Artificial Neural Network Model to Predict the Fatigue Endurance Limit for Asphalt Concrete Pavement

Artificial Neural Networks (ANN) are the highly interconnected structures which have strong computational and pattern recognition abilities utilizing simple processing units (artificial neurons) that have the ability to carry out multiple parallel computations. Fatigue is one of the major distresses occurring in asphalt concrete pavement caused by repeated traffic loading that results in escalated structural damage with the formation of cracks. These cracks allow the moisture to seep inside the pavement layer that results into potholes. Potholes and cracks results in damage of vehicles (tires, suspension, steering and body), increases the fuel consumption, increases vehicle delay cost and maintenance cost while lowering the quality of ride. Hence, this paper puts forward a model to predict endurance limit strain values by using ANN in MATLAB along with a standalone equation for predicting endurance limit strain value by using uniaxial tension-compression fatigue test results conducted under NCHRP Project 9-44 A.