model is constructed by combining fuzzy theory with Bayesian networks (BN), with building design standards used as evaluation criteria to enhance the accuracy of risk assessment.Building A is selected as the actual research subject for fire risk assessment.Additionally, to optimize the internal evacuation paths of Building A, BIM technology is integrated with fire simulation and personnel evacuation planning.Numerical simulation and visualization analysis methods are employed to conduct personnel evacuation simulations for a specific floor area of Building A. A fire personnel evacuation path planning method based on an improved Floyd algorithm is proposed.After comparison, the optimized evacuation path was shortened by 25.6 meters, evacuation time was reduced by 33.8 seconds, and evacuation efficiency was improved by 26%.The methods proposed in this paper effectively enhance fire evacuation efficiency and improve fire emergency management levels.