Hydroplaning risk evaluation plays a pivotal role in highway safety management. It is also an important component in the intelligent transportation system (ITS) ensuring human driving safety. Water-film is the widely accepted vital factor resulting in hydroplaning and thus continuously gained researchers’ attention in recent years. This paper provides a new framework to evaluate the hydroplaning potential based on emerging 3D laser scanning technology and water-film thickness estimation. The 3D information of the road surface was captured using a vehicle-mounted Light Detection and Ranging (LiDAR) system and then processed by a wavelet-based filter to remove the redundant information (surrounding environment: trees, buildings, and vehicles). Then, the water film thickness on the given road surface was estimated based on a proposed numerical algorithm developed by the two-dimensional depth-averaged Shallow Water Equations (2DDA-SWE). The effect of the road surface geometry was also investigated based on several field test data in Shanghai, China, in January 2021. The results indicated that the water-film is more likely to appear on the rutting tracks and the pavement with local unevenness. Based on the estimated water-film, the hydroplaning speeds were then estimated to represent the hydroplaning risk of asphalt pavement in rainy weather. The proposed method provides new insights into the water-film estimation, which can help drivers make effective decisions to maintain safe driving.

In constructing rapid rock identification models for measurement while drilling (MWD) via neural network methods, collecting actual drilling data to train the model is extremely time-consuming and labor-intensive. This requires extensive drilling experiments in various rock types, resulting in limited neural network training data for rock identification that covers a limited range of rock types. To suitably address this issue, a dynamic numerical simulation model for rock drilling is established that generates extensive drilling data. The input parameters for the simulations include torque, drill bit rotation speed, and drilling speed. A neural network model is then developed for rock classification using large datasets from dynamic numerical simulations, specifically those of granite, limestone, and sandstone. Building upon this model, transfer learning is appropriately applied to store the knowledge obtained in the rock identification based on the neural network model. Further training through transfer learning is conducted with smaller datasets obtained during actual drilling, making the model suitable for practical rock identification and prediction in the drilling processes. The neural network rock classification model, incorporating dynamic numerical simulation and transfer learning, achieves a prediction accuracy of 99.36% for granite, 99.53% for sandstone, and 99.82% for limestone. This reveals an enhancement in prediction accuracy of up to 22.94% compared to the models without transfer learning.