Search

Rear-end (RE) crashes are notably prevalent and pose a substantial risk on freeways. This paper explores the correlation between speed difference among the following and leading vehicles (Δν) and RE crash risk. Three joint models, comprising both uncorrelated and correlated joint random-parameters bivariate probit (RPBP) approaches (statistical methods) and a cross-stitch multilayer perceptron (CS-MLP) network (a data-driven method), were estimated and compared against three separate models: Support Vector Machines (SVM), eXtreme Gradient Boosting (XGBoost), and MLP networks (all data-driven methods). Data on 15,980 two-vehicle RE crashes were collected over a two-year period, from January 1, 2021, to December 31, 2022, considering two possible levels of injury severity: no injury and injury/fatality for both drivers of following and leading vehicles. The comparative performance analysis demonstrates the superior predictive capability of the CS-MLP network over the uncorrelated/correlated joint RPBP model, SVM, XGBoost, and MLP networks in terms of recall, F-1 Score, and AUC. Significantly, numerous shared variables influence the injury severity outcomes for the following and leading vehicles across both statistical and data-driven approaches. Among these factors, the following vehicle (a truck) and the leading vehicle (a passenger car) demonstrate contrasting effects on the injury severity outcomes for both vehicles. Furthermore, the SHapley Additive exPlanations (SHAP) values from the CS-MLP network visually show the relationship between Δν and injury severity, revealing non-linear trends unlike the average effects shown by statistical methods. They indicate that the least injury outcomes for both following and leading vehicles occurs at a Δν of 0 to 10 mph, matching observed patterns in RE crash data. Additionally, a marked variation in the trend of SHAP values for the two vehicles is noted as the speed difference increases. Therefore, the findings affirm the superior performance of joint model development and substantiate the non-linear impacts of speed difference on injury outcomes. The adoption of dynamic speed control measures is recommended to mitigate the injury outcomes involved in two-vehicle RE crashes., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Ltd. All rights reserved.)

Vehicles equipped with automated driving capabilities have shown potential to improve safety and operations. Advanced driver assistance systems (ADAS) and automated driving systems (ADS) have been widely developed to support vehicular automation. Although the studies on the injury severity outcomes that involve automated vehicles are ongoing, there is limited research investigating the difference between injury severity outcomes for the ADAS and ADS equipped vehicles. To ensure a comprehensive analysis, a multi-source dataset that includes 1,001 ADAS crashes (SAE Level 2 vehicles) and 548 ADS crashes (SAE Level 4 vehicles) is used. Two random parameters multinomial logit models with heterogeneity in the means of random parameters are considered to gain a better understanding of the variables impacting the crash injury severity outcomes for the ADAS (SAE Level 2) and ADS (SAE Level 4) vehicles. It was found that while 67 percent of crashes involving the ADAS equipped vehicles in the dataset took place on a highway, 94 percent of crashes involving ADS took place in more urban settings. The model estimation results also reveal that the weather indicator, driver type indicator, differences in the system sophistication that are captured by both manufacture year and high/low mileage as well as rear and front contact indicators all play a role in the crash injury severity outcomes. The results offer an exploratory assessment of safety performance of the ADAS and ADS equipped vehicles using the real-world data and can be used by the manufacturers and other stakeholders to dictate the direction of their deployment and usage., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Ltd. All rights reserved.)

Lane change behavior disrupts traffic flow and increases the potential for traffic conflicts, especially on expressway weaving segments. Focusing on the diversion process, this study incorporating individual driving patterns into conflict prediction and causation analysis can help develop individualized intervention measures to avoid risky diversion behaviors. First, to minimize measurement errors, this study introduces a lane line reconstruction method. Second, several unsupervised clustering methods, including k-means, agglomerative clustering, gaussian mixture, and spectral clustering, are applied to explore diversion patterns. Moreover, machine learning methods, including Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), Attention-based LSTM, eXtreme Gradient Boosting (XGB), Support Vector Machine (SVM), and Multilayer Perceptron (MLP), are employed for real-time traffic conflict prediction. Finally, mixed logit models are developed using pre-conflict condition data to investigate the causal mechanisms of traffic conflicts. The results indicate that the K-means algorithm with four clusters exhibits the highest Calinski-Harabasz and Silhouette scores and the lowest Davies-Bouldin scores. With superior classification accuracy and generalization ability, the LSTM is used to develop the personalized traffic conflict prediction model. Sensitivity analysis indicates that incorporating the diversion patterns into the LSTM model results in an improvement of 3.64% in Accuracy, 7.15% in Precision, and 1.34% in Recall. Results from the four mixed logit models indicate significant differences in factors contributing to traffic conflicts within each diversion pattern. For instance, increasing the speed difference between the target vehicle and the right preceding vehicle benefits traffic conflict during acceleration diversions but decreases the likelihood of traffic conflicts during deceleration diversions. These results can help traffic engineers propose individualized solutions to reduce unsafe diversion behavior., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Ltd. All rights reserved.)

The rise of Express Lanes also known as High Occupancy Toll (HOT) Lanes and Managed Lanes, signifies a major leap in traffic management and transportation funding. Despite their increased deployment to ensure reliable travel times through dynamic tolling during peak traffic periods, a comprehensive evaluation of their safety impact is notably lacking. Presently, the Crash Modification Factors Clearinghouse, a vital resource, only lists two case studies related to Express Lanes, one of which is our own research. This lack of data highlights the critical need for more extensive studies to thoroughly assess the safety benefits of Express Lanes and to improve their application. This study aims to rigorously evaluate the safety impact of express lanes on freeways, presenting a first-of-its-kind, in-depth analysis of their specific effects on both Express Lanes and General-Purpose Lanes (GP-Lanes) individually. The analysis utilized data from 55 miles of Express Lanes across various locations in Florida, comparing them to High Occupancy Vehicle (HOV) lanes. The results demonstrate that converting HOV lanes to Express Lanes or introducing new ones does not compromise overall freeway safety. In fact, safety within Express Lanes improves, as evidenced by a decrease in crash occurrence and Crash Modification Factors for Express lanes, which are below "1" across all crash categories. This underscores the effectiveness of Express lanes in enhancing roadway safety. In contrast, incidents in GP-Lanes have increased, indicating a shift of crashes to these lanes, and thus making Express lanes relatively safer. This underlines the importance of continued research into the safety impact of express lanes and calls for further studies to refine traffic management strategies, aiming at enhancing travel efficiency while ensuring traffic safety, especially for the GP-Lanes amid the expansion of express lanes., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Ltd. All rights reserved.)

Despite the recent advancements that Autonomous Vehicles have shown in their potential to improve safety and operation, considering differences between Autonomous Vehicles and Human-Driven Vehicles in accidents remain unidentified due to the scarcity of real-world Autonomous Vehicles accident data. We investigated the difference in accident occurrence between Autonomous Vehicles' levels and Human-Driven Vehicles by utilizing 2100 Advanced Driving Systems and Advanced Driver Assistance Systems and 35,113 Human-Driven Vehicles accident data. A matched case-control design was conducted to investigate the differential characteristics involving Autonomous' versus Human-Driven Vehicles' accidents. The analysis suggests that accidents of vehicles equipped with Advanced Driving Systems generally have a lower chance of occurring than Human-Driven Vehicles in most of the similar accident scenarios. However, accidents involving Advanced Driving Systems occur more frequently than Human-Driven Vehicle accidents under dawn/dusk or turning conditions, which is 5.25 and 1.98 times higher, respectively. Our research reveals the accident risk disparities between Autonomous Vehicles and Human-Driven Vehicles, informing future development in Autonomous technology and safety enhancements., (© 2024. The Author(s).)

