Investigation of Different Percentages of Recycled Asphalt Pavement in Moisture Susceptibility Performance of Warm Stone Matrix Asphalt Mixture by Adaptive Neuro Fuzzy Inference System (ANFIS)

In recent years, regarding the insufficiency of new material and significance of environmental conservation, using recycled material such as Recycled asphalt pavement (RAP) has increased due to the decreased production of environmental pollutants resulting from the preparation of asphalt mixtures. Also, to decrease the energy consumption rate, new technologies have been developed to produce and perform warm asphalt mixtures. These asphalt mixtures have lower temperatures than conventional asphalt mixtures. In this research, RAP has been used in building stone matrix asphalt (SMA) mixture to carry out the moisture susceptibility test by adding warm asphalt additives such as zycotherm, sasobit and Topcel fibers to prevent bitumen draindown of this type of asphalt mixture. The purpose of this paper is to develop a model based on adaptive neuro-fuzzy inference system (ANFIS) for prediction of moisture susceptibility of stone matrix asphalt mixtures (SMA) containing various percentages of RAP and warm mix additives. Various percentages of RAP containing warm mix additive are the parameters for input layer and the ratio of saturated specimen’s strength to that of dry specimen’s strength is the model output. Results indicated high accuracy of the model with a coefficient of determination (R2) of 1 and 0.982 for training and testing data sets and 0.774 for evaluation data-set, respectively. Also, the warm mixtures containing 50% RAP have shown more suitable behavior.