A comprehensive approach to evaluate presence and propagation of uncertainty in asphalt binder mastercurves

Abstract

ABSTRACT Due to various practical issues, researchers have resorted experimental determination of dynamic shear modulus (), phase angle () and conversion into unit response (shear creep compliance, and shear relaxation modulus, ) mastercurves through interconversion process. Despite strict quality control, scatter can be expected in these measured/predicted viscoelastic properties. Thus, it is inevitable to encounter significant uncertainty in unit response mastercurves. This work proposes a comprehensive framework to quantify, propagate and separate uncertainty in the finalized unit response mastercurves. For the demonstration of this uncertainty evaluation framework, a set of nine asphalt samples were taken from the same container, short term aged and tested for its viscoelastic properties. Subsequently, unit response mastercurves were constructed directly (using experimentally determined and values), and through numerical technique using interconversion approach. Further, uncertainty in mastercurves was evaluated using several statistical indicators. The numerical values of these statistical indicators reflected that higher uncertainty existed at extreme time/frequencies when compared to intermediate locations. Subsequently, uncertainty in and mastercurves was separated into epistemic and aleatoric uncertainty. The numerical values of statistical indicators reflected that uncertainty in mastercurves was dependent on testing condition, construction technique, chosen distribution function and sample size.