Leveraging Artificial Intelligence for Predictive Maintenance and Condition Rating of Off-System Bridges
Source abstract
Abstract
Off-system bridges are critical components of the United States’ transportation infrastructure, providing essential access to rural communities and enabling residents to reach vital services such as employment, education, and healthcare. Many of these bridges are structurally deficient, functionally obsolete, and unmaintained. This disproportionately hinders the mobility of underserved populations, worsening socioeconomic disparities. Despite existing research, there is insufficient focus on the unique challenges posed by off-system bridges, including handling the class imbalanced nature of the bridge condition rating dataset. This paper predicts bridge deck conditions by using Generative Adversarial Networks with Focal Loss (GAN-FL) to generate synthetic data which enhances precision–recall balance in imbalanced datasets. Results show that integrating GAN-FL with random forest (RF) classifiers significantly enhances the performance of minority classes, improving their precision, recall, and F1 scores. The study finds that augmenting training data using GAN-FL greatly enhances minority class prediction, thereby assisting in accurate bridge condition modeling.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Infrastructure Maintenance and Monitoring · Concrete Corrosion and Durability · Non-Destructive Testing Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Architecture, College of Architecture, Planning and Public Affairs, The University of Texas at Arlington, Arlington, TX 76000-76099, USA
Department of Planning and Public Affairs, College of Architecture, Planning and Public Affairs, The University of Texas at Arlington, Arlington, TX 76000-76099, USA
Funding
Dean’s Interdisciplinary Research Initiative Grant from the College of Architecture, Planning and Public Affairs · University of Texas at Arlington · Interdisciplinary Research Program at University of Texas at Arlington
References (67)
Show source-provided references
1. Cost-effective structures for off-system bridges (NCHRP Synthesis of Highway Practice Project 20-5, Topic 32-08). Draft report10.3141/1819b-50 ↗
2. Wood in transportation program: An overview10.3141/1696-32 ↗
3. Rural Roads and Bridges, 1994–2000: How Did the South Fare?10.1017/S1074070800021234 ↗
4. Sharkey, J. (2024, October 10). Infrastructure Investment and Jobs Act (Iija)/Bipartisan Infrastructure Law (Bil), Road School, Purdue University. Available online: https://docs.lib.purdue.edu/cgi/viewcontent.cgi?article=4550&context=roadschool.
5. US Congress (2024, October 10). Bipartisan Infrastructure Law (BIL) Infrastructure Investment and Jobs Act (IIJA) Infrastructure Investment and Jobs Act, H.R.3684, Available online: https://www.phmsa.dot.gov/legislative-mandates/bipartisan-infrastructure-law-bil-infrastructure-investment-and-jobs-act-iija.
6. Load-capacity rating of bridge populations through machine learning: Application of decision trees and random forests10.1061/(ASCE)BE.1943-5592.0001103 ↗
7. Quadri, M.W. (2024). Statistical Analysis on the Effects of Environmental Factors on Bridge Condition Rating in Texas. [Master’s Thesis, University of Texas].
8. Bridge infrastructure management system: Autoencoder approach for predicting bridge condition ratings10.1061/JITSE4.ISENG-2123 ↗
9. An In-Depth Analysis of the National Bridge Inventory Database Utilizing Data Mining, GIS and Advanced Statistical Methods
10. Probabilistic models for life—cycle performance of deteriorating structures: Review and future directions10.1002/pse.180 ↗
11. FHWA (Federal Highway Administration) (1995). Recording and Coding Guide for the Structure Inventory and Appraisal of the Nation’s Bridges.
