Characterization of Macro-Scale and Meso-Scale Performance of Asphalt Concrete Mixtures Under Compression
MetadataShow full item record
Researchers at Texas A&M University have developed the Pavement Analysis using Nonlinear Damage Approach (PANDA) for predicting the performance of asphalt concrete mixtures. PANDA offers substantial improvements in mechanistic modeling and simulation of pavement performance over other existing approaches. However, in order to facilitate the use of PANDA, there is a need to develop a systematic approach for determining the input parameters of its constitutive models. In this dissertation, a well-designed experimental testing protocol is developed to characterize the resistance of asphalt concrete mixtures to permanent deformation. This approach involves conducting two experimental tests in order to extract the PANDA model parameters: the dynamic modulus test (DMT) and repeated creep-recovery test at various stresses (RCRT-VS). Then, a systematic analytical approach is used to determine the linear viscoelastic, nonlinear viscoelastic, and viscoplastic PANDA model parameters for different types of asphalt mixtures and at different temperatures, air void contents, and aging levels. The analytical method employs DMT data to determine the long-term linear viscoelastic properties and time-temperature shift factors, and it employs the RCRT-VS data to determine the nonlinear viscoelastic and viscoplastic properties. A significant part of this dissertation focuses on the implementation of the global sensitivity analysis (GSA) approach to determine the sensitivity of the asphalt mixture performance to the PANDA’s input parameters. This analysis is performed in order to reduce the output uncertainty to input uncertainty, focus the experimental methods on evaluating the key parameters that influence performance, and simplify the analytical approach to extract significant model parameters from experimental data. The GSA results show that the viscoelastic nonlinearity parameter (g2), viscoplastic hardening function parameters (k1 and k2), and viscoplasticity-relaxation time (1/ Гvp) are the most significant and sensitive parameters. The PANDA constitutive modeling framework is used to efficiently simulate and predict the viscoelastic and viscoplastic responses of asphalt pavements. Two different scales of asphalt mixture performance are investigated: macro-scale (full dense-graded mixture, DGM) and meso-scale (fine aggregate matrix, FAM, and coarse aggregate matrix, CAM). The computational results show that the FAM controls the viscoelastic response of asphalt mixtures, while the CAM properties primarily influence the viscoplastic response of asphalt mixtures.
Awed, Ahmed Mohamed Metwaly (2016). Characterization of Macro-Scale and Meso-Scale Performance of Asphalt Concrete Mixtures Under Compression. Doctoral dissertation, Texas A&M University. Available electronically from