PSI - Issue 78

Available online at www.sciencedirect.com

ScienceDirect

Procedia Structural Integrity 78 (2026) 1609–1616

© 2025 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of XX ANIDIS Conference organizers Keywords: Seismic assessment; Seismic residual capacity; Post-earthquake; Reinforce Concrete; CNN. Abstract Assessing the seismic residual capacity of damaged structures and infrastructures after a major earthquake is critical to sup port decision-making on re-occupancy, repair, or demolition. This task typically requires collecting information on the observed earthquake-related damage to structural components and performing additional safety and economic loss evaluations for the dam aged structure. While an expert engineering judgment remains crucial to assess damage severity, modern machine-learning tech niques such as Convolutional Neural Networks (CNNs) can represent a valuable supporting tool to automate and standardize this process. In this context, this paper presents and discusses a framework for post-earthquake seismic residual capacity assessment of damaged buildings, integrated with and enhanced by a CNN-based damage detection tool. The framework employs nonlinear static analyses performed through a simplified analytical / mechanical procedure and capacity reduction factors for damaged struc tural components. The level of earthquake-related damage is automatically evaluated from images using a CNN based on a Visual Geometry Group Network 16 (VGG16) architecture, trained on 5,000 images sourced from existing databases and re-labelled by experts. The proposed framework is implemented for a case-study reinforced concrete structure for illustrative purposes. The seismic performance before and after the damaging earthquake is evaluated in terms of a capacity-to-demand safety index, consid ering also the possible uncertainties in the CNN-based damage classification. The results highlight the potential of the CNN-based framework to support emergency planning and decision-making, particularly for large building portfolios. XX ANIDIS Conference Integrating Machine Learning Techniques into Post-Earthquake Residual Capacity Assessment of Reinforced Concrete Structures Livio Pedone a, ∗ , Michele Matteoni a , Simone Saquella b , Michele Scarpiniti c , Stefano Pampanin a a Department of Structural and Geotechnical Engineering, Sapienza University of Rome, via Eudossiana 18, 00184 Rome, Italy b Department of Astronautical, Electrical and Energy Engineering, Sapienza University of Rome, via Eudossiana 18, 00184 Rome, Italy c Dept. of Information Engineering, Electronics and Telecommunications, Sapienza University of Rome, via Eudossiana 18, 00184 Rome, Italy

1. Background and motivations

After a damaging earthquake, rapid post-earthquake surveys are fundamental to assess the safety level of damaged buildings. Substantial earthquake-related damage to buildings can lead to a loss of their lateral-force resisting capacity,

∗ Corresponding author. Tel.: + 39-333-9101890. E-mail address: livio.pedone@uniroma1.it

2452-3216 © 2025 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of XX ANIDIS Conference organizers 10.1016/j.prostr.2025.12.205