Optimizing performance and sustainability: Predictive modeling of RAB effects on CRMB in paving applications

In recent years, asphalt recycling and cost-effective, cleaner production methods have become crucial for enhancing binder properties in the paving industry. This study aims to determine the optimal percentage of reclaimed asphalt binder (RAB) in crumb rubber modified binder (CRMB) by investigating both rheological and viscoelastic properties. The results demonstrate a direct correlation between CRMB viscosity and RAB percentage, emphasizing the need to limit RAB to mitigate potential increases in pollution and energy consumption. While RAB negatively impacts low-temperature performance, up to 40 % RAB does not affect the low performance grade (PG). Increasing RAB beyond 60 % improves rutting resistance by counteracting temperature and stress effects. Classification using PG and MSCR methods identified that CRMB with RAB60 and RAB80, both rated at PG76S-16, exhibits optimal performance. The LAS test results indicate that as strain increases, higher RAB percentages (above 50 %) adversely affect fatigue performance; however, binders with greater CRM proportions demonstrate superior fatigue resistance due to the presence of a polymer network. Variance analysis reveals that test temperature, RAB percentage, and stress level significantly influence the rutting resistance and Jnr parameter of rubberized asphalt. Both the regression and neural network models exhibit strong predictive performance, with the regression model slightly outperforming the neural network. Based on the results of response surface methodology (RSM) analysis, the optimal percentage of RAB in CRMB was determined to be 50 %, 69 %, 33 %, and 55 % for temperate regions with standard traffic, tropical regions, cold regions, and temperate regions with very heavy traffic, respectively. The simultaneous use of a softer binder (PG52–28), RAB, and CRM provides optimal performance conditions, suitable temperature ranges, and high resistance to traffic. • The study finds 50 % RAB optimal for CRMB in various climates. • Excessive RAB lowers low-temp and fatigue resistance; above 60 % boosts rutting resistance. • Temperature, RAB %, and stress affect performance; MVR and ANN models predict well.