An efficient method for incorporating modeling uncertainties into collapse fragility of steel structures

Deterioration parameters that are commonly used to simulate nonlinear behavior of steel components were mainly calibrated based on the results from experiments beams. Recently, state knowledge for deterioration columns has been improved by experimental and analytical studies wide-flange columns. These deteriorating characteristics introduced as regression relationships in which associated uncertainties represented coefficient variation (COV). Accounting these estimating collapse fragility curves through incremental dynamic analysis (IDA) simulation-based reliability methods is impractical due large amount required computational effort. In this study, two main goals pursued. The first goal comprehensive evaluation main, interaction, quadratic effects modeling random variables capacity structures. second propose an efficient approach create response surface (RS), combination with Monte Carlo (MC) sampling method will be incorporate into fragility. This efficiency achieved employing screening design techniques reduce a RS also endurance time (ET) alternative less estimate structural responses. order develop reliable probabilistic model, Bayesian model inference applied account created model. proposed procedure performed both IDA ET prototype 5-story frame. Results indicate highly influenced strength beam well ultimate rotation column components. addition, considerable reduction costs provides comparable responses framework.