Traffic speed up by 21 percent, congestion index down by nearly 19 percent, green light extended by 1–15 seconds where there is a jam

As reported by CCTV+, Beijing is deploying intelligent traffic management systems under the city’s “Dual Intelligence” programme, using artificial intelligence to reduce congestion and improve traffic flow.

In the Haidian District, the system already optimises traffic light timing in real time at busy intersections. Thanks to AI control, traffic lights have learned to “think”. Instead of fixed schedules, green lights are now automatically adjusted according to road conditions, extended by 1 to 15 seconds when congestion occurs and shortened when traffic clears.

Driver Zhou Shiguang shared: “Before, I had to wait for two or three green lights to get through the Weigongcun intersection during the evening rush hour. But now, after the system was adjusted, I can get through on the first green and still catch the next green light. Traffic has become much smoother.”

The system is currently in use at 19 intersections in Haidian District. In the Sidaokou area, where 13 intersections have already been upgraded, average vehicle speed has increased by 21 percent, and the congestion index has fallen by nearly 19 percent.

Zhao Tianmi, a representative of the Zhongguancun Science City management committee, explained: “Using technologies such as 3D modelling, spatial intelligence and digital twins, we create digital twins of road users — vehicles and pedestrians. Road conditions are accurately captured in real time by AI cameras, and precise calculations are made using AI algorithms. Finally, traffic lights are controlled by AI, shifting from the situation where vehicles watch the traffic light to the situation where the traffic light watches the vehicles.”

The system has also upgraded traffic enforcement. Thanks to the intelligent transport platform, enforcement has moved from “single-point recording” to “full-field automatic tracking”. Multi-camera fusion technology allows the system to focus on a single vehicle based on its continuous trajectory, capture licence plates from different angles and store the entire violation process.

Li Guang, deputy head of the traffic management department of the Haidian branch of the Beijing Public Security Bureau, noted: “Since the launch of the intelligent traffic management system, there has been a qualitative leap in traffic control. Non-local enforcement within the intelligent transport system can effectively reduce police involvement and improve traffic enforcement efficiency. As part of the overall deployment, we will upgrade the system in four directions: model localisation and scaling, multi-line coordinated management, and intelligent operation and maintenance.”

The “Dual Intelligence” programme was launched in December 2020 and encourages the search for synergies between urban development and intelligent mobility. As one of the first pilot cities, Beijing is leading the country in the deployment of intelligent roadside infrastructure and the creation of intelligent technology application scenarios.

The “Dual Intelligence” programme is a Beijing initiative to integrate AI and the Internet of Things into urban infrastructure to create “smart” transport systems.

Beijing’s traffic lights no longer just flash on a schedule. They watch the road, see the jams and adapt. Instead of making drivers wait for an empty lane, the green light now appears exactly when it is needed. Speed has increased by a fifth, congestion has eased. Cameras no longer just record violations but track the entire trajectory of a vehicle. AI has not replaced the traffic controller – it has become his digital assistant. The question is not when traffic lights will learn to think. They already have. The question is when other cities will adopt this experience. While some are stuck in traffic, others are already driving non-stop. And a city that breathes more freely is not science fiction. It is Beijing. Today.