Railway signalling has seen many step changes over the years – from mechanical semaphores to coloured lights, relay interlockings to solid state, hard wired logic to computer based intelligence and others. Some of these took a long while to roll out but others occurred over a very short period of time. The process of evolvement never stops and step changes in both technology and application will happen in this decade.

An interview with Mark Southwell, signalling programme director for Network Rail, revealed the plans for Control Period 5 (CP5) covering the years 2014-19. Mark is a civil engineer who sits within the Network Rail Infrastructure Projects group, so perhaps has a wider breadth of understanding on overall railway requirements than a classically-trained signal engineer.

Recognising that professional and discipline knowledge is nonetheless required, he has a close relationship with Ed Rollins, the professional head of signalling responsible for policy and standards, and Andrew Simmons who is leading the development of new signalling technology – most notably the deployment of ERTMS. Another important relationship is with Andy Hudson, the head of Network Rail Telecoms who will provide the resilient transmission paths for the wide area signalling schemes.

Control Period 5

The advent of Control Periods has given a much needed focus to railway investment projects. Departments now know in broad terms what money will be available over the 5 year period and the ORR in turn knows what the expected outputs will be. The majority (90%) of expenditure for signalling will be renewal driven leaving only 10% for enhancement. This however can be misleading as the necessary renewals will be the catalyst for new technology and ways of working. The expected allocation in CP5 for signalling is £3.33 billion, a considerable sum. With the new devolved organisation structure introduced into Network Rail, much of this expenditure will be regionally focussed.

Knowing that skilled signalling manpower resources are in short supply, the renewals activity will be controlled by one central and four route-based teams. This takes into account the increasing difficulty in containing signalling projects within route based boundaries since the logical control areas, and particularly the acquisition of decision making information, will be much wider spread.

Critical to developing the projects within CP5 will be defining an accurate scope of work since this will form the basis upon which contracts are let and delivered. The GRIP process (Governance for Railway Investment Projects) and its sequential stages has brought focus to the key elements on how a project is planned. Mark and his teams will be responsible for seeing this used to its full potential.

So what are the signalling systems and technology plans for the CP5 years?

Railway Operational Centres (ROCs)

A number of these are being built and six are now operational. The vision is to have the entire railway controlled by these centres. The eventual number required is still not finally decided but the six in operation are at Didcot (Thames Valley), Cowlairs (West of Scotland), Gillingham (East Kent), Derby (East Midlands), Edinburgh (East of Scotland) and Cardiff (South Wales). None of them have yet reached their full operational area nor are they equipped with all their intended systems. Others are planned for Rugby (West Coast Main Line), York (East Coast Main Line), Romford (Anglia), Manchester (North West), Basingstoke (South West) and Three Bridges (South of England) and these are currently being built.

Debate continues as to whether the existing Saltley Signalling Centre will become a ROC and it is also likely that Upminster will be included since it controls a self-contained railway (the London Tilbury & Southend / C2C line) where the proposed traffic management opportunities can be tested.

Anyone who has visited a ROC may well wonder what all the hype is about. Is this not just another name for a modern generation power signalbox?

They will certainly contain this vital function but will do much more. Firstly, the traditional traffic control offices, both for Network Rail and the relevant TOCs, will be concentrated within these centres. Actions taken in the past have combined Network Rail and TOC controls on to a single site and yielded significant benefits. Why not therefore go a step further and integrate these with the people who actually control train movements? A further step will be to add site electrification control rooms in these centres as well.

Secondly, with all this integration, there will be a need for systems that make the decision making process easier and capable of optimisation. Under the generic term Traffic Management, this is becoming a discipline in its own right.

Whether or not the ROCs will eventually control the total railway remains to be seen. Certainly, it will not happen during CP5.

23FX9924 [online]

There have been predictions in the past as to where the spread of power box technology will end; remember the plan for covering the Southern Region of BR with 13 boxes. It never was fulfilled and the ROC program may stop short of every possible line. The business case gets more difficult as the schemes enter into rural territory. One factor that nowadays differs from times gone by is the existence of the FTN, thereby making transmission capacity available to even the remotest areas

Traffic management

The combining of signalled movements with operational decisions is loosely what traffic management is all about. This has, in reality, been carried out for decades but the decisions taken were based on human knowledge and experience. Attempts to bring in ‘expert systems’, with some form of computer based intelligence, made little progress. Junction Optimising Techniques (JOT), dating from the 1980s, and Automatic Route Setting systems have not been used in the quantities predicted. Latterly we have seen the introduction of Driver Advisory Systems (see issue 104 – June 2013) but at present these are more about fuel economy than improvements to rail operations. Network Rail has bitten the bullet and set up framework contracts with three companies – Thales, SSL and Hitachi – to firstly pilot and then introduce traffic management into the ROCs. The first deployment is expected in 2014 and this will require the development of a Remote Interface Facility (RIF). The three selected firms have products akin to this in use elsewhere and it remains to be seen how well these can be adapted for UK conditions. The aim of Traffic Management is to provide decision making tools to identify the optimum operational solution. How realistic this goal will be remains to be seen.

