Vulnerable road users' crash hotspot identification on multi-lane arterial roads using estimated exposure and considering context classification

The dynamic-static dual-branch deep neural network for urban speeding hotspot identification using street view image data

The visual information regarding the road environment can influence drivers' perception and judgment, often resulting in frequent speeding incidents. Identifying speeding hotspots in cities can prevent potential speeding incidents, thereby improving traffic safety levels. We propose the Dual-Branch Contextual Dynamic-Static Feature Fusion Network based on static panoramic images and dynamically changing sequence data, aiming to capture global features in the macro scene of the area and dynamically changing information in the micro view for a more accurate urban speeding hotspot area identification. For the static branch, we propose the Multi-scale Contextual Feature Aggregation Network for learning global spatial contextual association information. In the dynamic branch, we construct the Multi-view Dynamic Feature Fusion Network to capture the dynamically changing features of a scene from a continuous sequence of street view images. Additionally, we designed the Dynamic-Static Feature Correlation Fusion Structure to correlate and fuse dynamic and static features. The experimental results show that the model has good performance, and the overall recognition accuracy reaches 99.4%. The ablation experiments show that the recognition effect after the fusion of dynamic and static features is better than that of static and dynamic branches. The proposed model also shows better performance than other deep learning models. In addition, we combine image processing methods and different Class Activation Mapping (CAM) methods to extract speeding frequency visual features from the model perception results. The results show that more accurate speeding frequency features can be obtained by using LayerCAM and GradCAM-Plus for static global scenes and dynamic local sequences, respectively. In the static global scene, the speeding frequency features are mainly concentrated on the buildings and green layout on both sides of the road, while in the dynamic scene, the speeding frequency features shift with the scene changes and are mainly concentrated on the dynamically changing transition areas of greenery, roads, and surrounding buildings. The code and model used for identifying hotspots of urban traffic accidents in this study are available for access: https://github.com/gwt-ZJU/DCDSFF-Net.

Short-term Safety Performance Functions by Random Parameters Negative Binomial-Lindley model for Part-time Shoulder Use

Part-time Shoulder Use (PTSU) is a traffic management and operation strategy that allows the use of the left or right shoulder as a travel lane, typically during the peak hours of the day. Though PTSU is an effective strategy for increasing roadway capacity in congested traffic conditions, there is very limited quantitative information about PTSU design elements and operational strategy in the existing literature, which could impact the occurrence of crashes on freeways. This study contributes to the safety literature by analyzing various potential crash contributing factors related to PTSU operation and design elements through the development of short-term Safety Performance Functions (SPFs). A comparison of the estimated models demonstrated that by utilizing the mixed distribution and allowing the posterior parameter estimates of explanatory variables to vary from one observation to another, the Random Parameters Negative Binomial-Lindley (RPNB-L) model outperformed the traditional NB and fixed coefficient NB-L models. The results of the proposed RPNB-L model indicated that the PTSU implemented sections experienced a lower number of traffic crashes compared to the non-PTSU freeway sections. Among the attributes related to PTSU operation and design elements, the usage of the leftmost shoulder lane as PTSU, the presence of emergency rest areas for damaged vehicles, and adequate shoulder width would significantly reduce crash frequency for the PTSU implemented freeways. Moreover, investigation of the identified hotspots revealed that the transition areas (start/end locations of PTSU) are the most critical sections. The findings from this research could assist transportation agencies to take appropriate countermeasures for preventing and reducing crash occurrences on PTSU implemented freeways.

A data-mining study on the prediction of head injury in traffic accidents among vulnerable road users with varying body sizes and head anatomical characteristics

