PSI - Issue 78

Merani Margherita Gabriella Bruna et al. / Procedia Structural Integrity 78 (2026) 785–792

789

overall structural response.

Figure 3: The permanent monitoring system installed on Palazzo Bellevue: schematic of the building showing the location of the base and top floor sensors. Before these permanent systems can be used to detect damage, a detailed and reliable baseline of each building's reference state must be established. To do this, ambient noise measurements were acquired, and from this data, each structure's natural frequencies and mode shapes were estimated. This process must account for potential fluctuations caused by environmental factors, a known challenge in structural monitoring highlighted by experiences from networks like the Italian Seismic Observatory of Structures (OSS) (Sivori et al., 2025). Using the Enhanced Frequency Domain Decomposition (EFDD) technique, the modal parameters for each building were extracted, with the frequencies of the first two identified modes reported in Table 1. Table 1: Summary of the structural properties of the analyzed buildings and the baseline frequencies of their first two identified modes (f₁, f₂), extracted from ambient vibration data Building Material Number of storeys f 1 (Hz) f 2 (Hz) Palazzo Bellevue Masonry 5 3.05 3.35 Hotel Alexander Masonry 5 3.1 4 Hotel Marinella RC 5 3 3.7 Hotel Nazionale Masonry 7 2.9 4 3.1. From predictive to refined scenarios: a comparative application To demonstrate the impact of local data, we compare two damage scenarios for Sanremo for the 1887 earthquake: a standard predictive approach versus a refined methodology using site- and structure-specific information. The first scenario follows a conventional workflow. Seismic hazard is defined by amplifying the rock PGA (estimated from GMPEs) with a linear, frequency-independent site factor (FS) from the ITA18 law, derived from the national-scale shear-wave velocity maps ( ,30 ) of Forte et al. (2019). This rapid method treats all structures at a site identically, ignoring the dynamic interaction between soil and specific building typologies. The second scenario shows a more sophisticated, physically grounded approach. Here, we replace the generic FS factor with a frequency-dependent amplification factor, FA, derived from detailed Seismic Microzonation (MS) studies available for Sanremo. Specifically, they quantify local ground motion amplification through amplification factors defined across different frequency (period) ranges that reflect structural response (i.e., 0.1-0.5 s, 0.4-0.8 s, and 0.7-1.1 s). The key innovation is the integration of monitoring data to inform the selection of FA. The fundamental periods (T₁) identified for the key monitored buildings — Palazzo Bellevue, Hotel Alexander, Hotel Marinella, and Hotel Nazionale — are assumed to be representative of their respective building classes. For the other building typologies in the exposure model, for which a monitored T₁ was not available, the period was estimated using the simplified formula 1 = ∙ 3/4 proposed in the Italian building code (Ministero delle Infrastrutture e dei Trasporti, 2018). In this formula, T₁ is calculated as a function of the building's height (H) and a coefficient (C) that