Interactive and adaptive traffic management model at signalized intersections

The increasing urbanization rate along with the population inevitably increases the traffic density caused by the vehicles people use for transportation day by day. For this reason, traffic management has become one of the most critical elements in city life. The biggest problem on the roads people use for transportation is the signalized intersections where the roads intersect and prevent uninterrupted traffic flow. It is seen that signalization controls play a very big role in preventing delays in traffic. Ensuring the most efficient flow of traffic at these intersections is very important both for the country's economy by causing less fuel consumption and for the creation of a more sustainable environment by reducing carbon dioxide emissions caused by internal combustion vehicles. Within the scope of this thesis, studies that will make positive contributions to traffic flow were carried out and an adaptive traffic management model that can be operated on signalized intersections was developed. The adaptive model developed to efficiently provide and control traffic flow at intersections has phase operation in the minimum and maximum time range and phase skipping features according to traffic density. The most significant difference of this developed model from other signalized intersection controls is that it controls traffic flow at intersections not sequentially but with a flexible phase structure. In order to examine the results of the developed adaptive traffic management model and to solve possible problems, the adaptive model obtained as a result of the study was simulated one-to-one in the SUMO simulation program using real intersection data belonging to Heybe Intersection on the Kemer district road within the borders of Antalya province. The adaptive traffic management model obtained as a result of this simulation study was applied to this intersection and the obtained results were interpreted and evaluated. In this study, the improvement results obtained from the SUMO program were verified with different methods. Delay calculation was performed on the Heybe Intersection fixed-time signaling system data with Webster's theorem. In addition, the delay values of the vehicles were determined with the visual detection method via the camera and the delay parameter per vehicle was calculated. Finally, the Heybe Intersection was modeled in the AIMSUN simulation program and operated on the same traffic data with the SUMO simulation operation. Thus, three different verification methods were applied, namely Webster's theorem, the visual detection method and the AIMSUN simulation program. Similar results were obtained with the SUMO simulation output from all verification methods. Thus, the effect results of the designed adaptive intersection control model on the intersection were verified. Finally, a performance measurement sensor was designed in the study and placed at the Heybe Intersection. A system design that can perform real-time performance measurement at signalized intersections with Bluetooth signals was implemented. The performance measurement system, which can detect the delay values originating from the signaling system by measuring the travel times of the vehicles at the intersection arms via Bluetooth signals, was used to determine the effect of the developed adaptive intersection control model on the intersection all day long. The outcomes of the adaptive intersection control model developed in this study were assessed using a performance measurement system specifically designed for this purpose. According to the results, each subcomponent of the adaptive intersection control model positively impacts performance compared to the fixed-time signaling system. Key features, including minimum and maximum time interval phase time, phase skipping, the variable matrix structure, and adaptive phase configuration, were enhanced the efficiency of the signalized intersection. As a result of the study, the application of the developed adaptive traffic management model in the morning one-hour peak traffic at Heybe Intersection, where serves an average of 50,000 vehicles daily, reduced the average delay per vehicle by 28.1 seconds/vehicle, reduced the waiting time by 28.5 seconds/vehicle and increased the average speed by 0.91 kilometers/hour compared to the fixed-time signaling management. In addition, in the measurements made throughout the day with the designed performance measurement system, it was determined that the application of the developed adaptive traffic management model to Heybe Intersection provided an improvement of 38.42% in the average delay per vehicle compared to the fixed-time signaling management.