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

The study objective was developing an artificial neural network (ANN) model series for indirect tensile strength (ITS) and tensile strength ratio (TSR) prediction for various mixtures in which five input variables were considered, such as asphalt binder content, conditioning duration, anti-striping agents, aggregate source, and asphalt binder sources. Study results indicate that regardless of test conditions, ANN-based models are effective in mixture ITS and TSR value prediction and can be implemented in a spreadsheet easily, thus making application easy. Developed ANN models also can be used for ITS value prediction (or estimation) in mixtures used in other research projects. Results also show that the most important factors in the developed ANN models are asphalt binder source, aggregate source, and asphalt binder content, while conditioning duration is of relative unimportance (i.e., in comparison with other variables, it has less effect on the ITS values). Additionally, input variable sensitivity analysis indicated that ITS value changes are significant as the most important independent variable changes.

  • Availability:
  • Authors:
    • Gandhi, Tejash S
    • Xiao, Feipeng
    • Amirkhanian, Serji N
  • Publication Date: 2009-1


  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01140863
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Sep 21 2009 12:30PM