The primary objective of this research consisted of incorporating laboratory-determined viscoelastic material properties into a three-dimensional finite element model to accurately simulate the behavior of a perpetual pavement structure subjected to vehicular loading at different pavement temperatures and vehicular speeds. With this finite element model, statistical models that were based on Falling Weight Deflectometer testing were developed to predict the structural response of a perpetual pavement.
In this research, the dynamic modulus test was chosen to determine viscoelastic properties of hot-mix-asphalt materials in the laboratory. A 5-term Prony series was used to describe the viscoelastic behavior of hot-mix-asphalt materials. Resilient modulus tests were performed to measure resilient moduli of hot-mix-asphalt mixtures and subgrade soils. All these laboratory-determined material properties were inputted into the developed viscoelastic finite element model to predict pavement response.
The developed viscoelastic finite element model was validated by and calibrated to field-measured pavement responses collected at the U.S. 30 perpetual pavement constructed in Wayne County, Ohio. The results demonstrated that the developed viscoelastic finite element model can predict pavement responses accurately.
Parametric studies revealed that the developed viscoelastic finite element model performed better in pavement thickness design compared with perpetual-pavement-design-oriented software PerRoad which underestimated pavement responses. Layer modulus variation did not affect pavement response significantly. The ratio maximum-tensile-strain/load was independent of the axle load. The ratio maximum-tensile-strain/speed increased with decreasing in vehicular speeds. A nomograph was developed to correlate the maximum tensile strain to the pavement temperature depending on the thickness of the ODOT302 layer and the aggregate base.
Finally, the developed finite element model was tailored to work for Falling Weight Deflectometer tests. Statistical models were developed to estimate pavement response using the Falling Weight Deflectometer upon the completion of a perpetual pavement. These models are important in practice to assess pavement quality using the Falling Weight Deflectometer.
Outcomes of this research are significantly important to improve the accuracy of current design and analysis methods which are widely used in predicting pavement responses and to provide practical guidelines for perpetual pavement design and analysis.