The RILEM stiffness testing recommendation shows that the determination of the one-dimensional linear viscoelastic properties of bituminous mixtures is possible with good accuracy for a wide range of test methods and sample sizes, when sinusoidal cyclic tests are considered. Nevertheless,  it is a delicate measurement and some conditions should be respected. The most important sources of errors are underlined in this paper. They are not listed again but have very different origins:

• quality of the measurement devices,
• calibration procedure,
• specimen preparation, preconditioning and fitting,
• rheological properties of the material (linearity, dissipation,..).
  

In addition it appears that the temperature-time (or frequency) superposition principle holds with good accuracy for pure bitumen asphalt mixes. The master curve can be determined with Arrhenius or WLF equations.

Some aspects remains rather unknown and need further investigations, such as:

• multiaxial stress and strain distribution in the samples,
• theoretical modelling (non linearity,..),
• anisotropy, (Young and shear moduli [E* and G*] relationship,..).