AI video analytics for railway: forecasting safe train operation with reinforcement learning
Anson Ng 2 min read
GNS Technology has been applying AI video analytics and machine learning to railway operations in Hong Kong since 2021. This case study explains one of our flagship projects: forecasting the allowable time of train operation using open weather data from the Hong Kong Observatory (HKO) and data collected from SACEM dataloggers.
The challenge
Railway operation control centres (OCC) need to know how environmental conditions affect safe train operation. Strong wind, for example, can limit how long and where trains may run safely. GNS was engaged in a 3-year consultancy to implement AI video analysis for the railway, and designed a system that predicts the time limit that trains can achieve under different environmental thresholds.
What we built
- Collect real-time railway operation status from SACEM dataloggers installed on 2XX trains, plus historical data since 2020
- Ingest open weather data from HKO, including real-time wind speed
- Find the threshold conditions for train operation against environmental conditions on different lines
- Compare collected data and use a reinforcement learning model to predict the achievable time limit
- Expose results through an API for third parties and a cloud-based web application for public enquiries
Outcomes
- The trial system can be integrated into the existing SACEM monitoring system used by MTR, for direct evaluation by the OCC and design teams
- The model targets 80–90% accuracy within ±15 minutes, giving OCC teams an additional reference for operation time and shutdown planning
- After internal testing, results can be published through a public cloud web application
Learn more
This project is registered on the EMSD I&T solution platform as reference S-1662: Forecasting allowable time of train operation using open weather data from HKO and collected data from SACEM datalogger using reinforcement learning model.
This case study was first published on the EMSD I&T platform in January 2024.
Related reading
- How we select providers: how to choose a systems integrator in Hong Kong and how to choose an AI solution provider.
- Privacy-preserving 3D area monitoring registered on Smart LAB.
- Our AI & IoT solutions and project references.
FAQ
What does this railway AI forecasting system do?
It combines SACEM datalogger data with Hong Kong Observatory weather data to forecast the allowable time trains can operate under different environmental conditions, targeting 80–90% accuracy within ±15 minutes. Control-centre teams use it as an additional reference for operation time and shutdown planning.
Where does the data come from?
Real-time railway operation status is collected from SACEM dataloggers installed on 2XX trains, together with historical data since 2020, and open weather data from the Hong Kong Observatory including real-time wind speed.
Can it integrate with MTR's existing systems?
Yes. The trial system is designed to integrate with the SACEM monitoring system already used by MTR, so OCC and design teams can evaluate results directly; outputs are also exposed through an API and a cloud-based web application.
How can this project be verified publicly?
The project is registered on the EMSD I&T solution platform as reference S-1662, where the full solution description is published for public verification.
What other GNS railway AI projects exist?
GNS also delivers real-time analytics over 100 million train logs a day, predictive maintenance that reduced hardware-related incidents by over 30%, and WiFi deauth detection with CNN-based train localisation in tunnels.
