Prediction of the Dynamic Modulus of Hot Mix Asphalt Concrete Using Artificial Neural Network (Ann) Method

Under a wide range of temperature and loading conditions, which the pavement is exposed, the asphaltic mixtures behave as a viscoelastic material. The viscoelastic behavior of the asphaltic concrete can be well characterized by the dynamic modulus. The dynamic modulus is an indication of the resistance against permanent deformation of viscoelastic materials under dynamic loading. The new AASHTO mechanistic-empirical pavement design uses the dynamic modulus as the main mechanical property of the asphaltic mixtures of the pavement. Therefore, determination of the dynamic modulus of asphaltic concrete is important for the pavement design, and it can also be used for flexible pavement analysis. It can be determined by two methods, directly by testing or using a number of regression models, which relate the dynamic modulus to the other physical properties of the mixture. The main limitation of the testing is being tedious and costly, and the main limitation of the regression models are their lower accuracy. Therefore, finding an alternative method to accurately predict the dynamic modulus of the asphaltic concrete with a lower cost and time would be useful. Artificial neural networks are able to be trained, and like the human brain learn the laws and use them for generalization and prediction of new data. This make the ANN method to be a powerful tool to be used for modeling in N dimensions space. This method has been successfully used in many applications for solving engineering problems. In this research, using the ANN. ..