Unbound granular materials are generally used in road pavements as base and subbase layers. The granular materials provide load distribution through aggregate contacts to a level that can help the subgrade to withstand the applied loads. Several research studies have shown that unbound pavement layers exhibit anisotropic properties. Anisotropy is caused by the preferred orientation of aggregates and compaction forces. The result is unbound pavement layers that have higher stiffness in the vertical direction than in the horizontal direction. This behavior is not accounted for in the design and analysis procedures included in the proposed AASHTO 2002 design guide. One of the objectives of this study is to conduct a comparative analysis of flexible pavement response using different models for unbound pavement layers: linear isotropic, nonlinear isotropic, linear anisotropic and nonlinear anisotropic. Pavement response is computed using a finite element program. The computations from nonlinear isotropic and anisotropic models of unbound layers are compared to the AASHO field experimental measurements. The second objective is to analyze the influence of using isotropic and anisotropic properties for the pavement layers on the performance of flexible pavements calculated using the AASHTO 2002 models. Finally, a comprehensive sensitivity analysis of the proposed AASHTO 2002 performance models to the properties of the unbound pavement layers is conducted. The sensitivity analysis includes different types of base materials, base layer thicknesses, hot mix asphalt type and thickness, environmental conditions, and subgrade materials.

Unbound granular materials are generally used in road pavements as base and subbase layers. The granular materials provide load distribution through aggregate contacts to a level that can help the subgrade to withstand the applied loads. Several research studies have shown that unbound pavement layers exhibit anisotropic properties. Anisotropy is caused by the preferred orientation of aggregates and compaction forces. The result is unbound pavement layers that have higher stiffness in the vertical direction than in the horizontal direction. This behavior is not accounted for in the design and analysis procedures included in the proposed AASHTO 2002 design guide. One of the objectives of this study is to conduct a comparative analysis of flexible pavement response using different models for unbound pavement layers: linear isotropic, nonlinear isotropic, linear anisotropic and nonlinear anisotropic. Pavement response is computed using a finite element program. The computations from nonlinear isotropic and anisotropic models of unbound layers are compared to the AASHO field experimental measurements. The second objective is to analyze the influence of using isotropic and anisotropic properties for the pavement layers on the performance of flexible pavements calculated using the AASHTO 2002 models. Finally, a comprehensive sensitivity analysis of the proposed AASHTO 2002 performance models to the properties of the unbound pavement layers is conducted. The sensitivity analysis includes different types of base materials, base layer thicknesses, hot mix asphalt type and thickness, environmental conditions, and subgrade materials.