Body sizes and head anatomical characteristics play the major role in the head injuries sustained by vulnerable road users (VRU) in traffic accidents. In this study, in order to study the influence mechanism of body sizes and head anatomical characteristics on head injury, we used age, gender, height, and Body Mass Index (BMI) as characteristic parameters to develop the personalized human body multi-rigid body (MB) models and head finite element (FE) models. Next, using simulation calculations, we developed the VRU head injury dataset based on the personalized models. In the dataset, the dependent variables were the degree of head injury and the brain tissue von Mises value, while the independent variables were height, BMI, age, gender, traffic participation status, and vehicle speed. The statistical results of the dataset show that the von Mises value of VRU brain tissue during collision ranges from 4.4 kPa to 46.9 kPa at speeds between 20 and 60 km/h. The effects of anatomical characteristics on head injury include: the risk of a more serious head injury of VRU rises with age; VRU with higher BMIs has less head injury in collision accidents; height has very erratic and nonlinear impacts on the von Mises values of the VRU's brain tissue; and the severity of head injury is not significantly influenced by VRU's gender. Furthermore, we developed the classification prediction models of head injury degree and the regression prediction models of head injury response parameter by applying eight different data mining algorithms to this dataset. The classification prediction models have the best accuracy of 0.89 and the best R2 value of 0.85 for the regression prediction models.

The impact of target speed on pedestrian, bike, and speeding crash frequencies

This research aims to investigate the influence of adopting the target speed concept on different types of crashes including pedestrian, bike, and speeding-related crashes. The Target speed is the highest speed that vehicles should operate on a roadway segment in a specific context. Based on the reviewed literature, this is the first study to investigate the relationship between target speed and crash frequency. Hence, big data including probe-vehicle data, traffic characteristics, geometric features, and land use attributes were utilized to develop crash prediction models. The main contributions of this research are to quantify the impacts of target speed on traffic safety considering context categories and to conclude the potential recommendations to lower different types of crashes. The 85th percentile speed was calculated and utilized in the developed models. Three crash prediction models were developed for pedestrian, bike, and speeding-related crashes. They were used in the analysis to quantify the impact of adopting target speed on different crash types. The results showed a significant reduction in the three crash types when using the target speed. Most of the improvements took place in three context categories: C3C: Suburban Commercial Segments, C3R: Suburban Residential Segments, and C4: Urban General Segments. Hence, this research recommends adopting target speed specifically in urban and suburban areas. Further, it suggests considering some measures to lower vulnerable road users' and speeding-related crashes. Following the recommendations of this research would help to reduce different types of crash frequency, hence, improving the mobility and safety for all users in different context classifications.

Investigating surrogate safety measures at midblock pedestrian crossings using multivariate models with roadside camera data

This study aims to evaluate and compare Surrogate Safety Measures (SSMs) at five midblock Rectangular Rapid Flashing Beacons (RRFB) and two midblock Pedestrian Hybrid Beacons (PHB) sites in Florida using extensive video data collected over the study period of July to November 2021. Computer vision and data processing resulted in four pedestrian SSMs, namely spatial gap, temporal gap, relative time to collision (RTTC) and Post Encroachment Time (PET). An initial investigation of the SSMs using Mann-Whitney-Wilcoxon tests revealed significant differences in the SSM values across different treatment types and hours of the day. Additionally, univariate regression of spatial gap, and multivariate regression of temporal gap, RTTC and PET revealed significant differences of SSMs across RRFB and PHB sites. The study considered both linear and non-linear (gamma, inverse Gaussian and lognormal) regression models. After considering various traffic and operational parameters, the data were aggregated for each pedestrian-vehicle interaction on each lane to create a total of 395 observations. The SSMs included average spatial gap, temporal gap, RTTC and PET for each interaction of pedestrian and vehicle on each lane. The results indicated that non-linear models performed better than the linear models. Moreover, the presence of the PHB, weekday, signal activation, lane count, pedestrian speed, vehicle speed, land use mix, morning period and pedestrian starting position from the sidewalk have been found to be significant determinants of the SSMs. Results also suggest temporal SSMs increase at the PHB sites compared to the RRFB sites, indicating an improvement of traffic safety at PHB sites. However, the spatial gap decreased for PHB sites compared to the RRFB sites, which suggests that pedestrians tend to start to cross the RRFB sites when they perceive vehicles to be further away than at the PHB sites.

A hybrid clustering and random forest model to analyse vulnerable road user to motor vehicle (VRU-MV) crashes

The main goal of this study is to investigate the unobserved heterogeneity in VRU-MV crash data and to determine the relatively important contributing factors of injury severity. For this end, a latent class analysis (LCA) coupled with random parameters logit model (LCA-RPL) is developed to segment the VRU-MV crashes into relatively homogeneous clusters and to explore the differences among clusters. The random-forest-based SHapley Additive exPlanation (RF-SHAP) approach is used to explore the relative importance of the contributing factors for injury severity in each cluster. The results show that, vulnerable group (VG), intersection or not (ION) and road type (RT) clearly distinguish the crash clusters. Moto-vehicle type and functional zone have significant impact on the injury severity among all clusters. Several variables (e.g. ION, crash type [CT], season and RT) demonstrate a significant effect in a specific sub-cluster model. Results of this study provide specific and insightful countermeasures that target the contributing factors in each cluster for mitigating VRU-MV crash injury severity.

