LAUSR

Abstract Flexural toppling is a common mode of instability of anti-dip bedding rock slopes. However, there is currently no theoretical method available for determining the failure surface associated with this kind of failure. In this work, a limit equilibrium model combined with a genetic algorithm (LEM-GA) method was proposed for predicting the safety factors and failure surfaces of rock slopes subject to flexural toppling failure. A MATLAB program was subsequently developed to implement the proposed method which takes the geometric and mechanical parameters of the slope as inputs and the safety factor and potential failure surface associated with the slope as outputs. Two test models were used to check the validity of the LEM-GA method. The correspondence between the calculated and experimental results is good which demonstrates the feasibility of using the LEM-GA method to predict the stability of this kind of slope. The LEM-GA method is also extended to investigate the effect of non-uniform joint spacing on slope stability. The results indicate that when an average joint spacing is used to assess the stability of a non-uniform slope, the predictions made are conservative and err on the safe side. The LEM-GA method proposed in this work offers a sound basis for evaluating the stability of anti-dip bedding rock slopes and improving their design.