<p>The present study developed a numerical method to estimate resilient modulus from the result of repeated load CBR test. Direct shear test, CBR test, and repeated load CBR test were carried out in the laboratory at various physical states of soil. An artificial neural network model was employed to estimate the elastic modulus of the soil samples based on the results of laboratory tests. The computed elastic modulus was then used to develop a CBR model in a 2D finite element program. A correlation was established to determine the pressure against a given deformation under half-space conditions using the result of the CBR model. The elastic deformations of the CBR sample, measured in the repeated load CBR test, were incorporated into this correlation to compute the pressure under the half-space condition. The computed pressure was finally applied in the Boussinesq equation to compute the resilient modulus of the soil sample. The pressure computed using the resilient modulus from the CBR model was found to agree with that of the repeated load CBR test, with a coefficient of determination (<i>R</i>²) of 1.0. This validates the computation method for determining the resilient modulus from repeated load CBR test.</p>