Traffic accidents have emerged as one of the most public health safety matters, raising concerns from both the public and urban administrators. The ability to accurately predict traffic accident not only supports the governmental decision-making in advance but also enhances public confidence in safety measures. However, the efficacy of traditional spatio-temporal prediction models are compromised by the skewed distributions and sparse labeling of accident data. To this end, we propose a Sparse Spatio-Temporal Dynamic Hypergraph Learning (SST-DHL) framework that captures higher-order dependencies in sparse traffic accidents by combining hypergraph learning and self-supervised learning. The SST-DHL model incorporates a multi-view spatiotemporal convolution block to capture local correlations and semantics of traffic accidents, a cross-regional dynamic hypergraph learning model to identify global spatiotemporal dependencies, and a two-supervised self-learning paradigm to capture both local and global spatiotemporal patterns. Through experimentation on New York City and London accident datasets, we demonstrate that our proposed SST-DHL exhibits significant improvements compared to optimal baseline models at different sparsity levels. Additionally, it offers enhanced interpretability of results by elucidating complex spatio-temporal dependencies among various traffic accident instances. Our study demonstrates the effectiveness of the SST-DHL framework in accurately predicting traffic accidents, thereby enhancing public safety and trust., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Ltd. All rights reserved.)

To facilitate efficient transportation, I-4 Express is constructed separately from general use lanes in metropolitan area to improve mobility and reduce congestion. As this new infrastructure would undoubtedly change the traffic network, there is a need for more understanding of its potential safety impact. Unfortunately, many advanced real-time crash prediction models encounter an important challenge in their applicability due to their demand for a substantial volume of data for direct modeling. To tackle this challenge, we proposed a simple yet effective approach - anomaly detection learning, which formulates model as an anomaly detection problem, solves it through normality feature recognition, and predicts crashes by identifying deviations from the normal state. The proposed approach demonstrates significant improvement in the Area Under the Curve (AUC), sensitivity, and False Alarm Rate (FAR). When juxtaposed with the prevalent direct classification paradigm, our proposed anomaly detection learning (ADL) consistently outperforms in AUC (with an increase of up to 45%), sensitivity (experiencing up to a 45% increase), and FAR (reducing by up to 0.53). The most performance gain is attained through the combination of Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) in an ensemble, resulting in a 0.78 AUC, 0.79 sensitivity, and a 0.22 false alarm rate. Furthermore, we analyzed model features with a game-theoretic approach illustrating the most correlated features for accurate prediction, revealing the attention of advanced convolution neural networks to occupancy features. This provided crucial insights into improving crash precaution, the findings from which not only benefit private stakeholders but also extend a promising opportunity for governmental intervention on the express lane. This work could promote express lane with more efficient resource allocation, real-time traffic management optimization, and high-risk area prioritization., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Ltd. All rights reserved.)

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., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Ltd. All rights reserved.)

Despite awareness campaigns and legal consequences, speeding is a significant cause of road accidents and fatalities globally. To combat this issue, understanding the impact of a driver's visual surroundings is crucial in designing roadways that discourage speeding. This study investigates the influence of visual surroundings on drivers in 15 US cities using 3,407,253 driver view images from Lytx, covering 4,264 miles of roadways. By segmenting and analyzing these images along with vehicle-related variables, the study examines factors affecting speeding behavior. After filtering the images, to ensure an accurate representation of the driver's view, 1,340,035 driver view images were used for analysis. Statistical models, including hurdle beta and bivariate probit models with random driver effects as well as Machine Learning's eXtreme Gradient Boosting (XGBoost), were employed to estimate speeding behavior. The results indicate that factors within the driver's visual environment, weather conditions, and driver heterogeneity significantly impact speeding. Speeding behavior also varies across geographic locations, even within the same city, suggesting a connection between local context and speeding. The study highlights the importance of the driver's environment, showing that more open spaces encourage speeding, while areas with trees and buildings are associated with reduced speeding. Notably, this research differs from previous studies by utilizing real-time data from dash cameras, providing a dynamic and accurate representation of the driver's visual surroundings. This approach enhances the reliability of the findings and empowers transportation engineers and planners to make informed decisions when designing roadways and implementing interventions to address effectively excessive speeding. In addition to examining speeding behavior, the study also analyzes time-headway, a key factor affecting safety and risky driver behavior, to explore its relationship with speeding. The findings offer valuable insights into the factors influencing speeding and the driver's visual environment. These insights can inform efforts to create environments that discourage speeding (and close car following) and ultimately reduce severe accidents caused by excessive speed (and tailgating)., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Ltd. All rights reserved.)

Toll plazas are commonly recognized as bottlenecks on toll roads, where vehicles are prone to crashes. However, there has been a lack of research analyzing and predicting dynamic short-term crash risk specifically at toll plazas. This study utilizes traffic, geometric, and weather data to analyze and predict dynamic short-term collision occurrence probability at mainline toll plazas. A random-effects logit regression model is employed to identify crash precursors and assess their impacts on the probability of crash occurrence at toll plazas. Meanwhile, a Long Short-Term Memory Convolutional Neural Network (LSTM-CNN) network is applied for crash prediction. The results of random-effects logit regression model indicate that the flow standard deviation of downstream, upstream occupancy, speed difference and occupancy difference between upstream and downstream positively influence the probability of crash occurrence. Conversely, an increase in the proportion of ETC lanes negatively impacts the probability of crash occurrence. Additionally, there appears a higher likelihood of crashes occurring during summer at toll plaza area. Furthermore, to address the issue of data imbalance, Synthetic Minority Oversampling Techniques (SMOTE) and class weight methods were employed. Stacked Sparse AutoEncoder-Long Short-Term Memory (SSAE-LSTM) and CatBoost were developed and their performance was compared with the proposed model. The results demonstrated that the LSTM-CNN model outperformed the other models in terms of the Area Under the Curve (AUC) values and the true positive rate. The findings of this study can assist engineers in selecting suitable traffic control strategies to improve traffic safety in toll plaza areas. Moreover, the developed collision prediction model can be incorporated into a real-time safety management system to proactively prevent traffic crash., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2024 Elsevier Ltd. All rights reserved.)

The utilization of traffic conflict indicators is crucial for assessing traffic safety, especially when the crash data is unavailable. To identify traffic conflicts based on traffic flow characteristics across various traffic states, we propose a framework that utilizes unsupervised learning to automatically establish surrogate safety measures (SSM) thresholds. Different traffic states and corresponding transitions are identified with the three-phase traffic theory using high-resolution trajectory data. Meanwhile, the SSMs are mapped to the corresponding traffic states from the perspectives of time, space, and deceleration. Three models, including k-means, GMM, and Mclust, are investigated and compared to optimize the identification of traffic conflicts. It is observed that Mclust outperforms the others based on the evaluation metrics. According to the results, there is a variation in the distribution of traffic conflicts among different traffic states, wide moving jam (phase J) has the highest conflict risk, followed by synchronous flow (phase S), and free flow (phase F). Meanwhile, the thresholds of traffic conflicts cannot be fully represented by the same value through different traffic states. It reveals that the heterogeneity of thresholds is exhibited across traffic state transitions, which justifies the necessity of dynamic thresholds for traffic conflict analysis., (© 2024. The Author(s).)

Recent state-of-art crash risk evaluation studies have exploited deep learning (DL) techniques to improve performance in identifying high-risk traffic operation statuses. However, it is doubtful if such DL-based models would remain robust to real-world traffic dynamics (e.g., random traffic fluctuations.) as DL models are sensitive to input changes, where small perturbations could lead to wrong predictions. This study raises the critical robustness issue for crash risk evaluation models and investigates countermeasures to enhance it. By mixing up crash and non-crash samples under the traffic flow fundamental diagram, traffic flow adversarial examples (TF-AEs) were generated to simulate real-world traffic fluctuations. With the developed TF-AEs, model accuracy decreased by 8% and sensitivity dropped by 18%, indicating weak robustness of the baseline model (a convolutional neural network, CNN-based crash risk evaluation model). Then, a coverage-oriented adversarial training method was proposed to improve model robustness in highly imbalanced crash and non-crash situations and various crash risk transition patterns. Experiments showed that the proposed method was effective to improve model robustness as it could prevent 76.5% accuracy drops and 98.9% sensitivity drops against TF-AEs. Finally, the evaluation model outputs' stability and limitations of the current study are discussed., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023. Published by Elsevier Ltd.)

