Skip to main content
2026 ASCE T&DI Transportation Conferences Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.

Pavements 2026

Characteristics of Pavement Materials

POSTER RECEPTION (View list of posters here)

Comparative Evaluation of Predictive Models for Anisotropic Stiffness of Unbound Materials

Tuesday, June 30, 2026
4:45 PM - 6:00 PM East Coast USA Time
Abstract: Integrated into pavement design frameworks, the stiffness of unbound geomaterials is usually modeled with nonlinear predictive models. However, these models often assume isotropic behavior by using the same stiffness formulation for both vertical and horizontal directions, despite being initially based on vertical response characteristics. Previous research has demonstrated the anisotropic nature of unbound materials; however, the effectiveness of existing predictive models in predicting lateral stiffness has not been thoroughly tested. This study assesses the accuracy of five predictive fitting models (Bulk Stress, Uzan, UT-Austin, Universal, and NCHRP 1-28A) using data from Standard Resilient Modulus (SMR) and Advanced True-Triaxial Resilient Modulus (AMR) tests. For SMR and AMR-Z tests, predicted and measured values generally stayed within ±10%, while the lateral directions (AMR-X and AMR-Y) showed larger deviations, ranging from +70% to -50% and +50% to -30%, respectively. Among these, the NCHRP 1-28A formulation, which includes effective stresses and a tensile resistance parameter (k₆), provided the most consistent and logical predictions, meeting the Mechanistic Empirical Pavement Design Guide (MEPDG) precision criterion (R² ≥ 0.90). Conversely, the Bulk Stress, Uzan, and UT-Austin models displayed some physical or mathematical inconsistencies under certain stress conditions, especially for cohesive and stress-softening materials. Overall, the NCHRP 1-28A model performed better in predicting vertical resilient modulus behavior. However, the lower accuracy in lateral stiffness highlights the need for developing a dedicated predictive framework that captures the anisotropic response of unbound materials under multi-axial loading.

Learning Objectives:

Attendees can expect to learn the following from this session:
  • Upon completion, participants will be able to evaluate the performance of five nonlinear predictive models in estimating the resilient modulus of unbound geomaterials under both standard and true-triaxial testing conditions.
  • Upon completion, participants will be able to identify the limitations of existing isotropic models and recognize the need for developing frameworks that capture the anisotropic response of unbound materials.
  • Upon completion, participants will be able to interpret the relationship between predicted and measured resilient modulus values and assess how deviations across stress directions reflect model suitability and physical consistency.