Korean Journal of Construction Engineering and Management (한국건설관리학회논문집)

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

DOI QR Code

A Developing a Machine Leaning-Based Defect Data Management System For Multi-Family Housing Unit

기계학습 알고리즘 기반 하자 정보 관리 시스템 개발 - 공동주택 전용부분을 중심으로 -

  • Park, Da-seul (Department of Architectural Engineering, Ajou University) ;
  • Cha, Hee-sung (Department of Architectural Engineering, Korea University)
  • 박다슬 (아주대학교 스마트융합건축학과) ;
  • 차희성 (아주대학교 스마트융합건축학과)
  • Received : 2023.06.20
  • Accepted : 2023.09.11
  • Published : 2023.09.30
Download PDF
⟨ Previous Next ⟩

Abstract

Along with the increase in Multi-unit housing defect disputes, the importance of defect management is also increased. However, previous studies have mostly focused on the Multi-unit housing's 'common part'. In addition, there is a lack of research on the system for the 'management office', which is a part of the subject of defect management. These resulted in the lack of defect management capability of the management office and the deterioration of management quality. Therefore, this paper proposes a machine learning-based defect data management system for management offices. The goal is to solve the inconvenience of management by using Optical Character Recognition (OCR) and Natural Language Processing (NLP) modules. This system converts handwritten defect information into online text via OCR. By using the language model, the defect information is regenerated along with the form specified by the user. Eventually, the generated text is stored in a database and statistical analysis is performed. Through this chain of system, management office is expected to improve its defect management capabilities and support decision-making.

공동주택 하자 분쟁의 증가와 함께, 하자관리의 중요성 또한 커지고 있다. 그러나 기존의 연구는 '공용 부분'에 초점을 맞추어 진행되었다. 또한 하자관리의 주체인 '관리사무소'를 위한 시스템 연구도 부족한 실정이다. 이는 관리사무소의 하자관리 능력의 부족과 관리 품질의 저하를 초래한다. 따라서, 본 논문에서는 관리사무소를 위한 기계학습 기반의 하자 정보 관리 시스템을 제안한다. OCR과 NLP 모듈을 사용하여 관리상의 불편한 점을 해소하는 것을 목표로 한다. OCR을 통해 수기로 작성된 하자 정보를 디지털 문서로 변환한다. 이후 언어모델을 이용하여 사용자가 지정한 양식과 함께 하자 정보를 재생성한다. 최종적으로 생성된 텍스트를 데이터베이스에 저장하고 이를 기반으로 통계적 분석을 실행한다. 이러한 일련의 과정을 통해, 관리사무소의 하자관리 역량을 향상할 수 있도록 돕고, 의사결정을 지원할 수 있을 것으로 기대한다.

