In July, the first European Train Control System (ETCS) Business Readiness Congress took place over two days in London. This included topics such as driver and operator training, overcoming operational challenges, adopting new standards, new ways of testing, retro fitting ETCS to trains, and managing transition risks with implementing and operating ETCS.
ETCS is not just a new way of signalling trains, and to gain the full benefits of the system requires successful change management and a cultural transformation of how railways operate. This was made very clear by the speakers and the technology on display, including smart simulators.
Northern City Line lessons learned
Ollie Turner head of ERTMS at GTR, shared the lessons learned from implementing ETCS on the Northern City Line between Finsbury Park and Moorgate. From May, and after a year operating as ETCS overlay, this became the first commuter railway in Britain to run without conventional track side signals, and including Automatic Train Operations (ATO). Ollie explained how strong business engagement was achieved, including the use of specialist managers and close working with the trade unions.
An ETCS training course of five days duration was agreed versus the national assumption of 15 days. A monthly meeting with stakeholders including health and safety representatives was established for ETCS, completely separate from all other issues in the business. Union representatives were also upskilled with ETCS training, and visits to the Cambrian system were organised.
He stressed that it was important to be very clear about what was being tested during the proving trials and to make the best use of, and not lose, critical train access. It is important to start planning for systems proving earlier rather than later. The involvement of train planners is also vital. A 27-hour testing window was lost due to a possession planning error and more access was required than originally anticipated. The Office of Road and Rail (ORR) Authorisation for Placing In Service (APIS) is a thorough and tough process, but APIS does not equal an operational train ready for service, said Ollie.
Driver and test train resource need to be drawn from a pool of competent and available staff. Experienced drivers soon spot any problems arising from any incorrect design assumptions, and Ollie also mentioned that baseline upgrades should be avoided when driver training is underway.
To illustrate how important it is to manage ETCS as a system, Ollie explained that GSM-R is at the heart of ETCS and that adequate specialist telecoms resource is required. The Northern City Line project faced challenges due to constrained telecoms resources, as multiple top-priority initiatives competed for attention.
The railway has a set way of communicating with its frontline staff, which may not always be the best for a system change such as ETCS. Projects may need to think ‘outside the box’ for ways to communicate if they want to achieve results. Staff do not need to know the technical details, but they need to know the ‘why’ behind the new way of ETCS working.
Elizabeth line
Effective front-line communication was also a key message from Imran Chaudhry head of safety & assurance at GTS Rail which operates the Elizabeth line in London. GTS use a fleet of 70 Class 345 nine-car fixed formation Aventra units with AWS / TPWS, CBTC (including ATO) and ETCS Level 2 overlay. This includes operating the world’s only heavy rail auto-reverse system between Paddington and Westbourne Park.
Two near-identical speeding incidents were identified with driver’s wrongly managing the train speed to the higher intervention speed and not the line speed. Applying the company’s ‘Just Culture’ model and discussing the issue with the trade unions, revised instructions were developed and the failure mode raised with the drivers. Checks were added to the automated data feeds to look for similar instances, and a software update was provided to alert drivers to use the correct mode.
More issues came to light, which, given the complexity of the system, was to be expected explained Imran. The flow of information to drivers, trainers, assessors, and control teams became a concern, and the strong relationships that had been established with the trade unions through open, early, and transparent engagement was vital.
A driver was made available to the project to assist with updating key documents, such as the ‘Essentials Manual’. The language was changed to make sure the explanations resonated with drivers, trainers, assessors, and control teams. Stakeholders were also involved to ensure clear communications at all levels was in place. The lessons learned have been used as part of the businesses driver ‘RISK’ communication campaign. This has included podcast style communications, posters and pop-up banners, and interviews with key people involved with ETCS / CBTC operation. The campaigns regularly receive viewing figures of 80% of drivers.
Despite the problems driver turnover rates have been low at less than 3% (the predicted loss was 10%). Imran said the technology is “absolutely amazing” and is every bit as transformational as expected, and will probably be more so in time. Drivers and operators have found the system far easier to use than expected and it is very reliable. He finished by saying people are key, and to make sure they know who to go to when they need simple answers to complex questions.
