Geostatistical analysis in Continuous Compaction Control

Most of the modern vibrating compactors can be geared with in-built testing devices able to directly evaluate the stiffness level of the compacted layer down to a decimetric span. Setting a target stiffness value during compaction leads to the so called Automatic Intelligent Compaction, while measuring the stiffness with a constant drum setting produces CCC proofrolling records. Applying geostatistical methods to the analysis of CCC data have the advantage of quantifying spatial variability, which is not possible with the classical statistical analysis. Many studies have demonstrated the importance of the bearing capacity uniformity on foundations to achieve long term pavement performance. Spatial variability analysis performed using geostatistical tools, such as variograms and kriging, could be useful to quantify uniformity for acceptance. Authors demonstrated that non-uniform pavement subgrade, subbase or foundation stiffness increases localized deflections and causes stress concentrations which can lead to accelerated pavement distresses. Univariate statistics of CCC measurements alone do not characterize the spatial variability and specifically do not address the issue of uniformity from a spatial standpoint. A variogram model, in combination with univariate statistics, could be utilized to effectively address the issue of uniformity. This paper focuses on the use of statistical and geostatistical analysis on experimental data obtained from testing an embankment built with C&D materials. The final objective is to study the validity of this approach in Quality Control- Quality Assurance schemes.