PSI - Issue 78

Gianluca Fagotti et al. / Procedia Structural Integrity 78 (2026) 2070–2077

2075

account from the very beginning, therefore considered in the preliminary studies and currently evaluated in the drafting of the executive project. There are two relevant aspects: (1) whether the vertical component of the seismic action should be considered in the analyses and with what intensity, (2) whether the effects of the vertical action are so significant as to justify the use of an isolation system also for the vertical component. EC8 provides guidance on this only for bridges, stating that site spectra that take into account near-source effects should be used when the site is within 10 km of a known active fault capable of producing a seismic event of moment magnitude Mw > 6.5. Furthermore, in high seismicity areas the effects of the vertical component must be taken into account when the bridge is located within 5 km of an active fault or when the piers are subjected to high bending stresses due to the permanent vertical actions of the deck. Italian standards propose the same method provided in EC8 for constructing vertical acceleration spectra, but do not specify any distinction regarding near-fault conditions. For buildings with isolation they suggest to consider the vertical component of the earthquake only when the ratio between the vertical and horizontal stiffness of the isolation system is less than 800, condition that never occurs in reality. ASCE/SEI 7-22 provides the definition of a site as a near-fault site as a function of the distance, D, from the projection surface of a known active fault and of the moment magnitude (Mw) of the potentially generatable events: a near-fault site can be characterized by D  15 km with Mw ≥ 7 or by D  10 km with Mw ≥ 6. The standard also provides formulas for calculating vertical response spectra (at MCE level, corresponding to the European SLC level) to be applied when site-specific spectra are not used. However, it has to be noted that the prediction equations for the vertical component of the earthquake are derived from models developed by a research project (NGA- West2) studying seismic events in the western United States, certainly not compatible with the seismicity model of the Norcia area. Nonetheless, the use of vertical ground motion is considered optional since, particularly for base-isolated buildings, the high safety factors used in designing static loads are such that vertical seismicity does not pose a real problem, as it has been observed from the damage to structures resulting from the most significant seismic events worldwide. Ultimately, it is appropriate, as with any significant structure, to definitely consider the vertical component in the analyses because: (a) considering only the horizontal component could lead to underestimating the accelerations on the structural components at various levels; (b) some portions of the structure could be highly sensitive to amplifications of vertical accelerations; (c) some components could be sensitive to the combination of vertical and horizontal accelerations; (d) isolation devices subject to low gravitational loads (i.e. isolators in the perimeter zones of the structure) could potentially be subject to uplift or traction caused by the combination of high vertical movements and global overturning phenomena and these uplifts or tractions on the isolators could induce high impact forces on the substructure, risking isolator instability or failure. In reality, the listed situations are highly unlikely, almost impossible, given: (i) the very high frequencies characterizing the vertical seismic input; (ii) the resulting instant duration of the uplift or decompression pulses; (iii) the significant value of vertical forces associated with the mass of the elevated structures and the isolated slab; (iv) the mitigation associated with the stiffness/deformability and dissipation characteristics of the vertical behaviour of both devices and buildings and that shall be accurately considered in nonlinear dynamic analyses on the numerical model of the structure. On the other hand, the simulations carried out in the preliminary studies using both spectrum fitting generated and site-recorded acceleration time-histories showed the absence of the phenomena. Last, but not least, it must be considered that the records of the vertical component of earthquakes can be affected by uncertainties, inaccuracies and overestimations, due to the characteristics of the recording instruments, their support system, the interaction with the host structure, the characteristics of the soil where the control units are located, as per Castellaro et al. (2022) and Castellaro and Musinu (2023). 4. Development and implementation of the project The idea that is at the base of the solution applied in the project was initially conceived as an innovative reconstruction model for towns or city areas that were entirely destroyed in the 2016 Central Italy seismic sequence. Feasibility studies and simulations were carried out on sample areas of actual cities hit by the earthquakes as reported in Mezzi (2019), Fulco et al. (2019), Fulco and Mezzi (2020), obtaining optimum results in terms of: reduction of the seismic response; economic convenience; social value; safe re-proposal of the previous construction.