Skip to main content
ICTD 2023 Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.

Change to Poster Presentation at the Conference, with Full Paper in Conference Proceedings

Machine Learning & AI Applications in Pavements

Session: Meet the Author - Poster Session PS-3

1371695 - Implementing and Optimizing the Artificial Neural Network to Predict Fatigue Resistance of Asphalt Mixtures

Friday, June 16, 2023
8:30 AM - 10:00 AM CST
Room: Governor's Grand Ballroom
Salon ABC
TRACK: Poster Presentation with a Paper in the Proceedings

    Primary Author(s)

  • Wangyu Ma, n/a

    Project Manager
    The Transtec Group
    Mechanicsburg, Pennsylvania, United States

    • Presenter(s)

  • Wangyu Ma, n/a

    Project Manager
    The Transtec Group
    Mechanicsburg, Pennsylvania, United States

    • Co-Author(s)

  • Di Wang, n/a

    Postdoc Researcher
    Aalto University
    Espoo, Uusimaa, Finland

    • Co-Author(s)

  • CC

    Chen Chen, n/a

    Research Engineer
    National Center for Asphalt Technology at Auburn University, United States

    • Co-Author(s)

  • Qinrui Tang, n/a

    Research Associate
    German Aerospace Center, United States

  • Fan Gu, n/a

    Professor
    National Engineering Research Center for Highway Maintenance Technology, School of Traffic and Transportation Engineering, Changsha University of Science & Technology, United States

  • CC

    Chen Chen, n/a

    Research Engineer
    National Center for Asphalt Technology at Auburn University, United States

The fatigue resistance of asphalt mixtures is usually evaluated using various laboratory performance tests, which typically needs hours or even days to complete. As more agencies are moving to Balanced Mix Design (BMD), mixture performance tests will be required to assure the mixture quality during mix design and production, especially the mixture’s fatigue resistance. In this case, several candidate mixes with different mixture components need to be characterized for the fatigue resistance prior to finalizing the mix design, which is time-consuming and non-practical. To reduce the required laboratory tests, a quick prediction method to identify the mixtures with better fatigue resistance without need to complete all the time-consuming tests is required. The Artificial Neural Network (ANN) have been increasingly applied in different fields of pavement engineering, including pavement design, construction, monitoring, and maintenance. This study investigated the implementation of the ANN to predict the fatigue life of asphalt mixtures based on the material types, volumetric properties, and test setup parameters. The Overlay Test was selected for analysis since it has shown potential to better indicate the fatigue and reflective cracking resistance of asphalt mixtures compared to other cyclic tests in recent studies. The structure and parameters of the ANN were established and optimized, with predicting capability tested and evaluated. The predicting accuracy of the ANN was found comparable to the regression analysis when the size of training data was limited; and its accuracy was significantly improved as the size of training data increased, demonstrating a potential for preliminarily predicting the fatigue resistance of asphalt mixture in order to reduce the laboratory testing time and effort.

Learning Objectives:

Attendees can expect to learn the following from this session:
  • Understand the basics of applying ANN in asphalt mixture performance prediction.
  • List the advantages and limitations of ANN in prediction.
  • Identify asphalt mixture candidates with better performance during mix design using this prediction method.