A deep learning approach to predict Hamburg rutting curve

The Hamburg wheel tracking test (HWTT) is a widely used testing procedure designed to accelerate and simulate the rutting phenomena in laboratory. Rut depth, as one of outputs HWTT, dependent on number parameters related mix design conditions. This study introduces new model for predicting depth asphalt mixtures using deep learning technique - convolution neural network (CNN). A database containing 10,000 data points from comprehensive collection HWTT results was develop CNN-based machine prediction model. has been formulated terms known influencing mixture variables such binder high-temperature performance grade, type, aggregate size, gradation, content, total recycling parameters, including temperature passes. can be tool estimate rut when laboratory not feasible or cost-saving, pre-design trials.