Objective: A lower helmet-wearing rate and overspeeding in Pakistan are critical risk behaviors of motorcyclists, causing severe injuries. To explore the differences in the determinants affecting the injury severities among helmeted and non-helmeted motorcyclists in motorcycle crashes caused by overspeeding behavior, single-vehicle motorcycle crash data in Rawalpindi city for 2017-2019 is collected. Considering three possible crash injury severity outcomes of motorcyclists: fatal injury, severe injury and minor injury, the rider, roadway, environmental, and temporal characteristics are estimated., Methods: To provide a mathematically simpler framework, the current study introduces parsimonious pooled random parameters logit models. Then, the standard pooled random parameters logit models without considering temporal effects are also simulated for comparison. By comparing the goodness of fit measure and estimation results, the parsimonious pooled random parameters logit model is suitable for capturing the temporal instability. Then, the non-transferability among helmeted and non-helmeted overspeeding motorcycle crashes is illustrated by likelihood ratio tests and out-of-sample prediction, and two types of models provide robust results. The marginal effects are also calculated., Results: Several variables, such as age, cloudy and weekday indicators illustrate temporal instability. Moreover, several variables are observed to only show significance in non-helmeted models, showing non-transferability across helmeted and non-helmeted models., Conclusions: More educational campaigns, regulation and enforcement, and management countermeasures should be organized for non-helmeted motorcyclists and overspeeding behavior. Such findings also provide research reference for the risk-compensating behavior and self-selected group issues under overspeeding riding considering the usage of helmets.

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., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023 Elsevier Ltd. All rights reserved.)

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., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023 Elsevier Ltd. All rights reserved.)

Vulnerable road users (VRUs) involved crashes are a major road safety concern due to the high likelihood of fatal and severe injury. The use of data-driven methods and heterogeneity models separately have limitations in crash data analysis. This study develops a hybrid approach of Random Forest based SHAP algorithm (RF-SHAP) and random parameters logit modeling framework to explore significant factors and identify the underlying interaction effects on injury severity of VRUs-involved crashes in Shenyang (China) from 2015 to 2017. The results show that the hybrid approach can uncover more underlying causality, which not only quantifies the impact of individual factors on injury severity, but also finds the interaction effects between the factors with random parameters and fixed parameters. Seven factors are found to have significant effect on crash injury severity. Two factors, including primary roads and rural areas produce random parameters. The interaction effects reveal interesting combination features. For example, even though rural areas and primary roads increase the likelihood of fatal crash occurrence individually, the interaction effect of the two factors decreases the likelihood of being fatal. The findings form the foundation for developing safety countermeasures targeted at specific crash groups for reducing fatalities in future crashes., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023. Published by Elsevier Ltd.)

The application of Computer Vision (CV) techniques massively stimulates microscopic traffic safety analysis from the perspective of traffic conflicts and near misses, which is usually measured using Surrogate Safety Measures (SSM). However, as video processing and traffic safety modeling are two separate research domains and few research have focused on systematically bridging the gap between them, it is necessary to provide transportation researchers and practitioners with corresponding guidance. With this aim in mind, this paper focuses on reviewing the applications of CV techniques in traffic safety modeling using SSM and suggesting the best way forward. The CV algorithms that are used for vehicle detection and tracking from early approaches to the state-of-the-art models are summarized at a high level. Then, the video pre-processing and post-processing techniques for vehicle trajectory extraction are introduced. A detailed review of SSMs for vehicle trajectory data along with their application on traffic safety analysis is presented. Finally, practical issues in traffic video processing and SSM-based safety analysis are discussed, and the available or potential solutions are provided. This review is expected to assist transportation researchers and engineers with the selection of suitable CV techniques for video processing, and the usage of SSMs for various traffic safety research objectives., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023 Elsevier Ltd. All rights reserved.)

Traditional safety research mostly relies on accident data to analyze the precedents to a crash. Alternatively, surrogate safety measures have the potential to proactively evaluate safety events. The era of connected vehicles and smart sensing has brought about tremendous innovations in safety research. GPS data from such vehicles form a useful case of big data analytics where surrogate safety measures have largely been unexplored. In this paper, we propose time to collision estimation from connected vehicle GPS data. The vehicle dynamics such as speed, acceleration, yaw rate, etc. are then coupled with geometric and non-geometric roadway attributes to understand the contributing factors for a traffic conflict. The dataset contains 2,568,421 GPS points from 14,753 unique journeys. 1:4 ratio of conflict to non-conflict events was used to select 15,258 samples with 28 independent vehicle dynamics, geometric, and non-geometric variables. Binary logit model was used to investigate the relationship of these variables with conflicts. Model results showed that out of 28 independent variables, 6 independent variables and 7 interaction variables were found significant. The results showed some interesting and unique relations of these variables with conflicts. Based on these significant variables, k-means clustering was performed to understand the threshold for the significant values for which the number of conflicts is significantly increased. Results from k-means clustering and two sample binomial proportion t-tests revealed that when absolute acceleration crossed 0.8 m/s 2 , conflict probability increased by 8 percentage points.​ Moreover, when the yaw rate crossed 8 degrees/s, the conflict probability doubled. Besides, vehicles traveling at more than 140% of the recommended speed limit increased conflict probability by 7 percentage points., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023 Elsevier Ltd. All rights reserved.)

In the twentieth year of the twenty-first century, humanity is facing an unprecedented global crisis owing to the COVID-19 pandemic. It has brought about drastic changes in the way we live and work, as well as the way we move from one place to another, namely transportation. Previous studies have preliminarily found that mobility, travel behavior, and road traffic safety status experienced great changes after the outbreak of the COVID-19. The objective of this study is to explore how crash patterns have changed, as well as the contributing factors of such changes and the heterogeneity between counties in Florida. Thus, data of COVID-19 cases, crash, socioeconomic factors, and traffic volume of 2019 and 2020 are collected. Preliminary analyses show a considerable reduction from March to June. Substantial changes are shown in the proportions of crashes by time of occurrence and injury severity. Two types of statistical models are developed to identify factors of (1) changes in the percentages of crashes by type and (2) the numbers of crashes by type. The developed models reveal various demographic, socioeconomic, and travel factors. After controlling other factors, the total numbers of crashes are 14% lower after the outbreak. The most significant reductions are observed in peak-hour (22%), while no significant change is found in fatal crashes. The results show that the number of crashes has significantly decreased even after controlling the traffic volume, but some crash types (e.g., fatal) did not show a significant reduction. The findings are expected to provide some insights into better transportation planning and management to ensure traffic safety in a possible future epidemic., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023. Published by Elsevier Ltd.)

