Analysis and Study of Firefighter Firefighting and Rescue Decision-Making Based on K-means Clustering

With the accelerated development of urbanization, firefighters are facing increasingly severe challenges, with their personal safety being greatly threatened. This paper conducts a comprehensive analysis of firefighters' decision-making preferences at different stages of tasks based on a survey of 129 firefighters from the Changning Branch of the Shanghai Fire Brigade. By applying the K-means clustering method, firefighters are profiled into three categories: conservative, balanced, and aggressive. The study reveals individual differences in decision-making among firefighters during firefighting tasks and statistically analyzes the personal experiences and understanding of the fire scene among different types of firefighters. The research indicates that firefighters with more than ten years of work experience are mainly conservative, while aggressive firefighters mostly have work experience ranging from 2 to 5 years. Conservative firefighters have a better understanding of the fire scene, and there are significant differences in the understanding of the fire scene among balanced firefighters. Aggressive firefighters have a higher level of familiarity with the fire scene information, but they do not pay enough attention to it, and they also have the highest historical injury rate. This study deeply explores the diversity of firefighters' decision-making behavior and provides corresponding profiles, laying a solid foundation for future personalized training, real-time command and control, and the realization of intelligent firefighting through the integration of advanced technology and intelligent support means.