Damage Prediction and Countermeasure Proposals Due to Urban Flooding with Machine Learning and Network Analysis (I) - Development of Estimation Model for Disaster Victims and Application - 


Vol. 39,  No. 6, pp. 44-52, Dec.  2024
10.14341/JKOSOS.2024.39.6.44


PDF
  Abstract

Owing to climate change and urbanization, the frequency of urban flooding is increasing, highlighting the need to accurately predict the scale of disaster victims for effective disaster response strategies. Most methods for estimating the number of victims rely on probabilistic functions and physical indicators, which are limited in their ability to reflect the nonlinear interactions among various variables discussed in the literature. To address these limitations, a random forest model that incorporates flood-related and characteristic variables and effectively captures these nonlinear interactions was developed. Additionally, a polynomial regression model was partially introduced to improve the accuracy of victim estimation in cases where the random forest model had not sufficiently learned from the data. The model was applied to Gwanak-gu, Seoul, Korea, as a case study area, and it predicted a high number of disaster victims in major flood-affected areas. The results confirmed that factors such as population, terrain, and building types—beyond just the flood area—contributed to these predictions. With the incorporation of additional data for model training and factors such as the distance to evacuation facilities in the future, this model is expected to become a crucial tool for urban planning and disaster management strategy development.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

박형준, 송수민, 김동현, 이승오, "Damage Prediction and Countermeasure Proposals Due to Urban Flooding with Machine Learning and Network Analysis (I) - Development of Estimation Model for Disaster Victims and Application -," Journal of the Korean Society of Safety, vol. 39, no. 6, pp. 44-52, 2024. DOI: 10.14341/JKOSOS.2024.39.6.44.

[ACM Style]

박형준, 송수민, 김동현, and 이승오. 2024. Damage Prediction and Countermeasure Proposals Due to Urban Flooding with Machine Learning and Network Analysis (I) - Development of Estimation Model for Disaster Victims and Application -. Journal of the Korean Society of Safety, 39, 6, (2024), 44-52. DOI: 10.14341/JKOSOS.2024.39.6.44.

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)

pISSN : 1229 - 9197
eISSN : 1875 - 0052

Impact Factor(2022): 2.5