Statistical Analysis of the Sources of Flexible Pavement Variability

This project identifies, quantifies, and statistically characterizes the components of asphalt pavements that may lead to spatial variability of pavement performance. Researchers analyzed extensive data sets at the Minnesota Road Research Project (Mn/ROAD) project, including Falling Weight Deflectometer (FWD) data, Ground Penetrating Radar (GPR) data, core analysis, and loading history data on 22 asphalt test sections.

On the average, the as-built thickness is slightly thicker than the design. However, even at this carefully constructed site, there is a significant chance that the as-built pavement thickness can be in error by more than 2.5 cm ( one inch). Spatial statistics indicate that overly thin zones will occur in areas on the order of 101s of square meters, on the average.

While the more heavily traveled lane shows a consistent increase in the measures of pavement distress--for example, rutting depths--a quantitative and predictive conclusion is impossible from currently available data. At the time of this study, the pavement structures have not degraded sufficiently to clearly correlate long-term pavement performance with the available measures of pavement stiffness.