Fatigue cracking remains a primary mode of distress in asphalt pavements, particularly under high traffic loads. The complex viscoelastic behavior of asphalt rubber (AR) binders presents unique challenges for laboratory evaluation, as conventional fatigue parameters and standard criteria, such as the peak pseudo-strain energy (PSE) from the linear amplitude sweep (LAS) test-often fail to capture their true performance. This study systematically evaluates the fatigue performance of unmodified, polymer-modified, and AR binders, as well as their corresponding mixtures, to identify more reliable characterization methods. Five different LAS failure criteria were assessed under multiple aging and temperature conditions. Traditional fatigue parameters, such as |G*|sinδ from the Superpave classification and its variants, were also considered for comparison. The relationships between binder-level fatigue indexes and mixture performance were explored through rheological testing, and mechanical testing of asphalt mixtures using Brazilian materials. Results show that AR binders and mixtures consistently offer enhanced fatigue resistance, yet the standard LAS–PSE criterion frequently fails to represent this improvement. Instead, alternative LAS-based criteria, especially those involving maximum shear stress and the integrity curve, notably at 25°C, produced a fatigue factor of the binder (FFB) with the strongest correlation to the fatigue factor of the mixture (FFM). These findings support the use of more robust, mechanism-based fatigue metrics in performance specifications for rubber-modified asphalts. The integration of laboratory binder and mixture data provides a solid foundation for selecting and designing sustainable, fatigue-resistant asphalt mixtures for pavement applications.
Enríquez-León et al. (2026) studied this question.