AI helps analyze in-car video footage to identify scrap metal and materials along the sides of the railroad // Credit: Network Rail
Network Rail is using artificial intelligence software to find abandoned timber off the track so it can be removed faster and safer than current methods.
The Network Rail Southern region is testing One Big Circle’s video and AI technology to find old rail materials that can be reused or recycled.
Removing material from the sides of the tracks reduces the risk of injury to co-workers from slips, trips and falls, while also making the rails look cleaner.
On Network Rail’s Wessex line, “slips, trips and falls” are the number one cause of injury, especially since most of the work is done during dark hours, and trackside debris is a significant hazard.
A technology known as Automated Intelligent Video Review (AIVR) captures high-definition train point-of-view video from across the rail network and uploads it to the cloud for instant access. It then uses AI to analyze the video to find abandoned rails, sleepers, and bags of ballast, and uses GPS to map their locations. This allows maintenance teams to plan when and how items can be safely removed.
Network Rail chose the Wessex Line, the busiest of the railway networks, for its trials. The results of the trial will determine whether this technology will be deployed more broadly. The technology is expected to not only improve efficiency, but also improve safety, leading to economic savings.
Wayne Cherry, senior innovation engineer at Network Rail, said: “Technologies such as AIVR offer Network Rail a great opportunity to improve its efficiency as a business.
“Although AIVR has already been used in other parts of Network Rail, this is the first time the technology has been used in AI in this way and has the potential to be truly transformative.
“Scrap on the side of the railroad is not only unsightly, but also interferes with planned earthwork, blocks safe passage, and delays teams accessing parts of the railroad infrastructure for repairs during disruptions. There is a possibility.
“If we can identify and remove waste materials more safely and efficiently, we will not only ensure the safety of our colleagues, but also benefit the broader rail industry, passengers and taxpayers.”
Network Rail’s Safety and Health Advisor for Wessex Lines, Martin Shaftow, who is leading the project, said: The technology will play a key role in keeping frontline colleagues safe, finding and removing scrap more efficiently, and improving overall rail conditions for the benefit of passengers and local communities. I believe it will play an important role in
“Unfortunately, in recent years, railroads have become dumping grounds for discarded sleepers, scrap rails, surplus ballast bags and many other assets. There is no clear list of where the assets are located.
“The possibility of using high-definition video footage and AI to pinpoint the location of waste materials without my colleagues having to walk along the tracks presents a huge opportunity to improve safety.”
“Financially, not only can we recycle some of the scrap material and use the money generated to support railway operations, but we can also reuse some of the leftover material. For example, Bomac concrete sleepers. are no longer manufactured, but are still in demand as a replacement for sidings and some tracks.Thanks to this technology, 40 of these sleepers have been identified on a site between Yeovil and Weymouth, where It can be retrieved and stored for future use throughout the business, eliminating the need to purchase expensive new equivalents.”
“It is an exciting prospect to contribute to the potential cost savings for the industry by reusing or recycling this treasure trove of scrap materials, and we look forward to rolling it out even more broadly across our business, hopefully later this year.”
Emily Kent, co-founder and director of One Big Circle, said: “This is a very exciting AI application developed in collaboration with Network Rail experts.
“The AIVR system collects high-quality wayside imagery from across the UK and helps a wide variety of engineers and disciplines see what they need to see without being onsite.
“By adding further intelligence to that data, we are able to automatically detect and locate abandoned rails and other hazards along the track, allowing us to focus on specific issues and respond quickly and safely. increase.”