Time-specific Safety Performance Functions (SPFs) were proposed to achieve accurate and dynamic crash frequency predictions. Unfortunately, some states do not have or archive the needed high-resolution traffic data to develop time-specific SPFs. This study proposes a novel iterative imputation method to impute the 100% missing volume and speed data from different states with similar crash rates. First, this study calculated the crash rates for 18 states and applied the One-Way Analysis of variance (ANOVA) test to group the states with similar crash rates. Second, as an example FL and VA, which both have traffic data, were used to test the proposed iterative imputation method. The results indicated that the imputed traffic data could capture the same traffic pattern as the real-collected traffic data. Further, the Mean Absolute Error (MAE) between the imputed Ln Volume and the real-collected Ln Volume for FL is only 2.47 vehicles for each segment for three hours. The MAE between the imputed Ln AvgSpeed and the real-collected Ln AvgSpeed for FL is only 1.36 mph. The Mean Absolute Percentage Error (MAPE) between the imputed Ln Volume and the real-collected Ln Volume is 11.07%. Meanwhile, the MAPE between the imputed Ln AvgSpeed and the real-collected Ln AvgSpeed is 7.40%. Finally, this study applied the proposed iterative imputation method to develop time-specific SPFs for the state without traffic data and compared the results. The results illustrated that the time-specific SPFs developed by imputed traffic data perfectly reflected the significant variables for both morning and afternoon peak models, with a prediction accuracy of 87.1% for the morning peak model., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023 Elsevier Ltd. All rights reserved.)

This study develops Safety Performance Functions (SPFs) for freeway weaving segments. Due to the coexistence of three different movements including through, merging, and divering traffic, the probability of crashes in weaving segments is higher compared to other segment types. Further, the traffic flow in this section is the most unstable. Hence, to analyze detailed traffic conditions, this study utilized lane-level traffic data. The SPFs were developed using the Poisson Lognormal (PLN) regression model technique. The results showed that different traffic parameters were significant based on the types of crashes. For the rear-end crashes model, more general traffic conditions of the weaving segment were found to be significantly associated with the crash frequency such as the natural logarithm of average speed of through lanes. Nevertheless, for the sideswipe and angle crashes models, the traffic variables which are directly related to the weaving movements were selected as significant factors such as the off-ramp volume ratio, and standard deviation of speed of the rightmost lane. The results presented in this study can be meaningful in that they can serve as a basis for the weaving segments related safety evaluation studies. In addition, the developed models' results can be a great source to establish operational strategies to improve traffic safety on freeway weaving segments., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023 Elsevier Ltd. All rights reserved.)

Driving characteristics often vary between the different states of the signal. During red and yellow phase, drivers tend to speed up and reduce the following distance which in turn increases the possibility of rear end crashes. Intersection safety, therefore, relies on the correct modelling of signal phasing and timing parameters, and how drivers respond to its changes. This paper aims to identify the relationship between surrogate safety measures and signal phasing. Unmanned aerial vehicle (UAV) video data has been used to study a major intersection. Post encroachment time (PET) between vehicles was calculated from the video data as well as speed, heading and relevant signal timing parameters such as all red time, red clearance time, yellow time, etc. Random parameter ordered logit model was used to model the relationship between PET and signal timing parameters. Overall, the results showed that yellow time and red clearance time is positively related to PETs. The model was also able to identify certain signal phases that could be a potential safety hazard and would need to be retimed by considering the PETs. The odds ratios from the models also indicate that increasing the mean yellow and red clearance times by one second can improve the PET levels by 10% and 3%, respectively., (© 2023. The Author(s).)

In the diverging area of toll plazas, the absence of lane markings, the gradual widening of lanes, and the crossing of vehicles with different tolling methods increase the likelihood of collisions. This study proposed a concept of motion constraint degree to investigate traffic conflict risks in the toll plaza diverging area. On the basis of the motion constraint degree, a two-step method was developed, in which all potentially influencing factors were separated into two parts. The first part was used to analyze the association between the motion constraint degree and some factors, while the remaining factors were utilized for risk regression/prediction together with the motion constraint degree. The random parameters logit model was applied for regression analysis and four prevalent machine learning models were employed for risk prediction. Results indicate that (1) the proposed approach considering motion constraint degree outperforms the conventional direct method, no matter for conflict risk regression or prediction; (2) the motion constraint degree is not monotonically correlated with the risk level of vehicles; (3) due to the layout of the toll plaza, ETC vehicles are less likely to be at risk in the diverging area; and (4) lane-changing behaviors in the restricted space increase the conflict risk., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023 Elsevier Ltd. All rights reserved.)

To mitigate the risk of injuries, many countries recommend bicycle helmets. The current paper seeks to examine the effectiveness of bicycle helmets by performing a systematic review focusing on meta-analyses. First, the current paper explores the findings of studies that employ meta-analyses using bicycle crash data. Second, the results are discussed considering the findings from research analyzing bicycle helmet effectiveness in a laboratory using simulation, and then are complemented with key methodological papers that address cycling and the overall factors contributing to the injury severity. The examined literature confirms that wearing a helmet while cycling is beneficial, regardless of age, crash severity, or crash type. The relative benefit is found to be higher in high-risk situations and when cycling on shared roads and particularly preventing severe head injuries. The results from the studies performed in laboratories also suggest that the shape and size of the head itself play a role in the protective effects of helmets. However, concerns regarding the equitability of the test conditions were found as all reviewed studies used a fifty-percentile male head and body forms. Lastly, the paper discusses the literature findings in a broader societal context., (© 2023. The Author(s).)

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., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2022 Elsevier Ltd. All rights reserved.)

This paper evaluates the effectiveness of Rectangular Rapid Flashing Beacons (RRFB) on crash severity. The study used and compared XGBoost and Random Parameters Discrete Outcome Models (RPDOM) respectively. The dataset comprises of 312 pedestrian crossing locations, among which 154 treatment locations were provided with the Rectangular Rapid Flashing Beacons (RRFB) and 158 control locations without RRFB. These control locations have similar roadway, traffic, and land use characteristics of that of the treatment locations but are not treated with RRFB or other pedestrian crossing countermeasures. This study shows the impact of RRFB and other factors on severity of nighttime, pedestrian, total and rear-end crashes. Crash severity data was compiled from driver, vehicle, and event level data of each crash. Due to availability of larger number of observations for total (35,553), rear-end (15,675) and nighttime crashes (8,144) XGBoost was used, and due to less observations for pedestrian crashes (369), it was modeled using RPDOM. The results showed positive impact of RRFB for the reduction of nighttime crashes. It was noted that RRFB reduces the K and A nighttime crashes according to the SHAP values from the XGBoost model but does not have the desired significance for rear end and overall total crashes in the study area. From the RPDOM, it was seen that RRFB showed statistically significant reduction in injury severity of pedestrian crashes and nighttime crashes. To compare the two models, nighttime crashes were modeled using both the techniques, the prediction accuracy of XGBoost Model was 97% which was much greater than that of the RPDOM at 73.8% prediction accuracy. Thus, both XGBoost and the RPDOM model for showed positive impact of installing RRFB in reducing the severity of nighttime crashes., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2022 Elsevier Ltd. All rights reserved.)

Vehicles frequently change lanes at weaving segments, and there is a high probability of collision. To assess the risk of lane change, this study proposes a novel lane change risk assessment index (LCRAI), which takes both exposure time and conflict severity into consideration. First, the generalized time-to-collision (GTTC) and exposure-to-risk index (ERI) are proposed to evaluate conflicts during lane change, and then a spatiotemporal overlap analysis is applied to exclude the cases when two vehicles arrive at the conflict area at different times. Second, the change of velocity after a crash occurrence is calculated and used to evaluate the severity risk (SRI). Finally, ERI and SRI are combined to compute the LCRAI. A case study of a weaving segment is conducted, and the weaving segment is divided into four sections to compare their risks. Conflict frequency on the section close to the exit is the largest and it is smallest after passing the exit, while conflict severity is most serious just after entering the main line on the middle lane. As for both conflict frequency and severity, LCRAI shows that the middle section is the most dangerous while the last 100 m section is the safest in the weaving segment. The index proposed will contribute to the risk assessment of lane change in connected environments and improve safety in weaving segments of expressway combined with real-time safety countermeasures., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2022 Elsevier Ltd. All rights reserved.)

