Permanent deformation, manifested as rutting, remains one of the most critical threats to the structural integrity and functional performance of flexible pavements. The Mechanistic–Empirical Pavement Design Guide (MEPDG) includes rutting models that are highly sensitive to the dynamic modulus (E*) of asphalt mixtures—a parameter that can be determined experimentally or predicted by analytical models. In this study, the influence of E* prediction error on rutting estimation is systematically evaluated by comparing laboratory-measured E* values with those predicted by two models: NCHRP 1-37A and a locally calibrated model. The dynamic pavement behavior and rut depth predictions were determined using the finite layer program 3D-Move under standard traffic loads. Comparative analysis revealed that the NCHRP 1-37A model tends to underestimate E*, leading to significant overestimation of vertical strains and accumulated permanent deformation. In contrast, the locally calibrated model provided predictions that closely matched the laboratory measurements, resulting in minimal deviation in rut depth estimates. The results highlight the importance of local calibration and model selection to improve the reliability of mechanistic–empirical pavement predictions, enabling smarter pavement performance evaluation and supporting more sustainable pavement management practices, especially when laboratory testing is not feasible.
Georgouli et al. (Thu,) studied this question.