Rail Digital Twin brings Train Movements into Sharper Focus

A new berth-topology system developed by Emu Analytics is combining live GPS feeds with operational railway data to give control teams a more precise view of train movements across the network.

The company says the technology, developed for a major UK train operator, can show where trains are at berth level and provide greater context around congestion, delays and disruption.

Built within Emu Analytics’ Flo.w digital twin platform, the system is intended to replace fragmented or delayed operational information with a live, spatial view of the railway. According to the company, the approach has already helped the operator reduce delays by giving staff a clearer picture of what is happening on the network.

Turning raw location data into operational insight

Traditional rail performance systems often rely on schematic berth diagrams or reports showing when a train has moved between fixed points. While useful, these systems may provide limited information about what happens between those points.

GPS data can offer a more detailed view, but raw coordinates are not always reliable enough to support operational decisions. Signals may be incomplete or unstable, and a train’s location needs to be matched with the correct track, berth and surrounding infrastructure before the information becomes meaningful.

Emu Analytics’ system combines GPS feeds with railway topology and operational data. Rather than presenting berths as abstract blocks, it maps them to their real-world position on the track.

This allows teams to see more precisely where trains are, assess congestion and investigate incidents. It also demonstrates a wider principle in applied data science: the value of data depends not only on how much is collected, but on how effectively different sources are cleaned, combined and placed in context.

Supporting real-time decisions

The platform is designed to present live information through a web-based interface that can be used by operational teams.

Alongside train locations, the system can show nearby stations, roads, track access points and local conditions. Emu Analytics says this could help staff respond more quickly to incidents involving stranded trains, severe weather or access to the railway.

The technology can also support retrospective analysis. By examining historical train movements alongside operational events, teams may be able to identify recurring causes of delay, such as timetable pressure, station congestion or patterns associated with particular routes.

The same data could be used to support route familiarisation, staff training and the development of live congestion maps.

Building a railway digital twin

Digital twins create dynamic representations of physical systems by combining data from assets, processes and their surrounding environment.

In rail, a digital twin can provide more than a map. It can connect live train movements with infrastructure, historical performance and operating conditions, giving users a more complete model of the network.

Emu Analytics argues that this could help operators improve punctuality, manage capacity more effectively and reduce costs associated with passenger compensation.

For data scientists, the project brings together several technical challenges, including geospatial analytics, real-time data processing, data fusion and visualisation. It also highlights the importance of designing systems around the people who will use them, particularly when insights need to support time-critical decisions.

From reactive to predictive operations

The longer-term aim is to use the detailed spatial and historical data generated by the system to develop predictive tools based on artificial intelligence and machine learning.

These tools could potentially identify emerging patterns associated with congestion or disruption before passengers are affected. This would represent a shift from reacting to incidents towards anticipating them.

The predictive capability remains an ambition rather than a demonstrated result. However, the berth-topology system provides the structured data foundation such models would need.

As transport operators increasingly adopt digital twins and data-led planning, the project offers an example of how combining live data with operational context can turn a fragmented picture into practical decision support.


References

https://news.railbusinessdaily.com/berth-topology-the-next-generation-approach-creating-a-clearer-picture-of-operational-performance/

https://arxiv.org/abs/2204.04085

Join Data Science Connects

Join Data Science Connects

Join Today! As a member of Data Science Connects the door opens to a world of opportunities and resources that can propel your career forward.

Read more

Data Science Connects News

Data Science Connects News

Stay Ahead with the Latest in Data Science – Dive In Now!

Read more

Become A Volunteer With Us

Become A Volunteer With Us

Become a volunteer to support charities and support in raising awareness of operational research in your area.

Read more