Prediction of unsaturated soils e ective stress parameter using gene expression programming

Unsaturated soil shear strength can be determined using eective stressrelation that depends on the eective stress parameter. several models have been developedin the past few years to estimate this parameter. in this research, the gene expressionprogramming (gep) is used as an articial intelligence (ai) method for developing amodel to predict the eective stress parameter, using ecient parameters. the principaladvantage of the gep approach is its ability to generate powerful prediction equationswithout any prior assumption on the possible form of the functional relationship. theinput terminal set consists of net conning pressure, suction, soil water characteristiccurve (swcc) tting parameter, bubbling pressure, residual and saturated volumetricwater content. the output terminal set has one member, which is the eective stressparameter. an experimental database obtained from the literature is employed to developthe model. comparison of the model prediction with the actual data, as well as otherinvestigators, indicates a very good performance and ability of model. sensitivity andparametric analyses are conducted to verify the results. it is also shown that soil suctionis the most inuential parameter in the eective stress parameter of unsaturated soils.