Analytical versus observational fragilities: the case of Pettino (L’Aquila) damage data database

A damage data database of 131 reinforced concrete (RC) buildings, collected after 2009 L’Aquila (Italy) earthquake, is employed for the evaluation of observational fragility curves. The specific interpretation of damage data allowed carrying out fragility curves for slight, moderate, and heavy damage, (i.e., DS1, DS2, and DS3), defined according to EMS 98 macroseismic scale. Observational fragility curves are then employed for the calibration of FAST analytical methodology. FAST method is a spectral based approach, meant for the estimate of fragility curves of infilled RC buildings up to DS3, evaluated, again, according to EMS98. Kullback-Leibler divergence is employed to check the matching between analytical and observational fragilities. FAST input variables can vary in quite large ranges and the calibration provides a valuable suggestion for the application of the method in other cases in which field damage data are not available. Results showed that optimizing values, for the input variables calibrated, are in good agreement with typical values assumed in literature. Analytical results showed a very satisfactory agreement with observational data for DS2 and DS3, while systematical underestimation was found for the case of DS1.