LAUSR

Slope stability analysis usually requires the determination of the critical failure surface. In this paper, a new global optimization algorithm, based on radial movement optimization, is proposed to search the non-circular critical failure surface of a slope. Factors of safety of the slip surfaces are determined using the Spencer method. Nonlinear equations from the Spencer method are solved using the Newton–Raphson method. The data structure of the radial movement algorithm is further optimized to overcome the instability and inaccuracy of the original method. By considering the self-feedback of the particles, the particle swarm can inherit the information of its own particles. The proposed method is validated by examples taken from the literature. The results show that the improved radial movement optimization algorithm can be successfully applied to slope stability analysis, and it has better performance than other global optimization algorithms.