Automated pavement distress detection benefits road maintenance and operation by providing the condition and location of various distress rapidly. Existing work generally relies on manual labor or specific algorithms trained by dedicated datasets, which hinders the efficiency and applicable scenarios of methods. Street view map provides interactive panoramas of a large scale of urban roadway network, and is updated in a recurrent manner by the provider. This paper proposed a deep learning method based on a pre-trained neural network architecture to identify and locate different distress in real-time. About 20,000 street view images were collected and labeled as the training dataset using the Baidu e-map. Eight types of distress are notated using Yolov3 deep learning architecture. The scale-invariant feature transform (SIFT) descriptors combined with GPS and bounding boxes were applied to judge the deterioration of the distress. A decision tree was designed to evaluate the change of the distress over some time. A typical road in Shanghai was selected to verify the effectiveness of the proposed model. The images of the road from 2015 to 2017 were collected from the street view map. The results showed that the mean average precision of the proposed algorithm is 88.37%, demonstrating the vast potential of applying this method to detect pavement distress. 43 distress were newly generated, and 49 previous distress were patched in the two years. The proposed method can assist the authorities to schedule the maintenance activities more effectively.

In addition to closed merge lanes as physical bottlenecks of work zones, traffic oscillations caused by merging vehicles at multiple locations could reduce work-zone capacity. This study took a step-wise procedure to reveal spatial distributions of merging vehicles along work zones and their influence on speed-flow relationships of lane traffic flows. Field data showed that inserting vehicles from merge lanes could spread their influence over adjacent unclosed through lanes. Moreover, with increases in total volume, merging vehicles could choose their inserting positions further upstream of the work zone, which could induce oscillations near the insertion point. At the identified upstream bottlenecks, capacity drop was found in speed-flow diagrams of through-lane traffic, but it was not found in the diagrams of merge-lane traffic flows. Lack of sufficient demand and special merging behaviors on merge lanes could be attributed to the distinct speed-flow relationship. Two-part piecewise regression models were developed to fit the speed-flow relationships of uncongested and congested flows of through lanes. By comparing the estimated speed-flow models, it was found that when a queue is forming, the extent of the capacity drop and speed reduction is different for through lanes. Queue discharge uses different lengths of time on through lanes and multiple merging locations.

Low temperatures and icing in winter are significant factors that severely affect highway safety and traffic mobility. To enhance the precision and reliability of real-time winter road surface temperature (RST) prediction, a short-term prediction model is developed that harnesses both feature selection and deep learning. Leveraging meteorological data from a mountain highway in Yunnan, China, the key environmental variables affecting road surface temperature were first extracted using a random forest (RF) model for feature selection. These features were then combined with RST data to construct multiple groups of input variable combinations for the prediction model. A short-term prediction model with a 10-minute update frequency was built using a long short-term memory neural network (LSTM), namely RF-LSTM. The best input variable combination and preset parameters for the prediction model were determined through comparative testing with prevalent machine learning models, and the transferability of the prediction model was verified. The results showed that the best input variable combination for the RF-LSTM prediction model was road surface temperature and air temperature. The model recognised that the short-term RST was affected by long and short-term memory characteristics within a two-hour timeframe. When compared to the RF model, backpropagation (BP) neural network model and the standard LSTM model, the proposed model reduces prediction errors by 59.15%, 31.10% and 20.26%, respectively, while the prediction accuracy is 99.13% within an error margin of ±0.5℃. On the verification dataset, the proposed model maintains its time transferability with an average prediction absolute error of 0.0478. In all, the proposed model not only achieves a higher level of precision in real-time RST predictions but also ensures a more consistent and reliable performance under the challenging conditions of high altitude and mountainous terrain, offering enhanced support for traffic safety and road maintenance decision-making.

License plate restriction (LPR) policy presents the most straightforward way to reduce road traffic and emissions worldwide. However, in practice, it has aroused great controversy. This policy broke the original structure of the urban transportation mode, which needed some matching strategies to adapt to this change. Investigating this travel demand change is a challenging task because it is greatly influenced by features of the local built environment. Fourteen variables from four dimensions, location, land-use diversity, distance to transit, and street design, are used to depict the built environment; moreover, the severe collinearity underlies these feature variables. To solve the multicollinearity among the variables and high-dimensional problem, this study utilizes two different penalization-based regression models, the LASSO (least absolute shrinkage and selection operator) and Elastic Net regression algorithms, to achieve the variable selection and explore the impacts of the built environment on the change of travel demand triggered by the LPR policy. Travel demand changes are assessed by the relative variation in taxi ridership in each traffic analysis zone based on the taxi GPS data. Built environment variables are measured using the transportation network data and the Baidu Map Service points of interest (POI) data. The results show that regions with a higher level of public transportation service and a higher degree of the land mix have a stronger resilience to the vehicle restriction policy. Besides, the contribution rate of public transportation is stable as a whole, while the contribution rate of richness depends on specific types of land use. The conclusions in this study can provide in-depth insights into the influence of the LPR policy and underpin traffic complementary policies to ensure the effectiveness of LPR.

Emergency escape ramps (EERs) is an infrastructure of mountain freeways to stop runaway vehicles. As the last defence of vehicle occupants’ safety, the performance of EERs in reducing crash injury severity is a concern for stakeholders of road safety. Based on crash records collected on a mountain freeway equipped with five EERs, this study compared the injury severity of crashes on EERs and other road sections, and identified the factors that significantly influence injury severity in the two conditions. Estimations of the parameter coefficients and marginal effects of a random parameters ordered probit model were used to infer EER performance under the impacts of various factors. The results confirm the effectiveness of EERs on the reduction of crash injury severity. The protection function of EERs is weakened by nighttime, the rollover status of crashed vehicle, multi-vehicle collisions, improper design or installation of the roadside infrastructure, drivers’ unfamiliarity with local driving conditions, and crashed vehicle weight. The paper compares the findings with those of previous studies and proposes some recommendations to improve EER performance for occupant and property protection on mountain freeways.