Current ICT Project
Investigators:
  Prof. Erol Tutumluer (Primary), Prof. Marshal R. Thompson
Project Title:
  Nondestructive Pavement Evaluation Using ILLI-PAVE Based Artificial Neural Network Models
Start Date: 8/1/2004
End Date: 8/31/2008
Summary of Project:
  Evaluating structural condition of existing, in-service pavements is a part of the routine maintenance and rehabilitation activities undertaken at IDOT. In the field, the pavement deflection profiles gathered from the FWD test data are typically used to evaluate pavement structural conditions. This kind of evaluation requires the use of backcalculation type structural analysis to determine pavement layer stiffnesses and as a result estimate pavement remaining life. The recent use of artificial neural network models trained with ILLI-PAVE finite element solutions has proved to give much better results then the statistical algorithms currently in use.
Objective of Project:
  The objective is to develop Artificial Neural Network (ANN) models based on the ILLI-PAVE finite element solutions as a pavement evaluation toolbox for rapidly and more accurately back-calculating pavement layer properties and predicting critical responses of in-service pavements in real time from the measured FWD deflection data.
Expected Outcome:
  Currently, IDOT engineers use ILLI-PAVE based algorithms in FWD backcalculation analysis. Recent work at the University of Illinois showed that properly trained ANN models are capable of backcalculating the pavement layer moduli and predicting the maximum stresses and strains with very low average absolute errors of those obtained directly from ILLI-PAVE analyses. These error magnitudes are much smaller than the statistically formulated ILLI-PAVE algorithms currently in use by IDOT. Once validated, ANN models will provide more accurate and rapid (real-time) analyses of the collected FWD deflection data. The developed user-friendly software and the toolbox will be very helpful for assessing pavement condition, estimating pavement remaining life, and evaluating pavement rehabilitation alternatives by validating IDOTs mechanistic pavement analysis and design concepts with special emphases on extended life HMA designs.
Click here to check the status of this project.
Last Updated: 11/18/2009