Algorithms for increased efficiency of finite element slope stability analysis

Abstract

An algorithm was designed to improve the efficiency of a finite element slope stability program. The finite element software currently uses a bisectional algorithm to arrive at a factor of safety. However, several inefficiencies have been identified in the bisectional algorithm as it is applied to slope stability analysis. Among them are the high computational costs of failing to converge. Several algorithms were proposed to address these inefficiencies, and through optimization and verification, the most efficient of the proposed algorithms has been selected. The Farnsworth-Griffiths Adaptive (FG-Adaptive) algorithm adjusts the size of step it takes in the strength reduction factor based on the number of iterations it takes for the model to converge. This leads to fewer SRF trials and ultimately to faster solutions.