Older pedestrians are vulnerable on the streets and at significant risk of injury or death when involved in crashes. Pedestrians’ safety is critical for roadway agencies to consider and improve, especially older pedestrians aged greater than 65 years old. To better protect the older pedestrian group, the factors that contribute to the older crashes need to be analyzed deeply. Traditional modeling approaches such as Logistic models for data analysis may lead to modeling distortions due to the independence assumptions. In this study, Extreme Gradient Boosting (XGBoost), is used to model the classification problem of three different levels of severity of older pedestrian traffic crashes from crash data in Colorado, US. Further, Shapley Additive explanations (SHAP) are implemented to interpret the XGBoost model result and analyze each feature’s importance related to the levels of older pedestrian crashes. The interpretation results show that the driver characteristic, older pedestrian characteristics, and vehicle movement are the most important factors influencing the probability of the three different severity levels. Those results investigate each severity level’s correlation factors, which can inform the department of traffic management and the department of road infrastructure to protect older pedestrians by controlling or managing some of those significant features.

China's rapid urbanization and high traffic accident frequency have received many researchers’ attention. It is important to reveal how urban infrastructures and other risk factors affects the traffic accident frequency. A growing amount of research has examined the local risk factors impact on traffic accident frequency at certain time. Some studies considered these spatial influences but overlooked the temporal correlation/heterogeneity of traffic accidents and related risk factors. This study explores risk factors’ influence on urban traffic accidents frequency while considering both the spatial and temporal correlation/heterogeneity of traffic accidents. The study area is split into 100 equally sized rectangle traffic analysis zones (TAZs), and the urban traffic accident frequency and attributes in each TAZ are extracted. The linear regression model, spatial lag model (SLM), spatial error model (SEM) and time-fixed effects error model (T-FEEM) are established and compared respectively. The proposed methodologies are illustrated using ten-month traffic accident data from the urban area of Guiyang City, China. The results reveal that the time-fixed effects error model, which considers both spatial and temporal correlation/heterogeneity of traffic accidents, is superior to other models. More traffic accidents will happen in those TAZs that have more hospitals or schools. Moreover, hospitals have a greater influence on traffic accidents than schools. Because of the location in the margin of the city, those TAZs that have passenger stations have more traffic accidents. This study provides policy makers with more detailed characterization about the impact of related risk factors on traffic accident frequencies, and it is suggested that not only the spatial correlation/heterogeneity but also the temporal correlation/heterogeneity should be taken into account in guiding traffic accident control of urban area.

It is a dangerous behaviour for pedestrians and nonmotorized vehicles to cross intersections without waiting when they arrive at intersections during the red-light period. This paper investigates three typical signalized major-major intersections in the center of Beijing, by collecting and analyzing 1368 samples of pedestrians and nonmotorized vehicles. A random parameter logit model (RPLM) is established, with immediate red-light running (IRLR) behaviour as the dependent variable. The results show that the number of people waiting upon arrival, number of people crossing upon arrival, traffic mode, motor vehicle phase upon arrival, and speed change upon arrival have significant effects on IRLR behaviour. Accordingly, we suggest enforcing education administration on cyclists to reduce cyclists’ IRLR behaviour. Thus, people’s red-light running (RLR) behaviour will further decrease with fewer cyclists’ IRLR behaviour.

There are few studies on the violation of truck drivers, especially the hazmat truck driver, although truck driver’s violation may cause serious casualties. This paper aims to investigate hazmat truck drivers’ violation behavior and identify associated risk factors. Different data sources in intelligent transportation system (ITS) including hazmat transportation management system and traffic safety management system are extracted and emerged together. Three years (2016–2018) of violation data that comprised 11612 trip record in China are employed in this research. Based on Bayesian theory, this study proposes zero-inflated ordered probit (ZIOP) model and three alternative models to exploring the relationship between hazmat truck drivers’ violation frequency and the key risk factors. The results show that ZIOP model can handle excessive zero observation problem of violation data properly and differentiate between ‘always-zero group’ drivers and drivers who did not violate the traffic rules during research period but would do so in different surroundings. The results also indicate that the violation probability and the violation frequency level of hazmat truck drivers are influenced by driver characteristics, freight order attributes, and drivers’ violation records. This research provides guidance for driving training and safety education of hazmat truck drivers, and will be helpful in building better driving simulation models.

The investment of education capital and fixed assets play an important role in the economic growth of one country, while traditional regression analysis may underestimate or overestimate the contribution of the investment of education capital and fixed assets on economic growth. In this paper,...