ABSTRACT
Limit equilibrium (LE) slope stability analysis methods are typically combined with global search methods to locate a single critical slip surface, which corresponds to the minimum factor of safety for a topographical model. However, in complicated 3D models there often exist multiple modes of failure with similar factors of safety within the topography. In such cases, it is necessary to consider multiple slip surfaces in the design process rather than just a single surface. To achieve such a goal, this paper proposes two different methods: 1) probabilistic analysis with stochastic response surfaces (SRS), and 2) the Locally Informed Particle Swarm with Radius Filter (LIPS-R) niching algorithm. SRS is a very fast and effective alternative to Monte Carlo or Latin Hypercube sampling of rock and soil material parameters for probabilistic analysis. LIPS-R is a multi-modal optimization (MMO) algorithm based on a niching method called locally informed particle swarm (LIPS). The purpose of MMO algorithms is to output multiple local minima solutions for a given optimization problem. A 3D case study is presented in this paper where both methods are demonstrated to identify multiple failure modes for an open pit mine. Despite the differences between the proposed methodologies, the failure modes identified in both cases are shown to agree.
