The engineering disciplines within the railway do not always collaborate as much as they should. It was therefore good to attend a joint IRSE (Signalling) and IMechE (rolling stock) meeting in Glasgow on 20 April. There was no principal speaker, but a panel of experts gave an introductory five-minute talk, after which the meeting opened up as a discussion session. This turned out to be very illuminating.
Transport Scotland
Bill Reeve, the Director for Rail within the Scottish Government transport department, and a former traction and rolling stock engineer, looked forward to the beginning of CP7 and the impact this would have on the Scottish High Level Output Statement (HLOS) for rail services and investment.
Unlike CP6, which had an associated HLOS specified by the Scottish Government but proved difficult to get a centralised Network Rail to respond, CP7 had much more of a devolved team within Network Rail Scotland to shape the proposals, thus allowing local prioritisation on spending. In effect, Bill has the Scottish cheque book and will have much more freedom to spend the allocated money to obtain the best value investments. The amount to be spent is around £1.4 billion per year, and includes Network Rail, Scotrail, and Caledonian Sleepers. This is a considerable sum and, whilst Scotland’s railways are considered successful, getting the right level of competition, both into the railways and against the road and air competitors, is high on the list of priorities.
Services have to be delivered with a diminishing unit cost. Passenger operations have remained static in terms of cost even though the numbers of people travelling have doubled. Why should this be? There is a thrust for decarbonisation where transport is 37% of the total carbon emissions but rail represents only 1.2% of this. One can deduce that modal shift will be one of the objectives. Electrification will play a major part in this, leading to lower cost trains with more capacity and better performance. Diesel, whether by road or rail, will always be more expensive. Already, the Scottish electrification plans are moving forward much faster than in other parts of the UK.
Transport Scotland has major concerns about two parts of the rail supply base: ticketing and signalling. Whilst unit costs for the latter are better than in the whole of the UK, they are still too high. Signalling is regarded as a ‘distress purchase’ and, despite pressures to invest in ETCS, more electrification appears to offer better value. Signalling renewals will be progressed as to whatever turns out to be the cheapest and most practical. Even refitting traditional mechanical signalling is considered a realistic option. At present, the return on investment for ETCS takes too long.
Putting signalling into context
Lynsey Hunter, the Signalling Asset Manager for the Region has the challenging task of keeping Scotland’s signalling systems in good order, ensuring that renewals take place to ensure a safe railway is maintained. Signal engineers are very good at squeezing life out of an asset and the clever use of data to monitor infrastructure age, condition, and failures helps enormously. That said, the challenge in CP7 will be to avoid partial renewals and life extensions. These are not good value for money although it is recognised that some component replacements will happen.
The existing signalling has some technologies that will need replacing. The RETB control centres at Banavie and Inverness are out of date, tokenless block systems on some routes are difficult to maintain, and geographical interlockings of the 1970s era are not always in good fettle. The knowledge base for these systems is diminishing and only those engineers and technicians with 40-plus years of experience have the in-depth expertise to design and maintain the technology associated with the equipment.
There is a need to optimise the way forward and to create an overarching, market-led strategy. This will concentrate on a line of route concept, always remembering that Scottish signalling is a business. The central belt is where the highest traffic levels are encountered, and these must have the priority with the control centres constantly scrutinised as to whether their operation can be improved. In the remoter parts, attention needs to be given to the Stranraer line where a way forward is required. Maybe a form of on-board signalling will be the answer, with ETCS and RETB being possible solutions. Here again, cost will be a deciding factor.
Rail operations
Ross Moran, the regional Operations Director for Network Rail, asked whether the concept for national rail operations in the UK is right for Scotland. With a long railway career including signaller and shipping (CalMac), he has seen many operational disasters – trains that do not fit platforms, ships that don’t fit docks, and suchlike – and is always wary of new things that have not been properly thought through. He rejects the national plan to have only two Railway Operations Centres (ROCs) for Scotland. It is too complex and has too many associated high risks. He says the right number is probably around five which, as well as Edinburgh and Glasgow, will include Perth, Inverness, Aberdeen, and maybe Stirling. There is a need to give opportunities for the local workforces where local knowledge and expertise is all important.
The evolving signalling strategy must have operations at its heart and the line of route control is favoured. In parallel, other digital systems including Traffic Management Systems (TMS) and Driver Advisory Systems (DAS) have to be there to ensure operational performance is maximised. The TMS Luminate system is to be provided in Edinburgh as a trial and will be expanded elsewhere once the benefits are understood and used.
A signalling supplier viewpoint
The main supplier of signalling systems in Scotland is Siemens, so says Steve Wright, the company’s Scottish representative. Siemens partnership with Network Rail is widely considered to be working well but an open discussion is needed as to how this will work into the future.
