Project and Report (M.S.)--Virginia Polytechnic Institute and State University. M.S. 1992. / Vita. Abstract. Includes bibliographical references (leaf 77). Also available via the Internet.

Highway safety evaluation has traditionally been performed using crash data. However crash data based safety analysis has limitations in terms of timeliness and efficiency. Previous studies show that the use of surrogate safety data allows for earlier evaluation of safety in comparison to the significantly longer time horizon required for collecting crash data. However, the predictive capability of surrogate measures is an area of ongoing research. Previous studies have often resulted in inconsistent findings in the relationship between surrogates and crashes, one of the primary reasons being inconsistent definitions of a conflict. This study evaluated the effectiveness of certain surrogate measures (Acceleration-Deceleration profile, intersection entering speed of through vehicles, and Post Encroachment Time (PET)) in assessing the safety of opposing left-turn interactions at 4-legged signalized intersections by collection of time resolved video from eighteen selected intersections throughout Georgia. Overall, this research demonstrated that surrogate measures can be effective in safety evaluation, specifically demonstrating the use of PET as a surrogate for crashes between left-turning vehicles and opposing through vehicles. The analysis of data found that the selected surrogate threshold is critical to the effectiveness of any surrogate measure. For example, the required PET threshold was found to be as low as 1 second to identify high crash intersections, significantly lower than the commonly reported 3 second threshold. Non-parametric rank analysis methods and generalized linear modeling techniques were used to model PET with other intersection and traffic characteristics to demonstrate the degree to which these surrogates can be used to identify potential high-crash intersections without resorting to a crash history. Finally, the effectiveness of PET and its assistance to decision makers is also been demonstrated through an example that helped find errors in reported crash data.

Traffic platoons created by traffic signals may have impacts on the operations of downstream intersections because they change the arrival pattern and gap distribution of upstream traffic. There’s been a lot of research dealing with platoon effects on operations at signalized intersections, while very limited research has been done for that of unsignalized intersections. This research aims to develop a methodology for analyzing the platoon impacts on major-street left-turn (MSLT) delay at two-way stop-controlled (TWSC) intersections. The main idea is using a microscopic simulation tool to simulate different platoon scenarios in opposing through traffic, then applying regression models to capture the impacts of platoons on the delay of MSLT. Two platoon variables were adopted as a simplification of the complex platoon scenarios, making it practical to analyze the platoon effects on MSLT delay. The first two steps were to build simulation models for real-world unsignalized intersections and simulate scenarios with a combination of various factors related to platoons in VISSIM simulation. Calibrations of these simulation models based on field data were performed before simulation started. The next step was to define, derive and calibrate two platoon variables for describing the duration and intensity of platoon arrivals in the opposing through traffic, which effectively simplified the large combination of various factors. At last, the two platoon variables and their relationship with MSLT delay change factor were modeled with regression tools. A relationship between the two variables and the delay change factor was established, which indicated a positive effect by upstream platoons on MSLT delay and made it possible to quantify the impacts. The findings in this research could also be used for future research on left turn treatment regarding platoon or signal impacts.

Master of Science

This study evaluates the impact of pedestrian flow through vehicular delays in a proposed DLT intersection located in an urban area with a high vehicular and pedestrian traffic volume. In addition, it evaluates the pedestrian–vehicle conflict points, the effect on the other accesses after the implementation of the displaced left turns, and the consequences of reorganizing pedestrian flow. Three scenarios are evaluated through microsimulation; the existing one, which is a conventional cross-shaped intersection; a proposed DLT intersection with two displaced left turns on adjacent accesses; and a third one, similar to the last one, but with the inclusion of a pedestrian underpass. Results indicate that a proposed conventional DLT increases the vehicle delays of the intersection by 6.0% compared to the existing condition, mainly due to the high volume of pedestrians and the conflicts it generates. However, with the inclusion of a pedestrian underpass, vehicle delays are reduced significantly by up to 34.0%.

This research presents more realistic models for left-turn and through volume capacity by taking into account the probabilistic nature of the left-turn bay blockages and spillbacks at a signalized intersection under the leading phasing scheme with a short left-turn bay. Generally, the left-turn bay spillback situation has been overlooked in the leading left-turn signal because much attention has been given to the more common problem of left-turn blockage under the leading left signal. The left-turn spillback situation, however, might happen because the ratio of left-turning vehicle tends to be relatively high in the traffic after the occurrence of left-turn bay blockage. That is, left-turn bay blockage, spillback situations, left-turn capacity, and through capacity are closely connected with one another. Hence, this research estimates more precisely the capacity for through and left-turn movement by considering the left-turn bay blockage and spillback situations associated with left-turn bay under leading left-turn signal operations. In order to find general agreement between the results from this proposed model and a real-world situation, the developed capacity model is validated with the results from CORSIM simulations of a real-world signalized intersection. The binomial distribution is applied as the arrival distribution for through movement considering the characteristics of expected arrivals under heavy flow conditions. Finally, since left-turn bay blockage and spillback situation seem to have adverse impacts on each other, this research investigates if there are any dependent relationships between left-turn bay blockage and spillback. Here, this study confirmed that close relationships between left-turn bay blockage and spillback situations obviously exists.

