Comparative analyses of finite element and limit-equilibrium methods for heavily fractured rock slopes

Abstract

Limit-equilibrium method (LEM) and finite element method (FEM) with shear strength reduction (SSR) technique are the most widely used analysis tools in slope stability assessment. Recently, researchers have reported that both factor of safety (FOS) values and failure surfaces obtained from LEM and FEM are generally in good agreement except in some particular cases. On the other hand, the consistency between two methods has not been adequately discussed for heavily fractured rock mass models by employing Generalized Hoek–Brown Criterion (GHBC). In this study, the FOS values and failure surfaces derived from LEM and FE-SSR based on GHBC were compared concerning static and pseudo-static conditions, various overall slope angles, geological strength index (GSI) values, and various water table levels. In this context, three homogeneous, highly fractured rock slope models with irregular geometry and different slope heights were generated by two-dimensional Slide and Phase2 software. Limit-equilibrium (LE) analyses were performed by Bishop, Fellenius, Morgenstern–Price, and Spencer techniques. The comparisons of global minimum FOS values for 431 cases and the effects of variables on two methods were investigated by statistical analyses. Consequently, it was determined that the difference between the FOS values are statistically significant. However, if the seismic coefficient is higher than 0.1 g, slope angle is higher than 34°, and the slope is assumed to be fully saturated, Morgenstern–Price is the most well-matched technique with FE-SSR than the others. For the same cases, the failure surfaces detected by Fellenius is more similar to the ones detected by FE-SSR.