All Issue

2025 Vol.3, Issue 2 Preview Page

Research Article

CTGAN-Based Data Sampling for Prediction of Bridge Safety Rating

교량 안전등급 예측을 위한 CTGAN 기반 데이터 샘플링

Jisu Hong1*
,
Se-Jin Jeon2
홍 지수1*
,
전 세진2
1Graduate Student, Department of Civil Systems Engineering, Ajou University, Suwon 16499, Korea
2Professor, Department of Civil Systems Engineering, Ajou University, Suwon 16499, Korea
1아주대학교 건설시스템공학과 박사과정
2아주대학교 건설시스템공학과 교수
*Corresponding Author
31 December 2025. pp. 192-202
PDF XML Citation
Abstract

Dataset related to safety rating of bridges often suffers from limited reliability and severe class imbalance, which hinders the development of AI(Artificial Intelligence) models. To address these challenges, this study implemented a comprehensive data preprocessing focused on improving data quality and incorporating key factors related to safety rating of bridges. Inspection reliability was enhanced by integrating detailed results from in-depth and precise safety inspections, and additional variables representing maintenance history and environmental conditions were introduced to better capture factors affecting safety performance. Because the dataset exhibits a highly imbalanced class distribution, which is difficult to resolve using traditional sampling techniques, this study adopted the Conditional Tabular Generative Adversarial Network(CTGAN), a generative model specialized for tabular data. CTGAN was used to synthesize minority-class samples while preserving the original distribution of the majority class, enabling balanced learning without data loss. Overall, the results demonstrate that CTGAN-based data sampling offers an effective alternative for handling imbalanced dataset that contains both categorical and numerical variables, such as bridge safety ratings. The proposed approach is expected to contribute to improving the performance of future AI-based models for bridge safety assessment and related decision-making in infrastructure maintenance.

Keywords
artificial intelligence
CTGAN
bridge safety rating
data preprocessing
synthetic data

국내 교량 안전등급 데이터는 신뢰성의 한계와 등급 분포의 불균형 문제로 인해 인공지능 기반 예측 모델의 성능 확보에 어려움이 존재한다. 이 연구에서는 이러한 문제를 해결하기 위해 데이터 신뢰성 향상과 안전등급 관련 주요 인자의 확보에 중점을 둔 전처리 절차를 수행하고, 표 형식의 데이터 생성이 가능한 Conditional Tabular Generative Adversarial Networks(CTGAN) 기반 합성 데이터 기법을 도입하였다. 정밀안전점검 및 진단 결과를 반영하도록 하여 안전등급 결과를 개선하고, 유지관리 이력 및 환경 특성을 고려한 변수를 추가하여 교량 안전등급 데이터를 구축하였다. 교량의 안전등급 데이터의 경우 전통적 샘플링 기법을 통해 불균형 분포를 해결하기 어렵기 때문에, CTGAN 기반 합성 데이터 기법을 적용하여 소수 클래스의 분포를 실제 데이터와 유사한 형태로 재현하고, 다수 클래스의 본래 분포를 유지하여 데이터의 손실 없이 학습 데이터의 균형을 확보하였다. 따라서, 이 연구는 CTGAN 기반 샘플링이 교량 안전등급과 같이 범주형 및 수치형 변수가 혼재되고 불균형 분포인 데이터에서 효과적인 대안이 될 수 있음을 제시하였으며, 향후 교량 안전등급 예측과 같은 유지관리 의사결정 모델의 성능 향상에 기여할 수 있을 것으로 기대된다.

키워드
인공지능
CTGAN
교량 안전등급
데이터 전처리
합성 데이터
References
1

ASCE (2025) Report card for America’s infrastructure. ASCE (American Society of Civil Engineers), Reston, VA, USA.

2

Assaad, R. and El-Adaway, I. H. (2020) Bridge infrastructure asset management system: Comparative computational machine learning approach for evaluating and predicting deck deterioration conditions. Journal of Infrastructure Systems, 26(3), 04020032. https://doi.org/10.1061/(ASCE)IS.1943-555X.0000572.

10.1061/(ASCE)IS.1943-555X.0000572
3

Bayat, M. and Kharel, S. (2025) Leveraging artificial intelligence for predictive maintenance and condition rating of off-system bridges. Applied Sciences, 15(21), 11301. https://doi.org/10.3390/app152111301.

10.3390/app152111301
4

Bektas, B. A., Carriquiry, A. and Smadi, O. (2013) Using classification trees for predicting national bridge inventory condition ratings. Journal of Infrastructure Systems, 19(4), 425-433. https://doi.org/10.1061/(ASCE)IS.1943-555X.0000143.

10.1061/(ASCE)IS.1943-555X.0000143
5

Chávez, A. A., Gonzalez-Libreros, J., Wang, C., Capacci, L., Biondini, F., Elfgren, L. and Sas, G. (2024) Assessment of residual prestress in existing concrete bridges: The Kalix bridge. Engineering Structures, 311, 118194. https://doi.org/10.1016/j.engstruct.2024.118194.

