NEHRP Clearinghouse
 Title
 A Probabilistic Model for Seismic Slope Stability Analysis.
 File

PB301369.pdf
 Author(s)
 AGrivas, D.; Howland, J.; Tolcser, P.
 Source

National Science Foundation, Washington, DC. Engineering and Applied Science.,
June 1979,
95 p.
 Identifying Number(s)
 CE785
 Abstract
 A model for probabilistic stability analysis of earth slopes under earthquake loading is presented. Significant uncertainties associated with conventional pseudostatic methods of seismic stability analysis are rcognized and probabilistic tools are introduced for their description and amelioration. The proposed method of analysis accounts for: the variability of material strength parameters; the uncertainty in the exact location of potential failure surfaces; and the uncertainty in the value of the maximum slope acceleraton during an earthquake. The soil material comprised in the slope is assumed to be probabilistically homogeneous with strength parameters being identically distributed random variables with given statistical values. Potential failure surfaces are considered to have an exponential shape defined with the aid of three random variables. Slope safety is measured in terms of its probability of failure. The seismic load is introduced into the analysis through the maximum horizontal acceleration experienced by the slope during an earthquake. Two different attenuation relationships are employed to determine the maximum horizontal ground acceleration and the corresponding results are compared and discussed.
 Keywords
 Probability theory; Probability density functions; Mathematical models; Seismic risk; Earthquake engineering; Earthwork; Soil properties; Random variables; Slopes; Earthquakes