Estimating slope stability is an engineering problem that involves several parameters. The interactions between the factors affecting slope instability are complex and multifactorial, so it is often difficult to describe slope stability mathematically. This work proposes the use of a simple genetic algorithm without mutation as a heuristic search method for finding a slope stability model.A simple genetic algorithm proves to be superior to simple optimization routines, which can find false minima, and to "brute force" approaches, which require the analysis of a very large number of possibilities.The proposed method presents an effective procedure and a convergence at a number of generations lower than that of a simple genetic algorithm; the results being of the right combination of the type of initial population, of the selection, of the method of crossing neglecting the mutation because being a rare phenomenon in the nature and assigning to the child population the different characteristics of the parents, which is in our opinion an attempt to force the solution to converge.