GNGTS 2021 - Atti del 39° Convegno Nazionale

351 GNGTS 2021 S essione 2.2 A METHODOLOGY FOR ESTIMATING SEISMIC FRAGILITY CURVES OF URBAN DISTRICTS: FIRST RESULTS ON A CASE STUDY A. Sandoli, B. Calderoni, A. Prota Department of Structures for Engineering and Architecture, University of Naples Federico II, Napoli, Italy Due to high seismic hazard of its national territory and vulnerability of existing buildings, Italy is recognized prone to high seismic risk. Methods for assessing seismic vulnerability at large-scale can represent a useful tool to plane human and economic resources and hence to mitigate the seismic risk. During the last decade, different methodologies for seismic risk assessment have been developed by researchers based on empirical, mechanical (or analytical), expert-judgement or hybrid approaches (Zuccaro et al ., 2015). Basically, each method provides a classification of the buildings in different typological classes and defines the corresponding fragility curves aiming at predicting the seismic vulnerability and the expected damage scenarios after an earthquake of a certain area. In the case of empirical approaches both building’s classification and fragility curves derive from statistical analyses of data on seismic damaged buildings, collected in post-earthquake inspections (Rosti et al ., 2020). In the case of mechanic-based approaches, fragility curves are obtained from statistical elaboration of data derived from numerical structural analyses performed on case study buildings (real or simulated), while in the case of expert-judgment fragility curves come from statistical processing of information provided by a team of experts which estimate feasible damage level and structural behavior for structures with different characteristics. Hybrid methods, instead, combine two of the above-mentioned methodologies (Kappos et al ., 2006). Structural-typological classification of existing buildings Recently, the authors developed a hybrid method to define typological fragility curves for typical Italian Unreinforced Masonry (URM) and Reinforced Concrete (RC) buildings (Sandoli et al . 2018; Sandoli et al., 2021). The method combines expert-judgement with mechanical approach: the first identifies five classes of masonry buildings (indicated with the acronym fromURM-1 to URM-5) and four for RC buildings (indicated with the acronym from RC-1 to RC-4) representative of the Italian building stock, while the second provides all the parameters (e.g., statistical variables) derived from numerical analyses necessary for developing the fragility curves. The buildings were classified as a function of three main parameters: age of construction (with intervals defined by the emanation of seismic Codes), structural typology (e.g., arrangement of vertical walls, floors, wall-to-wall and wall-to-floor connection) and observed seismic damaging . Together, such parameters define the global seismic behavior of the construction, which represents the real discriminant of the buildings’ classification. In Tab. 1 are reported the building classes with the relative ages of construction, whereas more detailed information can be found in Sandoli et al ., 2021. For the sake of brevity, in Fig. 1 are depicted the typological fragility curves for masonry buildings only. Fragility functions have been represented by means of a Log-normal distribution, whose Probability Density Function (PDF) is given by the following formula: A METHODOLOGY FOR ESTIMATING SEISMIC FRAGILITY CURVES OF UR DISTRICTS: FIRST RESULTS ON A CASE STUDY A. Sandoli, B. Calderoni, A Prota Department of Structures for Engineering and Architecture, University of Naples Federico II, Napoli, Due to high seismic hazard of its ational territory and vulnerability of existing buildings, Italy is reco prone to high seismic risk. Methods for assessing seismic vulnerability at larg -sc le can represent a tool to plane human and economic r sou ces and e ce to mitigate the seismic risk. During the last d different methodologies for seismic risk assessment have been developed by researchers based on em mechanical (or analytical), expert-judgement or hybrid appr aches (Zuccaro et al ., 2015). Basicall method provides a classification of th buildings in different typological classes and defin corresponding fragility curves aiming at predicting the seismic vulner bility and the expected d scenarios after an earthquake of a certain area. In the case of empirical approaches both buil classification and fragility curves derive from statistical analyses of data on seismic damaged bui collected in post-earthquake inspections (Rosti et al ., 2020). In the case of mechanic-based appr fragility curves are obtained from statistical elaboration of data derived from numerical structural a performed on case study buildings (real or simulated), while in the case of expert-judgment fragility come from statistical processing of information provided by a team of experts which estimate f damage level and structural behavior for structures with different characteristics. Hybrid methods, i combine two of the above-mentioned methodologies (Kappos et al ., 2006). Structural-typological classification of existing buildings. Recently, the authors developed a method to define typological fragility curves for typical Italian Unreinforced Masonry (UR Reinforced Concrete (RC) buildings (Sandoli et al . 2018; Sandoli et al., 2021). The method combines judgement with mechanical approach: the first identifies five classes of masonry buildings (indicate the acronym from URM-1 to URM-5) and four for RC buildings (indicated with the acronym from R RC-4) representative of the Italian building stock, while the second provides all the parameters statistical variables) derived from numerical analyses necessary for developing the fragility curve buildings were classified as a function of three main parameters: age of construction (with intervals by the emanation of seismic Codes), structural typology (e.g., arrangement of vertical walls, floors, wall and wall-to-floor connection) and observed seismic damaging . Together, such parameters defi global seis ic behavior of the construction, which represents the real discriminant of the bui classification. In Ta . 1 are re rte t e ilding classes with the relative ages of construction, wherea detailed information can be found in Sand li et al ., 2021. For the sake of brevity, in Fig. 1 are depic typological fragility curves for masonry b ildings only. Fragility functions have been r presented by means of a Log-n rmal distribution, whose Probability Function (PDF) is give by the following formula: PDF = 1 IM μ √ 2 π exp [ − 1 2 ( ln IM − μ β ) 2 ] (1) where IM represents the Intensity Measure chosen for an earthquake (in this case the IM=PGA at Ul Limit State),  is the standard deviation and  the average value of the Log-normal distributio Cumulative Density Function (CDF), which define the fragility curves, is obtained by integrating the e (1) where IM represents the Intensity Measure chosen for an earthquake (in this case the IM=PGA at Ultimat Limit State), β is the stand rd deviation and μ the average value of the Log-normal dis ribution. The Cumula ve Density Function (CDF), w ich d fine the fragility curves, is obtained by integrating the eq. (1).