Influence of inelastic seismic response modeling on regional loss estimation

Abstract

This study investigates the implications of the use of alternative damage prediction methodologies to correlate ground shaking intensity to socio-economic losses in regional seismic loss analyses. In particular two methodologies are investigated: the over-damped Capacity Spectrum Method (CSM), and a modified version of a Parameterized Fragility Method (PFM) based on dynamic nonlinear analysis to reflect the generalized effects of degradation in hysteretic response. Investigations include comparisons of a variety of structural systems first assuming full hysteresis to minimize the discrepancies in modeling approaches, as well as parametric variations in ground motion intensity and degradation effects. The study culminates with an investigation of the influence of the application of each methodology on regional loss estimates, using a seismic risk analysis performed for the state of South Carolina as a case study, by comparing seismic loss estimates obtained using the over-damped CSM with estimates obtained by application of a modified PFM (MPFM). It is shown that the estimates obtained from the two methodologies are different, and that the over-damped CSM displays increased sensitivity to key analysis parameters relative to the MPFM method both in terms of displacement demands and estimated socio-economic losses.