PSI - Issue 84

Marianna Crognale et al. / Procedia Structural Integrity 84 (2026) 898–905

905

5. Conclusions A hybrid framework for deterioration-aware seismic assessment of RC bridges was presented and applied to a three-span RC viaduct. The methodology combines offline fiber-section analyses with machine-learning surrogates to characterize stiffness and strength degradation mechanisms, while retaining a conventional finite-element formulation for global pushover analysis. Section-level nonlinear moment–curvature simulations were employed to generate a structured dataset covering multiple concrete and steel degradation scenarios under different axial load levels. Machine-learning models were trained to predict physically meaningful degradation indicators for effective stiffness and strength as functions of curvature demand and axial force, demonstrating high fidelity with respect to the underlying fiber-based responses. At the structural scale, the trained ML surrogates were employed exclusively for post-processing and interpretation, enabling the evolution of section-level degradation to be quantified along the global pushover loading path without altering the nonlinear equilibrium solution. This strategy preserves the robustness and transparency of traditional pushover analysis, while providing additional insight into the progression of material deterioration that cannot be inferred from capacity curves alone. Overall, the proposed approach offers a computationally efficient and physically consistent enhancement to conventional seismic assessment procedures for existing RC bridges. The framework also provides a scalable foundation for future extensions, including uncertainty-aware assessment, data-driven model calibration, and integration within digital-twin-oriented monitoring and decision-support systems. References Banerjee, S., Shinozuka, M., 2007. Nonlinear static procedure for seismic vulnerability assessment of bridges. Computer-Aided Civil and Infrastructure Engineering 22(4), 293–305. Bernardini, D., Ruta, D., Di Re, P., & Paolone, A. (2024). A fiber-based deterioration modeling framework for reinforced concrete structures subject to spatially non-uniform corrosion patterns described by limited information. Structural Concrete, 26(5), 5650–5676. Chen, T., Guestrin, C., 2016. XGBoost: A Scalable Tree Boosting System. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, 13-17 August 2016, 785-794. Ciminelli, F., Bernardini, D., Lofrano, E., Paolone, A., 2025. Multi-risk assessment of bridges and viaducts according to classes and logical operators: Conceptual analysis and statistical paths. Journal of Civil Structural Health Monitoring 15(7), 2159–2182. De Domenico, D., Messina, D., Recupero, A., 2023. Seismic vulnerability assessment of reinforced concrete bridge piers with corroded bars. Structural Concrete 24(1), 56–83. Di Re, P., Ciambella, J., Lofrano, E., Paolone, A., 2024. Dynamic testing and modeling of span interaction in high-speed railway girder bridges. Measurement 226, 114078. Dizaj, E.A., Salami, M.R., Kashani, M.M., 2023. Seismic vulnerability analysis of irregular multi-span concrete bridges with different corrosion damage scenarios. Soil Dynamics and Earthquake Engineering 165, 107678. Feng, Y., Kowalsky, M.J., Nau, J.M., 2014. Fiber-based modeling of circular reinforced concrete bridge columns. Journal of Earthquake Engineering 18(5), 714–734. Gattulli, V., Lofrano, E., Paolone, A., Pirolli, G., 2017. Performances of FRP reinforcements on masonry buildings evaluated by fragility curves. Computers and Structures 190, 150–161. Ghazal, H., Mwafy, A., 2023. Comparative performance evaluation of retrofit alternatives for upgrading simply supported bridges using 3D fiber based analysis. Buildings 13(5), 1161. Mosleh, A., Jara, J., Varum, H., 2015. A methodology for determining the seismic vulnerability of old concrete highway bridges by using fragility curves. Journal of Structural Engineering and Geo-Techniques 5(1), 1–7. McKenna, F., Fenves, G.L., Scott, M.H., 2000. OpenSees (Version 3.7.1): Open System for Earthquake Engineering Simulation. University of California, Berkeley, CA, software framework. Available at http://opensees.berkeley.edu Paduano, I., Mileto, A., Lofrano, E., 2023. A perspective on AI-based image analysis and utilization technologies in building engineering: Recent developments and new directions. Buildings 13(5), 1198. Salvatore, W., et al., 2024. Application of Italian guidelines for structural-foundational and seismic risk classification of bridges: the FABRE experience on a large bridge inventory. Procedia Structural Integrity 62, 1–8.