PSI - Issue 84

Enrico Pasquale Zitiello et al. / Procedia Structural Integrity 84 (2026) 360–367

364

updating and optimisation. The first phase consists of creating the information model of the infrastructure, which forms the basis of the digital twin. The BIM model is developed from available design data or, in the case of existing bridges, through digital surveying techniques such as laser scanning and photogrammetry. The model includes the structural geometry, the mechanical properties of the materials, the functional subdivision of the elements and the relationships between the components. It provides the information and geometric reference for all subsequent phases, enabling the localisation of sensors, the management of technical information and integration with calculation models. Based on the information model and preliminary analyses, the structural monitoring network is designed. The layout of the sensors is determined based on structural criticalities, operating conditions and monitoring objectives, ensuring consistency between the behaviour predicted by the model and the parameters actually measured. The sensor network may include accelerometers for dynamic analysis, strain gauges for deformation measurement, inclinometers, displacement sensors, environmental sensors, and GNSS systems. The devices, connected via IoT protocols, constitute the physical layer of the Digital Twin, enabling the continuous acquisition of structural and environmental parameters. The data acquired by the sensors is transmitted to a centralised management infrastructure, where it is stored and synchronised in time. This phase includes the collection of real-time information flows, data normalisation, and management of communication protocols between devices and the digital platform. The data infrastructure represents the connection layer of the Digital Twin, ensuring interoperability between sensors, information models, and analysis systems. The collected data is then processed using signal processing techniques, statistical analysis, and numerical models.At this stage, the Digital Twin acts as a computational core, integrating real data with the BIM model and structural models, such as finite element models. Processing allows performance indicators to be extracted, anomalies to be identified and the status of the virtual model to be updated based on actual operating conditions. The digital model thus becomes a dynamic representation of the infrastructure, capable of reflecting structural changes over time. Based on the results of the processing, the system enables the diagnosis of structural conditions and supports the planning of maintenance interventions. The Digital Twin allows the simulation of alternative scenarios, the evaluation of the effectiveness of different strategies, and the estimation of the progression of degradation. This approach enables the adoption of predictive maintenance policies based on real data and structural behaviour models, reducing risks and optimising management costs. Once the interventions are complete, the monitoring system continues to collect data, allowing the effectiveness of the interventions to be verified. The new data is compared with the virtual model's predictions, allowing the Digital Twin to be calibrated and the predictive models to be updated (12). This phase closes the operational cycle and reactivates the monitoring process, configuring the system as an iterative cycle in which each data acquisition phase improves knowledge of the infrastructure and the quality of future decisions (13). The entire methodology is configured as an integrated system in which the information model, sensor network and computational core operate synergistically. The Digital Twin thus becomes an operational platform for continuous monitoring, structural diagnosis and intervention planning, enabling the transition from reactive maintenance strategies to data-driven predictive approaches. In the context of road infrastructure, this approach enables more efficient and safer management of the structure, improving knowledge of structural behaviour and optimising decisions throughout the entire life cycle (14). 4. Case study This chapter describes the operational application of the proposed methodology to two viaducts undergoing structural monitoring, for which an integrated system of sensors, data acquisition, and information modeling was implemented. The two case studies represent different contexts in terms of structural characteristics and monitoring objectives, but they share the same methodological workflow, which connects the digital model of the structure to the real-time data acquisition system. In both cases, the developed system is based on the integration of the information model of the structure and a network of sensors installed at the most structurally significant points, consistent with the phases described in the methodology. The digital model, created in a BIM environment, represents geometry, materials, constraints, and sensor locations, constituting the information reference for the Digital Twin. The data acquired in real time is associated with this model, creating a continuous correspondence between real and simulated behavior. The monitoring network is composed of various types of sensors, capable of detecting structural and environmental