William Scott Professor of Civil Engineering and Applied Mechanics, and Director, Geotechnical Research Centre, McGill University, Montreal, Quebec
Geotechnical project engineerAuxiliary professor, National Boring and Sounding Inc., Group LavalinMcGill University, MontrealMontreal, QuebecQuebec
Pages: 17 Published: Jan 1981
This paper discusses the random nature of shear strength of soils as opposed to the customary deterministic approach, where shear strength is thought of as a property that attains a specific value. In this approach uncertainty resulting from the intrinsic nature of the soil, as well as from the sampling and testing techniques (STT), is identified and supported by actual data.
Implications of the random character of shear strength are discussed insofar as they affect present concepts of soil plasticity. This paper advances the notion that currently used yield criteria are merely theoretical models which fit available data to some extent and could actually be violated.
An analytical model of mapping and prediction of shear strength is introduced, where one, two, or more sets of measurements corresponding to various STT are analyzed and could be combined to enable prediction of the mean, variance, and covariance of shear strength at desired locations.
soils, shear strength, random, modeling, mapping, prediction, design, statistical analysis, sampling, testing
Paper ID: STP28766S