Data for ETCS systematic failure analysis
There are many other benefits of ETCS data, as explained by Connor Philpott, ERTMS engineering subject matter expert at GTR. A Fully Automatic Selective Door Opening (FASDO) stopping window using Eurobalises was introduced at Drayton Park to trigger an automatic power changeover between overhead line and third rail when a Class 717 train came to a stand within a 10-metre window. This would allow the driver to focus on the passengers getting on and off the train and not have to worry about carrying out a manual power changeover. Platform 1 worked 90% of the time, however Platform 2 never worked. By analysing the data it was identified that the window had been placed too far forwards.
Using the data from the live system, a distribution curve was established to ensure a new value would capture as many instances as possible within the 10-metre window, while still allowing for slight overshoots. It was determined that moving the window 8 metres back would resolve the problem. Platform 1 was also adjusted slightly to get the same level of consistency.
The data played a key part in rectifying this problem and finding a solution. Connor explained a few more examples where ETCS data had been used to resolve problems, and he said that data is objective as it tells a story of exactly what happened, and it is recorded automatically. It allows analysis of trends to predict future failures and can provide key insights into complex behaviours to model solutions.
Dutch ERTMS
Thijs van Steen gave an overview of the Dutch ERTMS Programme and the Netherland’s migration and delivery strategy. He explained the changes that have been made to the programme and finished with some key learning points:
- Always keep the end in mind.
- No individual really grasps the whole challenge, and those who say they do are dangerous.
- System integration is a verb and requires strong management, as ERTMS requires changes in a complex ecosystem.
- Programmes must undo (or avoid) ‘the great divide’ and must rally operational disciplines and it may be necessary to slow signalling engineers down. It is important that ETCS is not seen as just a new way of signalling, as it requires changes in many aspects of operating a railway.
- Change is hard – it requires time, determination and stamina.
Daim Willemse of Pro Rail covered the testing of a simulation when preparing for ETCS. He explained how existing data and talent management was used by Pro Rail. Pro Rail’s existing signalling was ‘good’. It did all that was required and there was no need to replace it. “Why fix a system that still runs perfectly?” Daim said. This is often the case, and resistance to change a well-functioning legacy system can become a significant blocker to necessary change. However, Pro Rail was concerned how long it could support the existing technology, as knowledge and spares were quickly running out for example.
Daim stressed the importance of open engineering and knowledge transfer. Knowledge and the experience gained must be shared to achieve consistent implementation. A key part of the safety case for ETCS implementation must include the need to recruit and train the future workforce.
The railway industry is inherently action orientated and less good at planning, so to achieve successful ETCS implementation it is vital to use data driven processes, to set achievable goals, work with an actual problem, and expand on the experience of learning. Daim also covered the concept of a Personal Protective Strategy (PPS). People need competency and confidence to use a new system such as ETCS. It’s a case of use it or lose it. ETCS ‘traps’ are different to conventional signalling. People do not need to know lots of information, but they need to know what to do when faced with a new trap.
He gave an example of drivers at certain platforms needing to open train doors on the other side of a train and a PPS solution to potentially opening the ‘wrong’ side was to encourage drivers to face the side of the train next to the platform.
East Coast Digital Programme
Jessica Linhart, ETCS project manager for Network Rail Operations (East Coast), explained the operational lessons learned from the initial stage of the East Coast Digital Programme (ECDP) covering the first 100 miles of the East Coast mainline in England. ETCS overlay and signals away were completed in May 2025 and, from the December 2025 timetable change, 19 trains per hour will operate through the overlay area.
Key to the preparation for ETCS was achieving stakeholder buy in from trade unions, frontline staff, safety and human factors, directors, and managers, explained Jessica. She recommended “making it simple before adding complexity” and to ensure stakeholders are involved in such things as panel design, route setting functions, and workspace design. Non-ETCS changes, such as line speed, level crossings, recontrols, and timetable changes also need to be subject to the same process and ongoing engagement with end users throughout design is essential.
The pre-deployment stage needs to cover the testing and understanding of the process changes, including operating principles, new documentation, and operational scenarios. Timings need to be clear to cover how deployment fits with changes to the role and workplace, and to address change overload. Implementation and bedding-in of any preparation work, such as route setting and panel changes, need addressing, along with visibility and communication. The operations manager must establish a presence with the end users and be involved with both formal and informal communication channels. Who needs training and to what level for day one needs to be clear.
