A comprehensive statistical analysis and review of the dataset was undertaken on the March 2015 LTPP database, including application of transformations on the skewed raw data.
Following a detailed analysis, it was found that the numerical data available to undertake a statistical analysis to identify factors that need to be present for accelerated condition trending was not sufficiently robust. It is not possible to find useful or significant correlations with this data set as it stands
A manual investigation by a person with extensive experience in road engineering and maintenance was undertaken. This investigated engineering explanations for the sites highlighted by the statistical analysis, which involved interpreting site photographs, notes and construction records.
This review was unable to identify any reliable data to show pavements displaying cracking are at a higher risk of failure. It does suggest, however, that the current selection of maintenance treatment type and the quality of maintenance and reconstruction practices may assist in making the level of service worse after maintenance, compared with sterilised sites where maintenance is restricted to emergency repair work only.
The maintenance practice of water cutting also needs to be carefully considered, as this may contribute to a more rapid condition deterioration.
Recommendations are provided on how the dataset could be restructured to provide additional benefit.
Keywords: asset management; long-term pavement performance (LTPP); road maintenance.