Empirical fragility curves from damage data on RC buildings after the 2009 L’Aquila earthquake

Abstract

The study analyses the data related to a database of 7597 private Reinforced Concrete buildings located in the city and the province of L’Aquila surveyed after the 2009 earthquake. Survey data were collected by the Italian Department of Civil Protection during post-earthquake usability inspections including information on building characteristics, level and extent of damage to structural and non-structural components. For each building, the Peak Ground Acceleration demand has been determined according to data available from the ShakeMap of the event and the georeferenced building location. The analysis of data highlights the key role played by the damage to non-structural components—namely, infills and partitions. Damage Grades according to the European Macroseismic Scale EMS-98 have been derived from damage data to single building components. Two building classes have been defined in the study in order to investigate the influence of number of storeys of buildings on the observed damage. Damage Probability Matrices have been derived for the assumed building classes and bins of Peak Ground Acceleration, and observed trends are analyzed. Different methodologies for estimating fragility functions from data on Damage Grades and Peak Ground Acceleration demand are illustrated, discussed and applied to the database, leading to the derivation of EMS-98-based fragility curves for the defined building classes. Finally, the proposed fragility curves are compared with main empirical fragility curves for RC buildings from literature studies

Empirical Damage and Actual Repair Costs on Masonry Private Buildings After L’aquila Earthquake

“…Furthermore, the relationship between DF and ARC is also reported in terms of an a-dimensional cost ratio obtained as a ratio between the actual repair costs related to the building and the average building demolition and reconstruction cost, Building Repair Cost Ratio (Cr). The unit costs for demolition and reconstruction of private residential buildings damaged by L'Aquila earthquake resulted 1,192.00 €/m 2 , [12,13]. Based on the regression analyses the following function has been determined:…”

“…By contrast, the damage grades D2-D3 and D4-D5 were split into the corresponding EMS-98 damage state depending only on the mean damage detected on VS for the building class in the selected location: the higher the mean damage, the higher the probability of being in the heaviest DS. A correlation between empirical damage derived by AeDES form and the six damage grades as defined by EMS-98 was provided also by [11] and [12]. According to [11] the maximum damage detected on structural components VS, F, and R is accounted for the correlation between empirical damage and the building DS.…”

“…According to [11] the maximum damage detected on structural components VS, F, and R is accounted for the correlation between empirical damage and the building DS. In [12], the damage on IP is taken into account as a key parameter to define the building damage state attained due to the earthquake; the maximum damage detected on VS or IP is accounted for the correlation between empirical damage and the building DS. In spite of the numerous studies revised carried out to analyze the empirical damage, a study to correlate the empirical damage to the cost computed by practitioners to repair the damage suffered by each structural and non-structural component, namely Actual Repair Costs, ARC, is still lacking.…”

A Class-Oriented Large Scale Comparison With Post-Earthquake Damage for Abruzzi Region

“…The definition of homogeneous classes of buildings will be carried out identifying the parameters that greatly influence their seismic fragility. In Del Gaudio et al [29], two classes of buildings have been defined as a function of the number of storeys (Ns): L , Low rise (1 ≤ Ns ≤ 3) buildings, and "MH", Medium-High rise (Ns ≥ 4) buildings.…”

“…Plan ratio is assumed as a uniform random variable between [1-2,5] according to statistics reported in (Del Gaudio et al [1,29]). …”