Objective: Several studies have confirmed the existence of a safety-in-numbers effect in relation to vulnerable road users. The safety-in-numbers effect refers to a phenomenon wherein the number of bicyclists/pedestrians on a road is higher, and consequently, the risk of each bicyclist/pedestrian being involved in a crash is lower. Nevertheless, the existence of the safety-in-numbers effect in the aspect of injury severity in traffic crashes has not yet been investigated. Thus, this study aimed to explore whether traffic injuries are more (less) severe with fewer (more) pedestrians/bicyclists at the county level., Method: Using two fractional split multinomial logit models, the relationships between the number of bicyclists/pedestrians and the proportion of crashes involving bicyclists/pedestrians based on crash severity were investigated at the county level using crash data from Florida. In other words, we explored whether differing number of bicyclists/pedestrians could change the distribution of traffic injury severity levels., Results: The modeling results clearly revealed a lower proportion of severe injuries caused to bicyclists/pedestrians at a higher level of daily bicycle/pedestrian flows, indicating existence of the safety-in-numbers effect. Several variables (e.g., the percentage of people aged 65 years and older, the percentage of commuters using public transportation, and the proportion of recreational land use) were found to have a significant effect on the distribution of traffic injury severity among bicyclists/pedestrians., Conclusion: This study proves that a safety-in-numbers effect exists in the aspect of injury severity among bicyclists and pedestrians., Practical Applications: These findings are expected to provide recommendations for promoting the use of active transportation, which will improve the safety of vulnerable road users in future., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Published by Elsevier Ltd.)

In the within-intersection area, vehicles from different approaches make turning movements resulting in many conflict points. Hence, drivers are more prone to make mistakes in that area, which leads to severe crash outcomes. In the current roadway system, the Closed-Circuit Television (CCTV) cameras could be a cost-effective sensor to monitor the safety condition in the within-intersection area. This study proposed a framework named "Near Miss Event Detection System (NMEDS)" for road safety diagnostics using video data collected from CCTV cameras. The proposed framework combined the Mask-RCNN bounding box detection and Occlusion-Net detection algorithm to reconstruct vehicles' key points in a 3D view. Vehicles' key points including right-front headlight, left-front headlight, right-back taillight, and left-back taillight could be identified and transformed into a 2D bird's-eye view (i.e., real-world coordinate system) for safety analysis. A method was proposed to modify the occluded key points, which could not be observed by cameras due the turning movements in the within-intersection area. The post-encroachment time (PET) was calculated by using the trajectory data in the 2D view. The proposed framework was compared with two counterparts (i.e., bounding box detection only and key point detection only) by conducting an empirical study at a 4-leg intersection. The results suggested that the proposed framework could obtain more accurate vehicle trajectory and better autocorrelation analytics was conducted to identify the significantly dangerous locations in the within-intersection area. It is expected that the proposed methods could help diagnose road safety problems using CCTV cameras. Moreover, the proposed method could be incorporated with Connected Vehicle Systems and provide information to nearby drivers based on Infrastructure-to-Vehicle (I2V) technologies., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2022 Elsevier Ltd. All rights reserved.)

Pedestrians' red-light crossing can present a threat to themselves and the safety at intersections, especially for the through vehicles since their speeds are higher compared to the turning vehicles. The automated emergency braking (AEB) system could actively detect pedestrians and react to avoid potential conflicts. This study contributes to evaluate the effectiveness of the AEB system under occlusion conditions. The braking algorithm was developed in the virtual simulator CARLA to control the ego vehicle. Three occlusion scenarios in which the sensor of the AEB system could not detect the pedestrian if the pedestrian is occluded by a stopping vehicle. The evaluation experiments were conducted at a typical 4-leg intersection considering different motion statuses of the ego vehicle and pedestrian. The effects of field of view (FOV) of the sensor and activation threshold of the AEB system were also explored. The study indicated that the effectiveness of the AEB system could be reduced by the occlusion time. A longer activation threshold is recommended if the pedestrian is potentially occluded for a long time. The effects of other factors such as the speed of the ego vehicle and pedestrian and scenarios were also identified., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2022 Elsevier Ltd. All rights reserved.)

Pedestrian deaths constitute 23% of road traffic deaths globally. Although several research papers have contributed to pedestrian safety analysis, they did not provide a comprehensive overview of the progress in the research domain and publication trends. This makes it difficult to identify trends and insights into the pedestrian research domain in light of the voluminous number of papers. This study fills this gap with a scientometric analysis of research on pedestrian safety analysis indexed in the Web of Science. The scope covers 2594 papers published between 2010 and 2021 in English. This study analyzed the annual publications and citation trends, top ten most cited papers, influential papers in their first three years after publication, contributing authors, funding agencies, and contributing journals. The regional gaps between the proportion of pedestrian deaths and research were also analyzed. The results showed low research productivity from low and middle-income countries although they have a high incidence of pedestrian deaths. Subsequently, the main keyword clusters or frontier topics were identified and topic analysis was employed to identify the evolution of studies. Four keyword clusters were identified, i.e., "vehicle-to-pedestrian crash and injury severity analysis", "pedestrian movement and decision simulation experiments", "improving the vehicle system towards reducing body region impact injuries", "pedestrian behavior in crosswalks and signalized intersections". This study contributes an integrated knowledge map of pedestrian safety analysis, publication trends, the evolution of studies, and under-researched topics to guide future research work in pedestrian safety analysis., (Copyright © 2022 Elsevier Ltd. All rights reserved.)

Secondary crashes usually occur within the spatio-temporal impact ranges of primary crashes, which could cause traffic disturbance and increase traffic safety problems. However, existing studies only focused on predicting the likelihood of crashes leading to secondary crashes without considering the likelihood of the occurrence of secondary crashes. In addition, previous studies did not consider the real-time implementation of secondary crash likelihood prediction models and included too many features that were not available in real-time. A real-time secondary crash likelihood prediction model aims to predict the likelihood of secondary crashes in a short period (e.g., 5-10 min) and update every minute. The main objective of this paper is to develop a machine learning model to predict secondary crash likelihood in real-time. Two XGBoost models were developed for predicting the likelihood of crashes leading to secondary crashes and the likelihood of the occurrence of secondary crashes, respectively. A hybrid model was proposed to combine the results from the two developed models. Results indicated that the proposed hybrid model significantly improved the accuracy of secondary crash likelihood prediction. The proposed model has the potential to be applied in proactive traffic safety management system and prevent the occurrence of secondary crashes. Moreover, experimental results suggested that several features related to real-time traffic flow conditions were crucial for predicting secondary crash likelihood, such as the average traffic volume and average occupancy., (Copyright © 2021 Elsevier Ltd. All rights reserved.)

Introduction: Studies have proven that the crash possibility and crash type are not the same among different expressway segment types. However, few studies have conducted real-time safety analysis considering different segment types. This study aimed to explore the crash mechanism's heterogeneity for different segment types (i.e., merge, diverge, weaving, and basic segments)., Method: To enable in-depth exploration, this study used detailed traffic data, which were 0-10 min before crash, at 1-min intervals, and from five detectors of both the upstream and downstream to the target segment. This study analyzed the crash mechanism's heterogeneity from the following aspects: crash characteristics, significant crash contributing variables, and variables' importance. Based on this, a variables selection method was proposed to solve the huge dimension scale in modeling. Then, a nested logit model was built, which could consider the crash mechanism's heterogeneity, to quantitatively analyze the impact of crash contributing factors on the crash risk., Results: The results revealed that there are statistically significant differences in crash characteristics between each segment type. Additionally, the sources of most crash contributing factors were found to be significantly different in the spatial-temporal dimension between each segment type. Moreover, this study found that the weather parameter, indicating pavement's wet condition, had a similar effect on crash risk between different segment types. However, the geometry and traffic parameters had significantly different impacts between different segment types. Moreover, when the number of target segments' upstream ramps increases or when the distance between ramps and the target segment decreases, the crash risk would increase. Practical Applications: This study can be applied in the intelligent transportation system to improve traffic safety performance, especially in active traffic management systems., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2021 National Safety Council and Elsevier Ltd. All rights reserved.)

