GNGTS 2022 - Atti del 40° Convegno Nazionale

GNGTS 2022 Sessione 2.2 267 The methodology considers the management of the different information through two informative systems, GIS and BIM, selected according to the details of the evaluation. Initially, the structural units were investigated by external visual inspections, collecting the data in a GIS platform. The seismic vulnerability of the different units was evaluated through the Giovinazzi and Lagomarsino modifiers (2004) based on the EMS-98 scale (Grunthal 1998). The latter allows the identification of a vulnerability index based on the structural features of each structure (Fig. 2-b). Later, in-depth cognitive studies have been executed through a suitable approach for the geometrical survey. The cadastral plans of each house were collected and combined according to a laser scanner survey executed from the outside (Fig. 2-c). This new information allowed the adoption of a different empirical methodology, i.e., the GNDT second level assessment. To use this method further information are required, e.g., the resistant area along the two directions and other internal features. In addition, the GNDT sheet has been used adopting the aggregate rows implemented by Formisano et al. (2011). The latter introduces 5 new scores to account on the effects of adjacent buildings within masonry aggregates, as the initial GNDT method was conceived for single structures. The comparisons within the two empirical methodologies in terms of vulnerability indexes and damage scenarios highlights the significance of deepen cognitive studies at the aggregate level (Cardinali et al., 2020). Specifically, the EMS-98 modifiers lead to more conservative results, highlighting a lower variability. On the other hand, the Formisano GNDT second level approach points out major differences, remarkable thanks to the deepen cognitive studies. In Fig. 2-d the Vulnerability curves adopting the second empirical approach are presented. The curves plot the Damage levels - DLi, with i=1,…,5 – (ordinate) according to different PGA values (abscissa). Namely, the DLs represent the performance point after which a Damage State occurs (DS). The DS are defined in accordance to EMS-98 scale (Grünthal, 1998): DS0 – no damage, DS1 – slight, DS2 – moderate, DS3 – heavy, DS4 – very heavy, and DS5 – collapse. The mean curve accounts the mean Vulnerability index coming from the different structural units of the center; in addition, the mean curve accounting on the standard deviation of the sample together with the min/ max curves coming from the urban center are plotted. For a seismic action with 475 years return period, the mean expected damage for the historical center is divided with a 29% of probability to reach DL3, the 13% for DL4 and 25 % for DL2. In Fig. 2-f the damage scenarios for 475 years return period action are shown. The new available contents also provided the usage of BIM platforms to collect the architectural and spatial information of the urban clusters (Fig. 2-e). Additional studies have been targeted at the execution of analytical evaluations on case studies representative of the urban stock. To this aim, out-of-plane mechanisms have been studied considering different uncertainties and possible variabilities within the facades of the aggregates (Nale et al., 2021). In addition, four case studies representative of the houses of the center of Scarperia have been investigated through equivalent frame modeling and nonlinear static analysis. The outcomes of the different phases have been discussed in terms of fragility curves (Fig. 2-g). Fragility curves define the probability of occurrence of a certain damage level as a function of the intensity measure of the ground motion (Matio and Tsionis, 2015). The latter have been defined according to different methodologies. The probability of the empirical ones was obtained by the damage distributions from the Vulnerability curves. The analytical ones have been derived by computing the threshold values and the relative dispersions (according to the Response Surface Method; Pinto et al., 2004). The comparisons within local and global analysis showed the intrinsic relationships within the two types of seismic response of masonry structures, where the two behaviours express complementary possibilities. The comparisons within the empirical curves and the ones coming from the nonlinear static analysis show relevant variabilities, both referring to the thresholds of the different DLs as to the dispersion’s computation (Chieffo and Formisano 2019). Further analytical studies are certainly encouraged in order to in-depth investigate the variabilities of the curves based on