The objective of this study was to provide an estimate of the protected dual left turn saturation flow rates in the Tucson area. The data were collected from dual left turn lanes at six intersections. The headway times were measured in order to calculate the saturation flow rates at each of the study sites. Observations were made at four intersections with left turn protected and at two intersections with permitted plus protected left turns. Saturation flow was measured from the third to the last vehicle in the queue. Observations were made during the time of expected peak traffic flow. A stop watch was used to measure the time intervals to the nearest tenth of a second. The saturation flow rates were then calculated and varied from 1621 to 2017 Vehicles Per Hour of Green Per Lane (VPHGL) for the inside lane to 1737 to 1802 VPHGL in the outside lane.

The objective of this study was to determine the safety impacts of right turn followed by U-turn movements (RTUT) at signalized intersections as well as median openings. RTUT movements are the most common alternatives to direct DLT movements(DLT). In order to achieve such data in a shorter amount of time, conflict analysis was chosen to be useful in this study as opposed to crash analysis. Additionally, data collection sites were divided dependent on certain geometric criterion and conflict data was recorded by the use of video recording equipment. Seven out the eleven conflict types used during the study were related to RTUT movements while the remaining observed conflicts were related to DLT movements. The safety comparison of right turns followed by U-turns to direct left turns at traffic signal sites indicated that DLT movements generated two times more conflicts per hour than RTUT movements. When the effects of traffic volumes have been taken into consideration, RTUT movements had a 5 percent higher conflict rate than DLT movements. At median opening sites, DLT movements generated 10 percent more conflicts per hour than RTUT movements. Furthermore, the other conflict rate, which takes the effect of traffic volumes into consideration, was 62 percent higher for DLT movements as compared to RTUT movements.Impacts of separation distance on safety of RTUT movements were investigated by a regression model. The model investigated impacts of U-turn bay locations and the number of lanes on major arterial on separation distance requirements. The model results indicated that U-turn bays located at signalized intersections and greater number of lanes on major arterials increases the minimum separation distance requirements. Finally, on four lane arterials U-turn distributions at median openings were analyzed to investigate how U-turns are accommodated at such locations. A u-turn regression model was developed to investigate impacts of median modifications on signalized intersection safety. The model results indicated that median modifications across the high volume driveways may cause safety problems at downstream signalized intersection.

This thesis reports the use of Paramics microscopic simulation software to model the differences between the performance of ten Christchurch intersections under the existing New Zealand road rule which requires left turning vehicles to give way to vehicles turning right into the same road, and a changed rule that would see the right turning vehicle have priority. Previous research concerning this issue is reviewed and the history of the existing road rule and recent moves to change it are discussed. At each of the ten intersections a range of traffic volume combinations was assessed and the journey times and queue lengths were compared. The ten intersections represent a range of different layouts and forms of control including give way signs, stop signs and traffic signals. The impact of a rule change on the use of shared lanes at intersections using a Paramics model of the Christchurch Central Business District, as developed for the Christchurch City Council, is also reported. Conclusions are drawn about which types of intersections and traffic volume combinations are likely to be affected by a rule change. The features of intersections that contribute to this susceptibility are identified and conclusions drawn about whether positive or negative effects are likely. It is concluded that there is no compelling intersection performance reason why the rule could not be changed. The successful implication of such change would require a review of the road network to identify critical intersections. Some monitoring and mitigation measures may also be required.

Augmented reality "AR" is a promising paradigm that can provide users with real-time, high-quality visualization of a wide variety of information. In AR, virtual objects are added to the real-world view in a real time. Using the AR technology can offer a very realistic environment for driving enhancement as well as driving performance testing under different scenarios. This can be achieved by adding virtual objects (people, vehicles, hazards, and other objects) to the normal view while driving in a safe controlled environment. In this dissertation, the feasibility of adapting the AR technology into traffic engineering was investigated. Two AR systems; AR Vehicle "ARV" system and Offline AR Simulator "OARSim" system were built. The systems' outcomes as well as the on-the-road driving under the AR were evaluated. In evaluating systems' outcomes, systems were successfully able to duplicate real scenes and generate new scenes without any visual inconsistency. In evaluating on-the-road driving under the AR, drivers' distance judgment, speed judgment, and level of comfort while driving were evaluated. In addition, our systems were used to conduct two traffic engineering studies; left-turn maneuver at un-signalized intersection, and horizontal visibility blockage when following a light truck vehicle. The results from this work supported the validity of our AR systems to be used as a surrogate to the field-testing for transportation research.