Intelligent Infrastructure – insight Prediction Modelling and Data Visualisation (DST)
Martin Mason, Network Rail - Programme Manager Development and R&D, Intelligent Infrastructure Programme
Remote Condition Monitoring for Train Doors
Christoph Schuessler, Data Scientist, Siemens Mobility
Track circuits and level crossings: Remote condition monitoring using Centrix
Dr Nick Wright AMIRSE Senior R&D Engineer, MPEC Technology Ltd
OptRail: FOAS for S&C Monitoring
Saeed Fararooy Director, RCM2 Ltd
REPAIR Project: Using data to enhance rail freight
Amar Ramudhin, Hull University and Chris Jones, Frazer Nash Consultancy
The morning event on Day 2, looked at how the vast quantity of data collected on the railways can be interpreted.
The second day began by hearing from Network Rail’s Martin Mason, Programme Manager Development and R&D Intelligent Infrastructure Programme and colleague Stephen Carpenter. The pair presented on their insight tool, which is a way to present intelligence to engineers using the intelligent monitoring techniques that Network Rail has deployed across its assets.
Martin explained that the key objective is to move maintenance planning earlier, meaning they can plan and conduct maintenance ahead of a potential failure. This has the potential to significantly reduce the costs and time needed for maintenance. On track for example, it requires up to 90 days to plan an access, or on an asset that could be up to 7 days to plan a visit.
On the tracks, they collect measurement on track geometry through measurement trains and ground penetrating radar. Ultimately, they create an actions inbox for the engineers to know what steps they need to take.
It also provides Network Rail and customers with trend data in how an asset trends. The most important is the decision support tool – it is all driven from frontline staff, so it is about presenting data in the way that they require it and it is most useful.
Christoph Schuessler from Siemens Mobility then covered their Railigent, providing a summary of their approach to monitoring train doors remotely, after they realized there are often issues with doors. Showing his passion for the area, Christoph said it was a “Really interesting task as a data scientist”.
The team are able to scan and monitor all doors of an entire fleet, 24/7. Following each abnormality detected, they alertthe maintenance teams at the depots to inspect and repair the trains. The upshots of identifying issues at such an early stage include reducing the downtime of trains and improving service reliability.