High-Precision Simulation of Asphalt Pavement Service Performance and Development of Intelligent Maintenance Platform

During the service life of asphalt pavement, its performance degrades rapidly, causing high maintenance costs. This paper gathers over 100,000 traffic data records from Guangxi and uses a spatiotemporal multi-scale data deep learning algorithm to simulate and predict the future pavement technical condition for the next 20 years. In the validation set, the mean PA values of the PCI and RQI service performance indicators exceed 90%, with P90 values also over 90%. This overcomes the limitations of existing methods in terms of low accuracy and high subjectivity. Based on this, a complete intelligent preventive maintenance platform has been developed, integrating multiple functions. During the service period, the average pavement performance enhanced by 14%. This platform simplifies maintenance operation, cuts costs and extends service life.