Evaluation of Balanced Asphalt Surface Mixtures with Conventional and High RAP Contents Using Laboratory and Accelerated Pavement Testing

Balanced Mix Design (BMD) represents an asphalt mixture design methodology that replaces certain traditional volumetric parameters with performance-based testing to address predominant distresses such as rutting and cracking. This approach offers an avenue to properly design and produce engineered asphalt mixtures, including those with high reclaimed asphalt pavement (HRAP) contents, recycling agents (RAs), fibers, and polymer-modified binders. Laboratory performance tests are essential to the BMD process, as they ensure the production of durable, high-performance materials. Beyond laboratory performance evaluation, accelerated pavement testing (APT) plays a crucial role in bridging the gap between laboratory material characterization and field pavement performance. This dissertation aimed to assess the BMD concept for designing durable, long-lasting surface mixtures in Virginia, with particular emphasis on higher RAP content mixtures (HRAP mixtures, i.e., exceeding 30% RAP). The study involved laboratory and APT testing of six surface mixtures featuring a range of RAP contents (both conventional and high), two binder grades (PG 64-22 and PG58-28), one RA, and one warm mix additive. Findings indicated that dense-graded, unmodified surface mixtures with higher RAP contents can be successfully designed using the current Virginia Department of Transportation (VDOT) BMD special provision. These mixtures can be produced in the plant with no significant deviations in aggregate gradation and asphalt binder content from the design specifications. The combined effect of variations in different volumetric properties during production may influence the primary performance of the mixtures, potentially resulting in an imbalance. As a consequence, the produced BMD mixture may fail to meet one or more performance thresholds. Additionally, the results underscored the effectiveness of BMD concept with incorporating RAs and/or a softer binder when designing HRAP surface mixtures. Importantly, the current selected BMD tests characterized the laboratory performance of mixtures, aligning with the performance observed under APT. This research provided a steppingstone towards the examination and validation of the VDOT BMD thresholds, which ensures satisfactory field performance. The study also indicated that while current BMD thresholds provided sufficient margins for satisfactory field cracking performance, rutting resistance may become a concern for overly designed BMD HRAP mixtures. For instance, mixtures with excessively high asphalt binder content may exhibit compromised rutting resistance. Furthermore, to address the challenges uncovered during BMD test analysis—issues like the constraints of traditional pair-wise comparisons, risks of repetitive design processes, and the difficulty in pinpointing critical factors in mixture production—this dissertation proposed innovative solutions to enhance BMD application and streamline the evaluation process. First, a novel Composite Performance Index (CPI), visualized through a 3D plot, captured the "balance" status of various mixtures. Second, a machine learning-enhanced BMD framework was introduced, offering intelligent optimization throughout the design and production phases. The integration of these two tools offers significant potential for simultaneously improving multiple performance indices of asphalt mixtures. Finally, this research demonstrated that the performance of higher RAP content mixtures can exceed that of lower RAP content mixtures through the application of BMD approaches. This dissertation not only advanced the implementation of BMD for surface mixtures but also contributed to the sustainable and performance-driven evolution of asphalt mix design. The insights gained from this study provided practical guidance and strategic recommendations for enhancing asphalt mixture design, production, and performance monitoring.