EGU25-5495, updated on 14 Mar 2025
https://doi.org/10.5194/egusphere-egu25-5495
EGU General Assembly 2025
© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.
Poster | Friday, 02 May, 10:45–12:30 (CEST), Display time Friday, 02 May, 08:30–12:30
 
Hall X3, X3.34
Linking Disaster Risk Assessment at the Building Unit Level to Risk Reduction and Management
Han Kyul Heo1, Taehwan Hyeon1, and Jun Woo Kim2
Han Kyul Heo et al.
  • 1Architecture Urban Research Institute, Big Data Research, Sejong, Korea, Republic of
  • 2Human and environmental design, Cheongju University, Cheongju, Korea, Republic of

Disasters that occur in buildings and urban spaces have a profound impact on people's daily lives, often exacerbating anxiety. In the case of buildings, the financial repercussions of damage can be substantial, and there is also a possibility of physical injuries and fatalities among occupants. Consequently, there is an imperative to assess the risk of disasters in buildings and to devise measures that ensure safety.
The building risk analysis model developed in this study was designed to meet three criteria. First, it should be capable of responding to various disaster types, allowing the addition or removal of disaster categories as needed. Second, it should be able to present disaster risk at the building level. Third, the results must be easily comprehensible so that countermeasures can be prepared based on the assessed risk. To this end, we developed a building disaster risk analysis model to evaluate building-level risks for individual disaster types and to link these results. A multidimensional matrix was employed to assess fire, flood, and landslide risks at the building level.
We then proceeded to analyze the fire, flood, and landslide risks of buildings in a sample area and linked the results. Machine learning and deep learning techniques were applied to the risk analysis. The integration of these three risk categories resulted in the classification of buildings into eight distinct categories—ranging from “very risky” to “safe”—based on the number of high-risk disaster types. A total of 32,079 buildings were assessed in the target area, of which 48 buildings (0.15%) were identified as being at high risk of both fire and flood, primarily situated along rivers and boulevards. Conversely, 47 buildings (0.15%) were at high risk of both fire and landslide, mainly located in forested areas. No buildings were found to be at high risk for all three disaster types. A total of 95 buildings (0.3%) were determined to be at high risk for two or more disaster types.
A comprehensive approach to disaster risk mitigation necessitates the establishment of a building-level disaster risk check system. This system would be informed by the risk characteristics of each disaster type and the regional distribution of high-risk buildings. By leveraging this information, it would be possible to delineate inspection areas and items, as well as prepare countermeasures to ensure building safety. The establishment of a service that can assess disaster risk on a building-by-building basis will empower residents and users to proactively identify safety concerns and implement countermeasures, thus transitioning from a passive reliance on government assistance to a more autonomous and proactive approach to disaster mitigation.

How to cite: Heo, H. K., Hyeon, T., and Kim, J. W.: Linking Disaster Risk Assessment at the Building Unit Level to Risk Reduction and Management, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-5495, https://doi.org/10.5194/egusphere-egu25-5495, 2025.