Probabilistic comparative investigation on introduced performance-based seismic design and assessment criteria

Despite all valuable progresses resulted by modern seismic design approaches from reducing fatalities to prevent catastrophic failures, still higher demands are imposed by society. Therefore, another generation of methodologies so-called Performance-Based Seismic Design (PBSD) was introduced. Satisfying all objectives of this method requires developing precise predicting tools and introducing reliable thresholds. These criteria not only should represent structural/non-structural damage states at desired hazard level, but also should strongly correlate with economic aspects, residence safety and functionality. In this regard, a variety of recommendations were proposed in the literature which can be classified into two general groups, i.e. at section/component level (local) or at story/building level (global). In this article, the priority of these two groups regarding each other, their relation and reliability of proposed recommendations are investigated. In this regard, diversity of outcomes related to the employed analysis method, type of the subjected ground motion record, building's height and desired performance level is reported. On the other hand, key parameters such as the influence of allowed percentage of elements to surpass from local limits on maximum experienced inter-story drift and amount of reserved capacity at components when global criteria controls the performance are investigated. Later, analytical approaches are followed to extract global limits and compare with those from the literature. Finally, the probability of exceeding local threshold at any desired inter-story drift/performance level and their safety indices is studied. In this regard, previously obtained outcomes and conventional safety levels are used to evaluate reliability of presented conclusions. Considering all, it is concluded that the global criteria controls performance of the building at all levels; but using local thresholds on higher levels may lead to less conservative designs. Therefore, it seems essential for global thresholds to be revised in development of PBSD to make it more compatible with its objectives.