Competence in the industry is a concern. Network Rail has to be assured that supply industry designers and implementers have the right skill sets. The people who were trained in the British Rail era are coming to the end of their careers and the onus for competence has shifted towards the suppliers. With the change in technology evolving towards the digital railway, there is a need to attract people from different backgrounds. Within the supply chain, the purchase of materials is becoming increasingly difficult. Microchips are a real problem, with manufacturers uninterested in supplying small quantities. It is recognised that better integration between signalling, telecommunications, and rolling stock is needed and a move away from silo thinking is required as well as the recognition that signalling is the enabler for operations.
The rolling stock situation
Whilst concerns are raised about interoperability and interchangeability in signalling, it comes as something of a surprise that similar concerns exist for rolling stock. In the British Rail days, the specifications for the various types of multiple units designed in the 1980s required that they could all couple together, for example 153s, 155s, 156s, and 158s for DMUs and similar for EMUs. With the huge numbers of new trains having been purchased over the past decade, regrettably the specifications for these did not call for the same inter-coupling requirement from train builders. This has seriously impacted on operational flexibility.
Graham Taylor from CAF remarked that a third of Scotland’s operating costs is spent on the provision of trains and this is a sixth of the total of Scotland’s railway costs. Not enough work is carried out to ascertain what is really needed. In other industries, this is known as Front End Engineering Discipline (FRED), where, for example, BP as an example spends £25 million per annum on such investigations. Rail companies spend precious little on rolling stock research and getting the resultant product wrong can result in horrendous costs. It must be remembered that there are lots of alternative ways of travelling and that road vehicles are catching up fast with electric technology. If rail is to produce the modal shift that everyone seems to think would be good, getting the technology right is going to be critical.
Audience participation
The discussion session threw up some difficult questions that don’t have ready answers:
- Why are we buying so many trains from abroad? Although Brexit has happened, procurement law still exists on an international basis which is supposed to open up markets and see fair play.
- Why do we not specify interoperable trains? Unfortunately, trains tend to be built as standard products by manufacturers. The closer the trains remain to this standard, the cheaper they are. Any deviation from the standard design causes a cost increase which can be considerable. One manufacturer has offered to have a height adjustment to the coupler used. This has not been taken up!
- When Solid State Interlockings (SSIs) were introduced, a standard communications protocol between the control centre and the outstations was part of the specification. This, however, was itself non-standard in terms of telecom transmission bandwidth allocations so it needed to be modified. Since privatisation, different interlockings have adopted different protocols although some elements of the original have been maintained. It is not a satisfactory situation.
- Will signalling standards ever be harmonised so as to get away from ‘lock in’ to a particular manufacturer’s system? Although this has been talked about and various harmonisation initiatives have taken place, the suppliers have concerns about progressing this. They are not against it in principle but regard the cost of converting the proprietary technology as significant and not very good value for money.
- Are specifications deficient in not asking for common standards in signalling? Back in the British Rail days, detailed technical specifications were issued but this was considered to be stifling innovation. As a result, technical specs were replaced by functional specifications that detailed the output required without detailing the technology. This gave the suppliers more freedom of design and was supposed to cheapen the overall cost.
- Are the arrangements for taking possessions and managing site safety too restrictive? This is recognised as a problem with often less than half the possession time resulting in work actually being performed. There should be improved methods for taking a possession using digital technology and communications. The problem has resulted in a blockade mentality creeping in with lines often completely closed for days, even weeks, on end with resulting disruption to passenger travel. The GBR statement of putting passengers first will be a challenge to get a revised possession management regime in place.
Final thoughts
The meeting exposed many of the challenges faced by Network Rail and the train operating companies. Although the focus was on Scotland, the issues raised will be recognised by other parts of the UK including devolved administrations. The current buzzword of innovation, whilst aimed at generating new ideas and working methods, has to be balanced against the drawbacks of not having standardised systems and technology to allow interworking.
Obtaining value for money is vital and it is clear that Transport Scotland has severe reservations as to whether ETCS will achieve this. Clearly, the cost of signalling is seen as a major issue and, according to press reports elsewhere, the ORR has similar concerns.
Electrification is important in the decarbonisation agenda but there must never again be the same cost and timescale overruns that occurred on the GW main line and to some extent on Edinburgh Glasgow. When supplying trains, a return to standards whereby rolling stock from different train builders could couple together would be a welcome step forward. This may have to be at an international level as the same problems exist in Europe and beyond.
It was good to see signalling and rolling stock engineers debating these challenges in the same room as it seems inevitable that signalling will be increasingly train borne into the future. The influence of the customer must be made more decisive.
