Asphalt pavement rutting depth prediction based on FGM(1, 1│sin) model

With the rapid development of urban transportation, pavement performance is directly related to the safety of transportation and environmental protection. To effectively model the rutting depth of asphalt pavements this study is based on the FGM(1, 1│sin) model, which aims to improve the prediction accuracy of the rutting disease evolution process of asphalt pavements. The model combines the fractional-order accumulation operator and the sinusoidal operator to accommodate the characteristics of asphalt pavement rutting depth data, such as long sampling periods, small data samples, and the presence of oscillations. By introducing time weights, the model enhances the ability to mine the rut depth data of different pavements and describes the rut development law more accurately. In addition, the model was tested on 19 different asphalt pavements at the RIOHTrack experimental site and demonstrated higher prediction accuracy compared to other existing algorithms, providing strong data support for structural design, material use, and maintenance decisions of asphalt pavements.