Analysis and Research on Fire Safety of University Dormitory Based on Bayesian Network

In recent years, with the continuous increase in the number of college students, a large number of students live together intensively, so there are many potential safety hazards. Dormitory fire threatens the safety of students’ lives and property. Quantitative and qualitative analysis and evaluation of the fire safety of college student dormitories are of vital importance for strengthening the fire safety management of colleges and universities. Therefore, this paper proposes a Bayesian network Analysis model of the fire in the dormitory of colleges and universities. Combining expert knowledge and a large number of related documents to analyze fire safety hazards, establish a relevant index system, build a Bayesian model for fire risk analysis, verify its accuracy based on relevant data in the past ten years, and analyze various variables through node backward reasoning, Extract the key factors of hidden dangers. The analysis results show that unsafe human behaviors, including illegal use of electrical appliances and smoking, are the main influencing factors leading to dormitory fires. Therefore, in order to effectively avoid accidents, it is necessary to strengthen the investigation of illegal electrical appliances in the student dormitory and improve the management system.