Jonathan Schofield

Said the following about Intelligent Infrastructure Track Team:

“Network Rail’s Intelligent Infrastructure (II) Track Team

A fantastic, hard-working team that has designed a digital tool that will revolutionise the way track teams carry out maintenance work of the track asset. For the first time, track teams in the routes will be able to access all the information they need, from one digital source, to carry out the right work, at the right time, in the right place.

The team was tasked with creating a predictive single-interface tool that would tell engineers where, and most importantly when, a track fault was going to occur.

The tool - the Track Decision Support Tool (TDST) - will shift Network Rail from century-old ‘fix on failure’ regimes to ‘predict and prevent’ regimes – allowing work to be planned ahead and reducing speed restrictions and delays for passengers. It will transform safety for Network Rail’s (NR) track engineers by reducing the need for repeat manual inspections and through better targeted work - sending teams to use their expertise where it's required.

The team is a collaboration of Network Rail colleagues from engineering, business change and analytics disciplines, working with Atkins, NR’s partner and systems integrator.

Why was it needed?

As the operator of Britain’s railway infrastructure, NR manages a highly complex, multi-dimensional and ageing railway consisting of 20,000 miles of track, 23,000 switches and crossings. Add a further one billion passengers by 2030 and the challenge for NR to run a safe, high performing railway meant it had to look at new, innovative ways of working to meet current and future demands.

Historically, track teams have carried out maintenance based on fixed timescales or when an asset breaks, rather than intelligence-led data. Data was held in multiple systems, often in paper form and had to be pulled together manually to create work orders.

Testing testing

Collating, cleaning and streamlining data, the TDST has been built over the last year in direct collaboration with track teams in the routes. Every step has been tested with engineers on the frontline through workshops and views fed back to the team for further refining. NR is a data-rich organisation, but as one engineer said: “data, data everywhere and not a drop to use.” It was time to make data work; to collate it, isolate it, present it to engineers and allow them to make sense of it; to turn data into intelligence and into wisdom, allowing teams to carry out the ‘right work, at the right time, in the right place’.

How does it work?

Containing a range of key data sets, the TDST delivers a maintenance dashboard, underpinned by decision support capability for track geometry (TG). It aligns with track geometry outputs from NR’s measurement trains identifying rates of degradation and highlighting sites at risk of reaching Alert Limit (AL) and Intervention Limit (IL). The TDST allows engineers to filter information by TG Parameters, Engineer’s Line References, Track ID, Start & End Mileage, Sleeper Type and Track Category, allowing users to target the precise information they need.

An algorithm aligns trace data from the Track Geometry Measurement Trains to provide maintenance teams with run-on-run trace that can be used to analyse and demonstrate the deterioration of an asset over time. The TDST is completely mobile responsive on both Android and iOS devices. It allows engineers access to the data they need from any location at any time across the network.

What are the benefits?

- Drive ‘predict and prevent’ maintenance

- Algorithmic data that gives a predictive date for asset faults

- Data that shows the degradation rates of track assets*

- Avoid sending teams onto track for repeat or unnecessary work

- Allow complete reprioritisation of maintenance schedules

- Reduce ‘boots on ballast’ – teams carrying out visual inspections

- Prioritise work orders based on level of risk – right work, right time, right place

- Reduction in line speed restrictions and blocked lines due to geometry defects

- Clear data on maintenance work already carried out on a site to see effectiveness and dictate future work

*Graphs provide teams with degradation rates for a specific location over time, allowing them to carry out effective and timely work

A new digital direction

The team has travelled the length and breadth of the country, meeting engineers in their depots to gather the knowledge they needed to design the tool. Once the prototype tool was built they packed their suitcases again to start an intensive training and testing campaign across Britain to fine tune the final details of the tool.

Ian Dean, Network Rail’s principal engineer, said:

“The TDST will give us prediction analysis and is role specific. As an engineer I need to work out what maintenance and renewals I need. The TDST will highlight this and let me know exactly where, when and why I need to use my engineering skills.”

At its core the TDST is a step change to creating the digital railway engineer of the future. Engineers who have access to accurate, up-to-date, trusted data that will increase certainty across the railway. Engineers who can use their expertise to carry out evidence-based work where and when it matters. These predictive tools will drive the data-driven railway, an intelligence-led railway that predicts faults, and will continually build information about assets which is both concrete and analysable.

Conclusion

The track team, part of NR’s Intelligent Infrastructure Programme, is at the forefront of a digital data transformation programme that will deliver predictive data to help meet the challenges of Britain’s railway in the 21st century. The team has worked tirelessly, travelled relentlessly, toiled beyond their working hours to meet design, training and delivery dates of this new tool.“