Estimating Indirect Tensile Strength of Mixtures Containing Anti-Stripping Agents Using an Artificial Neural Network Approach

Abstract

The objective of this study was to develop a series of artificial neural network (ANN) models to predict the indirect tensile strength (ITS) and tensile strength ratio (TSR) of various mixtures considering five input variables such as asphalt binder source, aggregate source, anti-striping agents (ASA), conditioning duration, and asphalt binder content. The results indicate that ANN-based models are effective in predicting the ITS and TSR values of mixtures regardless of the test conditions and can easily be implemented in a spreadsheet, thus making it easy to apply. In addition, the developed ANN models can be used to predict (or estimate) the ITS values of the mixtures used in other research projects. Furthermore, the results also show that the asphalt binder source, aggregate source, and asphalt binder content are the most important factors in the developed ANN models while the conditioning duration is relatively unimportant (i.e., it has less effect on the ITS values in comparison with other variables). In addition, the sensitivity analysis of input variables indicated that the changes of ITS values are significant as the changes of the most important independent variables.