A task-level emergency experience reuse method for freeway accidents onsite disposal with policy distilled reinforcement learning

A large number of freeway accident disposals are well-recorded by accident reports and surveillance videos, but it is not easy to get the emergency experience reused from past recorded accidents. To reuse emergency experience for better emergency decision-making, this paper proposed a knowledge-based experience transfer method to transfer task-level freeway accident disposal experience via multi-agent reinforcement learning algorithm with policy distillation. First, the Markov decision process is used to simulate the emergency decision-making process of multi-type freeway accident scene at the task level. Then, an adaptive knowledge transfer method named policy distilled multi-agent deep deterministic policy gradient (PD-MADDPG) algorithm is proposed to reuse experience from past freeway accident records to current accidents for fast decision-making and optimal onsite disposal. The performance of the proposed algorithm is evaluated on instantiated cases of freeway accidents that occurred on the freeway in Shaanxi Province, China. Aside from achieving better emergency decisions performance than various typical decision-making methods, the result shows decision maker with transferred knowledge owns 65.22%, 11.37%, 9.23%, 7.76% and 1.71% higher average reward than those without in the five studied cases, respectively. Indicating that the emergency experience transferred from past accidents contributes to fast emergency decision-making and optimal accident onsite disposal.

Time-specific hierarchical models for predicting crash frequency of reversible and high-occupancy vehicle lanes

Time-specific Safety Performance Functions (SPFs) were proposed to achieve accurate and dynamic crash frequency predictions. This study contributes to the literature by developing time-specific SPFs for freeways that include reversible lanes (RL) and freeways that include High-Occupancy Vehicle lanes (HOV) using Microwave Vehicle Detection System (MVDS) data from Virginia, Arizona and Washington States. Variables that capture the time-specific traffic turbulence were prepared and considered in the developed SPFs. Moreover, two different hierarchical models were proposed to identify factors associated with the different crash types or severity in crash frequency prediction. The results indicated that the variables representing the volume difference between reversible and general-purpose lanes (GPL) were positively associated with crash frequency. Further, the variable that indicated the design of the access point of the reversible lane was positively associated with crash frequency. The models comparison results showed that the hierarchical models outperformed the corresponding Poisson lognormal model with lower AIC and MAE values. This study also tested the proposed hierarchical models on High-Occupancy Vehicle freeway sections and reached the same conclusion on model comparison results. The significant variables representing the logarithm of volume were found to be significant and positive with crash frequency. Moreover, the difference in average speed between the HOV lanes and GPL was also found to be positive and significant with the crash frequency. In general, this study successfully identified the factors associated with the different crash types or severity in crash frequency prediction models.

Risk-Compensation Trends in Road Safety during COVID-19

The COVID-19 pandemic has had a global impact, disrupting the normal trends of our everyday life. More specifically, the effects of COVID-19 on road safety are still largely unexplored. Hence, this study aims to investigate the change in road safety trends due to COVID-19 using real-time traffic parameters. Results from the extensive analyses of the 2017 to 2020 data of Interstate-4 show that traffic volume decreased by 13.6% in 2020 compared to the average of 2017–2019’s volume, whereas there is a decreasing number of crashes at the higher volume. Average speed increased by 11.3% during the COVID-19 period; however, the increase in average speed during the COVID-19 period has an insignificant relationship with crash severities. Fatal crashes increased, while total crashes decreased, during the COVID-19 period; severe crashes decreased with the total crashes. Alcohol-related crashes decreased by 22% from 2019 to 2020. Thus, the road-safety trend due to the impact of COVID-19 has evidently changed and presents a unique trend. The findings of the study suggest a larger need for a more in-depth study to analyze the impact of COVID-19 on road safety, to minimize fatalities on roads through appropriate policy measures.