Seismic capacity and demand assessment in bridge-specific fragility analysis

During the last three decades, a large number of analytical methodologies have been developed for the seismic assessment of bridges and the derivation of bridge fragility curves. However, recently, the research focus has shifted on the derivation of bridge-specific fragility curves, recognizing the effect of different geometry, structural system, component and soil properties, on the seismic assessment results. In this context, a new, component-based methodology for the derivation of bridge-specific fragility curves has been proposed by the authors, with a view to overcoming the inherent difficulties in assessing all bridges of a road network and the drawbacks of existing methodologies, which propose usage of the same group of fragility curves for bridges classified within the same category. Therefore, the main issue addressed in this paper is to analytically determine capacity and demand in bridge-specific fragility analysis in the frame of the methodology previously proposed by the authors, aiming at application to realistically sized bridge stocks. Capacity (resistance) is calculated individually for each critical component with the aid of inelastic pushover analysis, quantifying damage in displacement terms based on the capacity curve and the correlation of global to local damage. Bridge piers, abutments and bearings are considered as the (seismically) critical components of a bridge system; piers of different types and characteristics, and different types of abutments and bearings are analysed and included in a database that provides case-specific limit state thresholds of component capacity. The effect of component parameters on limit state thresholds is therefore assessed, highlighting the differences according to the limit state and component considered. Demand is also calculated for each component using either inelastic response-history or elastic dynamic analysis, depending on the application scale (single bridge vs bridge stock), while uncertainties in both capacity and demand are quantified. Case-specific capacity and demand estimations using the proposed methodology are given here for three bridges having different structural systems, and the derived bridge-specific fragility curves are compared with fragility curves for all bridges classified within the same category, to assess the degree of over- or under- estimation of the probability of damage when generic bridge classes are used.