Statistical investigation on physical–mechanical properties of base and polymer modified bitumen using Artificial Neural Network

Improving bitumen properties by using polymeric materials such as ethylene butyl acrylate, styrene–ethylene–butylene–styrene, ethylene vinyl acetate and styrene–butadiene–styrene has been reported by the researchers. To avoid practical measuring difficulties, Artificial Neural Network has been employed to predict the properties. The presented study, by employing practical data obtained by the researchers and using MATLAB program, has suggested models to predict physical–mechanical properties of modified bitumen. In the process, an appropriate architecture has been obtained to predict each of desirable properties by using try and error method. Comparing the model results with experimental data showed the high acceptability of the suggested models.

► ANN method was employed to model polymer modified bitumen properties.
► Fraass breaking point, creep stiffness and creep rate were the desirable properties.
► An appropriate architecture has been obtained to predict desirable properties.
► Accuracy and authenticity of simulation process was confirmed.