evidoria

← Back to browse

Good practice Imported

Hangzhou ET City Brain — China's AI platform for smart city governance and traffic management

China · Hangzhou · See the China profile · See the Hangzhou profile

Evidence: Observational / pre–post Top 66% 53/100 · Ask Evidence Copilot about this practice

Hangzhou's ET City Brain, co-developed by the city government and Alibaba Cloud, uses computer vision and ML to optimise traffic signals and route emergency vehicles. By 2020, Hangzhou fell from 5th to 57th most congested city in China; ambulance response times fell 50%.

15 %
Average vehicle speed increase (pilot corridor) (2017-2018)
5th to 57th place
Hangzhou congestion ranking (among Chinese cities) (by 2020)
~50 %
Ambulance response time reduction (covered districts)

Details

Maturity
Scaling
Promoter
Hangzhou Municipal Government; Alibaba Cloud
Period
2017–present
Keywords
urban mobility, traffic management, emergency services, smart city

Context

Launched in 2017 as a joint initiative between the Hangzhou Municipal Government and Alibaba Cloud, ET City Brain integrates feeds from over 4,500 traffic cameras, traffic sensors, and emergency dispatch systems across the city's road network in Hangzhou, China. The platform uses computer vision and machine learning to dynamically adjust traffic signal timing, detect incidents such as accidents or stalled vehicles, and pre-clear routes for ambulances and fire trucks. By 2019 the system had expanded to more than 15 Chinese cities and to Kuala Lumpur, Malaysia, with an upgraded City Brain 3.0 released in March 2025.

Objectives

The platform aims to reduce urban traffic congestion and improve emergency response by using computer vision and machine learning to dynamically adjust traffic signals, detect incidents in real time, and pre-clear routes for emergency vehicles.

Activities

ET City Brain integrates live feeds from more than 4,500 traffic cameras, sensors, and emergency dispatch systems, using AI to dynamically retime traffic signals, detect incidents such as accidents or stalled vehicles, and pre-clear routes for ambulances and fire trucks across the city's road network.

Results

A peer-reviewed paper in IET Smart Cities (2019) reported that by 2018 average vehicle speeds on the pilot corridor had increased by 15%, and that Hangzhou's congestion ranking among Chinese cities fell from 5th to 57th place by 2020. Ambulance response times in covered districts dropped by approximately 50%.

Conclusions

The reported gains come mainly from a single peer-reviewed study of one pilot corridor and city-level congestion rankings; despite this limited evaluation scope, the platform has continued to expand, reaching more than 15 Chinese cities and one international deployment, with a third-generation version (City Brain 3.0) released in 2025.

Implementation

Indicative cost
High (€500k–€5M) — No official budget figure is disclosed in available sources; classified as high cost given the scale of infrastructure involved (4,500+ integrated cameras/sensors) and multi-year, multi-city expansion in partnership with a major cloud provider.
Time to results
Long (> 3 years) — Continuously operated since 2017, with expansion to 15+ cities by 2019 and a major platform upgrade (City Brain 3.0) in 2025; classified as long given nearly a decade of sustained deployment.
Staffing & skills
Hangzhou Municipal Government (public-sector lead), Alibaba Cloud (technology and cloud-computing partner)

Conditions for success

  • Dense network of traffic cameras and sensors integrated with a high-capacity cloud-computing partner
  • Direct integration with emergency dispatch systems for ambulance/fire-truck route clearance
  • Sustained multi-year government-industry partnership enabling iterative upgrades (City Brain 3.0 by 2025)

Common failure modes

  • Publicly reported outcome data comes almost entirely from a single 2019 peer-reviewed study of one pilot corridor, limiting independent verification of citywide or multi-city claims.

Where it fits

Governance type
centralized municipal government in partnership with a private technology company
Scale
citywide, expanded to 15+ cities and one international deployment
Income level
upper-middle-income (China)

Commonly funded by

National / regional programmes

Indicative funding routes for practices of this type — always check each programme's current calls and eligibility rules.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Similar practices you may find useful