Development, Calibration, and Verification of a New Mechanistic-Empirical Reflective Cracking Model for HMA Overlay Thickness Design and Analysis
Academic Article
Overview
Research
Identity
Additional Document Info
Other
View All
Overview
abstract
The purpose of this paper is to present a new mechanistic-empirical (ME) reflective cracking model developed for hot-mix asphalt (HMA) overlay thickness design and analysis. After reviewing existing models, the reflective cracking model based on Paris' law of fracture mechanics was considered as the state of the practice and was selected as the basis for the ME model development in this study. The model consists of stress intensity factor and fracture properties (A and n) as the fundamental input parameters for modeling reflective crack propagation caused by both traffic loadings (bending and shearing) and thermal effects (temperature variations). For practical application, 32 SIF regression equations were developed for HMA overlays with three levels of load transfer efficiencies (10, 50, and 90%) at joints/cracks under various traffic loading spectrums (bending and shearing) based on more than 1.6 million finite element simulations and computations. For the thermal induced reflective cracking, a "hybrid" approach, similar to the Strategic Highway Research Program low temperature cracking model, was proposed. In this hybrid approach, the viscoelastic properties of the HMA mixes were considered through the thermal stress at the "far field" (VE-far), which then ties with the stress intensity factor (Kthermal) determined through regression equations. Also, the required fracture properties (A and n) can easily be determined in the laboratory using the overlay tester that has unique features with respect to specimen size, specimen preparation, and relatively short testing time (about 15 min). Additionally, the proposed reflective cracking model was preliminarily calibrated using three HMA overlay field-case studies, and then verified using California's heavy vehicle simulator test results. Currently, more calibration is underway, but more field performance data are definitely needed for further model calibration and verification. 2010 ASCE.
