Key points are not available for this paper at this time.
Identifying critical slip surfaces in three-dimensional limit equilibrium analysis of soil and rock slopes remains a difficult global optimization problem, particularly for slopes with multiple competing failure mechanisms. This paper presents the Intelligent Search algorithm, an enhanced search framework that builds upon spline-based slip surface representation and introduces a probabilistic, multi-stage optimization strategy. The approach combines physics-informed sampling of candidate slip surfaces, unsupervised clustering to organize preliminary solutions into coherent groups, and a multi-swarm particle swarm optimizer that improves spatial coverage and reduces sensitivity to initialization. Optional local surface-altering refinement further improves the most promising surfaces. An open pit case study with several anisotropic materials demonstrates that the method consistently identifies lower factors of safety compared to conventional single-swarm search, while capturing a wider range of plausible failure modes. The results highlight the value of integrating physics-guided modeling with modern, population-based optimization techniques for robust 3D slope stability analysis.
Salvalaggio et al. (Fri,) studied this question.