Evaluation of Resilient Modulus Prediction Models for Cohesive and Noncohesive Soils

Geomaterials are often in unsaturated condition during their service life and their resilient modulus is influenced by several factors including moisture content, density, void ratio, plasticity, and etc. There are various constitutive models to predict the nonlinear resilient modulus of unbound materials as a function of the aforementioned factors-particularly moisture and stress states based on empirical equations or theoretical unsaturated soil mechanics concepts. In this study, seven existing constitutive models and two predictive models were evaluated using eight different soil databases with different properties including cohesive and non-cohesive soils. The constitutive models were calibrated based on the data at optimum moisture content and maximum dry density. Consecutively, the calibrated models were used to predict the resilient modulus at other moisture or density conditions. The models were compared in terms of their rationality, accuracy of prediction, and applicability to the widest range of soils.