ABSTRACT
The search for critical failure surface of a general soil slope is difficult as the objective function of the factor of safety is non-convex and multiple minima exist in general. For this problem, real-valued genetic algorithm is proposed and the constraints of the problem are transformed to the determination of the dynamic upper and lower bounds of the control variables. The critical failure surface can then be located with high precision during reasonable computer time under the present proposal. The method is shown to be superior to simple optimization routines, which can find false minima. An approach to implementing the genetic algorithm is defined and refined.
