Probability theory and seismic design

Abstract

Uncertainties always exist in engineering analysis and design. Even when they can be ignored, designers need to account for them by using safety factors. Uncertainty is the difference between what is known and what is reality. Of course, reality is only approximated by designers via tests and experiments. Uncertainty can be aleatory (random) and epistemic (subjective). Aleatory uncertainty refers to the inherent variability of a physical system, which can be modeled as a random variable or a random process, as described below. When data are scarce, probability distributions cannot be obtained, and epistemic uncertainty approaches are used. Epistemic uncertainty is defined as the lack of knowledge or information about some aspect of the system or model. Epistemic uncertainty can be reduced with the collection of more information. Sometimes, physical processes are so complex that, even with additional data, they are modeled as though they are inherently probabilistic. Examples of this include aerodynamic forces and seismic forces. In such cases, probability is used as an organizing principle. A review of reliability under epistemic uncertainty was provided by Kang et al. [1].