2021 journal article

Development of healing model and simplified characterization test procedure for asphalt concrete

CONSTRUCTION AND BUILDING MATERIALS, 271.

By: M. Ashouri, Y. Wang n, Y. Choi* & Y. Kim n

co-author countries: Korea (Republic of) 🇰🇷 United States of America 🇺🇸
author keywords: Asphalt; Fatigue; Healing; Viscoelastic continuum damage model; Mechanistic model
Source: Web Of Science
Added: February 15, 2021

Fatigue cracking is one of the major distresses in asphalt pavement. Numerous and significant efforts have been undertaken to predict the fatigue life of pavements. One of the mechanisms that affects fatigue life is healing, and thus, including healing in fatigue performance prediction models is necessary. For this study, twelve healing tests at three temperatures and four rest periods were conducted to evaluate the healing characteristics of asphalt materials. The percentage of healing (%Hs) used in this study is defined as the ratio of the internal state variable (S) before the rest period to the incremental internal state variable due to the rest period. A %Hs mastercurve was constructed by applying the time–temperature superposition (t-TS) principle to the %Hs versus rest period curves and thereby proved that the t-TS principle works for healing. The %Hs mastercurves at different damage levels were shifted vertically to construct one reference %Hs mastercurve. The amount of vertical shifting is referred to as the vertical healing shift factor. The reference %Hs mastercurve with the vertical healing shift factor and t-TS principle led to the proposed shift healing model as a function of rest period, temperature, and damage level. A test protocol to calibrate the model also is suggested in this work. The protocol requires only three specimens, and thus, testing can be completed within a day. The proposed shift healing model characterized by the suggested protocol test was implemented successfully in twelve healing tests to predict the healing behavior of the test specimens, which indicates the prediction capability of the suggested model and test protocol.