A Master of Science thesis in Civil Engineering by Mahmoud Ahmed Taha entitled, "Impact of Using Indirect Left-turns on Signalized Intersections," submitted in June 2015. Thesis advisor is Dr. Akmal Abdelfatah. Soft and hard copy available.
Most developing areas are suffering from traffic congestion problems. The majority of traffic delays in urban areas occur at traffic signals. Over the past few decades, several transportation agencies have been applying geometric changes, through adding more lanes, to signalized intersections in an effort to reduce delays. Because of the limited availability of right-of-way, many transportation agencies have started using unconventional traffic control systems for intersections to improve signal efficiency and reduce overall delays. Common unconventional left-turn control types such as the right-turn followed by a U-turn (RTUT) and a U-turn followed by a right-turn (UTRT) basically eliminate direct left-turn (DLT) movements at the intersection by rerouting left-turning vehicles away from the main junction. Following any of these alternatives reduces the number of phases and the average delay per vehicle at the main junction. However, it adds some additional travel time for left-turning vehicles and some delays at the U-turn locations. This thesis presents a parametric study to evaluate the impact of replacing direct left-turns with U-turns (either RTUT or UTRT). The main goal of this study is to determine the traffic operational performance of each alternative under different traffic conditions. Traffic signal evaluation (Synchro) and simulation (Vissim) tools were utilized in this study to determine the optimized signal timings and evaluate intersection performance for each left-turn control type, respectively. Many parameters were considered, such as the total traffic volume on the intersection, the percentage of vehicles on each approach, the turning percentage for each movement, and the U-turn locations. It was concluded that unconventional left-turn control types have less delay and travel time compared to the DLT, when the U-turn locations are 200 meters away from the main intersection. Also, the right-turn followed by a U-turn showed superior performance over the other left-turn control types, when the U-turn locations are 100 meters away from the main intersection. Furthermore, it is not recommended to have the U-turns at a distance less than 100 meters when using unconventional left-turn control types because of the queue spillback effect. Finally, both conventional and unconventional control types have comparable vehicle kilometers travelled (VKT).