Project:  Evaluation and Characterization of Uncertainty in Geotechnical Parameters

PI(s):  Pedro Arduino, and Steve Kramer

Sponsor:  PEER-NSF

Objective:   The development of performance-based earthquake engineering (PBEE) will require explicit consideration of uncertainties. There are numerous sources of uncertainty in the various components that make up the "systems" for which performance evaluations are required. Among the most significant of these sources of uncertainty are those associated with the geotechnical components of the systems of interest. Several PEER research projects are developing deterministic procedures and tools for prediction of the response of geotechnical components. To maximize the utility of this research with respect to PBEE development, it is important that uncertainties in component behavior be quantified.

The objective of this project is to characterize uncertainty in the geotechnical parameters that most strongly influence the performance of geotechnical components. The results of this research project will allow other PEER researchers to use their deterministic tools to quantify uncertainties in component performance. These uncertainties will then be used, along with uncertainties in input motions and in structural response, to evaluate system performance.

Research Approach:   The research approach will involve identification of geotechnical parameters to which component performance is most sensitive, a literature review for general information on uncertainty and spatial variability of those parameters, a careful review of available subsurface data from the field laboratory and other sites, and generation of appropriate stochastic fields. The stochastic fields will be developed for the field laboratory sites and for generic, hypothetical sites.

The stochastic fields will be provided to PEER researchers to use as input to their computational models of geotechnical components. Because these components exhibit nonlinear response, Monte Carlo analyses using these stochastic fields will illustrate and allow quantification of the uncertainty in component performance due to uncertainty and spatial variability in geotechnical parameters. This component uncertainty will play an important role in estimating the reliability of overall system performance.

For more information send E-Mail to: parduino@u.washington.edu, or kramer@u.washington.edu