A computer simulation study was conducted to investigate the viability of harmonising rut depth measurements from different automated rut depth measurement profilometers.
A computer program was written which allowed for a standard reference transverse profile to be analysed by any number of profilometer sensors and spacings. The software generated a variety of rut depth statistics.
The software was used to investigate the effect of the number of sensors on the predicted rut depth. It was found that the accuracy of the rut depth was proportional to the number of sensors and that this sampling effect results in an underestimation of 2–4 mm for the profilometers used in NZ.
The pseudo-rut model was found to be inappropriate for predicting rut depth. It did not prove possible to test the wire model due to the shape of the reference profiles.
There are limited differences between the rut depths when referencing perpendicular to the straight-edge as opposed to perpendicular to the elevation datum, so this factor does not need to be considered.
Strong linear relationships were found between the rut depth measurements from all instruments, both with each other and with the reference profiles. It is therefore practicable to develop a standard methodology and transfer functions for rut depths.
The lateral variation of profilometers has a significant impact on the accuracy of rut depth measurements. Some profilometer configurations appear to have inadequate coverage towards the kerb and so may miss important data if the first sensor measures outside the pavement area.
The variation in rut depth which arises from lateral placement can be greater than the change due to pavement deterioration between years, which may account for the problems found when trying to use profilometer rut depth data for monitoring pavement deterioration trends.
Keywords: rut depth, profilometer, straight-edge, pavement condition, pavement monitoring, simulation