Prediction of geogrid-reinforced flexible pavement performance using artificial neural network approach

Fan Gu*, Xue Luo, Yuqing Zhang, Yu Chen, Rong Luo, Robert L. Lytton

*Corresponding author for this work

Research output: Contribution to journalArticle

Abstract

This study aimed to develop a methodology to incorporate geogrid material into the Pavement ME Design software for predicting the geogrid-reinforced flexible pavement performance. A large database of pavement responses and corresponding material and structure properties were generated based on numerous runs of the developed geogrid-reinforced and unreinforced pavement models. The artificial neural network (ANN) models were developed from the generated database to predict the geogrid-reinforced pavement responses. The developed ANN models were sensitive to the change of base and subgrade moduli, and the variation of geogrid sheet stiffness and geogrid location. The ANN model-predicted geogrid-reinforced pavement responses were then used to determine the modified material properties due to geogrid reinforcement. The modified material properties were finally input into the Pavement ME Design software to predict geogrid-reinforced pavement performance. The ANN approach was rapid and efficient to predict geogrid-reinforced pavement performance, which was compatible with the Pavement ME Design software.

Original languageEnglish
JournalRoad Materials and Pavement Design
Volumein press
Early online date17 Mar 2017
DOIs
Publication statusE-pub ahead of print - 17 Mar 2017

Bibliographical note

This is an Accepted Manuscript of an article published by Taylor & Francis in Road Materials and Pavement Design on 17/3/17, available online: http://www.tandfonline.com/10.1080/14680629.2017.1302357

Keywords

  • artificial neural network
  • finite element model
  • geogrid-reinforced flexible pavement
  • pavement ME design

Fingerprint Dive into the research topics of 'Prediction of geogrid-reinforced flexible pavement performance using artificial neural network approach'. Together they form a unique fingerprint.

  • Cite this