Transport Department deploys AI for traffic and licensing

AI traffic signals, licence approvals and anti‑bot systems aim to boost efficiency across the city's transport network.

The Transport Department is deploying artificial intelligence (AI) across licensing and traffic management, including adaptive traffic signals at about 50 junctions and automated approval systems for vehicle licence renewals.

The department is implementing the Real-time Adaptive Traffic Signal System (RTATSS) at suitable signalised junctions across the city. A Smart Traffic Signals pilot project is also scheduled to roll out in the first quarter of 2027 at the junction of Gloucester Road and Great George Street in Causeway Bay.

RTATSS sensors will detect real-time vehicular and pedestrian traffic flows, with AI computing signal timings to dynamically allocate more green time to pedestrians and reduce maximum pedestrian waiting times by 15 to 20 seconds.

The department will launch the Easy Licence Application Approval project by the end of 2026, aiming to reduce vehicle licence renewal processing times from up to 10 working days to one.

“Using advanced large language model technology, this system will automatically recognise, cross-check, and verify application documents for online vehicle licence renewals,” the department said.

The application approval project will also cover two cross-boundary driving schemes: Northbound Travel for Hong Kong Vehicles and Southbound Travel for Guangdong Vehicles.

The department targets raising the proportion of cross-boundary driving permit applications automatically approved by AI to more than 50%, cutting manual processing times by about 20%.

An AI-enabled anti-bot solution and dynamic identification technology was deployed in late June for the online appointment booking system for the direct issue of Hong Kong Full Driving Licences.

The system identifies and blocks automated bot scripts and mitigates non-human operations to ensure fair and orderly access.

“As a result, the maximum waiting time for applicants successfully entering the booking system dropped significantly from up to two hours to within 45 minutes,” the department said.

The department is also studying an AI-powered Traffic Management Platform to identify traffic bottlenecks and formulate contingency plans to respond rapidly to special traffic conditions.

The platform aims to minimise delays during major events, adverse weather, or unexpected incidents.