Prediction of hydraulic conductivity for soil–bentonite mixture

Owing to its low hydraulic conductivity, soil and bentonite mixture is applied as a liner material. However, the experimental determination of hydraulic conductivity, which is controlled by various physical, chemical and mineralogical factors, requires an expensive and time-consuming setup. In the present work, multigene symbolic, genetic programming was used to model functional relationships for hydraulic conductivity. The developed model was able to generalize highly nonlinear variations in data as well as predict system behavior from experimental observations. It was found that the predictions obtained from developed model agree well with experimental observations.