Korean Journal of Construction Engineering and Management (한국건설관리학회논문집)

Korea Institute of Construction Engineering and Management (한국건설관리학회)
권호 표지
DOI QR코드

DOI QR Code

Affinity Analysis Between Factors of Fatal Occupational Accidents in Construction Using Data Mining Techniques

데이터마이닝 기법을 활용한 건설 중대 재해요인 간 연관성 분석

  • Lim, Jiseon (Department of Civil and Environmental Engineering, Incheon National University) ;
  • Han, Sanguk (Department of Civil and Environmental Engineering, Hanyang University) ;
  • Kang, Youngcheol (Department of Architecture and Architectural Engineering, Yonsei University) ;
  • Kang, Sanghyeok (Department of Civil and Environmental Engineering, Incheon National University)
  • 임지선 (인천대학교 건설환경공학과) ;
  • 한상욱 (한양대학교 건설환경공학과) ;
  • 강영철 (연세대학교 건축공학과) ;
  • 강상혁 (인천대학교 도시환경공학부)
  • Received : 2021.04.12
  • Accepted : 2021.08.09
  • Published : 2021.09.30
Download PDF
⟨ Previous Next ⟩

Abstract

Governments and companies are trying to reduce occupational accidents in the construction industry; however, the number of disasters are not decreasing significantly. This study aims to identify the correlation between factors affecting construction disasters quantitatively. To this end, 1,197 cases of serious disasters provided by Korea Occupational Safety and Health Administration (KOSHA) were analyzed using affinity analysis, one of the data mining techniques. The data from KOSHA were preprocessed and analyzed with variables of accident type, project type, activity type, original cause materials, sensory temperature, time of the accident, and fall height, and the association rules were derived for fall accidents and the others. For fall accidents, 64 association rules with lift ratios of 1.38 or greater were derived, and for the other accidents, 59 association rules with lift ratios of 1.54 or greater were derived. After analyzing the derived association rules focusing on the relationship among accident factors, this study presented the significance of applying the affinity analysis to address the study's limitations. The significance of this study can be found in that the correlation among factors affecting construction accidents is presented quantitatively.

정부와 기업이 건설업의 산업재해를 줄이기 위해 지속적으로 노력하고 있지만, 재해는 크게 줄어들지 않고 있다. 본 연구는 건설 재해에 영향을 미치는 요인들 간의 연관성을 정량적으로 규명하고자 하였다. 산업안전공단에서 공개한 중대재해 사례 1,197건을 대상으로, 데이터마이닝 기법 중 하나인 연관성 분석을 이용하여 연구를 수행하였다. 산업안전공단에서 제공하는 데이터와 외부 변수를 포함하여 재해 발생 형태, 건설업종, 작업내용, 기인물, 체감온도, 사고 시간대, 추락높이의 변수로 아이템을 구성하여 분석하였으며, 떨어짐 재해와 그 외의 재해로 구분하여 연관규칙을 도출하였다. 떨어짐 재해의 경우 향상도가 1.38 이상인 64개의 연관규칙을 도출하였으며, 떨어짐을 제외한 재해의 경우 향상도가 1.54 이상인 59개의 연관규칙을 도출하였다. 도출된 연관규칙을 재해요인 간의 연관성에 초점을 두고 해석한 후, 고찰에서 연구의 한계와 건설재해 요인 간의 관련성을 파악할 때 연관성 분석 기법을 적용함에 있어 유의사항을 제시하였다. 본 연구는 건설 재해에 영향을 미치는 요인들 간의 연관성을 정량적인 수치로 제시하여 추후 근로자들과 현장관리자가 건설현장에서 적절한 안전대책을 마련하는 기초자료를 제공하였다는 점에서 의미를 찾을 수 있다.

Keywords

Acknowledgement

본 연구는 2019년도 과학기술정보통신부 한국연구재단 신진연구지원사업의 연구비 지원 (과제번호: NRF-2019R1C1C1009979)을 받아 수행되었습니다.

References

  1. Cho, Y.R., Kim. Y.C., and Shin. Y.S. (2017). "Prediction Model of Construction Safety Accidents using Decision Tree Technique." Journal of the Korea Institute of Building Construction, 17(3), pp. 295-303. https://doi.org/10.5345/JKIBC.2017.17.3.295
  2. Choi, J.H., and Ryu, H.G. (2019). "Analysis of Occupational Injury and Feature Importance of Fall Accidents on the Construction Sites using Adaboost." Journal of the Arcotectiral Institute of Korea Structure & Construction, 35(11), pp. 155-162.
  3. Kim, Y.C., Yoo. W.S., and Shin. Y.S. (2017). "Application of Artificial Neural Networks to Prediction of Construction Safety Accidents." The Journal of the Korean Society of Hazard Mitigation, 17(1), pp. 7-14. https://doi.org/10.9798/KOSHAM.2017.17.1.7
  4. Korea Occupational Safety and Health Administration (KOSHA). (2016). "Guidelines on occupational accidents recorded and classified." KOSHA GUIDE G - 83 - 2016.
  5. Korea Occupational Safety and Health Administration (KOSHA). (2020). "2019 industrial accident status analysis."
  6. Lee, G.H., Lee, C.S., Koo, C.W., and Kim, T.W. (2020). "Identification of Combinatorial Factors Affecting Fatal Accidents in Small Construction Sites: Association Rule Analysis." Korean Journal of Construction Engineering and Management, KICEM, 21(4), pp. 90-99. https://doi.org/10.6106/KJCEM.2020.21.4.090
  7. Lee, G.Y. (2020). "Association Rule Mining Approach to Extracting Relationships of Accident Factors in Construction Sites." MS Thesis, Pukyong University.
  8. Lee, K.H. (2016). "An Analysis on the Correlation Between Climate Factors and Fatal Accident in Construction Site." MS Thesis, Semyung University.
  9. Leem, Y.M., and Choi, Y.H. (2005). "Selection of an Optimal Algorithm among Decision Tree Techniques for Feature Analysis of Industrial Accident in Construction Industries." Journal of the Korea Safety Management & Science, 7(5), pp. 1-8.
  10. Min, Y.E., Cho, S.J., Ji, H.J., Yoo, W.S., and Shin, Y.S. (2018). "Predicting Model for Occupational Disease using Complex Analysis in Construction Site." The Korean Society of Science & Art, 36, pp. 109-120. https://doi.org/10.17548/ksaf.2018.12.30.109
  11. Ministry of Land, Infrastructure and Transport (MoLIT). (2020). "Enforcement Decree of the Framework Act on Construction Industry." Law No. 31516.
  12. Shin, D.P., Son, C.B., and Lee, D.E. (2012). "Association Analysis of Construction Accident Attributes Causing Fatalities." Journal of the Architectural Institute Of Korea Structure & Construction, 28(2), pp. 87-94. https://doi.org/10.5659/JAIK_SC.2012.28.2.87
  13. Shmueli, G., Bruce, P.C., Yahav, I., Patel, N.R., and Lichtendahl Jr. K.C. (2018). "Data Mining for Business Analytics", Wiley.
  14. Son, K.Y., and Ryu, H.G. (2019). "Association Rules Analysis of Safe Accidents Caused by Falling Objects." Journal of the Korea Institute of Building Construction, 19(4), pp. 341-350. https://doi.org/10.5345/JKIBC.2019.19.4.341