Assessment of fragility models based on the Sept 19th, 2017 earthquake observed damage

Seismic vulnerability evaluations of buildings and critical infrastructure in densely populated cities strongly rely on fragility curves to estimate the probability of reaching or exceeding a given state of damage. Currently, however, the predicting capabilities of these models is limited. This paper presents a critical study of the accuracy of fragility curves for buildings, urban bridges, tunnels, overpasses, main roads, avenues and metro stations, aiming at establishing its advantages and shortcomings, when using them in seismic vulnerability and resilience assessments. Data gathered from one of the most damage areas during the 7.1 Mw September Mexico-Puebla 2017 earthquake was used in this study. Measured peak ground accelerations, PGA, and permanent ground deformations, PGD, along with appropriate fragility curves were used to establish the probability of having minor, moderate, large damage or collapse of buildings and critical infrastructure components. The damage on the study area documented during field reconnaissance was compared to that which would have been predicted using the fragility curves. Although in some cases, there was a good agreement between predicted and observed damage, overall it was clearly identified the need for developing fragility models that account for soil-structure interaction, and specific structural and ground conditions, to ensure a more reliable risk characterization of the City building stock and infrastructure during an earthquake.