Predictive Modeling of Graphene-Modified Warm-Mix Asphalt: Integrating Rheology and Stress Sensitivity

This study evaluates graphene oxide (GO) and reduced graphene oxide (RGO) as bitumen modifiers in warm-mix asphalt (WMA), focusing on durability and Marshall properties. Testing five concentrations (0.1%–0.5%), 0.3% GO/RGO demonstrated optimal performance. Rutting resistance, moisture susceptibility, fracture, and raveling loss assessments revealed enhanced WMA durability. Rheological analysis via dynamic shear rheometer showed RGO-modified bitumen (RGOMB) outperformed GO-modified bitumen (GOMB), with higher recovery percent and lower nonrecoverable creep compliance in multiple stress creep recovery tests. Both modifiers improved Marshall stability and indirect tensile strength (ITS) while boosting resistance to distress. Modified bitumen also exhibited reduced aging effects compared with virgin bitumen. Predictive modeling executed using principal component regression (PCR) and partial least-squares regression (PLSR) achieved high accuracy (R2=0.86, 0.92, 0.99, and 0.99 for PCR; 0.87, 0.92, 0.99, and 0.99 for PLSR), with PLSR proving more reliable for forecasting Marshall stability, ITS, rut depth, and Cantabro loss. Results highlight RGOMB’s superior performance, enhancing WMA mix longevity and mechanical properties.