International Journal of
Advanced Multidisciplinary Application

Volume 3, Issue 7

• Author(s)Ananya Krishnamurthy
• AffiliationDepartment of Civil Engineering and Centre for Data-Driven Infrastructure Systems, Indian Institute of Technology Hyderabad, Telangana, India
• Page No.35-40
• Volume, Issue & YearVolume 3, Issue 7,
• Published On2026/07/04
• JournalInternational Journal of Advanced Multidisciplinary Application (IJAMA)
• ISSN No.3048-9350

Abstract

Ageing steel highway bridge infrastructure across rapidly motorising economies faces a structural performance assessment bottleneck: visual inspection protocols mandated at 2-4 year intervals cannot detect fatigue cracking, corrosion-induced section loss, or connection deterioration between scheduled inspections, while continuous instrumentation generates data volumes that exceed manual interpretation capacity. This study addresses the integration gap between structural health monitoring (SHM) sensor networks and automated damage diagnosis by developing and field-validating a multidisciplinary machine learning framework that fuses vibration, strain, and electrochemical corrosion-potential data streams for bridge-level predictive maintenance.
A sensor network comprising 24 accelerometers, 18 strain gauges, and 12 half-cell corrosion-potential probes was deployed across two instrumented steel girder bridges (a 42 m simply-supported control structure and a 58 m structure with a known fatigue crack at Girder 3) on the Hyderabad Outer Ring Road corridor, generating 729 days of continuous monitoring data sampled at 200 Hz for dynamic channels and hourly for electrochemical channels. Four machine learning architectures — a hybrid convolutional neural network-long short-term memory (CNN-LSTM) model, a stacked LSTM, a random forest classifier, and a support vector machine — were trained on time-frequency and statistical features extracted from the fused sensor streams to perform four-class damage severity classification (none, minor, moderate, severe) and remaining useful life (RUL) regression.
The CNN-LSTM hybrid achieved the highest damage classification accuracy (96.8%) and area-under-curve of 0.974 for binary damage detection when all three sensor modalities were fused, compared to 0.887-0.901 AUC for single-modality models, confirming a substantial sensor-fusion advantage. The framework detected the onset of progressive natural frequency drift in the fatigue-cracked bridge 38 days before the deviation would have been flagged by routine biennial inspection, with RUL predictions achieving a root-mean-square error of 11.4 days across 20 field-validated bridge segments. Lifecycle cost analysis indicates that ML-driven predictive maintenance reduces annualised per-bridge costs by 60.7% relative to reactive run-to-failure maintenance and by 36.6% relative to fixed-interval scheduled maintenance, driven predominantly by avoided downtime and disruption costs. SHAP-based feature attribution identifies modal frequency shift, daily strain range, and corrosion potential as the three dominant predictors of damage severity, providing an interpretable basis for sensor network prioritisation in resource-constrained monitoring deployments.

Keywords: structural health monitoring, predictive maintenance, machine learning, sensor fusion, CNN-LSTM, steel bridges, remaining useful life, damage detection, corrosion monitoring, fatigue, infrastructure asset management

References

1. [1] Avci, O., Abdeljaber, O., Kiranyaz, S., Hussein, M., Gabbouj, M., & Inman, D. J. (2021). A review of vibration-based damage detection in civil structures: From traditional methods to machine learning and deep learning applications. Mechanical Systems and Signal Processing, 147, 107077.
2. [2] Bao, Y., Tang, Z., Li, H., & Zhang, Y. (2019). Computer vision and deep learning-based data anomaly detection method for structural health monitoring. Structural Health Monitoring, 18(2), 401-421.
3. [3] Cha, Y. J., Choi, W., & Büyüköztürk, O. (2017). Deep learning-based crack damage detection using convolutional neural networks. Computer-Aided Civil and Infrastructure Engineering, 32(5), 361-378.
4. [4] Farrar, C. R., & Worden, K. (2012). Structural Health Monitoring: A Machine Learning Perspective. Wiley.
5. [5] Figueiredo, E., Park, G., Figueiras, J., Farrar, C., & Worden, K. (2011). Structural health monitoring algorithm comparisons using standard data sets. Los Alamos National Laboratory Report LA-14393.
6. [6] Gordan, M., Razak, H. A., Ismail, Z., & Ghaedi, K. (2017). Recent developments in damage identification of structures using data mining. Latin American Journal of Solids and Structures, 14(13), 2373-2401.
7. [7] Lundberg, S. M., & Lee, S. I. (2017). A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems, 30, 4765-4774.
8. [8] Mishra, M., Lourenço, P. B., & Ramana, G. V. (2022). Structural health monitoring of civil engineering structures by using the internet of things: A review. Journal of Building Engineering, 48, 103954.
9. [9] Ministry of Road Transport and Highways. (2023). Guidelines for Inspection and Structural Rating of Bridges. Government of India.
10. [10] Reddy, K. S., & Vijayan, D. S. (2021). Bridge health monitoring using IoT sensors: An Indian infrastructure perspective. Materials Today: Proceedings, 43, 1297-1303.
11. [11] Salehi, H., & Burgueño, R. (2018). Emerging artificial intelligence methods in structural engineering. Engineering Structures, 171, 170-189.
12. [12] Seo, J., Hu, J. W., & Lee, J. (2016). Summary review of structural health monitoring applications for highway bridges. Journal of Performance of Constructed Facilities, 30(4), 04015072.
13. [13] Sony, S., Laventure, S., & Sadhu, A. (2019). A literature review of next-generation smart sensing technology in structural health monitoring. Structural Control and Health Monitoring, 26(3), e2321.
14. [14] Zhang, Y., Miyamori, Y., Mikami, S., & Saito, T. (2019). Vibration-based structural state identification by a 1-dimensional convolutional neural network. Computer-Aided Civil and Infrastructure Engineering, 34(9), 822-839

Explore Our Related Journals

Looking for the right journal for your next manuscript? Explore our international peer-reviewed journals covering engineering, management, computer science, artificial intelligence and multidisciplinary research.