PSI - Issue 84

Available online at www.sciencedirect.com

ScienceDirect

Procedia Structural Integrity 84 (2026) 288–295

© 2026 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 the scientific committee of the Conference Keywords: Bridges, Damage Assessment, Model-based, Neural Network, Surrogate Models, Transfer Learning Abstract This work presents a novel approach to structural damage assessment designed to support network-level monitoring of bridges. The increasing deployment of permanent Structural Health Monitoring systems has led to a rapid growth in available data, offering new opportunities to develop algorithms that exploit the knowledge acquired from the analysis of real-world structures to enhance the condition assessment of other similar structures. In this context, the study introduces a model-based transfer learning methodology that enables knowledge sharing across a network of bridges by means of neural network-based surrogate models. Specifically, deep neural networks are employed to construct surrogate models, replacing traditional non-intrusive response surface models commonly used in the literature. The proposed approach allows a surrogate model trained on one bridge to be transferred and adapted to another with comparable characteristics, paving the way for typology-driven surrogate modeling. By embedding shared knowledge of damage mechanisms across structurally similar systems, the proposed framework facilitates scalable, data informed, and model-aware monitoring strategies, contributing to more efficient infrastructure management at the network scale. 1. Introduction The safety and durability of bridge infrastructure are essential for the resilience of transportation networks and public safety. Recent surveys reveal that a significant portion of bridges worldwide exhibit structural deficiencies, III Fabre Conference: Existing Bridges, Viaducts, and Tunnels: Research, Innovation, and Applications Model-Based Learning for Multiple Bridge Monitoring in Transportation Networks: A Preliminary Study Elisa Tomassini a, *, Enrique García-Macías b and Filippo Ubertini a a Department of Civil and Environmental Engineering, University of Perugia. Via G. Duranti, 93 - 06125 Perugia, Italy. b Department of Structural Mechanics and Hydraulic Engineering, University of Granada, Av. Fuentenueva sn, 18002 Granada, Spain.

* Corresponding author. E-mail address: elisa.tomassini@dottorandi.unipg.it

2452-3216 © 2026 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 the scientific committee of the Conference 10.1016/j.prostr.2026.06.038