10.1016/j.engstruct.2024.118194
6

Chung, S., Lim, S. and Chi, S. (2016) Developing an estimation model for safety rating of road bridges using rule-based classification method. Journal of KIBIM, 6(2), 29-38. https://doi.org/10.13161/kibim.2016.6.2.029. (In Korean)

10.13161/kibim.2016.6.2.029
7

FMS (Facility Management System) (2023) Facilities management and maintenance services portal. Accessed December 11, 2023. (In Korean)

8

Hong, J. and Jeon, S. J. (2023) Prediction of safety grade of bridges using the classification models of decision tree and random forest. KSCE Journal of Civil and Environmental Engineering Research, 43(3), 397–411. https://doi.org/10.12652/Ksce.2023.43.3.0397. (In Korean)

10.12652/Ksce.2023.43.3.0397
9

Hong, J. and Jeon, S. J. (2025) Efficient decision tree-based classification models to predict safety rating for bridge maintenance. Journal of Infrastructure Systems, 31(1), 04024031. https://doi.org/10.1061/JITSE4.ISENG-2524.

10.1061/JITSE4.ISENG-2524
10

Jayathilaka, S. T., Phares, B. M. and Liu, Z. (2023) Implementation of a mathematical model for the prediction of the future condition rating for bridge components. Transportation Research Record, 2677(3), 1700-1714. https://doi.org/10.1177/03611981221127003.

10.1177/03611981221127003
11

KCI (2009) Standard specification for concrete. KCI (Korea Concrete Institute), Seoul, Korea. (In Korean)

12

Kim, S. J. and Yoon, M. O. (2018) A study on the improvement program of bridge safety management through public-private governance. Journal of the Korean Society of Hazard Mitigation, 18(1), 145-156, http://doi.org/10.9798/KOSHAM.2018.18.1.145. (In Korean)

10.9798/KOSHAM.2018.18.1.145
13

Kim, Y. J. and Queiroz, L. B. (2017) Big data for condition evaluation of constructed bridges. Engineering Structures, 141, 217-227. https://doi.org/10.1016/j.engstruct.2017.03.028.

10.1016/j.engstruct.2017.03.028
14

Kobayashi, K., Kaito, K. and Lethanh, N. (2012) A statistical deterioration forecasting method using hidden Markov model for infrastructure management. Transportation Research Part B: Methodological, 46(4), 544-561. https://doi.org/10.1016/j.trb.2011.11.008.

10.1016/j.trb.2011.11.008
15

Lee, J. H., Lee, K. Y., Ahn, S. M. and Kong, J. S. (2018) Proposal of maintenance scenario and feasibility analysis of bridge inspection using bayesian approach. KSCE Journal of Civil and Environmental Engineering Research, 38(4), 505-516. https://doi.org/10.12652/Ksce.2018.38.4.0505. (In Korean)

10.12652/Ksce.2018.38.4.0505
16

Lei, X., Xia, Y., Komarizadehasl, S. and Sun, L. (2022) Condition level deteriorations modeling of RC beam bridges with U-Net convolutional neural networks. Structures, 42, 333-342. https://doi.org/10.1016/j.istruc.2022.06.013.

10.1016/j.istruc.2022.06.013
17

Maaten, L. V. D. and Hinton, G. (2008) Visualizing data using t-SNE. Journal of Machine Learning Research, 9(86), 2579-2605.

18

Maćkiewicz, A. and Ratajczak, W. (1993) Principal components analysis (PCA). Computers & Geosciences, 19(3), 303-342. https://doi.org/10.1016/0098-3004(93)90090-R

10.1016/0098-3004(93)90090-R
19

MOLIT (Ministry of Land, Infrastructure and Transport) (2023a) National bridge standard data. Accessed September 12, 2023. (In Korean)

20

MOLIT (Ministry of Land, Infrastructure and Transport) (2023b) Yearbook of road bridge and tunnel statistics. Accessed September 12, 2023. (In Korean)

21

MOLIT (Ministry of Land, Infrastructure and Transport) (2024a) Guidelines for safety and maintenance of facilities. Sejoing City, Korea. (In Korean)

22

MOLIT (Ministry of Land, Infrastructure and Transport) (2024b) Special act on the safety and maintenance of infrastructure. Sejoing City, Korea. (In Korean)

23

Nakashima, M. and Nagai, K. (2014) An investigation of road bridge maintenance system in Japan in developed society. Society for Social Management Systems Internet Journal, 9.

24

NARS (2018) A study on infrastructure maintenance systems in major overseas countries. National Assembly Research Service (NARS), Seoul, Korea. (In Korean)

25

Xu, L., Skoularidou, M., Cuesta-Infante, A. and Veeramachaneni, K. (2019) Modeling tabular data using conditional gan. Advances in Neural Information Processing Systems, 32.

Information
  • Publisher :Korean Institute of Bridge and Structural Engineers
  • Publisher(Ko) :한국교량및구조공학회
  • Journal Title :Journal of Structure Research and Practice
  • Journal Title(Ko) :한국교량및구조공학회 논문집
  • Volume : 3
  • No :2
  • Pages :192-202
  • Received Date : 2025-12-11
  • Revised Date : 2026-01-02
  • Accepted Date : 2026-01-02
  • DOI :https://doi.org/10.22725/JSRP.2025.3.2.192
Journal Informaiton Journal of Structure Research and Practice Journal of Structure Research and Practice
Online Submission & Review System
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • KCI 문헌 유사도 검사
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close