Digital twins: Helping West Coast trains to run on time?

Digital twins: Helping West Coast trains to run on time?

Construction giant Laing O’Rourke is using digital twin technology to begin solving problems that many people face in their daily lives: inefficient railways.

Speaking at our Internet of Manufacturing event in London, Dr Graham Herries, Laing O’ Rourke’s director of Digital Technologies, explained how the construction company is using its Optimised digital twin platform to rework the operations of the West Coast Mainline – one of the UK’s busiest and most important rail corridors, currently operating as Virgin West Coast.

Digital twins use real-time data to build a model of a real-world asset in order to manage and optimise use of an asset, sometimes using 3D modelling techniques to create a virtual representation of it as well. Linked with supply chain and enterprise asset management tools, digital twins can bring about more efficient use of assets, and enable predictive maintenance techniques.

Herries said that Laing O’Rourke’s digital twin of the network is able to model the movement of its trains, and coordinate when they need to return to depot for maintenance.

With each train needing to be serviced at least every 20,000 miles, it’s a complex logistical challenge to ensure that each train finishes where it needs to each day, is serviced regularly, corrective works performed – and, critically for passengers, keeps running to the network’s timetable.

Carried out by a human being, this scheduling work typically takes three hours a day and can only plan one day ahead. This tends to result in trains being over-serviced.

By modelling the network as a digital twin and using a heuristic maintenance scheduling algorithm, Herries said that the process can be reduced to just 19 seconds in Microsoft Azure –though, at present, it then undergoes 30 mins of precautionary human validation.

Using digital twins, Herries said that it’s possible to forward-plan for up to 23 days, significantly reducing over-maintenance. Rolling stock usage is also balanced across the fleet to ensure that certain trains aren’t subjected to more journeys than others.

The result, claimed Herries, should be a more reliable network, with fewer last-minute cancellations or platform swaps, reduced maintenance costs, and more balanced train usage. But is that happening?

Internet of Business says

The adoption of digital twins and other Internet of Things (IoT) technologies to track network usage could be a critical component in Britain’s railways at a time when emergency timetables, soaring fares, and accusations of the government bailing out private operators with public money are rife.

The gauntlet is certainly down for the controversial Virgin West Coast route. According to Network Rail performance statistics, the network is currently performing just above the national average, with a public performance measure (PPM) of 87.6 percent against an average of 87.4 percent – a year-on-year PPM drop of 2.1 percent.

According to Network Rail’s figures, every network in the UK is performing significantly worse year on year, with the sole exception of the Caledonian Sleeper service, which has shown a PPM uptick of four percent to 94.6 percent.

The UK’s best performing networks are TfL Rail (95.8 percent), London Overground (95.6 percent), and Merseyrail (95.5 percent). Virgin East Coast is the worst performing network, with a PPM of just 74 percent.

If digital twins can help to turn around failing networks, then that will be a significant proof of concept for other organisations to follow.

Additional reporting and analysis: Chris Middleton.