Journal of the Korean Society for Railway (한국철도학회논문집)

The Korean Society for Railway (한국철도학회)
권호 표지
DOI QR코드

DOI QR Code

Study on Accident Prediction Models in Urban Railway Casualty Accidents Using Logistic Regression Analysis Model

로지스틱회귀분석 모델을 활용한 도시철도 사상사고 사고예측모형 개발에 대한 연구

  • Jin, Soo-Bong (Department of Railway Electrical and Signaling Engineering, Graduate School of Railway, Seoul National University of Science and Technology) ;
  • Lee, Jong-Woo (Department of Railway Electrical and Signaling Engineering, Graduate School of Railway, Seoul National University of Science and Technology)
  • Received : 2017.08.17
  • Accepted : 2017.08.25
  • Published : 2017.08.31
Download PDF
⟨ Previous Next ⟩

Abstract

This study is a railway accident investigation statistic study with the purpose of prediction and classification of accident severity. Linear regression models have some difficulties in classifying accident severity, but a logistic regression model can be used to overcome the weaknesses of linear regression models. The logistic regression model is applied to escalator (E/S) accidents in all stations on 5~8 lines of the Seoul Metro, using data mining techniques such as logistic regression analysis. The forecasting variables of E/S accidents in urban railway stations are considered, such as passenger age, drinking, overall situation, behavior, and handrail grip. In the overall accuracy analysis, the logistic regression accuracy is explained 76.7%. According to the results of this analysis, it has been confirmed that the accuracy and the level of significance of the logistic regression analysis make it a useful data mining technique to establish an accident severity prediction model for urban railway casualty accidents.

본 연구는 사고심각도 분류 및 예측을 위한 철도사고조사 통계기법에 관한 연구이다. 그동안의 선형 회귀분석은 사고 심각도 분석에 어려움이 있었으나 로지스틱회귀분석은 이를 보완할 수 있었다. 데이터마이닝 기법인 로지스틱회귀분석을 활용, 서울지하철(5~8호선) 역사 내 전도사고 중 에스컬레이터 전도사고 발생에 영향을 주는 사고예측 모형 변수는 사고자 연령, 음주여부, 사고 당시상황 및 행동, 핸드레일 잡음 여부였다. 분석의 정확도는 76.7%로 설명되었고 분석방법 결과에 따르면 정확도와 유의수준 측에서 로지스틱회귀분석 방법이 도시철도 사상사고 예측모형을 개발하는데 유용한 데이터마이닝 기법으로 판단된다.

Keywords

References

  1. www.seoul.go.kr (Accessed 29 June 2015)
  2. T.H. Kim (2015) Accident risk by subject and line, SMRT, The report of risk analysis Seoul Metro 5-8 line.
  3. C.W. Park, J.B. Wang, M.S. Kim, D,B. Choi et al. (2009) Development of risk assessment model for railway casualty accidents, Journal of Korean Society For Railway, 12(2), pp.190-198
  4. J.P. Lee (2009) A Study on the passenger's accidents in the subway, MS Thesis, University of Seoul
  5. S.G. Kim, I.H. Park, J.K. Oh, Y.K. Kim et al. (2014) A Factor analysis of urban railway casualty accidents and establishment of preventive response systems, Journal of Korean Society Of Civil Engineers, 34(3), pp.1017-1022 https://doi.org/10.12652/Ksce.2014.34.3.1017
  6. S.K. Kang (1995) Developing an accidents prediction model for railroad - highway grade crossings, Journal of Korea Transportation Research Society, 13(2), pp.43-58
  7. J.H. Park, S.K, Kim (2012) Development an accidents forecasting models in freeway using multiple linear regression analysis, The Journal of The Korea Institute of Intelligent Transport Systems, 11(6), pp.145-154 https://doi.org/10.12815/kits.2012.11.6.145
  8. H.S. Park, B.S. Son, H.J. Kim (2007) Development of accident prediction models for freeway interchange ramps, Journal of the Korean Society of Road Engineers, 25(3), pp.123-135
  9. C. V. Zegeer, J. Hummer, D. Reinfurt, L. Herf and W. Hunter (1986) Safety effects of cross-section design for two-lane roads, Federal Highway Administration, FHWA-RD-87-008
  10. H.S. Lee, J.H. Lym (2012) SPSS 18.0 Manual, Jyp Hyun Jae, Seoul, pp.342-347
  11. K.Y. Kim, M.S. Jeon, H.C, Kang, S.K. Lee (2009) Regression analysis by example, Free Academy, Seoul, pp.334-357
  12. H.Y. Lee, S.C. No (2009) Advanced statistical analysis (Theory and practice), Bum Mun Sa, Seoul, pp.357-395