This study aims to investigate the impacts of merge strategies of a ramp CAV on mainline human drivers. Previous studies evaluated CAV merge strategies mostly based on either the simulation or the restricted field testing, which lacks consideration of realistic driving behaviors in the merging scenario. To deal with this research gap, this study developed a multi-driver simulator system and considered realistic driving behaviors in the validation of merge strategies. Four CAV merge strategies were evaluated regarding their impacts on driving safety and comfort of the mainline human drivers. A set of driving safety and comfort metrics was adopted to verify the merge strategies. The results show that these algorithms might not have consistent performance when evaluated by different safety and comfort metrics. In addition, results revealed significant variations of the algorithm influences between the mergingand the following periods. Moreover, the AHS and GFM may have some superiority when evaluated at specific dimensions in terms of driving safety and comfort; nevertheless, the AHS may outperform other merge strategies in more scenarios. Findings suggest that the CAV merge strategy should not only ensure the ramp vehicle’s merging task but also consider mainline vehicles’ driving performance.

At an intersection, a crash between a pedestrian and a vehicle may occur under the occluded condition. An automated emergency braking (AEB) system could be utilized to actively detect pedestrians and react to avoid potential conflicts. This study contribution is evaluating 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.

The pedestrian-to-vehicle (P2V) technology is expected to reduce pedestrian crashes and improve roadway safety. Utilizing the smartphone as a communication platform could make the P2V more applicable for old cars without having additional retrofits. In UCF’s previous work, the effectiveness of a general P2V design has been demonstrated. However, the influence of different P2V designs remains uncertain. This research focuses on the influence of P2V designs in different scenario conditions and uncovers some insights about potential variations between drivers, for the sake of better informing drivers about potential pedestrian risk situations in the upcoming automation era. Two aspects of P2V design, i.e., the warning display mode and warning content, were tested in six pedestrian pre-crash scenarios. The warning display mode is categorized into a gradually changed warning and an emergency warning; and the warning content is referred to whether having specific distance information as a supplement or not. Thirty- six valid participants were tested in the simulator. The results demonstrate that the gradually changed warning and considering additional information would be better in terms of safety and driving performance. In addition, the effectiveness of the P2V design can be further improved when considering the scenario and drivers’ features.

The visual environment could have effects on the performance of automated vehicles within the V2X technology regarding traffic safety. This research aims to explore the effects of visual environment on traffic safety for the development of virtual simulation and driving simulator experiments. Both the effects on the speeding crashes and the severity of single-vehicle crashes were explored. To obtain the data of drivers’ visual environment in the real world, a framework was proposed to obtain the Google street view (GSV) images. Deep neural network and computer vision technologies were applied to obtain the clustering and depth information from the GSV images. To reflect drivers’ visual environment in the real world, the coordinate transformation was conducted, and several visual measures were proposed and calculated. Three different tree-based ensemble models (i.e., random forest, adaptive boosting (AdaBoost), and eXtreme Gradient Boosting (XGBoost)) were applied to estimate the number of speeding crashes and the comparison results showed that XGBoost could provide the best data fit. The explainable machine learning method were applied to explore the effects of drivers’ visual environment and other features on speeding crashes. The results validated the visual environment data obtained by the proposed method for the speeding crash analysis. It was suggested that the proportion of trees in the drivers’ view and the proportion of road length with trees could reduce speeding crashes. In addition, the complexity level of drivers’ visual environment was found to increase the crash occurrence.

Alternative intersection designs have been proposed due to their theoretical expected ability to simultaneously enhance traffic safety and operation as a result of reducing the number of conflict points and signal phases. However, this was only achieved at very limited intersection designs which have a very low number of conflict points and under certain traffic conditions. For example, the restricted crossing U-turn (RCUT) intersection, which has the lowest number of conflict points among other proposed intersection designs, has operational advantages at extremely unbalanced traffic volumes. Our shifting movement (SM) intersection design, which has the same number of conflict points as the RCUT intersection, has been proposed to replace the RCUT implementation at intersections with medium to high minor traffic volumes. It was proven that it outperforms the RCUT intersection which has medium to high minor traffic volumes in terms of average delay and throughputs. This study aimed to investigate the safety aspects of this intersection design by utilizing the driving simulator. The effectiveness of using infrastructure to vehicle (I2V) communication for mitigating the confusion at alternative intersections was also investigated in the study. The results indicated that RCUT and SM intersections have similar safety performance and crossing them is safer than crossing the conventional intersection. However, there is a need to improve drivers’ knowledge about the SM intersection, especially regarding the major left-turn movement. Most participants have found that using I2V communication is helpful in understanding the unconventional movement patterns.