Managing the delivery stage needs visibility and communication, with ongoing support and engagement in place from day one. Reviewing and managing performance also needs to be in place along with reviewing processes and procedures, such as the management of degraded situations based on percentage of ETCS services. Training and briefing of staff must continue to address any skill fade and provide continuous improvement and updates.
Jessica’s key ‘lessons learned’ are to avoid the bespoke and reach agreement at a national level for design and processes. Work with regions and national programmes for such things as the management of fleet fitment, reference design, and areas for deployment. Identify opportunities with sharing lessons for future deployments, including efficiencies – even if non ETCS. Handing over needs to be a collaborative process and best practice needs to be embedded into business-as-usual.
Australian ETCS
Paul Nicholson, principal engineer for Rio Tinto AutoHaul System, joined live from Australia to provide a fascinating example of ETCS in action. Rio Tinto relies on shipping iron ore from mining operations in Tom Price to the port of Cape Lambert as a critical 24/7 operation. The average return journey of the 2.4km-long trains is 800km, taking approximately 40 hours.
For the last eight years ETCS has been used as part of a system to control autonomous trains, consisting of three locomotives and carrying around 28,000 tonnes of iron ore, monitored remotely by operators in Perth more than 1,500km away. Previously when the trains were driven, every time there was a driver change the train needed to stop. On a typical journey this would take place three times, adding more than an hour to the journey. The driver change points were also in very remote areas and drivers needed to be transported 1.5 million kilometres each year as they changed shift. So, there was also a strong safety benefit for the change.
Paul explained how the system had been implemented, including the processes for operational readiness, training, maintenance, and safe working practices. There were no forced redundancies of drivers. Many were reaching retirement, some transferred to other roles, and others took voluntary redundancy. Some drivers were retained as ‘rescue drivers’ to recover trains in the event of an ETCS failure. However, the system has been so reliable that special arrangements have had to be implemented to run some trains manually to help retain competency.
Software defined solution
The cost, size, and difficulty of fitting ETCS equipment to trains is a concern to many in the industry. Jarlath Lally from The Signalling Company, based in Belgium, explained its software-defined ETCS solution for trains which could be a way forward. The company was founded in 2019 as a joint venture between ERTMS Solutions and Lineas SA, to revolutionise how on-board safety systems are designed, deployed, and maintained, and now operates as part of the Skoda Group.
Jarlath explained that traditional ETCS solutions for trains are too big, too heavy, too complex, too bespoke, and costs too much. The Signalling Company is developing a solution known as RailOS. Similar to what Apple IOS has achieved for the mobile telecoms industry, this aims to deliver the same benefits for rail, RailIOS is based on using a single powerful ‘off the shelf’ processor and software, with a high safety integrity to support many applications. The solution will be lighter, smaller, and more energy efficient than traditional on-board systems, and with open Ethernet communications. Jarlath said it will be scalable for possible track side use and for train ETCS could be deployed in two weeks and at a tenth or less cost than traditional on board ETCS solutions.
Conclusion
These are only a few of the 30 presentations, case studies, and extended Q&As, which are available from the strategy engineering research group. There were some common themes, such as communications, control rooms, and driver training behind the ready-for-service schedules.
Presenters said it was difficult for some operators to fully commit to a new system when the old one was still operating, but things can change quickly and the old system can become a liability. One solution was to create a ‘cliff edge’, with a firm cut-off date. However, in his post-conference report, Chair Steve Thomas said it was important to balance cliff-edge deployment with confidence-building. People may need a firm go-live date to incentivise action, but they also need tools and support to make things happen. There were clear examples of how to align training timelines with roll out phases. Train too early, and people forget. Train too late, and they get confused.
Another key message was that design with human factors in mind is required from day one. Integrated ergonomics, cognitive modelling, and operator interface testing early on were stressed as important as they can mean the difference between success and failure. Training systems, including simulations, were a regular topic and discussion. The point was made that this should also include engineers and non-drivers to build confidence in the system.
Degraded mode procedures need to be established as many said degraded mode working is inconsistently defined and implemented.
The organisations making real progress are the ones doing the basics well, with practical, thoughtful, simple strategies, delivered with consistency and care. We also learned where this approach is working, such as on the Elizabeth line and sections of the Northern City Line.
The ETCS Business Readiness Congress 2026 will be held in Brussels on 12-13 March 2026.
Image credit: iStockphoto.com