International conference on construction engineering and project management (국제학술발표논문집)

  • 2024.07a
  • /
  • Pages.351-358
  • /
  • 2024
  • /
  • 2508-9048(eISSN)
Korea Institute of Construction Engineering and Management (한국건설관리학회)
권호 표지
DOI QR코드

DOI QR Code

Enhancing Bridge Deterioration Prediction Using Element Adjacency Graphs by OCR-Processed Drawings: A Case Study of Girder Bridges in Japan

  • Shogo INADOMI (Department of Civil Engineering, Faculty of Engineering, The University of Tokyo) ;
  • Pang-jo CHUN (Department of Civil Engineering, Faculty of Engineering, The University of Tokyo)
  • Published : 2024.07.29
Download PDF
⟨ Previous Next ⟩

Abstract

In Japan, bridge inspections are mandated every five years. The inspection database for bridges under the jurisdiction of the Ministry of Land, Infrastructure, Transport, and Tourism enables the acquisition of damage progression data for each structural element. This study develops a methodology for predicting the deterioration of girder bridges, employing a novel approach where inspection drawings are processed using Optical Character Recognition (OCR) to extract element numbers and their spatial relationships, subsequently creating a comprehensive graph of these elements. A key feature of this prediction methodology is its ability to consider the adjacency relationships between different bridge members, made possible by the detailed analysis of drawing information and a Graph Transformer model. The research examines and compares the accuracy of predictions made with and without considering adjacency relationships, highlighting the effectiveness of incorporating detailed structural information in the predictive analysis of bridge deterioration.

Keywords

References

  1. P. A. M. Basheer, S. E. Chidiac, A. E. Long, "Predictive models for deterioration of concrete structures", Construction and Building Materials, vol. 10, no. 1, pp. 27-37, 1996.
  2. F. Ogawa, Y. Chikata, "A study on the statistical deterioration prediction based on inspection data", Journal of Structural Engineering A, vol. 64A, pp. 120-128, 2018.
  3. M. Ilbeigi, M. Ebrahimi Meimand, "Statistical Forecasting of Bridge Deterioration Conditions", Journal of Performance of Constructed Facilities, vol. 34, no. 1, 2020.
  4. Y. Tsuda, K. Kaito, K. Aoki, K. Kobayashi, "Estimating Markovian transition probabilities for bridge deterioration forecasting", Structural Engineering, vol. 23, no. 2, 2006.
  5. T. Tamakoshi, Y. Yoshida, Y. Sakai, S. Fukunaga, "Analysis of damage occurring in steel plate girder bridges on national roads in Japan", Proc. 22th US-Japan Bridge Engineering Workshop, Seattle, Washington, USA, 2006, pp. 1-14.
  6. T. Yamane, P. Chun, R. Honda, "Detecting and localising damage based on image recognition and structure from motion, and reflecting it in a 3D bridge model", Structure and Infrastructure Engineering, 2022.
  7. R. Assaad, I. H. El-adaway, "Bridge Infrastructure Asset Management System: Comparative Computational Machine Learning Approach for Evaluating and Predicting Deck Deterioration Conditions", Journal of Infrastructure Systems, vol. 26, no. 3, 2020.
  8. P. Miao, H. Yokota, Y. Zhang, "Deterioration prediction of existing concrete bridges using a LSTM recurrent neural network", Structure and Infrastructure Engineering, vol. 19, no. 4, pp. 475-489, 2023.
  9. G. Morcous, "Performance Prediction of Bridge Deck Systems Using Markov Chains", Journal of Performance of Constructed Facilities, vol. 20, no. 2, pp. 146-155, 2006.
  10. T. Minami, M. Fujiu, J. Takayama, "Analysis of Bridge Damage Discovered By the Bridge Surface Condition Using Bridge Inspection Data", Journal of Japan Society of Civil Engineers Ser. D3, vol. 74, no. 5, pp. 339-348, 2018.
  11. L. Gao, K. Yu, P. Lu, "Considering the Spatial Structure of the Road Network in Pavement Deterioration Modeling", Transportation Research Record, 2023.
  12. W. L. Hamilton, R. Ying, J. Leskovec, "Inductive representation learning on large graphs", Advances in Neural Information Processing Systems, vol. 2017-Decem, pp. 1025-1035, 2017.
  13. K. Liu, N. El-Gohary, "Ontology-based semi-supervised conditional random fields for automated information extraction from bridge inspection reports", Automation in Construction, vol. 81, pp. 313-327, 2017.
  14. C. Ying, T. Cai, S. Luo, S. Zheng, G. Ke, D. He, Y. Shen, T. Y. Liu, "Do Transformers Really Perform Bad for Graph Representation?", Advances in Neural Information Processing Systems, vol. 34, pp. 28877-28888, 2021.
  15. A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser, I. Polosukhin, "Attention is all you need," Advances in Neural Information Processing Systems, vol. 2017-Decem, pp. 5999-6009, 2017.
  16. J. Yi, M. Saito, K. C. Sinha, "Bridge performance prediction model using the Markov chain", Transportation Research Record, no. 1180, pp. 25-32, 1988.