A deterministic approach of slope stability, which is generally corresponding to the model of a simple non-linear function for slopes, is problematic in that it does not account the versatile characteristics of ground layers in an effective way. To resolve this problem, this study proposes a new way of analyzing slope stability, so-called “genetic algorithm method, ” so as to reflect some particular conditions pertaining to the grounds under concern. Similarities and differences in slope stability that may exist between homogeneous and multiple ground layers are examined in a competitive manner, Overall, though similarities deemed a little bit salient, the algorithm method turned out to be very applicable to estimating the validity of slope stability. Furthermore, an additional effort to consider long-standing sequential and dynamic changes in both the amount of rainfall and the underground water level is made in order to improve the results.