12. Fault diagnosis using variational autoencoder GAN and focal loss CNN under unbalanced data10.1177/14759217241254121 ↗
13. Bridge-condition assessment and load rating using nondestructive evaluation methods10.3141/1696-48 ↗
14. Zinno, R., and Obrien, E.J. (2023). State-of-the-Art Structural Health Monitoring in Civil Engineering. Appl. Sci., 13.10.3390/app132111609 ↗
15. Identification of environmental categories for Markovian deterioration models of bridge decks10.1061/(ASCE)1084-0702(2003)8:6(353) ↗
16. Multivariable proportional hazards-based probabilistic model for bridge deterioration forecasting10.1061/(ASCE)IS.1943-555X.0000534 ↗
17. Estimating the future condition of highway bridge components using national bridge inventory data10.1061/(ASCE)1084-0680(2004)9:1(16) ↗
18. Moomen, M., Qiao, Y., Agbelie, B.R., Labi, S., and Sinha, K.C. (2016). Bridge Deterioration Models to Support Indiana’s Bridge Management System, Indiana Department of Transportation and Purdue University.10.5703/1288284316348 ↗
19. Estimation of infrastructure transition probabilities from condition rating data10.1061/(ASCE)1076-0342(1995)1:2(120) ↗
20. Deterioration prediction of timber bridge elements using the Markov chain10.1061/(ASCE)CF.1943-5509.0000311 ↗
21. (2021). Chencho; Li, J.; Hao, H.; Wang, R.; Li, L. Development and application of random forest technique for element level structural damage quantification. Struct. Control Health Monit., 28, e2678.10.1002/stc.2678 ↗
22. Deep learning enhanced principal component analysis for structural health monitoring10.1177/14759217211041684 ↗
23. Presenting a new method and X-index based on Choi-Williams distribution and matrix density methods to detect damage in concrete beams
24. A new index based on short time Fourier transform for damage detection in bridge piers
25. Reliability based seismic fragility analysis of bridge
26. Xu, Z., Shen, D., Nie, T., and Kou, Y. (2020). A hybrid sampling algorithm combining M-SMOTE and ENN based on Random forest for medical imbalanced data. J. Biomed. Inform., 107.10.1016/j.jbi.2020.103465 ↗
27. Boruta–a system for feature selection10.3233/FI-2010-288 ↗
28. Kumar, S.S., and Shaikh, T. (2017, January 6–7). Empirical evaluation of the performance of feature selection approaches on random forest. Proceedings of the 2017 International Conference on Computer and Applications (ICCA), Dubai, United Arab Emirates.10.1109/COMAPP.2017.8079769 ↗
29. Zeng, M., Zou, B., Wei, F., Liu, X., and Wang, L. (2016, January 28–29). Effective prediction of three common diseases by combining SMOTE with Tomek links technique for imbalanced medical data. Proceedings of the 2016 IEEE International Conference of Online Analysis and Computing Science (ICOACS), Chongqing, China.10.1109/ICOACS.2016.7563084 ↗
30. SMOTE: Synthetic minority over-sampling technique10.1613/jair.953 ↗
31. Qi, Y., Su, L., Gu, J., and Li, K. (2022, January 27–30). CE-CGAN: Classification enhanced conditional generative adversarial networks for bearing fault diagnosis. Proceedings of the 12th International Conference on Quality, Reliability, Risk, Maintenance, and Safety Engineering, Emeishan, China.
32. ICoT-GAN: Integrated Convolutional Transformer GAN for Rolling Bearings Fault Diagnosis Under Limited Data Condition10.1109/TIM.2023.3271729 ↗
33. Lin, T., Goyal, P., Girshick, R., He, K., and Dollár, P. (2017, January 22–29). Focal loss for dense object detection. Proceedings of the IEEE International Conference on Computer Vision (ICCV), Venice, Italy.10.1109/ICCV.2017.324 ↗
34. Li, L., and Talwalkar, A. (2020). Random search and reproducibility for neural architecture search. arXiv.
35. Generative adversarial network: An overview of theory and applications
36. Liu, X., and Hsieh, C.J. (2019, January 27–28). Rob-gan: Generator, discriminator, and adversarial attacker. Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seoul, Republic of Korea.10.1109/CVPR.2019.01149 ↗
37. Liu, H., Wan, Z., Huang, W., Song, Y., Han, X., and Liao, J. (2021, January 20–25). Pd-gan: Probabilistic diverse gan for image inpainting. Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Nashville, TN, USA.10.1109/CVPR46437.2021.00925 ↗
38. Class-discriminative focal loss for extreme imbalanced multiclass object detection towards autonomous driving10.1007/s00371-021-02067-9 ↗
39. Imbalanced data learning by minority class augmentation using capsule adversarial networks10.1016/j.neucom.2020.01.119 ↗
40. Mullick, S.S., Datta, S., and Das, S. (November, January 27). Generative adversarial minority oversampling. Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision, Seoul, Republic of Korea.10.1109/ICCV.2019.00178 ↗
41. Bjorck, N., Gomes, C.P., Selman, B., and Weinberger, K.Q. (2018). Understanding batch normalization. Advances in neural information processing systems. arXiv.
