AI-integrated train deployment and crowd management systems adopted by Hong Kong's MTR
In a significant stride towards modernising Hong Kong's public transportation infrastructure, the MTR Corporation has implemented two AI-powered systems for train deployment and crowd control at Kai Tak Sports Park. These innovative tools, developed in collaboration with the Hong Kong University of Science and Technology, have proven to be instrumental in managing crowds and train deployment for major events.
The cornerstone of these AI-driven systems is the ridership prediction model. This sophisticated tool analyses extensive data, including billions of datasets from government surveys, MTR operations, and historical passenger flow during past major events at Hong Kong Stadium and the Hong Kong Coliseum. By simulating and forecasting passenger distribution and travel directions after an event, the model allows MTR to adjust train frequency and deployment efficiently, ensuring that large crowds are managed effectively.
The model's predictions are cross-checked with the analysis by the operations team to determine the frequency of trains needed to disperse crowds. This collaborative approach has led to nearly 90 percent accuracy during events such as Coldplay concerts, demonstrating the model's effectiveness.
The second AI-powered system, the intelligent crowd diversion system, uses real-time AI analysis of CCTV footage to monitor passenger density at critical points like exits and platforms. This enables dynamic crowd control measures and efficient dispersal of attendees from events, improving safety and passenger flow during over 20 major events at the park.
Together, these AI-driven systems help optimise train deployment and crowd management at Kai Tak Sports Park, supporting the handling of up to 50,000 spectators in large-scale concerts and sports events.
The MTR Corporation's chief of operations for engineering service and innovations, Chan Hing-keung, made a statement about the model's capabilities, expressing his satisfaction with the systems' performance and the positive impact they have had on the management of major events at Kai Tak Sports Park. The corporation also plans to consider expanding the applications of these AI systems in the future.
The implementation of these AI tools represents a significant step towards modernising Hong Kong's public transportation infrastructure, demonstrating the MTR Corporation's commitment to innovation and efficiency.
[1] MTR Corporation Press Release, July 2024. [2] Hong Kong University of Science and Technology Press Release, July 2024.
- The AI-driven ridership prediction model, a collaboration between the MTR Corporation and the Hong Kong University of Science and Technology, analyzes extensive data from various sources to simulate and forecast passenger distribution after events, enabling the MTR to adjust train frequency and deployment efficiently.
- The intelligent crowd diversion system, another AI-powered tool, uses real-time analysis of CCTV footage to monitor passenger density at critical points, ensuring dynamic crowd control measures and efficient dispersal of attendees from events at Kai Tak Sports Park.
- Leveraging artificial intelligence in these systems has significantly optimized train deployment and crowd management at Kai Tak Sports Park, supporting the handling of up to 50,000 spectators in large-scale concerts and sports events, demonstrating the MTR Corporation's commitment to innovation and efficiency.
