Indirect Estimation of Swelling Pressure of Expansive Soil: GEP versus MEP Modelling

In this article, detailed trials were undertaken to study the variation in genetic parameters order formulate more robust predictive models using gene expression programming (GEP) and multigene (MEP) for computing swelling pressure of expansive soils (Ps-ES). A total 200 datasets with ten input (i.e., clay fraction CF, liquid limit w L , plastic id="M2"> P plasticity index IP, specific gravity Gs, swell percent Sp, sand content, silt maximum dry density ρdmax, optimum water content id="M3"> o mathvariant="normal">p mathvariant="normal">t ) one output variable, i.e., Ps-ES are collected from literature, which comprises 120 internationally publications. The effect contributing has been validated Pearson correlation (r), sensitivity analysis (SA), as well a parametric study. results reveal that GP-based techniques correctly characterize characteristics ES, thus leading reasonable prediction performance; however, MEP model yielded relatively better performance. Also, proposed compared widely used AI (ANN, ANFIS, RF, GB-T, DT, SVM). ANN performed better; it is recommended use GEP due blackbox nature ANN. Other exhibited inferior SA revealed different importance by models, its confirmed moisture significantly affect Ps-ES. changes attributes further corroborated literature. Hence, can be deployed efficiently lessens laborious time-consuming testing.