Stability Analysis of Inhomogeneous Slopes in Unsaturated Soils Optimized by a Genetic Algorithm

Abstract

Most theoretical analysis for the assessment of the stability of soil slopes is commonly performed under completely dry or saturated and homogeneous conditions, the effect of suction and soil inhomogeneity are generally ignored in stability assessments. This paper presents an analytical framework to investigate the stability of three-dimensional (3D) inhomogeneous slopes in unsaturated soils under one-dimensional steady flow. Based on the kinematic limit analysis method, a 3D rotational failure mechanism is adopted, and three possible failure mechanisms of soil slopes (e.g., toe, face, and base failure) are considered. A closed-form solution for the factor of safety is derived by the energy balance equation, which takes the effects of the suction stress, effective unit weight, and inhomogeneity of soil simultaneously into account. To improve optimize efficiency, a genetic algorithm (GA), which has the advantages of high efficiency and good accuracy, is applied to search for the minimum of the factor of safety of the slope. This methodology is well validated through comparison with existing solutions and numerical simulation. Parameter analyses are performed to investigate the effects of different parameters on slope stability. The results of the present study indicate that the stability of the slope will be underestimated if the suction stress and the change in effective unit soil weight are not considered. The inhomogeneity of soil can reduce the stability of slopes.