Critical fatigue strain calculations of flexible pavements in the mechanistic-empirical pavement design method, shortcomings, and improved methodologies

Abstract

Fatigue Performance is considered one of the most effective measures for designing and maintaining flexible pavements. AASHTO proposed a Pavement Mechanistic-Empirical (PME) design framework in which the fatigue strains are calculated by the mechanistic part and then turned to fatigue cracking by the empirical part. However, it fails to compute actual fatigue strains since it utilizes multilayered linear elastic analysis and predefined calculation points. This paper aims to discuss the PME approach's deficiency in calculating fatigue strains and to develop simplified methodologies for addressing this issue. This study presents three solutions by comparing the fatigue life estimated from the multilayered linear elastic and viscoelastic analysis. The first solution modifies Asphalt Institute's model by suggesting adjusting factors for each axle load configuration. The second one proposes new response locations instead of predefined ones. The last solution involves developing an Artificial Neural Network (ANN) model to predict pavement fatigue life more accurately.

KEYWORDS:

• Fatigue performance; pavement mechanistic-empirical design (PME); viscoelastic (VE) analysis
• bottom-up fatigue cracking
• artificial neural network (ANN) model

Disclosure statement

No potential conflict of interest was reported by the author(s).