The rapid growth in interest in probabilistic methods applied to geotechnical engineering is relatively recent. Papers by eminent academic/practitioners such as Whitman (2000) and Duncan (2001), have raised the profile of probabilistic concepts, by explaining the methods in familiar terminology. A recent UEF conference held in Graz, Austria in 2002 on "Probabilistics in Geotechnics: Gechnical and Economic Risk Estimation" has brought these concepts to a wider audience, including the insurance industry. The first textbook dedicated to reliability and statistics in geotechnical engineering has recently been published by Baecher and Christian (2003).
Intellectual Merit: This research brings together random Field theory and elasto-plastic finite elements in the "Random Finite Element Methods" (RFEM). This approach has been developed by the PI and co-workers for a range of geotechnical applications involving highly variable soils. The methodology proposed in this research represents one of the very few techniques available for meaningful quantitative analysis of geotechnical boundary value problems. The impact of soil variability on design will be investigated. Different methods of generation of spatially correlated random fields will be examined.
Good statistical characterization of soil properties is crucial for practical application of the modelling tools that will be developed in this research. Digitized site-investigation data, particularly from CPT will be gathered enabling relevant parameters such as the mean, variance and correlation properties to be extracted. In addition to using these site-specific properties in the random field models, it will be possible to develop more robust guidelines for engineers indicating typical ranges of soil variability for use in design.
Broader Impact: This research seeks to promote the use of advanced proba- bilistic tools in geotechnical design by applying them to the classical geotechnical stability problems of bearing capacity and slope stability. These particular application are chosen because of their practical importance and the serious consequences of failure. It is expected that this research will lead to improved understanding and quantification of risk and reliability in geotechnical design. In addition to graduate training, the proposal will also provide research experience for undergraduates under the REU program. The project has potential to attract minorities and women, groups that are traditionally under-represented in engineering. The Colorado School of Mines has higher percentages of minority and women students in engineering as compared with local and national averages. This is in part due to the success and activity of the Minority Engineering and Women in Science, Engineering and Mathematics Programs that will help in recruitment for this project.
