Prediction of Resilient Modulus of Stabilized Aggregate Base Subjected To Wet-Dry Cycles Using Adaptive Neuro-Fuzzy Inference System (ANFIS)

One of the most important input parameters for pavement design using empirical mechanistic method is the resilient modulus of various materials. Resilient modulus is commonly determined by dynamic triaxial loading test which is very costly and time-consuming and requires special laboratory facilities.The purpose of this paper is to develop a model based on adaptive Neuro-Fuzzy inference system (ANFIS) for prediction of resilient modulus of stabilized base materials subjected to wet– dry cycles. For this purpose an experimental database consists of 704 records were used. In this study, the number of wet-dry cycles (WDC), the ratio of free lime to SAF (CSAFR), the ratio of maximum dry density (in kN/m3) to the optimum moisture content (in %) (DMR), the confining pressure ( ) and the deviator stress ( ) were considered as ANFIS input parameters. Results showed high accuracy of model with coefficient of determination (R2) of 0.9669 and 0.9625 for training and testing data sets and 0.9655 for overall data set, respectively. In addition, the result of sensitivity analysis showed that DMR and WDC have the most and least effectiveness on resilient modulus of stabilized bases, respectively. In addition, it was concluded that resilient modulus is more dependent on the value of variable compared with , and results of parametric studies also confirm this issue.