Optimal performance-based seismic design of structures using approximate performance estimation methods

We propose a new approach for the performance-based seismic design of buildings using a deterministic and a reliability-based structural optimization framework. To overcome the increased computing cost of Incremental Dynamic Analysis (IDA) we adopt an approximate seismic performance estimation tool, known as Static Pushover to IDA (SPO2IDA). The SPO2IDA tool is nested within the framework of a Genetic Algorithm resulting to an efficient seismic design procedure able to consider uncertainty. The Genetic Algorithm steps towards designs of improved performance, locating the most efficient design in terms of the minimum weight of the structure. Reliability-based constraints are considered in terms of the mean annual frequency of preset limit-states not being exceeded. A three-storey steel moment resisting frame is used to demonstrate the design algorithm proposed. The methodology presented leads to efficient real-world building designs within reasonable computing time, directly considering the seismic risk. Athanasia Zacharenaki, Michalis Fragiadakis and Man olis Papadrakakis 2