Identification of Key Risk Factors for Resident Evacuation Based on BC‐FRAM

Abstract The community evacuation process is a complex system, coupled and linked by multiple interactions, and traditional safety engineering methods are unable to explain how multiple causes can lead to an incident. By introducing the concept of mediated centrality and constructing the BC‐FRAM critical risk factor identification model for residential evacuation, the BC‐FRAM model solves the problem that traditional functional resonance methods can ignore risk modules with low performance fluctuations but high impact levels in complex networks. The ability of the BC‐FRAM model to identify key risk factors in community evacuation is improved by a factor of 1 and the ability to identify failed functional connections is improved by 44.4% over the traditional FRAM model.