Application of the partial healing model on laboratory fatigue results of asphalt mixture

• Specimen size significantly influence the inhomogeneous fatigue test results.
• The PH model is able to describe the fatigue process for a unit volume.
• Based on the model parameters δγ1 and δγ2, the endurance limit can be predicted.
• By the weighing procedure, the 4PB results are predicted by the UT/C results.

Fatigue characterization of an asphalt mixture is commonly estimated by laboratory fatigue tests. Based on the classical fatigue analysis, fatigue lives obtained from different test methods are not comparable even though the test conditions are the same. One of the main reasons causing the difference in fatigue results is the difference in stress–strain distribution of the different specimens. The stiffness measured from the inhomogeneous test is not a material property but a specimen property, which depends on the geometry and dimension of the specimen. With regard to the homogenous tests, the stress–strain field within the specimen is uniform in theory. The measured stiffness corresponds to a material property and is not influenced by the specimen dimension. The objective of this research is to find a correlation between the different fatigue test results by means of the partial healing (PH) model.In this paper, a homogenous test, the uniaxial tension and compression (UT/C) fatigue test, and a inhomogeneous test, the four-point bending (4PB) fatigue test were conducted. For each type of test, specimens with different sizes were tested to explore the size effect on the fatigue results. It is found that for the standard specimen size, the model parameters deduced from the UT/C and 4PB are comparable. It indicates that the PH model is able to describe the fatigue process in the 4PB test based on the UT/C fatigue results.