Connected vehicle technology is expected to reduce crashes and improve roadway safety overall despite its effect being dependent on the content of crash scenarios. The reason behind this is that the heterogeneity between crash scenarios may cause variation in a driver’s perception and interpretation of the crash scenarios. Further, the heterogeneity may lead to different driver behaviors and evasive strategies. Consequently, both the benefits and influence of connected vehicle technology are affected. This project aimed to identify the variation of the performance of connected vehicle technology between different crash scenarios. Specifically, two types of connected vehicle technologies, forward collision warning (FCW) technology and pedestrian-to-vehicle (P2V) technology, were tested in four rear-end crash scenarios and three pedestrian crash scenarios, respectively. The results showed promising effectiveness of FCW and P2V technologies to reduce the possibility of a crash. Specifically, FCW reduced rear-end crashes by 56.6%-69.8%, and P2V reduced pedestrian crashes by 89.2%-97.2%. More importantly, the results captured a significant variation in the performance of FCW and P2V between crash scenarios. In different scenarios, the technologies aroused different driver brake operations, and, consequently, the technologies achieved different safety benefits. In addition, the interaction effects between technologies and driver features were affected by crash scenarios. Age, gender, crash/citation experience, and driving experience were found to affect the warning effect in different scenarios. This study has practical implications for the understanding of how heterogeneity of crash scenarios can affect connected vehicle technology.

The main objective of this study was to investigate the effect of different CV lane configurations and various market penetration rates on the safety and operation of the MLs network. Additionally, work will be done for studying the lower levels of automated vehicles (Level 1/Level 2) in a CV environment in the MLs network and determining the optimal market penetration rates of automated vehicle in the network under CV environment. This ongoing project is composed of four sections. Chapter 2 provides a brief review of previous studies of MLs, studies related to microsimulation and analyzing traffic conflicts, and studies related to connected and automated vehicles. Chapter 3 describes the microsimulation process for the studied corridor, which mainly included network building, calibration and validation, and CV scenario design. It also presents results and findings. Chapter 4 provides a description of the impact of dedicated lanes for CV platooning on expressways.

The recent advent of connected vehicles (CV) technologies could bring unprecedented opportunities to improve road safety, especially under reduced-visibility conditions. Reduced-visibility conditions increase the probability of rear-end crash occurrences and their severity. Moreover, slow traffic may be formed due to bottlenecks on freeways. This phenomenon may lead to higher rear-end crash risk when vehicles approach slow traffic, since drivers might not notice front vehicles’ speed reduction in time to respond.For the abovementioned reasons, this research investigates the CV crash warning systems that have the potential to improve vehicle safety by alerting drivers of imminent situations so they can take timely crash-avoidance action(s). This study provides a driving simulator study to evaluate the effectiveness of the head-up display (HUD) warning system and the audio warning system on drivers’ crash-avoidance performance when the lead vehicle makes an emergency stop under fog conditions. Drivers’ throttle release time, brake transition time, perception response time, brake reaction time, minimum modified time-to-collision, and maximum brake pedal pressure are analyzed. According to the results, the crash warning system could help decrease drivers’ reaction time and reduce the probability of rear-end crashes in a CV environment. In addition, the effects of fog level and driver characteristics, including gender and age, are investigated in this study. The findings of this study could help car manufacturers design rear-end crash warning systems that enhance the effectiveness of the system’s application under fog conditions.Furthermore, this study also aims to develop an integrated variable speed limit (VSL) and CV control strategy to reduce the rear-end crash risk at freeway bottlenecks under fog conditions. Based on the car-following model, the VSL control algorithm is developed considering the different relationships between gap and visibility distance. Then, a feedback control framework is developed to combine the VSL and CV control. The proposed VSL strategy is tested for a freeway section with a bottleneck through VISSIM, and the Intelligent Driver Model (IDM) is employed to build the CV environment. Finally, two measurements, time-to-collision at braking (TTC_brake) and total travel time (TTT), are employed to evaluate the effectiveness of the proposed control strategy. The results demonstrate that the VSL control played an important role in reducing the rear-end crash risk. The CV control could also enhance traffic safety by increasing the traffic homogeneity. Moreover, the combination of VSL and CV control (VSL&CV) could further enhance traffic safety and diminish the increase in travel time due to VSL.