Keywords

References

  1. Brown, T.B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A, Shyam, P., Sastry, G., Askell, A., Agarwal, S., Herbert-Voss, A., Krueger, G., Henighan, T., Child, R., Ramesh, A., Ziegler, D.M., Wu, J., Winter, C., Hesse, C., Chen, M., Sigler, E., Litwin, M., Gray, S., Chess, B. Clark, J., Berner, C., McCandlish, S., Radford, A., Sutskever, I., and Amodei, D. (2020). "Language models are fewshot learners." arXiv preprint arXiv:2005.14165.
  2. Devlin, J., Chang, M.W., Lee, K., and Toutanova, K. (2019). Bert: Pretraining of deep bidirectional transformers for language understanding, In NAACL-HLT (1), arXiv:1810.0480.
  3. Go, S.S., and Lee, H.M. (2016). "Modeling of Apartment Defect Management System applying UML." Journal of the Architectural Institute of Korea Structure & Construction, 22(7), pp. 123-130.
  4. Jang, H.S. (2013). "A Study on the Building of Defect Information DB Management System of Apartment House for Defect Prevention." The Korean Society of Disaster Information, 9(3), pp. 300-314.
  5. Jeon, K., Lee, G., Yang, S., and Jeong, H.D. (2022). "Named entity recognition of building construction defect information from text with linguistic noise." Automation in Construction, 143, 104543.
  6. Kang, H.W., Park, Y.H., and Kim, Y.S. (2019). "Improvement Model of Defect Information Management System for Apartment Buildings." Korean Journal of Construction Engineering and Management, KICEM, 20(4), pp. 13-21.
  7. Kang, J.Y. (2021). "Residents' Needs for Housing Unit Maintenance in a Apartment Building." Journal Of The Korean Housing Association, 32(6), pp. 111-118. https://doi.org/10.6107/JKHA.2021.32.6.0111
  8. Kim, B.Y. (2022). Hybrid-Convolution Neural Network-Loss Distribution Approach-Association Rule Mining (HCLA) Defect Management Model for Automatic Classification and Characteristic Analysis of Apartment Defects. a domestic doctoral dissertation. Hanyang University.
  9. Kim, E.H., Ji, H.G., Kim, J.N., Park, E.I., and Ohm J. Young. (2021). "Classifying Sub-Categories of Apartment Defect Repair Tasks: A Machine Learning Approach." KIPS Transactions on Software and Data Engineering, KTSDE, 10(9), pp. 359-366.
  10. Kim, O.G., and Choe, J.H. (2010). "Current Status and Problems of Housing Defective Disputes." Korean Journal of Construction Engineering and Management, KICEM, 11(2), pp. 5-8.
  11. Kim, S.W. (2019). "Legal Matters on the Construction Defect and Management Neglect of Apartment Houses." The Korean Society Of Property Law, 36(1), pp. 185-206.
  12. Lee, D.W. (2022). Analysis of Defect Types and Management Methods of Apartment Houses Through Case Studies, a domestic master's thesis. Hanyang University.
  13. Lee, J.B. (2006). "Methods for improving efficiency in Apartment Maintenance Management using on Analytic Hierarchy Process." Korean Journal of Construction Engineering and Management, KICEM, 7(4), pp. 69-77.
  14. Lee, J.H. (2019). "Global research trends based on natural language processing (NLP) using unstructured text data in the construction industry." Korean Journal of Construction Engineering and Management, KICEM, 20(2), pp. 62-66.
  15. Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., Levy, O., Lewis, M., Zettlemoyer, L., and Stoyanov, V. (2019). Roberta: A robustly optimized bert pretraining approach. ICLR 2020.
  16. Lewis, M., Liu, Y., Goyal, N., Ghazvininejad, M., Mohamed, A., Levy, O., Stoyanov, V., and Zettlemoyer, L. (2020). "BART: Denoising sequence-tosequence pretraining for natural language generation, translation, and comprehension." In Association for Computational Linguistics, ACL, pp. 7871-7880.
  17. Oh, J.H., Song, Y.W., Choi, Y.K., and Lim, H.C. (2019). "Business Process Improvement of Defect Management in Apartment Housing Project." Korean Journal of Construction Engineering and Management, KICEM, 10(5), pp. 16-27.
  18. Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W., and Liu. P.J. (2020). "Exploring the limits of transfer learning with a unified text-to-text transformer." JMLR, arXiv:1910.10683.
  19. Sajadfar, N., Abdollahnejad, S., Hermann, U., and Mohamed, Y. (2019). "Text detection and classification of construction documents. In ISARC." Proceedings of the International Symposium on Automation and Robotics in Construction, IAARC, 36, pp. 446-452.
  20. Son, B.K. (2010). "A Handwritten Document Digitalization Framework based Defect Management System in Educational Facilities." Korean Institute of Sustainable Design and Educational Environment, 9(2), pp. 1-11.
  21. Subramani, N., Matton, A., Greaves, M., and Lam, A. (2020). A survey of deep learning approaches for ocr and document understanding. arXiv, arXiv:2011.13534.
  22. Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, L., and Polosukhin. I. (2017). "Attention is all you need," Advances in neural information processing systems, 30, pp. 5998-6008.
  23. Yang, D.U., Kim, B.Y., Lee, S.H., Ahn, Y.H., and Kim, H.Y. (2022). AutoDefect: Defect text classification in residential buildings using a multi-task channel attention network. Sustainable Cities and Society, 80(5), 103803.
  24. Yang, Z., Dai, Z., Yang, Y., Carbonell, J.G., Salakhutdinov, R., and Le, Q.V. (2019). "XLNet: Generalized autoregressive pretraining for language understanding." In Advances in Neural Information Processing Systems.
  25. Yu, B.J, Bang, H.S., and Kim, O.K. (2022). "Improvement of Defect Checklists for the Quality Inspection of Apartment Houses." Proceedins of the Korean Institute of Building Construction Conference, 22(1), pp. 33-34.
  26. Yun, S.H., Park, C.W., and Jeong, S.Y. (2008). "A Study on the Integrated Construction Defects Assessment of Apartments for Systematic Asset Management." Journal of the Architectural Institute of Korea Structure & Construction, JAIK, 24(8), pp. 179-186.