Introduction: Safety performance functions (SPF) are employed to predict crash counts at the different roadway elements. Several SPFs were developed for the various roadway elements based on different classifications such as functional classification and area type. Since a more detailed classification of roadway elements leads to more accurate crash predictions, multiple states have developed new classification systems to classify roads based on a comprehensive classification. In Florida, the new roadway context classification system incorporates geographic, demographic, and road characteristics information., Method: In this study, SPFs were developed in the framework of the FDOT roadway context classification system at three levels of modeling, context classification (CC-SPFs), area type (AT-SPFs), and statewide (SW-SPF) levels. Crash and traffic data from 2015-2019 were obtained. Road characteristics and road environment information have also been gathered along Florida roads for the SPF development., Results: The developed SPFs showed that there are several variables that influence the frequency of crashes, such as annual average daily traffic (AADT), signalized intersections and access point densities, speed limit, and shoulder width. However, there are other variables that did not have an influence in crash occurrence such as concrete surface and the presence of bicycle slots. CC-SPFs had the best performance among others. Moreover, network screening to determine the most problematic road segments has been accomplished. The results of the network screening indicated that the most problematic roads in Florida are the suburban commercial and the urban general roads. Practical Applications: This research provides a solid reference for decision-makers regarding crash prediction and safety improvement along Florida roads., (Copyright © 2021. Published by Elsevier Ltd.)

Active traffic management (ATM) strategies are useful methods to reduce crash risk and improve safety on expressways. Although there are some studies on ATM strategies, few studies take the moving vehicle group as the object of analysis. Based on the crash risk prediction of moving vehicle groups in a connected vehicle (CV) environment, this study developed various ATM safety strategies, that is, variable speed limits (VSLs), ramp metering (RM), and coordinated VSL and RM (VSL-RM) strategies. VSLs were updated to minimize the crash risk of multiple moving vehicle groups in the next time interval, which is 1 min, and the updated speed limits were sent directly to the CVs in the moving vehicle group. The metering rate and RM opening time were determined using mainline occupancy, the crash risk of upcoming moving vehicle groups, and the predicted time at which moving vehicle groups arrived at the on-ramp. The VSL-RM strategy was used to simultaneously control and coordinate traffic flow on the mainline and ramps. These strategies were tested in a well-calibrated and validated micro-simulation network. The crash risk index and conflict count were utilized to evaluate the safety effects of these strategies. The results indicate that the ATM strategies improved the expressway safety benefits by 2.84-15.92%. The increase in CV penetration rate would promote the safety benefits of VSL and VSL-RM. Moreover, VSL-RM was superior to VSL and RM in reducing crash risk and conflict count., (Copyright © 2021. Published by Elsevier Ltd.)

The automated enforcement system (AES) is an effective way of supplementing traditional traffic enforcement, and the traffic violation data from AES can also be effectively used for safety research. In this study, traffic violation data were used to analyze the influencing factors associated with traffic violations and to predict the probability of violations at intersections. The potential factors influencing violations include 24 independent factors related to time, space, traffic and weather. Results from a logistic model showed that the midday period, weekends, residential districts, collector roads, congested traffic conditions, high traffic flow, lower wind speed and low temperature would increase the probability of traffic violations. The probability of violations was predicted by the random forest algorithm, which was proven to be the best traffic violation prediction model among logistic regression, Gaussian naive Bayes, and support vector machine. Moreover, the proximity weighted synthetic oversampling technique (ProWSyn) method was applied to reduce the impact of the imbalance ratio (IR) and improve the model's prediction performance. The receiver operating characteristics (ROC) curves and Precision-Recall (PR) curves illustrated that the random forest algorithm using oversampling data had the best classifier prediction performance than undersampling data. The area under curve (AUC) and out-of-bag (OOB) error with IR = 1 reached 0.914 and 0.0787, which showed the better performance of the random forest algorithm using ProWSyn in dealing with imbalanced traffic violation data., (Copyright © 2021 Elsevier Ltd. All rights reserved.)

Speeding is one of the major contributing factors to traffic fatalities. Various speed management strategies have been proposed to encourage drivers to select more appropriate speeds. This study aims to explore the different effects of the speed management strategies on the speeding proportions at urban and suburban arterials. Probe speed data was used to calculate the speeding proportions. To overcome the variability of probe speed data caused by the signalized intersections, a new method was suggested to calculate the speeding proportion, and a fractional split model was estimated to adjust the probe speed data. A Beta regression model was developed to analyze the speeding proportion. A grouped random parameter modeling structure was adopted to realize the different effects of speed management strategies and other road attributes on speeding proportions by different road types. Besides, a fixed beta model was developed for the comparison. The results suggested the grouped random parameter model could provide better performance over the counterpart and could realize the different effects of road features and other contributing factors on the speeding of different roads. It is expected that the findings could help inform more appropriate road design in order to reduce speed limit violations on urban and suburban arterials., (Copyright © 2021 Elsevier Ltd. All rights reserved.)

This research develops safety performance functions and identifies the crash hotspots based on estimated vulnerable road users' exposure at intersections and along the roadway segments. The study utilized big data including Automated Traffic Signal Performance Measures (ATSPM) data, crowdsourced data (Strava), Closed Circuit Television (CCTV) surveillance camera videos, crash data, traffic information, roadway features, land use attributes, and socio-demographic characteristics. It comprises an extensive comparison between a wide array of statistical and machine learning models that were developed to estimate pedestrian and bike exposure. The results indicated that the XGBoost approach was the best to estimate vulnerable road users' exposure at intersections as well as bike exposure along the roadway segments. Afterwards, the estimated exposure was utilized as input variables to develop crash prediction models that relate different crash types to potential explanatory variables. Negative Binomial approach was followed to develop crash prediction models to be consistent with the Highway Safety Manual. The results show that the exposure variables (i.e., AADT, bike exposure, and the interaction between them) have significant influences on the two types of crashes (i.e., crashes of vulnerable road users at intersections and bike crashes along the segments). Further, the results indicated that the context classification is significantly related to crashes. Based on the developed models, the PSIs were calculated and the hotspots were identified for the two crash types. It was found that hotspots were more likely to be located near the city of Orlando. Coastal roadways were classified as cold categories regarding bike crashes. Further, C4 roadway segments were found to be significantly related to the increase of vulnerable road users' crashes at intersections and bike crashes along the segments., (Published by Elsevier Ltd.)

As the era of intelligent connected vehicles (ICVs) is approaching, a number of studies have investigated the potential benefits of ICVs, including the safety effects. Although previous studies agree that ICVs would significantly improve traffic safety, its quantified safety effects at different stages are still being debated. This study aims to estimate the ICVs' safety effects by market penetration rate (MPR) adopting a meta-analysis approach. The results from the meta-analysis indicate that the number of conflicts is exponentially reduced as the MPR goes up. For example, compared to the environment without ICVs, 4.2% and 17.4% of conflicts would decrease at the MPR of 10% and 50%, respectively. The effects are more obvious at higher MPR-43.4% of conflicts are expected to decrease at the MPR of 90%. From the case study in the United States based on the meta-analysis, it is expected that the MPR would reach 17-20% in the near future (2025) and 40-48% in 2035. The anticipated reduction in the number of fatal crashes would be 5% and 13%, in 2025 and 2035, respectively. The findings from this study will be useful for both public and private sectors to establish strategic plans to promote ICVs and identify their benefits at different MPRs., (Copyright © 2021 The Author(s). Published by Elsevier Ltd.. All rights reserved.)

