Effect of Stress‐Dependent Base Layer on the Superposition of Flexible Pavement Solutions

Abstract

Flexible pavement structural analysis for design usage must consider (as a minimum) multiple wheel/axle loading configurations, seasonal variations of material layer properties, and the nonlinear behavior of unbound materials. Although these requirements are all easily within the capabilities of three‐dimensional finite element analysis, the required computation times may be impracticably long for routine design. Compromises between analytical rigor (e.g., three‐dimensionality) and analysis features (e.g., multiple wheels, seasonal property variations, material nonlinearity) must be made. One compromise is to retain seasonal property variations and material nonlinearity within an axisymmetric single wheel finite element model and to approximate multiple wheel effects via superposition. Although this superposition of nonlinear solutions is undeniably invalid from a rigorous theoretical viewpoint, the errors may be well within acceptable magnitudes for practical design. The paper investigates this issue by comparing superimposed nonlinear solutions against computationally rigorous three‐dimensional nonlinear solutions and evaluating the discrepancies in key pavement response quantities. The results suggest that the errors from superimposing nonlinear solutions are acceptably small for key pavement response quantities. Moreover, these errors are substantially smaller than those resulting from neglect of nonlinear unbound material behavior, a modeling compromise that is common in pavement structural analysis today.