Journal of Industrial Convergence (산업융합연구)

DAEHAN Society of Industrial Management (대한산업경영학회)
권호 표지
DOI QR코드

DOI QR Code

A Spatial Analysis of Seismic Vulnerability of Buildings Using Statistical and Machine Learning Techniques Comparative Analysis

통계분석 기법과 머신러닝 기법의 비교분석을 통한 건물의 지진취약도 공간분석

  • Seong H. Kim (Department of GIS Engineering, Namseoul University ) ;
  • Sang-Bin Kim (Department of GIS Engineering, Namseoul University) ;
  • Dae-Hyeon Kim (AI Research Team, Data Science Lab, Korea Electric Power Corporation)
  • 김성훈 (남서울대학교 공간정보공학과) ;
  • 김상빈 (남서울대학교 공간정보공학과 대학원 ) ;
  • 김대현 (한국전력 데이터사이언스연구소)
  • Received : 2022.11.15
  • Accepted : 2023.01.20
  • Published : 2023.01.28
Download PDF
⟨ Previous Next ⟩

Abstract

While the frequency of seismic occurrence has been increasing recently, the domestic seismic response system is weak, the objective of this research is to compare and analyze the seismic vulnerability of buildings using statistical analysis and machine learning techniques. As the result of using statistical technique, the prediction accuracy of the developed model through the optimal scaling method showed about 87%. As the result of using machine learning technique, because the accuracy of Random Forest method is 94% in case of Train Set, 76.7% in case of Test Set, which is the highest accuracy among the 4 analyzed methods, Random Forest method was finally chosen. Therefore, Random Forest method was derived as the final machine learning technique. Accordingly, the statistical analysis technique showed higher accuracy of about 87%, whereas the machine learning technique showed the accuracy of about 76.7%. As the final result, among the 22,296 analyzed building data, the seismic vulnerabilities of 1,627(0.1%) buildings are expected as more dangerous when the statistical analysis technique is used, 10,146(49%) buildings showed the same rate, and the remaining 10,523(50%) buildings are expected as more dangerous when the machine learning technique is used. As the comparison of the results of using advanced machine learning techniques in addition to the existing statistical analysis techniques, in spatial analysis decisions, it is hoped that this research results help to prepare more reliable seismic countermeasures.

최근 지진 발생 빈도가 증가하고 있는 반면 국내 지진 대응 체계는 취약한 현실에서, 본 연구의 목적은 통계분석 기법과 머신러닝 기법을 활용한 공간분석을 통해 건물의 지진취약도를 비교분석 하는 것이다. 통계분석 기법을 활용한 결과, 최적화척도법을 활용해 개발된 모델의 예측정확도는 약 87%로 도출되었다. 머신러닝 기법을 활용한 결과, 분석된 4가지 방법 중, Random Forest의 정확도가 Train Set의 경우 94%, Test Set의 경우 76.7%로 가장 높아, 최종적으로 Random Forest가 선정되었다. 따라서, 예측정확도는 통계분석 기법이 약 87%, 머신러닝 기법이 76.7%로, 통계분석 기법의 예측정확도가 더 높은 것으로 분석되었다. 최종 결과로, 건물의 지진취약도는 분석된 건물데이터 총 22,296개 중, 1,627(0.1%)개의 건물데이터는 통계분석 기법 사용 시 더 위험하다고 도출되었고, 10,146(49%)개의 건물데이터는 동일하게 도출되었으며, 나머지 10,523(50%)개의 건물데이터는 머신러닝 기법 사용 시 더 위험하게 도출되었다. 기존 통계분석 기법에 첨단 머신러닝 기법활용결과가 추가로 비교검토 됨으로써 공간분석 의사결정에 있어서, 좀더 신뢰도가 높은 지진대응책 마련에 도움이 되길 기대한다.

Keywords

References

  1. Park, S. J. (2007). Tectonic movement in the Korean peninsula (II): A geomorphological interpretation of the spatial distribution of earthquakes. Journal of the Korean Geographical Society, 42(4), 488-505.
  2. Chun, Y. (2017). A Study on Earthquake Hazard Mapping using Risk Factors. Proc. of Korean Society for Geospatial Information Science, Seoul, Korea, May, 25-26.
  3. Koh, J. H., Kwon, J. H., & Choi, Y. S. (2005). Error Assessment of Attitude Determination Using Wireless Internet-Based DGPS. Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography, 23(3), 239-249.
  4. Hwang, C. H. (2012). A Study on Urban Damage Assessment by Earthquake Damage Assessment System. Seoul Venture University.
  5. Han, J., & Kim, J. (2019). A GIS-based seismic vulnerability mapping and assessment using AHP: A case study of Gyeongju, Korea. Korean Journal of Remote Sensing, 35(2), 217-228. https://doi.org/10.7780/KJRS.2019.35.2.2
  6. Sun, C. G. (2009). Seismic zonation on site responses in Daejeon by building geotechnical information system based on spatial GIS framework. Journal of the Korean Geotechnical Society, 25(1), 5-19. https://doi.org/10.7843/KGS.2009.25.1.5
  7. Federal Emergency Management Agency (US) (Ed.). (2015). Rapid visual screening of buildings for potential seismic hazards: a handbook. Government Printing Office.
  8. Federal Emergency Management Agency (US) (Ed.). (2015). Rapid visual screening of buildings for potential seismic hazards: a handbook. a Handbook (Third Edition).Government Printing Office.
  9. Catlin, A. C., & Pujol, S. (2015). NIST Disaster and Failure Studies Data Repository: The Chile Earthquake Database-Ground Motion and Building Performance Data from the 2010 Chile Earthquake-User Manual.
  10. S. B. Kim. (2019). A Development of a Seismic Vulnerability Model and Relative Spatial Analysis for Buildings. Paper of Master's Degree. Namseoul University, Cheonan.
  11. Kim, S. B., & Kim, S. H. (2020). A Development of a Seismic Vulnerability Model and Spatial Analysis for Buildings. Journal of the Korea Convergence Society, 11(10), 9-18. https://doi.org/10.15207/JKCS.2020.11.10.009
  12. He, Q., Shahabi, H., Shirzadi, A., Li, S., Chen, W., Wang, N., ... & Ahmad, B. B. (2019). Landslide spatial modelling using novel bivariate statistical based Naive Bayes, RBF Classifier, and RBF Network machine learning algorithms. Science of the total environment, 663, 1-15. https://doi.org/10.1016/j.scitotenv.2019.01.329
  13. Bui, D. T., Hoang, N. D., & Samui, P. (2019). Spatial pattern analysis and prediction of forest fire using new machine learning approach of Multivariate Adaptive Regression Splines and Differential Flower Pollination optimization: A case study at Lao Cai province (Viet Nam). Journal of environmental management, 237, 476-487.