Traditional laboratory rutting tests are performed at a constant temperature by neglecting pavement temperature variation. The mechanical properties of asphalt are susceptible to temperature variation. This sensitivity to temperature variations significantly influences the performance and durability of asphalt pavements. Following this purpose, a stepwise temperature-controlled rutting test method was proposed to investigate the rutting development of double-layer asphalt pavement (DLAP) under variable temperature. A time-hardening model was developed and employed to evaluate the rutting performance of DLAP under variable temperature. Results indicate that the rutting development of DLAP exhibits a stepwise variation when subjected to variable temperatures. Within a specific constant temperature range, rutting development can be fitted using a power function of load cycles. The rutting deformation of DLAP predominantly occurs at 20 °C; once the temperature exceeds 50 °C, the rutting development accelerates and becomes difficult to stabilize. The time-hardening model effectively captures the rutting development under variable temperature. The predicted values align closely with field values, which demonstrates the model’s feasibility in calculating rutting deformation under variable temperature. Under actual service conditions, the rutting development of DLAP follows a periodic S-shaped growth, yet this trend can still be represented by a power-law function. DLAP exhibits satisfactory durability and structural stability, effectively addressing the challenges posed by traffic loads and high temperatures in test sections.

To investigate the temperature dependence of the modulus of LSAM-50 flexible base asphalt pavement (LSAM-50 pavement) materials, specifically SMA-13, AC-20, and LSAM-50. The effects of temperature on the modulus of LSAM-50 pavement materials were investigated, and a temperature-dependent model of resilient modulus was established. A dynamic modulus master curve was constructed based on a generalized logarithmic Sigmoidal model. The correlation between the resilient modulus and dynamic modulus was studied, and a multiple linear regression model was developed to describe the relationship between the dynamic modulus and resilient modulus, temperature, and loading frequency. The results show that the resilient modulus and dynamic modulus gradually decrease with the increase in temperature and then tend to stabilize. The resilient modulus of LSAM-50 is higher than that of SMA-13 and AC-20 in the entire temperature range, and the dynamic modulus of LSAM-50 is higher than that of SMA-13 and AC-20 in the high-temperature range. The correlation coefficients (R²) of the established resilient modulus and dynamic modulus estimation models are greater than 0.97 and 0.94, respectively.