A PDEM-enhanced global factor method for reliability-based design and assessment of RC framed buildings against progressive collapse

Progressive collapse analysis of reinforced concrete (RC) structures under accidental loads is computationally demanding as it involves significant material and geometric nonlinearities. The global factor method (GFM) allows estimating the design value of structural resistance using a limited number of nonlinear analyses, while accounting for uncertainties in structural response on the system scale and thereby conducting global safety verification. However, the applicability of the GFM to progressive collapse of RC frames still needs to be investigated. This study presents an improved GFM for progressive collapse design and assessment of RC frame buildings under column loss scenarios. The influence of both aleatory and epistemic uncertainties is incorporated into the global safety factor, thereby preserving GFM's simplicity and ease of use. Moreover, a GFM-Probability Density Evolution Method (PDEM)-based approach is proposed to efficiently calibrate the correction factors of the mean value and coefficient of variation of global structural resistance. To illustrate the proposed method, six planar RC frames are considered for calibrating the correction factors, encompassing twenty-four column loss cases. Lognormal distributions are fitted to both of the resulting correction factors. Different values and selected quantiles for the correction factor of coefficient of variation are adopted to compute global safety factors corresponding to different target reliability levels. Ultimately, progressive collapse safety assessment is performed by comparing the design value of global structural resistance with design loads. Results show that the PDEM-enhanced GFM could be a promising, performance-based, and efficient approach for global-reliability-based progressive collapse design and assessment.