Modern large complex buildings, with their vast expanses and multipurpose usability, present diverse fire-spread scenarios, necessitating precise awareness of fire situations. Despite numerous studies integrating building information models with IoT and AI technologies to provide contextual information, there may be limitations in recognizing potential risks under dynamic fire behaviors during urgent situations. This paper presents a fire emergency management system based on an informative digital twin (IDT) to provide instantaneous fire situational information. Utilizing IDT, the system gathers building-specific fire scenario knowledge for immediate fire recognition. Leveraging this knowledge, the IDT assesses building fire risks and shares safety information on egress routes and zones with fire-related applications. A simulation case study revealed a fire recognition processing time of 0.88 s, acceptable for fire emergencies. Additionally, the IDT’s risk assessment reduced the total evacuation time by 65 s and potential injuries by 406, demonstrating the system’s effectiveness in fire emergency management.
Read full abstract