PSI - Issue 44

Marco Gaetani d’Aragona et al. / Procedia Structural Integrity 44 (2023) 1760–1767 Marco Gaetani d’Aragona et al./ Structural Integrity Procedia 00 (2022) 000–000

1763

4

§3.1 the image-processing based techniques adopted to derive information regarding building main geometrical features are evidenced. Further, the ground motion bin selected to account for uncertainties in seismic input definition, the derivation of the hazard at the site, and the scaling and rotation of ground motion components are described. Next, after synthetic description of the structural analysis approach (§3.2) in §3.3 and §3.4 the procedure proposed to estimate the direct economic losses is extended within a probabilistic Monte Carlo approach to account for uncertainties in response quantities, quantification of damageable components, the definition of damage states, and unit repair costs.

§3.2

§3.1

SEISMIC HAZARD (SHAKE-MAP)

Non-linear Time History Analysis

§3.3 -§4

GIS DATA

Damage assessment

Loss assessment

Building and site characterization

Structural Analysis

• Building heigth • In-plan dimensions • Orientation

Probabilistic damage and loss analysis

Stick-IT model

Fig. 2. Loss assessment framework

3.1. Building and site characterization For the database of 120 RC buildings considered in this paper, the geo-referenced position was available (fig. 3(a)). Based on the position, information regarding the building geometry is obtained starting from rapid image-based processing techniques. The building in-plan dimensions are obtained by regularizing the building footprint shapefile (Geoportale Nazionale - http://www.pcn.minambiente.it/, fig. 3(b)) of each building by assuming regular planar layout. The building height (i.e., number of stories) is obtained via elaboration of Geographic Information System data such as the Digital Elevation Model and the Digital Surface Model, see fig. 3(c), which are territorial raster discretized into a two-dimensional grid of points. Other parameters required to characterize the Stick-IT model, are the infill panel consistency and opening percentage, which cannot be obtained by using this procedure. For this reason, the infill panel consistency, represented by the infill thickness and shear modulus is assumed to correspond to the mean value considered in the original model Gaetani d’Aragona et al. (2020).

( )

DEM

DSM

(a)

Building position

(b)

Building footprint

(c)

Elevation models

Fig. 3. Building characterization: (a) Location of selected buildings (green dots) and accelerometric stations (red triangles), (b) Building footprint, (c) Digital Elevation Model and Digital Surface Model

Regarding the infill opening percentage, a mean value of 20% is considered for infill opening percentage for every building and assumed constant along the height and equal for both the directions. This represent a strong assunption, since the adoption of different opening percentages along the height may significantly affect the building response and the distribution of lateral deformations (Gaetani d’Aragona et al. 2017, 2019b, 2021b). The story mass is obtained by multiplying the surface area by a unit load of 1.1 tons/m 2 for intermediate stories, and 0.9 tons/m 2 for buildings with