Seismic reliability of steel-concrete composite and RC buildings using vector-valued intensity measure

Abstract

This study presents a novel comparative seismic reliability framework to quantify the performance advantage of steel–concrete composite (CM) structures over conventional Reinforced Concrete (RC) structures using vector-valued Intensity Measures (IMs). In seismic reliability analysis, accurate prediction of structural response critically depends on the appropriate selection of IMs and Damage Measures (DMs). In the proposed framework, the first-mode spectral acceleration, Sa(T1, 5%), and a spectral-shape parameter, Sa(T2)/Sa(T1), are employed as IMs, while the maximum inter-story drift ratio is adopted as the DM. Incremental Dynamic Analysis (IDA) is carried out on representative CM and RC building models using ten ground motion records. Although IDA is a well-established analysis tool, this study introduces a refined post-processing methodology based on an optimal window length (50% of the record set) to generate continuous median IDA surfaces and capacity contours from discrete ground motion data. Drift hazard curves are subsequently developed using a reliability-based framework that combines logistic regression for collapse modeling and linear regression for non-collapse demand estimation. The resulting limit-state exceedance probabilities are evaluated and compared for CM and RC structures. The results demonstrate that composite buildings exhibit significantly lower drift demands and substantially reduced probabilities of exceeding critical limit states compared with RC-framed structures, highlighting the superior seismic reliability and performance of composite construction.