Impact of changes in fleet composition and axle loading on pavement profile characteristics

Over the past two decades, heavy vehicle (HV) fleet in Australia experienced progressive changes in their composition and in the regulatory framework of their operation. This research project studied the effects of changes in HV fleet composition and axle loading on pavement profile characteristics. Efforts were also made to separately evaluate the affect of higher mass limit (HML) scheme on pavement performance. The sites selected include different loading scenarios, pavement types, and soil-climate compositions to facilitate evaluation of the interaction among factors. Phase one of the study included observational and statistical analysis of vehicular and axle loading parameters and their influences on pavement performance. Phase two involved analysing the changes in longitudinal profile characteristics to evaluate the progression of roughness in two wavebands. The latter include short and medium wavelength roughness (SWMR) and long wavelength roughness (LWR). The major findings from the study are outlined below: There was an increase in HV counts for all sites but with lower growth rate after 2000. The slow growth rate was caused by the increase in take-up of longer vehicles and the reduction in smaller articulated vehicles; Axle load distribution by individual axle groups varied over time but their peaks never crossed the maximum relevant legal General Mass Limit (GML). However, overloading increased after 2000 for most of the sites with tandem and tri-axle groups being the main contributors; Annual traffic loading in terms of Standard Axle Repetitions (SAR) showed an increase over time for all sites. The variation between the different sites and the fluctuation for each site over time could be explained by the variations in axle group counts, AG/HV and their overloading patterns; and LWR progressed at a higher rate than SMWR. The rate was higher in reactive soil areas within problematic climate zones and increased with soil reactivity level. The rate of progression of LWR also increased with increased traffic loading which could be attributed to the effect of dynamic wheel loading. Fleet composition and axle load spectra from WIM data proved to be very useful in explaining the variations in traffic loading over time and subsequent progression of pavement deterioration. Discrete wavelet transformation has proven to be effective in identifying segments with distinct profile changes in certain roughness wavebands that might cause the pavement to fail locally. Such analytical approach would be useful for performance investigation at project level management.