In this paper, we present a data augmentation technique to reproduce crash data. The dataset comprising crash and non-crash events are extremely imbalanced. For instance, the dataset used in this paper consists of only 625 crash events for over 6.5 million non-crash events. Thus, learning algorithms tend to perform poorly on these datasets. We have used variational autoencoder to encode all the events into a latent space. After training, the model could successfully separate crash and non-crash events. To generate data, we sampled from the latent space containing crash data. The generated data was compared with the real data from different statistical aspects. t-Test, Levene-test and Kolmogrove Smirnov test showed that the generated data was statistically similar to the real data. It was also compared to some of the minority oversampling techniques like SMOTE and ADASYN as well as the GAN framework for generating data. Crash prediction models based on Logistic Regression (LR), Support Vector Machine (SVM) and Artificial Neural Network (ANN) were used to compare the generated data from the different oversampling techniques. Overall, variational autoencoder (VAE) showed excellent results compared to the other data augmentation methods. Specificity is improved by 8% and 4% for VAE-LR and VAE-SVM respectively when compared to SMOTE while the sensitivity is improved by 6% and 5% when compared to ADASYN. Moreover, VAE generated data also helps to overcome the overfitting problem in SMOTE and ADASYN since there is flexibility in choosing the decision boundary., (Copyright © 2020 Elsevier Ltd. All rights reserved.)

Safety Performance Functions (SPFs) have been widely used by researchers and practitioners to conduct roadway safety evaluation. Traditional SPFs are usually developed by using annual average daily traffic (AADT) along with geometric characteristics. However, the high level of aggregation may lead to a failure to capture the temporal variation in traffic characteristics (e.g., traffic volume and speed) and crash frequencies. In this study, SPFs at different aggregation levels were developed based on microscopic traffic detector data from California, Florida, and Virginia. More specifically, five aggregation levels were considered: (1) annual average weekday hourly traffic (AAWDHT), (2) annual average weekend hourly traffic (AAWEHT), (3) annual average weekday peak/off-peak traffic (AAWDPT), (4) annual average day of the week traffic (AADOWT), and (5) annual average daily traffic (AADT). Model estimation results showed that the segment length and volume, as exposure variables, are significant across all the aggregation levels. Average speed is significant with a negative coefficient, and the standard deviation of speed was found to be positively associated with the crash frequency. It is noteworthy that the operation of the high occupancy vehicle (HOV) lanes was found to have a positive effect on crash frequency across all the aggregation levels. The model results also showed that the AAWDPT and AADOWT models consistently performed better (the improvements range from 3.14%-16.20%) than the AADT-based SPF, which implies that the differences between the day of the week and peak/off-peak periods should be considered in the development of crash prediction models. The model transferability results indicated that the SPFs between Florida and Virginia are transferrable, while the models between California and the other two states are not transferrable., (Copyright © 2021 Elsevier Ltd. All rights reserved.)

Pedestrian and bicycle safety is a key component in traffic safety studies. Various studies were conducted to address pedestrian and bicycle safety issues for intersections, road segments, etc. However, only a few studies investigated pedestrian and bicycle safety for bus stops, which usually have a relatively larger volume of pedestrians and bicyclists. Moreover, traditional reactive safety approaches require a significant number of historical crashes, while pedestrian and bicycle crashes are usually rare events. Alternatively, surrogate safety measures could proactively evaluate traffic safety status when crash data are rare or unavailable. This paper utilized critical bus driving events extracted from GPS trajectory data as pedestrian and bicycle surrogate safety measures for bus stops. A city-wide trajectory data from Orlando, Florida was used, which contains around 300 buses, 6,700,000 GPS records, and 1300 bus stops. Three critical driving events were identified based on the buses' acceleration rates and stop time; hard acceleration, hard deceleration, and long stop. The relationships between critical driving events and crashes were examined using Spearman's rank correlation coefficient. All three events were positively correlated with pedestrian and bicycle crashes. Long stop event has the highest correlation coefficient, followed by hard acceleration and hard deceleration. A Bayesian negative binomial model incorporating spatial correlation (Bayesian NB-CAR) was built to estimate the pedestrian and bicycle crash frequency using the generated events. The results were consistent with the correlation estimation. For example, hard acceleration and long stop events were both positively related to pedestrian and bicycle crashes. Moreover, model evaluation results indicated that the proposed Bayesian NB-CAR outperformed the standard Bayesian negative binomial model with lower Watanabe-Akaike Information Criterion (WAIC) and Deviance Information Criteria (DIC) values. In conclusion, this paper suggests the use of critical bus driving events as surrogate safety measures for pedestrian and bicycle crashes, which could be implemented in a proactive traffic safety management system., (Copyright © 2020 Elsevier Ltd. All rights reserved.)

Pedestrian protection is an important component of road safety. Intersections are dangerous locations for pedestrians with mixed traffic. This paper aims to predict potential traffic conflicts between pedestrians and vehicles at signalized intersections. Using detection and tracking techniques in computer vision, pedestrians' and vehicles' features are extracted from video data. An LSTM (Long Short-term Memory) neural network is proposed to predict the pedestrian-vehicle conflicts 2 s ahead. The established model reaches an accuracy of 88.5 % at one signalized intersection. It is further tested at a new intersection, reaching the accuracy of 84.9 %, while the new data merely takes up 30 % of the training data set. This indicates that the proposed model is promising to be implemented at different locations. Moreover, the proposed model can also be applied to develop collision warning systems under the Connected Vehicles' environment., (Copyright © 2020 Elsevier B.V. All rights reserved.)

Pedestrian safety plays an important role in the transportation system. Intersections are dangerous locations for pedestrians with mixed traffic. This paper aims to predict the near-accident events between pedestrians and vehicles at signalized intersections using PET (Post Encroachment Time) and TTC (Time to Collision). With automated computer vision techniques, mobility features of pedestrians and vehicles are generated. Extreme Value Theory (EVT) is used to model PET and minimum TTC values to select the most appropriate threshold values to label pedestrians' near-accident events. A Gated Recurrent Unit (GRU) neural network is further used to predict these events. The established model reaches an AUC (Area Under the Curve) value of 0.865 on the test data set. Moreover, the proposed model can also be applied to develop collision warning systems under the Connected Vehicle environment., (Copyright © 2020 Elsevier Ltd. All rights reserved.)

Real-time crash potential prediction could provide valuable information for Active Traffic Management Systems. Fixed infrastructure-based vehicle detection devices were widely used in the previous studies to obtain different types of data for crash potential prediction. However, it was difficult to obtain data in large range through these devices due to the costs of installation and maintenance. This paper introduced a novel connected vehicle (CV) emulated data for real-time crash potential prediction. Different from the fixed devices' data, CV emulated data have high flexibility and can be obtained continuously with relatively low cost. Crash and CV emulated data were collected from two urban arterials in Orlando, USA. Crash data were archived by the Signal for Analytics system (S4A), while the CV emulated data were obtained through the data collection API with a high frequency. Different data cleaning and preparation techniques were implemented, while various speed-related variables were generated from the CV emulated data. A Long Short-term Memory (LSTM) neural network was trained to predict the crash potential in the next 5-10 min. The results from the model illustrated the feasibility of using a novel CV emulated data to predict real-time crash potential. The average and 50th percentile speed were the two most important variables for the crash potential prediction. In addition, the proposed LSTM outperformed Bayesian logistics regression and XGBoost in terms of sensitivity, Area under Curve (AUC), and false alarm rate. With the rapid development of the connected vehicle systems, the results from this paper can be extended to other types of vehicles and data, which can significantly enhance traffic safety., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2020 Elsevier Ltd. All rights reserved.)

