Department of Information and Tourism, Taipei City Government

The Taipei City Fire Department was invited to participate in the "AI Disaster Preparation and Relief Needs and Technology Aspirations Workshop" organized by the AI Research Center on December 27th,2023. The AI Research Center was jointly established by National Center for Research on Earthquake Engineering (NCREE) and National Taiwan University Department of Civil Engineering. The representative of the fire department shared the recent collaboration experience with the NCREE in enhancing the preciseness of earthquake loss estimation in Taipei City and the future application of AI.

The Taipei City Fire Department explained that due to the narrow and densely populated nature of Taipei City, strong earthquakes could lead to the collapse of buildings or bridges, causing road interruptions. This could significantly impact the dispatching of rescue capabilities, distribution of relief supplies, and the command of disaster relief efforts. Therefore, starting from 2020, the department collaborated with the NCREE. Based on the assumed earthquake scenarios in Taipei City's disaster preparation and rescue plan, it pioneered a nationwide initiative by using a finer "street" scale (currently "village" scale) to build a database of building seismic attributes and conduct disaster damage analysis.

In the "AI Disaster Preparation and Technology Aspiration Workshop" held on December 27, in addition to presenting Taipei City's recent earthquake preparation achievements, the NCREE announced "The Results of Applicating Large Language Models in Disaster Preparation." This presentation utilized data and estimation results established through recent collaboration with Taipei City, applying them to assist in intelligence gathering, spatiotemporal analysis, early disaster damage estimation, and disaster exposure warning. These insights serve as references for post-earthquake damage assessment, vehicle patrols, and other execution purposes.

The Taipei City Fire Department's ongoing "Disaster Response Collaboration Platform" project will further incorporated the above achievements. It is expected that through the system and AI interface, it will provide disaster damage assessment, disaster relief resource allocation, warning of blocked routes, planning of alternative routes, incident command post location and other immediate disaster relief information, aiming to serve as a reference for post-earthquake relief decision-making.