Search

Gene expression programming (GEP) models, a robust variant of genetic programming, are developed in this study to correlate resilient modulus with routine properties of subgrade soils and state of stress for pavement design applications. A database used for building the model was developed that contained grain size distribution, Atterberg limits, standard Proctor, unconfined compression, and resilient modulus results for 97 soils from 16 different counties in Oklahoma. Of these, 63 soils (development data set) are used in training, and the remaining 34 soils (evaluation data set) from two different counties are used in evaluation of the models developed. Two different correlations were developed using different combinations of the influencing parameters. The proposed constitutive models relate the resilient modulus of routine subgrade soils to moisture content , dry density , plasticity index (PI), percent passing a No. 200 sieve (), unconfined compressive strength , deviatoric stress , and bulk stress . The results are compared with those from artificial neutral network (ANN) models. Overall, GEP models show good performance and are proven to be better than ANN models. The GEP-based design equations can be readily used for pavement design purposes or may be used as a fast check on solutions developed by more in-depth deterministic analyses. [ABSTRACT FROM AUTHOR]