Modular signalling

The programme to re-signal secondary and rural routes using modular signalling has already begun. Two competing firms have developed modular systems – Invensys (now Siemens Rail Automation) for Crewe to Shrewsbury and Signalling Solutions Ltd (SSL) for Ely to Norwich. The specification for such systems emerged as a very high level document with the consequence that the two systems are entirely different and incompatible with each other. In many ways this is a disappointment as it means the modules for doing the same thing are not interchangeable. Nevertheless, modular signalling looks like being able to significantly reduce the cost of re-signalling the less busy lines. One big plus factor is that they are entirely infrastructure based so there is no complication with having to fit kit on to rolling stock.

Modular means just that; there is a module for every piece of external item, be it point machine, signal head, AWS or TPWS track fitting, axle counter or level crossing. The modules can be mass produced and assembled in the factory to replicate the line where implementation is to happen. Basic testing can therefore be done off-site.

Early results show promise but the temptation of signal engineers from both sides of the supplier / client divide to dabble and ‘adapt’ modules for a particular application are threatening the cost savings that can be made. Mark Southwell is determined to resist this ‘bespokery’ otherwise the price for delivering a project will not be contained. A library of standard designs will gradually emerge and signal engineers will be duty bound to keep to these.

Level crossings

Another advance in CP5 will be to make level crossings safer. Several high profile accidents in recent times demand that technology improvements should be explored to achieve this. AHBs (automatic half barriers) and Open crossings carry the highest level of risk. The development of automated four-barrier crossings using obstacle detector technology but without CCTV monitoring is broadly completed, although there are teething problems with the performance of the LIDAR (Light Detection and Ranging) equipment that detects any objects near to the ground. It is reliability rather than safety that causes the difficulty since, if equipment fails to produce a crossing clear condition, the barriers will not raise and the crossing stays closed to road traffic. Keeping the lenses clean will require changes to the maintenance routines. As Mark Southwell commented, one has to persevere with these problems in order to make progress and a significant roll out of this crossing type will happen in the five year period.


The introduction of the European Rail Traffic Management System (or more succinctly the European Train Control System) into main line operation is probably the biggest challenge in CP5. With valuable lessons learned from the Cambrian Line deployment, the next step is the integration trial about to commence on the Hertford Loop line, known as ENIF (ETCS National Integration Facility). Here, the four main contenders for ETCS business – Siemens (formerly Invensys), SSL, Ansaldo and Infrasig (a consortium of Bombardier Rail Control Solutions and Carillion) – will be required to provide both ground and train-based equipment and demonstrate that interoperability between the different supplier offerings can be achieved. Testing of the system using a specially adapted Class 313 EMU will begin in September, which will include testing of the GPRS (General Packet Radio Service) as part of the GSM-R element.

Assuming all goes well, planning can then start with confidence to equip the London end of the GW main line in readiness for the Crossrail opening in 2018 and then moving westwards to Reading, Bristol, Cardiff and beyond. All of this has to go hand in glove with the ROC deployment to ensure that compatibility with signalling systems that already exist in these centres is achieved.

Another ETCS challenge will be on the central section of Thameslink, this being declared as the chosen train control system but with the requirement to add in an overlay Automatic Train Operation (ATO) package, which has still to be developed.

Routine investment

It must not be forgotten that many re-signalling schemes using traditional systems and equipment (SSI and its successor computer based interlockings) will continue. The Signalling Framework contracts let to Invensys (now Siemens), SSL, and Atkins were based upon Network Rail’s confidence in these firms to competently deliver in the UK environment. Atkins, being essentially a consultancy and design organisation with no manufacturing facility, is different to the other two. Not locked in to any particular product, it will be free to deploy equipment from other manufacturers some of whom will be from overseas. This will require obtaining product approval, which can be a slow process but the advantage of being able to use ‘best in class’ equipment from across the world is potentially useful.

The increasing content of signalling equipment that is train borne requires some organisational decisions to be made as to who does what. Train fitment will be carried out by the TOCs and ROSCOs but where will responsibility for system performance lie? It is anticipated that Network Rail will be accountable for system safety. There has been talk in the past of creating a ‘Systems Authority’ but this step has yet to be taken. Network Rail will look carefully at what is happening within European railways to see what practice will suit the UK best.

Signalling in CP5 is going to be an exciting time but, above all, Network Rail has a desire to share knowledge and ensure that no restrictive practices creep in. Good relationships with ATOC and the TOCs are in place, such that project control boards have been set up and signalling requirements may be written into future Franchise agreements. Close contact is kept with the ERTMS Users’ Group in Brussels and with the professional Engineering Institutions in the UK. It is confidently predicted that there will be much to write about in the coming months and years.