Predicting compaction of cohesionless soils using ANN

Compaction of soils is aimed at improving their mechanical properties to fulfil the requirements of earthwork projects. In this paper an artificial neural network (ANN) model was developed to predict the two main compaction parameters: the maximum dry unit weight (γ dry max ) and the optimum moisture content (w omc ). The study was performed based on the results of modified Proctor tests (ASTM D 1557). Based on the ANN model, empirical equations were developed to predict the compaction characteristics of graded cohesionless soils. The predicted values using the ANN model and the empirical equations were compared with a set of laboratory measurements. A parametric study on the developed equations was performed to present the control parameters that set the values of γ dry max and w omc .