Safety issues in school zone areas have been one of the most important topics in the traffic safety field. This research project assesses the safety effects of different roadway countermeasures in school zone areas. Although many studies have evaluated the effectiveness of various traffic control devices (e.g., sign, flashing beacon, speed monitoring display), there is a lack of studies exploring different roadway countermeasures that might have significant impacts on the school zone safety. In this research project, the most crash-prone school zone was identified in Orange and Seminole Counties, Florida, based on crash rate, which is defined as crash per thousand daily vehicle miles traveled. The results showed that Westridge Middle and Sadler Elementary schools were the top two crash-prone school zones. Afterward, a microsimulation network was built in VISSIM to test different roadway countermeasures in the school zones. Before applying different countermeasures, the network was calibrated and validated by traffic volume and travel time in order to replicate the real field. Three different countermeasures—two-step speed reduction, decreasing the number of driveways, and converting the two-way left-turn lane (TWLTL) to a raised median—were implemented in microsimulation and compared with the field condition. For each countermeasure, we also ran different sub-scenarios. In two-step speed reduction, we analyzed three sub-scenarios that were defined by the maximum speed limit on the main roadway. The number of driveways was reduced by 25%, 50%, 75%, and 100%, so four sub-scenarios were used to analyze in this countermeasure. We replaced TWLTL with a raised median, so all the left-turning vehicles made left turns either at the intersection or median. Therefore, two sub-scenarios, intersection U-turn and median U-turn, were analyzed. Surrogate safety measures are widely used as indicators to evaluate crash risk in the microsimulation software as it cannot directly measure the traffic crashes. In this research project, three surrogate safety measures were used; two of them were developed from time-to-collision (TTC) notations. Three surrogate safety measures—time-exposed time to collision (TET), time-integrated time to collision (TIT), and (3) time-exposed rear-end crash risk index (TERCRI) —were utilized in this research project as indicators for safety evaluation. The higher value of surrogate safety measures indicates higher crash risk. The results showed that all the sub-scenarios in two-step speed reduction and decreasing driveway access reduced TET, TIT, and TERCRI values significantly compared to the base condition. Moreover, the combination of two-step speed reduction and decreasing driveway access countermeasures outperformed their individual effects as well as the base condition. The one-way ANOVA analysis showed that all the sub-scenarios were significantly different from each other. Sensitivity analysis was also conducted to capture the impact of different sub-scenarios for different values of TTC threshold. The results show that all the sub-scenarios in two-step speed reduction and decreasing the number of driveway access reduced TET, TIT, and TERCRI values significantly for different values of TTC threshold, which ranged from 1 to 3 s. Conversely, for converting the TWLTL to the raised median, the crash risk was higher than the base condition because the value of TET, TIT, and TERCRI was much higher than the base condition. Therefore, the results of this research project provide useful insights for transportation and safety planners.

Cycling is encouraged in countries around the world as an economical, energy-efficient, and sustainable mode of transportation. Simulation is an important approach to analyzing the safety of cycling by identifying the effects of different factors. To ensure the success of a simulation study, it is essential to know the factors that have significant effects on bicycle safety. Although many studies have focused on analyzing bicycle safety, they lack bicycle exposure data, which could introduce biases for the identified factors. This study represents a major step forward in estimating safety performance functions for bicycle crashes at intersections by using crowdsourced data from STRAVA. Several adjustments considering the population distribution and field observations were made to overcome the disproportionate representation of the STRAVA data. The adjusted STRAVA data that includes bicycle exposure information was used as input to develop safety performance functions. The functions are negative binomial models aimed at predicting frequencies of bicycle crashes at intersections. The developed model was compared with three counterparts: a model using the un-adjusted STRAVA data, a model using the STRAVA data with field observation data adjustments only, and a model using the STRAVA data with adjusted population. The results revealed that the STRAVA data with both population and field observation data adjustments had the best performance in bicycle crash modeling. The results also addressed several key factors (e.g., signal control system, intersection size, bike lanes) that are associated with bicycle safety at intersections. It is recommended that the effects of these identified factors be explored in simulation studies. Additionally, the safety-in-numbers effect was acknowledged when bicycle crash rates decreased as bicycle activities increased. The study concluded that crowdsourced data is a reliable source for exploring bicycle safety after appropriate adjustments.