Probabilistic Sensitivity Analysis for Seismic Risk Based on Stochastic Sampling and Focusing on Stochastic Ground Motion Modeling

University Of Notre Dame, Notre Dame IN

Investigators

Abstract

Uncertainties in the seismic hazard parameters such the epicentral distance and intensity of future earthquake events as well as in the stochastic models used to describe the corresponding ground motion time-histories have significant impact on the calculated seismic risk for structural systems. The objective of this research is to develop a computationally efficient new framework to quantify the importance of such risk factors on the structural system performance and the design ground motion selection. To accomplish this goal, a probabilistic simulation-based methodology will be formulated to calculate the seismic risk for structural systems. Numerically efficient approaches based on stochastic sampling and recent advances in computer and computational science will be developed for sensitivity analysis to quantify the influence of the uncertain model characteristics on seismic risk. These approaches will be applied to a variety of structural systems with diverse performance definitions. The project will provide a methodology for efficient probabilistic sensitivity analysis and for quantifying the importance of various risk factors in the seismic structural design practice. The research will significantly advance our understanding of the impact of various seismic-risk factors and is expected to influence the future earthquake risk mitigation strategies. This research effort will educate the future generation of researchers and professionals about the potentials of simulation-based engineering science for handling uncertainties. This study will be used as a resource to improve the undergraduate curriculum by offering a new elective course in seismic design. The project will provide advanced training to graduate students and every effort will be made to recruit these students from the underrepresented groups by leveraging the success of existing programs at Notre Dame. Undergraduate researchers will also be involved in the project to develop web-based tools for wide and effective dissemination of the work.