Seismic Loss Analysis of a Modern Rc Building Accounting for Uncertainty of Infill Strut Modeling Parameters

Abstract

Probabilistic seismic risk analysis is affected by several sources of uncertainty. Investigating the influence of uncertainties on loss analysis results is important to understand their impact on computed risk. Even though the topic has been the focus of many previous studies, however, only a few studies focused on the effects of modeling uncertainty of often-neglected non-structural elements, such as masonry infill walls in reinforced concrete (RC) moment frame buildings. This paper explores the effect of uncertainty related to modeling parameters of masonry infill walls on the seismic loss estimates for a modern code designed infilled RC frame building. The unreinforced masonry (URM) infill walls are modeled using equivalent strut models and probability density functions (PDFs) are assigned to selected infill strut parameters, based on existing literature on analytical and experimental studies. Using a latin hypercube sampling (LHS) method to sample the PDFs, a set of two-dimensional models are generated and are subjected to nonlinear response history analyses (NRHAs) at increasing seismic intensities. The building seismic performance in terms of damage and direct losses is evaluated for two performance models, the first one considering only the median values of infill strut parameters, the second one considering the uncertainties of such parameters. For both performance models, element fragility characterization is completed by state-of-the-art fragility and consequence functions for clay brick infill walls. Results are presented for an intensity-based risk analysis as well as for complete life-cycle analysis, which outputs expected annualized losses.