| Each year, the annual expected losses due to earthquakes increase almost exponentially. The most effective way to arrest this rapid growth rate is to improve the performance of structural and non-structural systems that suffer from earthquakes. New structures can withstand earthquakes better through improved design codes and by giving special attention to detailing and construction techniques. However, improving the performance of existing structures is far more complicated. Given budget limitations, it is essential to identify the most vulnerable structures or group of structures and prioritize their mitigation with the objective of maximizing the reduction in expected losses due to earthquakes. Choosing between different mitigation techniques with different installation and maintenance costs is another challenge. This thesis proposes a new methodology for assessing the cost-effectiveness of different mitigation measures for structural and non-structural systems.;The study investigates different aspects of seismic hazard, seismic fragility, and loss estimation methodologies, and it introduces new representations for each. For seismic hazard, a new algorithm has been developed to generate random samples of seismic activities at the site based on regional seismicity. A model based on non-homogeneous cellular automata has been developed to numerically simulate regional seismicity. The main feature of the proposed seismic fragility representation is using two ground motion parameters, magnitude and epicentral distance of the earthquake event, to model the seismic intensity at the site. Fragility information takes the form of fragility surfaces that give the probability of reaching or exceeding different levels of damage for a given seismic intensity. The concept of lifecycle earthquake potential losses is introduced and employed to compare different mitigation techniques. All proposed methodologies have been implemented in computer programs and are illustrated through numerical examples. |
