Comparison of fatigue failure criterion in flexural fatigue test

• This study combines damage to stiffness to better define asphalt fatigue life.
• The stiffness based fatigue life is conservative.
• The effect of strain amplitude on fatigue is mix specific.

This study compares traditional stiffness and energy based fatigue failure criteria with the fatigue failure criterion based on the viscoelastic continuum damage (VECD) approach. In traditional approach, fatigue failure is defined as the number of cycles at which the stiffness of a material reduces by 50% (Nf50). In energy based approach, fatigue failure is defined by the number of cycles at the maximum energy ratio or Rowe’s maximum stiffness defined by stiffness multiplied by the corresponding number of the cycle (E * N). In VECD approach, fatigue failure is defined by the number of loading cycles at the inflection point of the normalized pseudostiffness (C) versus damage variable (S) curve. It is shown that a correlation exits between traditional criteria and VECD criteria. It is shown that maximum energy ratio or Rowe’s maximum stiffness based fatigue life is higher than the traditional fatigue life (Nf50). This indicates the traditional approach is conservative. A strong correlation of fatigue was observed between the VECD fatigue criterion and energy ratio based fatigue criteria. However, the fatigue life by VECD approach is always less than the fatigue life by energy ratio or Rowe’s maximum stiffness.