Bayesian updating of seismic fragility curves through experimental tests

Fragility curves, commonly derived using analytical methods, are important ingredients of seismic risk analysis structures in the framework performance-based earthquake engineering. Hence, accurate estimation realistic fragility functions is a decisive *: Bayesian updating considering diffuse prior step reliable assessment. This paper proposes procedure applied to curves reinforced concrete (RC) based on data from experimental tests, namely shaking table tests. The latter performed by progressively increasing intensity input motion same test specimen. In this regard, maximal benefit output sought here, aiming convert n sequential stages single virgin specimen into equivalent tests that specimens. modifying during stage-wise testing damage index coefficient. parametric studies validate objective reveal approach more suitable for simple compared large or complex structures. ATC-58 and Markov Chain Monte Carlo (MCMC) approaches also closely examined compared. proposed RC structure, where incremental dynamic updated results. three state models, HAZUS, homogenized strain-based damages states. work highlights pitfalls limited sample less priors. Besides, MCMC-based shown be robust presence fragilities than approach.