Development of a systematic method for intelligent compaction data analysis and management

Intelligent compaction (IC) on hot-mix-asphalt (HMA) is an emerging, yet still evolving technology for the constructions of highway system, airfield, and parking lots. IC produces massive amounts of geospatial data with 100% coverage of the compacted area in real time, which needs to be effectively analyzed and managed for quality control and acceptance (QC/QA). Accordingly, in this paper a systematic method using both the univariate and geo-statistical modeling techniques was developed for the IC data analysis and management. A data extraction method was proposed to categorize and extract IC data on different layer levels. Consequently, linear regression was performed to correlate IC Measurement Values (ICMVs) with random spot measurements. The semivariogram model was studied to evaluate the compaction uniformity, and the compaction curve was developed to identify the optimum number of roller passes. The systematic method with multiple statistical models was coded for numerical solutions, and demonstrated for eight HMA IC projects. Results indicate that compaction uniformity improves “from the ground, up”: subbase, HMA base, and then surface course. The compaction curve can help set the compaction target for QC/QA. ICMV has shown consistent linear correlations with spot measured deflections and material modulus, but it has inconsistent correlations with densities. Multivariate correlations indicate that multiple factors, including the ICMV of underlying layers and temperatures of HMA, affect the ICMV of the HMA layer.

► We develop a systematic method to analyze and manage intelligent compaction geospatial data.
► We evaluate the compaction uniformity with semivoriogram modeling.
► Intelligent compaction measurement value has inconsistent correlation with density measurement.
► Compaction uniformity improves from subbase to base, and then surface of pavement.