Helping Network Rail detect lineside vegetation to reduce the cost of maintenance

Network Rail

We enabled Network Rail to accurately spot where vegetation encroached onto the tracks, reducing maintenance costs and meeting safety regulations.

95%

Foliage accurately detected to reduce hazards across miles of rail

Background

Assessing rail infrastructure including platforms, signals and tunnels are critical for passenger safety. With over 20,000 miles of rail, it’s time-intensive and costly. Network Rail sought analytics to understand potential network disruption and optimise maintenance schedules.

Solution

We built a lineside vegetation encroachment detection tool using geospatial analytics including video footage, 3D reconstruction and distance measurements. The model identifies structures like platforms, signals, tunnels and vegetation, measures track geometry and precise location, and uses these inputs to evaluate whether there is sufficient clearance.

Impact

The project supported the Digital Railway transformation programme, identifies foliage with 95% accuracy, and significantly reduces the cost of unnecessary maintenance.

background placeholder
Powerful

“AI presents a powerful capability to improve asset inspection using commodity hardware. This first-of-a-kind project with Network Rail has helped to demonstrate the potential of deploying this technology at scale.”

David Bholat, Customer Director

Faculty

capabilities.