Optimization of target safety levels based on community-level objectives

Although buildings are designed in accordance with code provisions and local construction practices of the time, earthquakes have caused significant economic losses to communities. This is because of the fact that codes have focused on ensuring life safety of the buildings but not the socioeconomic impact on communities from the damage or loss of functionality of the buildings. This study aims at developing a framework for optimizing the target safety levels of buildings based on community-level objectives. Target safety optimization process requires structural analyses of multiple possible designs of different buildings in a community. This task is computationally demanding if it is done for good accuracy. In lieu of such time-consuming structural analyses, neural networks are used for estimating the responses. The framework is demonstrated with a seismic safety target optimization for a class of mid-rise office buildings with steel moment-resisting frames located in Los Angeles, CA. The obtained optimal target reliability indices are compared with the code recommended values, which are found to be lower than the code recommended values.