This study presents a new highly nonlinear model for the prediction of the compression index of fine-grained soils utilizing a robust evolutionary computational technique, called multi expression programming (MEP). The proposed model relates the soil compression index to its liquid limit, plastic limit and void ratio. The best model was selected after developing and controlling several models. The experimental database used for developing the models was established upon several consolidation tests conducted on different soils sampled from different construction sites in Iran. To verify the applicability of the derived model, it was employed to estimate the compression index of portions of test results that were not included in the analysis. Further, the generalization capability of the model was verified via several statistical criteria. The parametric and sensitivity analyses were performed and discussed. The results indicate that the MEP approach accurately characterizes the soil compression index leading to a very good prediction performance. The correlation coefficients between the experimental and predicted soil compression index values are equal to 0.935 and 0.901 for the calibration and testing data sets, respectively. The developed model has a significantly better performance than the existing empirical equations.

Keywords