In order to deal with dynamic traffic flow, adaptive traffic signal controls using reinforcement learning are being studied. However, most of the related studies are difficult to apply to the real field considering only mathematical optimization. In this study, we propose a reinforcement learning-based signal optimization model with constraints. The proposed model maintains the sequence of typical signal phases and considers the minimum green time. The model was trained using Simulation of Urban MObility (SUMO), a microscopic traffic simulator. The model was evaluated in the virtual environment similar to a real road with multiple intersections connected. The performance of the proposed model was analyzed by comparing the delay and number of stops with a reinforcement learning model that did not consider constraints and a fixed-time model. In a peak hour, the proposed model reduced the delay from 3 min 15 s to 2 min 15 s and the number of stops from 11 to 4.7 compared to the fixed-time model.

The increasing traffic demand in urban areas frequently causes traffic congestion, which can be managed only through intelligent traffic signal controls. Although many recent studies have focused on reinforcement learning for traffic signal control (RL-TSC), most have focused on improving performance from an intersection perspective, targeting virtual simulation. The performance indexes from intersection perspectives are averaged by the weighted traffic flow; therefore, if the balance of each movement is not considered, the green time may be overly concentrated on the movements of heavy flow rates. Furthermore, as the ultimate purpose of traffic signal control research is to apply these controls to the real-world intersections, it is necessary to consider the real-world constraints. Hence, this study aims to design RL-TSC considering real-world applicability and confirm the appropriate design of the reward function. The limitations of the detector in the real world and the dual-ring traffic signal system are taken into account in the model design to facilitate real-world application. To design the reward for balancing traffic movements, we define the average delay weighted by traffic volume per lane and entropy of delay in the reward function. Model training is performed at the prototype intersection for ensuring scalability to multiple intersections. The model after prototype pre-training is evaluated by applying it to a network with two intersections without additional training. As a result, the reward function considering the equality of traffic movements shows the best performance. The proposed model reduces the average delay by more than 7.4% and 15.0% compared to the existing real-time adaptive signal control at two intersections, respectively.

Transit signal priority (TSP) plays an essential role in reducing traffic jams in a multimodal road traffic environment. Traffic experts expect that the demand for cars will be transferred to buses as the bus-service quality improves. This study proposes a multimodal environment TSP model with dedicated median bus lanes to simultaneously increase bus travel speed at signalized intersections and reduce car delay times. The proposed TSP model includes three optimization stages. The TSP is first optimized using a MAXBAND-based bandwidth model after a partitioned TSP control group is defined relative to a median bus stop to support non-stop bus movement. The optimal combination between these groups is selected using the delay-difference-of-offset technique. A kinematic wave-based offset optimization model is proposed to reduce the car delay time. We analyze the performance of this model in improving two-way bus speeds and reducing the car delay time in Seoul’s bus rapid transit system. The proposed TSP model decreases the stop ratio of buses by analyzing the bus trajectory from the micro-simulation. This model can be used as a center-based control to improve the mobility of buses and cars in a multimodal environment.

Automated buses are attracting increasing attention for enhancing the efficiency of existing bus systems. However, operational challenges persist for two-way automated buses on arterial roads, particularly in ensuring their on-time performance. This study proposes a center-based cooperative control between transit signal priority and vehicle guidance. The proposed model establishes an interactive structure between transit management and signal control systems within a traffic management center, which simultaneously optimizes traffic signal operation and bus trajectories. The control generates planned bus schedules and traffic signal timing plans through fixed-time strategies. When buses deviate from their schedules, the proposed model supports schedule recovery through reactive priority and vehicle guidance, including departure control at bus stations and adaptive speed adjustments. A numerical simulation was conducted in Sejong, Korea, under various boarding and alighting times and headway conditions. The results reveal significant improvements in bus route schedule deviations by 51.8–57.3 % compared to active priority and speed guidance with active priority. The proposed approach can enhance public transit reliability and urban traffic flow by effectively integrating automated buses with traffic signal operation.

We developed a congestion index for traffic management that uses link travel time based on global positioning system probe vehicle data. The congestion index can be used to evaluate the performance of an arterial network under oversaturated conditions. It considers congestion intensity and congestion duration, unlike existing congestion indices. After investigating current big traffic data, we selected the average link travel speed that could be collected at all times over a wide range to develop the congestion index. We used extra travel time as a new measure of effectiveness. The congestion criterion for monitoring oversaturated conditions was set as extra travel time with three congestion levels when traffic flow exceeded capacity. We devised the new congestion index by reflecting the three congestion levels to monitor oversaturated conditions macroscopically and to examine congestion characteristics microscopically. A spatiotemporal congestion recognition index (SCRIN) was defined based on congestion intensity and congestion duration. We analyzed SCRIN to monitor the traffic congestion of Seoul city using the data on travel speeds from the seoul transport operation and information service (TOPIS) for 2016.