A combined experimental numerical statistical framework to characterize fracture behavior of asphalt mixtures

Abstract

This study investigates the fracture and viscoelastic behavior of asphalt mixtures through a combined experimental, numerical, and statistical approach. Semi Circular Bend (SCB) tests were performed on specimens with notch lengths of 10, 20, and 30 mm at two temperature levels (0 degrees C and room temperature) to determine fracture toughness (K-Ic), fracture energy, and stiffness parameters. The Extended Finite Element Method (XFEM) was employed to simulate crack initiation and propagation, providing a close match with experimental SCB results under various thermal and geometric conditions. To further interpret the effects of temperature and crack geometry, the Response Surface Methodology (RSM) was applied to the experimental data, enabling the identification of key interaction effects and optimization of fracture parameters. The results demonstrated that decreasing temperature and increasing notch length notably reduced fracture resistance, while XFEM accurately represented the nonlinear crack propagation observed in laboratory testing. Moreover, RSM provided predictive equations that were in strong agreement with XFEM outcomes. The proposed integrated SCB-XFEM-RSM framework presents an efficient and reliable methodology for evaluating and optimizing the fracture performance of asphalt mixtures, offering valuable insights for the design of durable and sustainable pavement systems under diverse environmental conditions.