Porous ultra-thin asphalt overlay (PUAO) is widely used in maintenance and new road surface layers. Since the paving thickness of PUAO is only about 20 mm, it is necessary to study its own performance and the overall performance of the double-layer material formed by the ultra-thin overlay and the under-layer mixture. PUAO and PUAO/asphalt concrete double-layer specimens with different air voids were prepared and the performance of the specimens were studied. The results showed that as the air voids increased from 15.8% to 24.4%, the PUAO’s high-temperature rutting dynamic stability decreased, the tensile strength ratio first decreased and then increased and the acoustic absorption coefficient peak value increased gradually, but the peak value of frequency moved from 400 Hz to 1250 Hz. The dynamic stability and low-temperature bending strength of the PUAO/AC-13 double-layer specimens were higher than those of the PUAO single-layer specimens. However, the double-layer specimen had a smaller permeability coefficient than the single-layer specimen. The International Friction Index showed that after 600,000 abrasion cycles, the F60 values of the five PUAOs were higher than those of the DUAO, indicating that the PUAO material had excellent skid resistance performance.

Many image processing techniques (IPTs) were proposed to inspect pavement defects for improving the precision and efficiency of the on-site inspections by humans. However, the various pavement conditions led to the unacceptable stability of IPTs. Therefore, an application of convolutional neural networks (CNNs) is presented for pothole detection using digital images in this study. Two CNNs, called conventional CNN and pre-pooling CNN, were designed, trained, and tested using 96,000 pavement images. Additionally, a stability study and comparative study were conducted based on the testing results. The main difference between the two CNNs was that a pre-pooling layer was used in the pre-pooling CNN to pre-process pavement images before the first convolutional layer. The results demonstrated that the optimised pre-pooling CNN had the 98.95% recognition precision in the testing. The stability study indicated that the optimised CNN model had the robustness in various real-world situations (e.g. light conditions and pavement materials). Compared with the traditional IPT methods, the CNN had a higher precision for extracting pothole features autonomously.

In recent years, asphalt pavements have been widely used due to their desirable properties: safety, smoothness, and durability. However, the aging of bitumen during its service period is inevitable, so the performance of asphalt pavement will decline. Thus, it is necessary to investigate the aging mechanism systematically to offer a new way of illustrating changes during the aging process, and the investigation of the short-term aging mechanism could be a first step. This study explored the variations of rheological properties and molecular weight distribution during the aging process, and the data obtained by gel permeation chromatography (GPC) were analyzed by principal component analysis. The experiments in this research include preparing aged bitumen samples under five aging temperatures and six aging durations, and then tests were performed using a dynamic shear rheometer (DSR), bending beam rheometer (BBR), and GPC. The results indicate that the aging increased complex modules ($G^{*}$ ), creep stiffness ($S$ ), and fatigue resistance index ($N_f$ ) but decreased phase angle ($\delta$ ) and $m$ -values. In addition, the transformation from medium molecular size to large molecular size was found to be the most significant phenomenon during the aging process. Moreover, during the aging process of bitumen, the molecular weight distribution index polydispersity index was found to have the best correlation with rheological properties.