All Issue

2025 Vol.21, Issue 3 Preview Page

Original Article

A Study on a Deep Learning-Based Task Risk Assessment Model in Plant Construction

플랜트 건설에서 딥러닝 기반 작업 위험성평가 모델 구축 연구

정진근1ㆍ정병철2ㆍ박교식3*

Jinkeun Jung1, Byeongcheol Jeong2, Kyoshik Park3*

1Doctor’s Course, Department of Safety & Health Convergence Engineering, Soongsil University, Seoul, Republic of Korea
2Doctor’s Course, Department of Safety & Health Convergence Engineering, Soongsil University, Seoul, Republic of Korea
3Professor, Department of Safety & Health Convergence Engineering, Soongsil University, Seoul, Republic of Korea
30 September 2025. pp. 661-672
PDF Citation
Abstract
Purpose: This study aims to develop a deep learning-based task risk assessment model capable of systematically identifying and evaluating hazardous and risk factors at plant construction sites. The model is specifically designed systematically to be easily utilized by workers with limited experience or knowledge, thereby addressing the issues associated with traditional experience-based assessment methods. Method: A total of 976,140 risk assessment records from domestic construction Company A. were cleansed and pre-processed. A risk assessment model was then constructed using LDA-based topic modeling and a deep neural network(DNN). The model was trained and tested using a dataset of 18,000, with final validation conducted on 1,026 new risk assessments. Result: The developed model systematically recommends hazardous and risky factors and types of accidents based on working conditions, with more than 95.6% of the recommendations evaluated as appropriate by experts. In the final validation, the model recommended 6,380 items for 1,026 assessments across eight work types, all achieving suitability score of above 80, thereby meeting the model design criteria and demonstrating consistent recommendation quality and high applicability across all construction domains. Conclusion: By structuring large-scale unstructured safety data into a data-driven risk assessment model, this study overcomes the limitations of traditional subjective evaluation methods. The application of AI enhances consistency and efficiency in hazard identification, improving the effectiveness of risk assessments.
Keywords
Plant construction
Risk assessment
Topic modeling
Deep learning
Data-driven
연구목적: 본 연구는 플랜트 건설현장에서 유해·위험요인을 체계적으로 식별하고 평가할 수 있는 딥러닝 기반 작업 위험성평가 모델을 구축하는 데 목적이 있다. 특히, 경험과 지식이 부족한 근로자도 쉽게 활용할 수 있도록 체계적인 시스템을 설계하여, 기존의 경험기반 평가 방식의 문제를 해결하고자 한다. 연구방법: 국내 플랜트 건설사인 A사의 976,140건의 위험성평가 데이터를 기반으로 데이터 정제 및 전처리를 수행하고, LDA 기반 토픽 모델링과 심층 신경망(DNN)을 활용하여 위험성 평가 모델을 구축하였다. 데이터셋 18,000개를 사용하여 학습 및 테스트를 진행했으며, 최종 검증은 1,026건의 신규 위험성평가 데이터를 활용하였다. 연구결과: 개발된 모델은 작업조건에 따라 유해·위험 요인과 재해 유형 등을 체계적으로 추천하였고, 추천 결과의 95.6% 이상이 전문가에 의해 적합하다고 평가되었다. 최종검증을 통해 8개 공종의 위험성평가 1,026개에 대해 추천된 6,380개 전체가 추천 적합도 80 이상으로 모델 설계 기준을 만족하였고, 전 공종에서 일관된 추천 품질과 높은 적합성을 지님을 입증하였다. 결론: 대규모 비정형 안전 데이터를 정형화하여 데이터 기반의 체계적인 위험성평가 모델을 구축함으로써, 기존의 경험 및 주관적인 평가 방식에 한계를 보완하였다. AI 기술을 활용을 통해 위험 요소 도출의 일관성과 효율성을 제고하고 위험성 평가의 실효성을 높였다.
키워드
플랜트 건설
위험성평가
토픽모델링
딥러닝
데이터 기반
References
  1. Bok, H., Kim, H.-T., Kang, K.-S. (2014). “Construction work by type risk factors increased levels proposed model for th PCRA(Pre Construcion Risk Assessment) - Focusing on building construction-.” Journal Title. Journal of Korea Safety Management & Science, Vol. 16, No. 3, pp. 9-22. 10.12812/ksms.2014.16.3.9
  2. Jeon, H., Park, S., Son, S., Lee, J., An, J., Choi, G., Lim, Y. (2024). “Challenges of lane detection using deep neural networks in severe heavy rain: A synthetic evaluation dataset based on the CARLA simulator.” Journal of Auto-Vehicle Safety Association, Vol. 16, No. 4, pp. 92-101. 10.22680/KASA2024.16.4.092
