KSCE Journal of Civil and Environmental Engineering Research (대한토목학회논문집)

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

DOI QR Code

A Deep Learning Model to Predict BIM Execution Difficulty Based on Bidding Texts in Construction Projects

건설사업 입찰 텍스트의 BIM 수행 난이도 추론을 위한 딥러닝 모델

  • Kim, Jeongsoo (Korea Institute of Civil Engineering and Building Technology) ;
  • Moon, Hyounseok (Korea Institute of Civil Engineering and Building Technology) ;
  • Park, Sangmi (Korea Institute of Civil Engineering and Building Technology)
  • 김정수 (한국건설기술연구원 구조연구본부) ;
  • 문현석 (한국건설기술연구원 미래스마트건설연구본부) ;
  • 박상미 (한국건설기술연구원 미래스마트건설연구본부)
  • Received : 2023.10.23
  • Accepted : 2023.11.08
  • Published : 2023.12.01
Download PDF
⟨ Previous Next ⟩

Abstract

The mandatory use of BIM(Building Information Model) in larger Korean public construction projects necessitates participants to have a comprehensive understanding of the relevant procedures and technologies, especially during the bidding stage. However, most small and medium-sized construction and engineering companies possess limited BIM proficiency and understanding. This hampers their ability to recognize bidding requirements and make informed decisions. To address this challenge, our study introduces a method to gauge the complexity of BIM requirements in bidding documents. This is achieved by integrating a morphological analyzer, which encompasses BIM bidding terminology, with a deep learning model. We investigated the effects of the parameters in our proposed deep learning model and examined its predictive validity. The results revealed an F1-score of 0.83 for the test data, indicating that the model's predictions align closely with the actual BIM performance challenges.

일정 규모 이상의 공공 건설 프로젝트에 대한 BIM(Building Information Model) 적용이 의무화됨에 따라 입찰단계에서부터 BIM 요구사항에 대한 관련 절차 및 기술에 대한 폭넓은 이해가 요구되고 있다. 그러나 대부분의 중소 시공 및 엔지니어링 기업은 BIM 수행역량이 낮고 관련 기존 업무 프로세스의 BIM 적용에 대한 이해 높지 않아, 입찰 요구사항에 대한 인지가 어렵고 입찰단계의 합리적인 의사결정이 쉽지 않다. 따라서 본 연구는 BIM 입찰문서 용어를 포함한 형태소 분석기를 딥러닝 모델에 결합하여 입찰문서의 BIM 요구사항의 난이도 판별 방법을 제시한다. 제안된 딥러닝 모델의 매개변수 영향이 조사되었으며 예측 결과의 타당성이 검토되었다. 그 결과, 제안된 모델이 시험 데이터에 대해 F1-score 0.83의 성능을 가지며, 모델의 판별 결과 또한 실제 BIM 수행 난이도를 타당하게 반영하고 있음을 보였다.

Keywords

Acknowledgement

This paper is supported by the Korea Agency for Infrastructure Technology Advancement(KAIA) grant funded by the Ministry of Land, Infrastructure and Transport (Grant RS-2022-00143371). The paper is also expanded from a conference paper presented at 2023 KSCE Convention held in Yeosu, South Korea on Oct. 18-20, 2023.

