The framework of a statistical acceptance plan based on the permanent deformation potential of asphalt concrete

The use of end-result specifications in highway construction is considered to be highly variable, which will create different risk implications for the contractor and the agency. In addition, recent research shows that control of compositional properties is not an adequate tool in controlling the expected performance of asphalt concrete. Therefore, there is a need to evaluate mechanical properties using performance tests, so that the acceptance procedure fairly determines compliance to the performance specifications. The purpose of the study was to develop a framework of statistical acceptance procedures for SHAs, based on the predicted permanent deformation potential, using performance tests. A life cycle costs analysis was conducted based on the expected performance and maintenance costs, in accordance with the FHWA guidance and recommendations. The proposed procedure provides significant contribution towards the existing acceptance procedures, including a uniquely constructed performance-based specification with thorough statistical consideration. The proposed specification modified the acceptance for bulk materials to accommodate the severity of pavement distress (Use of the 80th percentile of distress prediction using beta distribution). Application of the maximum permanent shear strain (mpss) as specific quality measure is presented. Another important benefit is the capability to define the risks between the agency and the contractor within the acceptance limits. Within the acceptance range, the risks of the contractor to be underpaid or the agency to overpay were obtained using simulation of the proposed plan with typical mean of mpss ranging from 1. 0% to 4. 8% and variance ranging from 0. 1% to 0. 4% for 8,000-paving lot, with 5 sublots each. The proposed plan involved a unique approach for performance-based specifications. Significant features include the association of acceptance to the expected performance of material, consideration of loss of function in the pay factor, progressive reward for an increasingly desirable process mean, a progressive reward for decreasing process variability, simplicity of use in design and construction, and the potential for application with new materials. Demonstration of the proposed plan using Superpave test data collected by the North East Superpave Center and the data from the Nevada project showed that the proposed plan worked reasonably well and was reliable.