PSI - Issue 44

Santa Anna Scala et al. / Procedia Structural Integrity 44 (2023) 267–274 Santa Anna Scala et al. / Structural Integrity Procedia 00 (2022) 000–000

268

2

1. Introduction

The recent seismic events have been a unique occasion to collect a huge amount of data about existing Italian building stock, highlighting its general weaknesses. In the last year, the availability of such data allowed the derivation of several empirical studies, especially about masonry buildings (that represent the greater part of the entire Italian portfolio). Most of these studies (Del Gaudio et al., 2019; Rosti et al., 2020; Zucconi and Sorrentino, 2022) focused on the influence of some building’s features, such as the structural type, the number of storeys or also the construction age, on the general damage attitude of the building. In the present study, after defining specific structural types, seismic fragility will be analysed varying the construction age and the number of storeys: thus, fragility curves will be derived based a very detailed building taxonomy. 2. Empirical data deriving from L’Aquila 2009 post-earthquake survey In this section, first a briefly description of the original data deriving from post-earthquake survey will be provided, then the study database will be defined addressing the issue of the completeness of the data (Rossetto et al., 2013). Lastly, the analysis of the main structural-geometric features (such as the quality and/or the texture of the vertical walls, the type of the horizontal structures, the presence of retrofit interventions, …) of considered masonry buildings portfolio will be shown, leading to the definition of adopted building taxonomy. 2.1. Original data deriving from the survey The original database is composed by 74.049 buildings subjected to reconnaissance field trips in the aftermath of the 6 th of April 2009 earthquake by groups of technicians coordinated by the Italian Department of Civil Protection, by means of the AeDES survey forms (Baggio et al., 2007). The aims of the latter are the identification ( Section 1 of the form) and the description ( Section 2 of the form) of the building, specifying the geometry (i.e., the number of storeys, the inter-storey height, the storey surface area, …), the use (residential, commercial, productive, …), the construction age of the inspected buildings. The Section 3 of the form handles with the building’s structural typology: vertical and horizontal structures, presence of tie rods and/or tie beams, structural regularity, type of roof. The damage description of the structural components ( Section 4 of the form) was based on (3+1) levels scale, going from null damage (D0) to very high damage (D4-D5), and on 3 damage extents. A detailed explanation of the database, in terms of constructions features was presented elsewhere (Dolce et al., 2019; Del Gaudio et al., 2021 e Scala et al., 2022). In this section, only the information relevant for developing the model proposed hereinafter are briefly recalled. 2.2. Revised data to define the study database Out of the original 74.049 inspected buildings collected after the earthquake, only the 57% has been considered in the present study, focusing only on masonry structures with residential use. A major part of those is sited in municipalities near to the epicenter (i.e., those subjected to a macro-seismic intensity value equal or greater than VI), where building-by-building surveys were done (Del Gaudio et al., 2019). Conversely, in the area further from the epicenter, the inspection was performed only if required by the building’s owners, thus likely only in case of damaged building. Such approach could cause an overestimation of the damage, because a great part of un-damaged buildings is systematically neglected from the collected database (Rossetto et al., 2013). To overcome this issue, a two-step mixed approach was adopted: • discarding from the original DB all buildings located in municipalities not completely surveyed (i.e., those with the percentage of inspected buildings less than the 91% of the total number obtained from census data); • adding to the original DB un-damaged buildings sited in not inspected and slightly inspected municipalities (i.e., those with the percentage of inspected buildings less than the 10%), using data deriving from Italian census. Such integration has been limited to the only municipalities seismically classified for the first time as seismic in the 1915, as for a major part of the inspected ones. Such data will be considered in the fragility curves derivation to prevent the