References

  1. Akinosho, T. D., Oyedele, L. O., Bilal, M., Ajayi, A. O., Delgado, M. D., Akinade, O. O. and Ahmed, A. A. (2020). "Deep learning in the construction industry: A review of present status and future innovation." Journal of Building Engineering, Elsevier, Vol. 32. https://doi.org/10.1016/j.jobe.2020.101827. 
  2. Bae, S., Ham, S., Lee, I., Lee, G. P. and Kim, D. (2019). "Deep learning based crack detection from tunnel cement concrete lining." Journal of Korean Tunnelling and Underground Space Association, KTA, Vol. 24, No. 6, pp. 583-598, https://doi.org/10.9711/KTAJ.2022.24.6.583 (in Korean). 
  3. Cahuantzi, R., Chen, X. and Guttel, S. (2023). "A comparison of LSTM and GRU networks for learning symbolic sequences." Proceeding of Science and Information Conference, London, UK, pp. 771-785, https://doi.org/10.1007/978-3-031-37963-5_53. 
  4. Choi, S. J., Choi, S. W., Kim, J. H. and Lee, E. B. (2021). "AI and text-mining applications for analyzing constractor's risk in Invitation to Bid(ITB) and contracts for engineering procurement and construction (EPC) projects." Energies, MDPI, Vol. 14, No. 15. https://doi.org/10.3390/en14154632. 
  5. Choi, S. W. and Lee, E. B. (2022). "Contractor's risk analysis of engineering procurement and construction(EPC) contracts using ontological semantic model and bi-long short-term memory (LSTM) technology." Sustainability, MDPI, Vol. 14, No. 11. https://doi.org/10.3390/su14116938. 
  6. Eom, S. H., Cha, G., Park, S. K., Park, S. and Park, J. (2023). "Analysis of potential construction risk types in formal documents using text mining." KSCE Journal of Civil and Environmental Engineering Research, KSCE, Vol. 43, No. 1, pp. 91-98, https://doi.org/10.12652/Ksce.2023.43.1.0091 (in Korean). 
  7. Kang, E. A., Kim, S. and Kim, S. (2022). "Quality control of reinforced concrete work using deep-learning based on object recognition." Journal of the Regional Association of Architectural Institute of Korea, AIKRA, Vol. 24, No. 2, pp. 17-24 (in Korean). 
  8. Kim, M. H. (2019). Application of computer vision based deep learning technique for detecting safety helmet of construction workers, Msc. thesis, Pukyong National University, Busan, Korea (in Korean). 
  9. Kim, S., Cha, G., Cho, M. and Park, S. (2022). "Text mining based analysis of construction accident causes and risk factors." Proceedings of the 2022 Spring Conference of the Korea Academia-Industrial Cooperation Society, Jeju, South Korea, pp. 272-273 (in Korean). 
  10. Kim, J., Lee, C. W., Park, S. H., Lee, J. H. and Hong, C. H. (2020). "Development of fire detection model for underground utility facilities using deep learning: Training data supplement and bias optimization." Journal of the Korea Academia-Industrial Cooperation Society, KAIS, Vol. 21, No. 12, pp. 320-330, https://doi.org/10.5762/KAIS.2020.21.12.320 (in Korean). 
  11. Kim, J., Park, S. and Hong, C. H. (2023). "A study on falling detection of workers in the underground utility tunnel using dual deep learning techniques." Journal of the Society of Disaster Information, KOSDI, Vol. 19, No. 3, pp. 498-509, https://doi.org/10.15683/kosdi.2023.9.30.498 (in Korean). 
  12. Kudo, T., Yamamoto, K. and Matsumoto, Y. (2004). "Applying conditional random fields to Japanese morphological analysis." Proceedings of the 2004 Conference on Empirical Methods in Natural Language Processing, ACL, Barcelona, Spain, pp. 230-237. 
  13. Lee, J. H. (2019). "Global research trend based on Natural Language Processing of irregular text data in the construction industry." Magazine of Construction Management, KICEM, Vol. 20, No. 2, pp. 62-65 (in Korean). 
  14. Lee, K., Kang, S. and Shin, Y. (2022). "A study on the application of object detection method in construction site through real case analysis." Journal of the Society of Disaster Information, KOSDI, Vol. 18, No. 2, pp. 269-279, https://doi.org/10.15683/KOSDI.2022.6.30.269 (in Korean). 
  15. Lee, D., Yeon, J., Hwang, I. and Lee, S. (2010)."KKMA : A tool for utilizing sejong corpus based on relational database." JKIISE Transactions on Computing Practices, KIISE, Vol. 16, No. 11, pp. 1046-1050 (in Korean). 
  16. Lee, D. H., Yoon, G. H. and Kim, J. J. (2019). "Development of ITB risk management model based on AI in bidding phase for oversea EPC projects." The Journal of the Institute of Internet, Broadcasting and Communication, IIBC, Vol. 19, No. 4, pp. 151-160, https://doi.org/10.7236/JIIBC.2019.19.4.151 (in Korean). 
  17. Moon, S., Chi, S. and Im, S. B. (2022). "Automated detection of contractual risk clauses from construction specifications using bidirectional encoder representations from transformers (BERT)." Automation in Construction, Elsevier, Vol. 142, 104465, https://doi.org/10.1016/j.autcon.2022.104465. 
  18. Moon, S., Lee, G. and Chi, S. (2021a). "Semantic text-pairing for relevant provision identification in construction specification reviews." Automation in Construction, Elsevier, Vol. 128, 103780, https://doi.org/10.1016/j.autcon.2021.103780. 
  19. Moon, S., Lee, G., Chi, S. and Oh, H. (2021b). "Automated construction specification review with named entity recognition using natural language processing." Journal of Construction Engineering and Management, ASCE, Vol. 147, No. 1. https://doi.org/10.1061/(ASCE)CO.1943-7862.0001953. 
  20. Park, S. (2021). Development of visualization system for deep learning-based progress comparison using the real image of construction site and 4D model, Ph.D thesis, Gyeongsang National University (in Korean). 
  21. Park, E. J. and Cho, S. Z. (2014). "KoNLPy: Korean natural language processing in Python." Proceedings of the 26th Annual Conference on Human & Cognitive Language Technology, KIISE, Kangwon, Korea, pp. 133-136 (in Korean). 
  22. Park, K. and Kim, H. (2021). "Analysis of seasonal Importance of construction hazards using text mining." KSCE Journal of Civil and Environmental Engineering Research, KSCE, Vol. 41, No. 3, pp. 305-316, https://doi.org/10.12652/Ksce.2021.41.3.0305 (in Korean). 
  23. Prabowo, Y. D., Warnars, H. L. H. S., Budiharto, W., Kistijantoro, A. I., Heryadi, Y. and Lukas (2018). "Lstm and simple rnn comparison in the problem of sequence to sequence on conversation data using Bahasa Indonesia." Proceedings of 2018 Indonesian Association for Pattern Recognition International Conference, IEEE, Jakarta, Indonesia, pp. 51-56, https://doi.org/10.1109/INAPR.2018.8627029. 
  24. Saitoh, K. (2018). Deep Learning from Scratch 2, Hanbit Media, translated by Gaeapmapsi (in Korean). 
  25. Shewalkar, A. N. (2018). Comparison of RNN, LSTM and GRU on Speech Recognition Data, Msc. thesis, North Dakota State University, North Dakota, USA. 
  26. Wiliams, T. P. and Gong, J. (2014). "Predicting construction cost overruns using text mining, numerical data and ensemble classifiers." Automation in Construction, Elsevier, Vol. 43, pp. 23-29, https://doi.org/10.1016/j.autcon.2014.02.014.