#### Date of Award

Summer 1992

#### Document Type

Dissertation - Restricted

#### Degree Name

Doctor of Philosophy (PhD)

#### Department

Civil, Construction, and Environmental Engineering

#### First Advisor

Claxton, Richard

#### Second Advisor

Matar, Joseph

#### Third Advisor

Wenzel, Thomas H.

#### Abstract

The purpose of this study is to develop a probabilistic approach for the stability analysis of earthen slopes. Typically, the safety of a slope is assessed in terms of a factor of safety. This factor gives an indication of the ratio of resisting forces developed in the soil mass to driving forces which arise due to gravity. These forces are calculated assuming limit-equilibrium conditions. This simplifies the analysis such that only peak shear strength parameters are required. For soils, peak shear strength is a function of two parameters, the cohesion c and the internal angle of friction "formula". The unit weight of soils 'Y is involved in calculating the driving force. These parameters are represented by single values chosen from scattered test results. Hence, the factor of safety which is a function of these parameters is an uncertain quantity. A procedure will be developed in which uncertainty is accommodated by treating the parameters as random variables. As a result, the factor of safety is also considered a random variable. The measure of the safety of a slope is now expressed in terms of a " probability of failure ". Two actual slope failures are investigated to validate the proposed procedure. These two cases are: 1. a naturally occurring land-slide at Selset, Yorkshire, U.K. 2. a slope failure at the Congress street open-cut, Chicago, IL. Both cases are well documented and have been studied by numerous investigators. The first case involves a long-term stability problem, while the second case is a short-term stability problem. Conclusions and recommendations are presented for the implementation and limitations of the proposed procedures. Areas of further research are also presented.