42. Asadi, B., and Jiang, H. (2020). On approximation capabilities of ReLU activation and softmax output layer in neural networks. arXiv.
43. Bock, S., Goppold, J., and Weiß, M. (2018). An improvement of the convergence proof of the ADAM-Optimizer. arXiv.
44. Xin, B., Yang, W., Geng, Y., Chen, S., Wang, S., and Huang, L. (2020, January 4–8). Private FL-GAN: Differential privacy synthetic data generation based on federated learning. Proceedings of the ICASSP 2020—2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Barcelona, Spain.10.1109/ICASSP40776.2020.9054559 ↗
45. Efficient medical diagnosis of human heart diseases using machine learning techniques with and without GridSearchCV10.1109/ACCESS.2022.3165792 ↗
46. Montavon, G., Orr, G.B., and Müller, K.-R. (2012). Early stopping—But when?. Neural Networks: Tricks of the Trade, Springer.10.1007/978-3-642-35289-8 ↗
47. Khoshgoftaar, T.M., Golawala, M., and Van Hulse, J. (2007, January 29–31). An empirical study of learning from imbalanced data using random forest. Proceedings of the 19th IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2007), Patras, Greece.10.1109/ICTAI.2007.46 ↗
48. To tune or not to tune the number of trees in random forest
49. The effect of splitting on random forests10.1007/s10994-014-5451-2 ↗
50. Genuer, R., and Poggi, J.M. (2020). Random Forests with R, Springer International Publishing.10.1007/978-3-030-56485-8 ↗
51. Predicting Important Features That Influence COVID-19 Infection Through Light Gradient Boosting Machine: Case of Toronto10.11648/j.ajmcm.20210603.11 ↗
52. Cross-validation methods10.1006/jmps.1999.1279 ↗
53. Area under the precision-recall curve: Point estimates and confidence intervals
54. MLCM: Multi-Label Confusion Matrix10.1109/ACCESS.2022.3151048 ↗
55. Dunphy, K., Fekri, M.N., Grolinger, K., and Sadhu, A. (2022). Data augmentation for deep-learning-based multiclass structural damage detection using limited information. Sensors, 22.10.3390/s22166193 ↗
56. Multiclass damage detection in concrete structures using a transfer learning—based generative adversarial networks10.1002/stc.3079 ↗
57. Few-shot learning augmented with image transformation for multiclass structural damage classification10.1061/JCCEE5.CPENG-6198 ↗
58. An Autoencoder-Based Machine-and Deep-Learning Approach for Predicting Bridge Deck Conditions in Texas10.1061/JSDCCC.SCENG-1799 ↗
59. Two-stage anomaly detection for imbalanced bridge data by attention mechanism optimisation and small sample augmentation10.1016/j.engstruct.2025.119613 ↗
60. An efficient 2D-3D fusion method for bridge damage detection under complex backgrounds with imbalanced training data10.1016/j.aei.2025.103373 ↗
61. Goodfellow, I.J., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., and Bengio, Y. (2020, January 18–22). Generative Adversarial Nets. Proceedings of the 27th International Conference on Neural Information Processing Systems, Bangkok, Thailand.
62. Frid-Adar, M., Klang, E., Amitai, M., Goldberger, J., and Greenspan, H. (2018, January 4–7). Synthetic data augmentation using GAN for improved liver lesion classification. Proceedings of the 2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018), Washington, DC, USA.10.1109/ISBI.2018.8363576 ↗
63. Imbalanced sample fault diagnosis of rotating machinery using conditional variational auto-encoder generative adversarial network10.1016/j.asoc.2020.106333 ↗
64. Damage identification of steel bridge based on data augmentation and adaptive optimization neural network10.1177/14759217241255042 ↗
65. Nonlinear modeling of temperature-induced bearing displacement of long-span single-pier rigid frame bridge based on DCNN-LSTM10.1016/j.csite.2023.103897 ↗
66. Missing measurement data recovery methods in structural health monitoring: The state, challenges and case study10.1016/j.measurement.2024.114528 ↗
67. Reconstruction of structural acceleration response based on CNN-BiGRU with squeeze-and-excitation under environmental temperature effects10.1007/s13349-024-00859-w ↗