KSCE Journal of Civil and Environmental Engineering Research (대한토목학회논문집)

Korean Society of Civil Engeneers (대한토목학회)
권호 표지
DOI QR코드

DOI QR Code

AI-based Construction Site Prioritization for Safety Inspection Using Big Data

빅데이터를 활용한 AI 기반 우선점검 대상현장 선정 모델

  • 황윤호 (국토안전관리원 디지털혁신추진단 빅데이터전략팀, 부산대학교 통계학과) ;
  • 지석호 (서울대학교 건설환경공학부) ;
  • 이현승 (부산대학교 통계학과) ;
  • 정현준 (국토안전관리원 디지털혁신추진단 빅데이터전략팀)
  • Received : 2022.05.06
  • Accepted : 2022.10.12
  • Published : 2022.12.01
Download PDF
⟨ Previous Next ⟩

Abstract

Despite continuous safety management, the death rate of construction workers is not decreasing every year. Accordingly, various studies are in progress to prevent construction site accidents. In this paper, we developed an AI-based priority inspection target selection model that preferentially selects sites are expected to cause construction accidents among construction sites with construction costs of less than 5 billion won (KRW). In particular, Random Forest (90.48 % of accident prediction AUC-ROC) showed the best performance among applied AI algorithms (Classification analysis). The main factors causing construction accidents were construction costs, total number of construction days and the number of construction performance evaluations. In this study an ROI (return of investment) of about 917.7 % can be predicted over 8 years as a result of better efficiency of manual inspections human resource and a preemptive response to construction accidents.

지속적인 안전관리에도 불구하고 매년 건설업 근로자 사망율은 줄어들지 않는 추세다. 이에 따라 건설현장 사고를 예방하기 위한 다양한 연구가 진행 중이다. 본 논문에서는 건설공사 비용 50억원 미만의 건설현장 중 건설사고가 발생할 것으로 예상되는 현장을 우선적으로 선별하는 AI기반 우선점검대상 선정 모델을 개발하였다. 특히, 적용한 AI 알고리즘 중 분류분석에서 가장 뛰어난 성능(사고발생예측 AUC-ROC 90.48 %)을 보인 랜덤 포레스트를 모델 개발에 활용하였으며, 건설사고를 유발하는 주요한 요인으로는 공사비, 총공사일수, 공사실적평가액이 확인되었다. 본 연구를 통해 점검인력 효율화와 건설사고에 대한 선제적 대응의 결과로 8년간 약 917.7 % ROI(투자수익률)를 기대할 수 있다.

Keywords

References

  1. Belgiu, M. and Dragut, L. (2016). "Random forest in remote sensing: A review of applications and future directions." ISPRS Journal of Photogrammetry and Remote Sensing, Vol. 114, pp. 24-31. https://doi.org/10.1016/j.isprsjprs.2016.01.011
  2. Breiman, L. (2001). "Random forest." Machine Learning, Vol. 45, No. 1, pp. 5-32. https://doi.org/10.1023/A:1010933404324
  3. Chan, J. C. W. and Paelinckx, D. (2008). "Evaluation of random forest and adaboost tree-based ensemble classification and spectral band selection for ecotope mapping using airborne hyperspectral imagery." Remote Sensing of Environment, Vol. 112, No. 6, pp. 2999-3011. https://doi.org/10.1016/j.rse.2008.02.011
  4. Chen, T. and Guestrin, C. (2016). "Xgboost: A scalable tree boosting system." In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, California, USA, pp. 785-794.
  5. 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, Vol. 17, No. 3, pp. 295-303 (in Korean). https://doi.org/10.5345/JKIBC.2017.17.3.295
  6. Choi, S. J., Kim, J. H. and Jung, K. H. (2021). "Development of prediction models for fatal accidents using proactive information in construction sites." Journal of the Korean Society of Safety, Vol. 36, No. 3, pp. 31-39 (in Korean). https://doi.org/10.14346/JKOSOS.2021.36.3.31
  7. Feurer, M. and Hutter, F. (2019). "Hyperparameter optimization." Automated machine learning, Springer, Cham, pp. 3-33.
  8. Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., Ye, Q. and Liu, T. (2017). "LightGBM: A highly efficient gradient boosting decision tree." In Advances in Neural Information Processing Systems, pp. 3149-3157.
  9. Korea Authority of Land & Infrastructure Safety (KALIS) (2021a). Submission of documents for participation in public big data analysis contest by MOIS in 2021, pp. 8 (in Korean).
  10. Korea Authority of Land & Infrastructure Safety (KALIS) (2021b). Submission of documents for participation in public big data analysis contest by MOIS in 2021, pp. 6 (in Korean).
  11. Korean Law Information Center (2022). Act on punishment of serious accidents, etc., Article 6, Available at: https://www.law.go.kr (September 23, 2022).
  12. Lim, J. R., Park, C. Y. and Yun, S. M. (2021). "A study on safety management measures for small and medium-sized construction sites-Focused on reinforced concrete construction." Journal of the Architectural Institute of Korea, Vol. 23, No. 6, pp. 197-204 (in Korean).
  13. Ministry of Employment and Labor (MOEL) (2021). 2020 Industrial accident and death statistics announced, pp. 13 (in Korean).
  14. Narkhede, S. (2018). "Understanding auc-roc curve." Towards Data Science, Vol. 26, No. 1, pp. 220-227.
  15. Tixier, A. J. P., Hallowell, M. R., Rajagopalan, B. and Bowman, D. (2016). "Application of machine learning to construction injury prediction." Article of Automation in Construction, Vol. 69, pp. 102-114. https://doi.org/10.1016/j.autcon.2016.05.016
  16. Yoon, Y. G., Lee, J. Y. and Oh, T. G. (2020). "Development of accident prediction model with construction accident report data." Korea Institute for Structural Maintenance and Inspection, Vol. 24, No. 2, pp. 6 (in Korean)