Managed lanes (MLs) have been implemented as a vital strategy for traffic safety and management improvements. Previous studies involving MLs didn't examine the impact of connected vehicles' (CVs) lane configuration on freeway facilities with managed toll lanes. CVs are quickly expanding in the transportation industry and are among the most recent promising developments in traffic and safety engineering. In this study, several scenarios were tested using microscopic traffic simulation to determine the optimal CV lane configuration strategies while taking into consideration the market penetration rate (MPR%) of CVs and traffic conditions (i.e., peak, off-peak). Both safety (i.e., conflict frequency, conflict reduction) and operational (i.e., average speed, average delay) performance measures were included in the analyses. A Negative Binomial model was developed for investigating the factors that affect the safety measures. Tobit models were used to evaluate traffic operation. The results of the safety and operational analyses suggested that an MPR% between 10 % and 30 % was recommended when the CVs were only allowed in MLs. By converting one of the general-purpose lanes (GPLs) to a managed lane, the MPR% could reach 60 %. It was also concluded that restricting CVs to only the CV lane was not recommended. Lastly, the findings suggested that by allowing CVs to use all the lanes in the network (MLs, GPLs and CV lane), the optimal MPR% could reach between 70 % and 100 %. This study has major implications for improving MLs by recommending the optimal CV lane configuration and market penetration rate for each design., (Copyright © 2020 Elsevier Ltd. All rights reserved.)

Adaptive traffic signal control (ATSC) systems improve traffic efficiency, but their impacts on traffic safety vary among different implementations. To improve the traffic safety pro-actively, this study proposes a safety-oriented ATSC algorithm to optimize traffic efficiency and safety simultaneously. A multi-objective deep reinforcement learning framework is utilized as the backend algorithm. The proposed algorithm was trained and evaluated on a simulated isolated intersection built based on real-world traffic data. A real-time crash prediction model was calibrated to provide the safety measure. The performance of the algorithm was evaluated by the real-world signal timing provided by the local jurisdiction. The results showed that the algorithm improves both traffic efficiency and safety compared with the benchmark. A control policy analysis of the proposed ATSC revealed that the abstracted control rules could help the traditional signal controllers to improve traffic safety, which might be beneficial if the infrastructure is not ready to adopt ATSCs. A hybrid controller is also proposed to provide further traffic safety improvement if necessary. To the best of the authors' knowledge, the proposed algorithm is the first successful attempt in developing adaptive traffic signal system optimizing traffic safety., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2020 Elsevier Ltd. All rights reserved.)

Methods based on crash data analysis are effective in identifying intersections with a potential for safety improvement. However, it is well recognized that crash data suffer from several shortcomings and that there are clues to safety other than crash occurrence. The systemic approach is an alternative method to address safety issues. In this approach, a transportation agency is able to identify priority locations based on the presence of risk factors rather than actual crashes. To promote wider use of the systemic safety approach, this paper aims at developing and validating a procedure to rank unsignalised urban intersections for safety improvement based on the evaluation of risk factors by road safety inspections. The procedure assesses a Safety Index (SI) that measures the safety performance of unsignalised urban intersections. The SI is formulated by combining two components of risk: the exposure of road users to road hazards (Exposure) and the risk factors, which increase the probability of involvement in crashes (Risk Index). The SI is made of two elements, one related to vehicles and one to pedestrians. Twenty-three detailed safety issues and ten general safety issues are assessed to compute the vehicle Risk Index and the pedestrian Risk Index. Safety issues were selected considering that they are common issues and that effective remedial measures exist and have already proven their effectiveness. Finally, criteria for identifying and ranking safety issues were defined. The SI has two main practical applications. High risk intersections, where safety measures that can reduce crash frequency exist, can be identified and ranked by the SI. Specific safety issues, that give more contribution to unsafety, are pointed out in order to give indication about more appropriate safety measures according to the systemic safety approach. The procedure was validated with a sample of eighty-nine urban intersections located in Orange County (Florida, U.S.). For these intersections, the SI scores, the Empirical Bayes (EB) safety estimates, and the potential for improvement (PFI) were compared. The correlation between the SI scores and the EB estimates was highly significant both for vehicles (R 2  = 0.66) and pedestrians (R 2  = 0.58) as well as for the total crashes (R 2  = 0.68). The results from the Spearman's rank-correlation analysis provided further validation for the SI indicating that ranking from the SI and the EB estimates agree at the 99.9% confidence level for vehicles (ρ s  = 0.78), pedestrians (ρ s  = 0.93), and total (ρ s  = 0.93)., Competing Interests: Conflict of interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2020 Elsevier Ltd. All rights reserved.)

Introduction: A pedestrian crash occurs due to a series of contributing factors taking effect in an antecedent-consequent order. One specific type of antecedent-consequent order is called a crash causation pattern. Understanding crash causation patterns is important for clarifying the complicated growth of a pedestrian crash, which ultimately helps recommend corresponding countermeasures. However, previous studies lack an in-depth investigation of pedestrian crash cases, and are insufficient to propose a representative picture of causation patterns., Method: In this study, pedestrian crash causation patterns were discerned by using the Driving Reliability and Error Analysis Method (DREAM). One hundred and forty-two pedestrian crashes were investigated, and five pedestrian pre-crash scenarios were extracted. Then, the crash causation patterns in each pre-crash scenario were analyzed; and finally, six distinct patterns were identified. Accordingly, 17 typical situations corresponding to these causation patterns were specified as well., Results: Among these patterns, the pattern related to distracted driving and the pattern related to an unexpected change of pedestrian trajectory contributed to a large portion of the total crashes (i.e., 27% and 24%, respectively). Other patterns also played an important role in inducing a pedestrian crash; these patterns include the pattern related to an obstructed line of sight caused by outside objects (9%), the pattern that involves reduced visibility (13%), and the pattern related to an improper estimation of the gap distance between the vehicle and the pedestrian (10%). The results further demonstrated the inter-heterogeneity of a crash causation pattern, as well as the intra-heterogeneity of pattern features between different pedestrian pre-crash scenarios. Conclusions and practical applications: Essentially, a crash causation pattern might involve different contributing factors by nature or dependent on specific scenarios. Finally, this study proposed suggestions for roadway facility design, roadway safety education and pedestrian crash prevention system development., (Copyright © 2020 National Safety Council and Elsevier Ltd. All rights reserved.)

This study investigates the traffic conflict risks at the upstream approach of toll plaza during the vehicles' diverging period from the time of arrival at the diverging area to that of entering the tollbooths. Based on the vehicle's trajectory data extracted from unmanned aerial vehicle (UAV) videos using an automated video analysis system, vehicles' collision risk is computed by extended time to collision (TTC). Then, two time-varying mixed logit models including time-varying random effects logistic regression (T-RELR) model and time-varying random parameters logistic regression (T-RPLR) are developed to examine the time varying effects of influencing factors on vehicle collision risk, and four models including the standard random effects logistic regression (S-RELR) model, standard random parameters logistic regression (S-RPLR) model, distance-varying random effects logistic regression (D-RELR) model and distance-varying random parameters logistic regression (D-RPLR) are developed for model performance comparison. The results indicate that the T-RPLR model has the highest prediction accuracy. Eight influencing factors including following vehicle's travel distance, following vehicle's initial lane, following vehicle's toll collection type, leading vehicle's toll collection type, distance between two vehicles' centroids, and following vehicle's speed, are found to have time-varying effects on collision risk. Meanwhile, the first six factors are found to exhibit heterogeneous effects over the travel time. Another important finding is that the vehicle that comes from the innermost lane has an increasing trend to be involved in traffic conflicts, whereas the collision risks of other vehicles decrease as the travel time increases. Moreover, vehicles with higher speed have a decreasing probability to be involved in crashes over the travel time. Interestingly, the results of D-RPLR model are similar with that of T-RPLR model. These findings provide helpful information for accurate assessment of collision risk, which is a key step toward improving safety performance of the toll plazas' diverging areas., (Copyright © 2020 Elsevier Ltd. All rights reserved.)