Abstract: A rutting predictive model utilized a power equation was developed with six factors,namely shear strength of asphalt mixture,shear stress of pavement structure,number of load repetitions,pavement temperature,vehicle speed and pavement depth,so as to develop a rutting predictive model with high predictive precision and extensive applicability.The rutting test with accelerated loading facility(ALF) was carried out on three types of pavement structures under three different pavement temperature and load pressure,and the rutting depth was measured at certain loading numbers.The simplify rutting predictive model setting the 20 km/h as the basic vehicle speed was determined through multi-fitting analysis of 117 sets data,including rutting depth,test temperature,shear stress,pavement thickness and loading numbers.The relation between pavement rutting and vehicle speed was determined by the time-hardening creep model and the function of load time,as a result,the simplify model was revised for the vehicle speed to the final rutting predictive model.Resultsshow that the rutting predictive model has comprehensive factors,extensive applicability and high predictive precision.