Traffic congestion after major sporting events and entertainment performances has always been a headache inducing issue. People will navigate through crowded streets, causing more congestion and accidents that have yet to be resolved.
The latest article from Google Research suggests that one way to solve this problem is to use simulation models, which are virtual replicas of real-world transportation networks (sometimes referred to as "digital twins"), attempting to capture every detail from the layout of streets and intersections to vehicle flow.
The research institute claims that its team uses these models to quantify the sustainability impact of routes, test evacuation plans, and display simulated traffic in immersive views.
Using these models can enable traffic experts to alleviate congestion, reduce accidents, and improve the experience of drivers, passengers, and pedestrians as much as possible.
The long-standing challenge in this field lies in the calibration and matching of traffic models. The availability of comprehensive transportation data, detailed road network data from Google Maps, advancements in transportation science, and calibration techniques are paving the way for efficient computation of transportation networks worldwide.
New model
Google has established a basic model through open-source software - Simulated Urban Traffic (SUMO), locking the area near T-Mobile Park and Lumen Field in Seattle.
SUMO based models can help describe traffic dynamics, such as how drivers make decisions on vehicle following, lane changing, or speed limit compliance.
In addition, the researchers also introduced data from Google Maps to draw a heatmap of the traffic network structure and various static segmentation attributes (such as the number of lanes, speed limits, and the presence of traffic lights) in the area.
(Note: There is no difference in the heat map between the day of the competition and the day of the competition)
Afterwards, the research team divided the heat map into multiple small areas and introduced a "user behavior model" and route suggestions provided by the Seattle Police Department, thus establishing a "traffic diversion" model that can allocate the best route.
Realistic application
In order to test this technology in the real world, Google Research Institute collaborated with the Seattle Department of Transportation (SDOT) to develop a virtual model-based traffic congestion plan.
The research institute points out that our goal is to help thousands of participants in large-scale sports and entertainment activities quickly and safely leave the stadium area. During large-scale events, the model re planned 30% of the transportation routes, reducing the average time for vehicles to leave the congested central area by 7 minutes.
Google claims that this study can demonstrate the potential of simulation models in transportation planning, thereby improving transportation efficiency in large-scale events; And it can enable road planners to understand the low utilization sections, thereby improving the overall traffic environment and achieving better flow distribution.