  3. Kang, H.J., Choi, S.J., Han, S.J., Kim, J.H., Lee, S.H. (2024). “How to build an AI safety management chatbot service based on IoT construction health monitoring.” Journal of the Society of Disaster Information, Vol. 20, No. 1, pp. 106-116. 10.15683/KOSDI.2024.3.31.106
  4. Kim, H.E., Kim, C.H., Kim, T.G., Park, J.J. (2023). “Verification of ground subsidence risk map based on underground cavity data using DNN Technique.” Journal of the Society of Disaster Information, Vol. 19, No. 2, pp. 334-343. 10.15683/KOSDI.2023.6.30.334
  5. Kim, J.-H. (2023). “An accurate forward head posture detection using human pose and skeletal data learning.” Journal of The Korea Society of Computer and Information, Vol. 28, No. 8, pp. 87-93. 10.9708/JKSCI.2023.28.08.087
  6. Kim, J.-M., Kim, I.-M., Kim, M.-S. (2022). “Multi-DNN acceleration techniques for embedded systems with tucker decomposition and hidden-layer-based parallel processing.” Journal of the Korea Institute of Information and Communication Engineering, Vol. 26, No. 6, pp. 842~849.
  7. Korea Enterprises Federation (2023). “Survey on risk assessment practices and policy improvement directions.” KEF E-magagzine, Vol,101, pp.12-13.
  8. KOSIS(KOrean Statistical Information Service), www.kosis.kr
  9. Lee, D.-Y., Kim, D.-E. (2021). “A study on the probabilistic risk analysis for safety management in construction projects.” Journal of The Korea Society of Computer and Information, Vol. 26, No. 8, pp. 139-147. 10.9708/JKSCI.2021.26.08.139
  10. Lee, H.-Y. (2021). A Study on Data Processing and Accident Prediction Modeling based on Machine Learning Methods for Occupational Accident Analysis in the Construction Industry. Ph.D. Dissertation, Incheon National University.
  11. Lee, J., Lee, S. (2023). “Construction site safety management: A computer vision and deep learning approach.” Sensor, Vol. 23, No.2, 944. https://doi.org/10.3390/s23020944 10.3390/s23020944 36679738 PMC9863726
  12. Lee, K.-T., Kim, M.-S., Kim, H.-J., Kim, J.-H. (2021). “A model to predict occupational safety and health management expenses in construction applying multi-variate regression analysis and deep neural network.” Journal of the Architectural Institute of Korea, Vol. 37, No. 9, pp. 217-226. 10.5659/JAIK.2021.37.9.217
  13. Ministry of Employment and Labor (2023). Guidelines on Workplace Risk Assessment. MOEL Notification No. 2023-19.
  14. Qian Y.R., Yu, J.H., Lee, S.K. (2024). “An Automated risk assessment based on GPT and RAG for small and medium-sized construction sites.” Korean Journal of Safety Culture, Vol. 35, pp. 169-179. 10.52902/kjsc.2024.35.169
  15. Seo, J.U., Song, C.H. (2024). “Big data analytics of construction safety incidents using text mining.” Korean Society of Industry Convergence, Vol.27, No.3, pp.581-590.
  16. Seo, Y.H., Woo, I.S., Jang, C., Hwang, M.H. (2015). “Research of effectiveness of risk assessment of construction industry by self administered questionnaires.” Journal of Risk Management, Vol. 26, No. 3, pp. 1-27. 10.21480/tjrm.26.3.201509.001
  17. Yoon, Y.G., Lee, J.Y., Oh, T.K. (2022). “Text mining-based data preprocessing and accident type analysis for construction accident analysis.” Journal of the Korean Society of Safety, Vol. 37, No. 2, pp. 18-27. 10.14346/JKOSOS.2022.37.2.18
Information
  • Publisher :The Korean Society of Disaster Information
  • Publisher(Ko) :한국재난정보학회
  • Journal Title :Journal of the Society of Disaster Information
  • Journal Title(Ko) :한국재난정보학회논문집
  • Volume : 21
  • No :3
  • Pages :661-672
  • DOI :https://doi.org/10.15683/kosdi.2025.9.30.661
Journal Informaiton Journal of the Society of Disaster Information Journal of the Society of Disaster Information
  • NRF
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close