Britain’s railways have gone through considerable upheaval in recent years but it is worth remembering that this is not for the first time. John Armitt reviews the history and present state of the railway network, ranging widely over engineering, management and environmental issues. He then looks forward to the way in which the network will be enhanced and expanded over the next five years.
This article is based on The Royal Academy of Engineering’s Hinton Lecture, given by John Armitt on 1 October 2002
The rail network
The development of Britain’s rail network helped to create, and then support, the Industrial Revolution of the nineteenth century. The founding engineers, such as George Stephenson and Brunel, fuelled by speculative equity, laid down track and developed engines in an orgy of raw energy.
Between 1820 and 1870, 15 000 route miles of track were thrown down and 100 000 tunnels, major viaducts and bridges erected in a highly competitive environment. While steam-driven construction plant was emerging, most of the work was hand-built by thousands of men known as navigators or ‘navvies’ whose industrial origins could be found in the building of the canal system, itself now consigned to the backwaters by the railway. The beginnings of the separation of the engineer and the contractor came as early engineers realised that they could not organise and finance the physical work, as well as survey, design, obtain parliamentary consents and keep their financiers happy.
Finance and politics were central to the process. Fortunes were made and lost in a frenzy of speculation. The engineers’ forecasts of the costs were like today’s, often underestimated, whilst the demand and revenue forecasts were equally optimistic.
The first Act of Parliament permitting the building and operation of a passenger railway was for the Stockton to Darlington in 1821. Thousands of further Acts were subsequently passed, with 272 in 1846 alone.
The major lines were developed for carrying goods as much as for passengers, taking over from the canals at a time when roads were useless for carrying the raw materials and finished products of the revolution.
The network, twice as long as the one we have today, was essentially complete by 1920. A series of mergers and acquisitions between the numerous owners created in 1923 the great railway companies of the interwar years. The Great Western, Southern, LNER and LMS dominated. Each employed tens of thousands of poorly paid workers within military style hierarchies. By the late 1930s competition from road transport was real and some companies were beginning to face hard times.
In 1946 the railway companies were nationalised into a single entity, their infrastructure exhausted by wartime demand with minimal upkeep. The nationalisers did not appreciate the demands the railway would put on the Exchequer, nor did they realise the scale of the infrastructure, which needed more than an uncertain annual allowance to allow for a proper maintenance and renewal programme. By 1964 reduced demand as motor car use expanded meant that a review of real need was required. The result was Dr Beeching’s programme, which saw 5000 miles of the network closed in five years along with 2000 stations closed, and 250 train services withdrawn.
In engineering terms the railway was changing. Electrification had first appeared on the underground as early as 1890. Above ground on the mainline railway the Southern had adopted the third rail 750 DC system in the 1920s, the prime motivation being low running costs and greater efficiency.
Steam, however, held on to the intercity routes until the 1960s, replaced by a diesel building programme over the decade from 1955.
Before the Second World War, work started on overhead electric schemes. The first – from London Liverpool Street and Shenfield – was completed in 1948 at 1500 V DC. In the 1950s overhead lines were standardised on alternating current at 50 Hz and the early lines converted to 6.25 kV, the maximum given small electrical clearances under bridges. As the Clean Air Acts reduced pollutants, the lines were further uprated to 25 kV within the same clearances between 1976 and 1987.
The block system
Behind the scenes the real controllers of the railway have always been the signallers and their control system.
The block system, whereby a train cannot enter a section of track until the preceding one has left the section, has developed from tokens given to the driver on entry, handed over on exit, to flags, to semaphore signals controlled by wires, to electrical control of lights. The principle of interlockings, whether by mechanical means or by computer programs, creating the block or safe routing to protect against collision has been consistent. Accidents are very rarely due to system failure but human error overriding or ignoring the systems.
The 1993 privatisation was not simply a transfer of ownership from the state to private equity and debt, but also for the first time a separation of track from rolling stock, with train access by commercial contract. Britain was not the first country to create this split. Sweden had created Banverket as infrastructure controller but it is state-owned. The arguments about separation continue. It is probably fair to say that, within the railway, the majority view is that efficient operation is through an integrated railway. Other countries have also separated but none to the same degree of complexity as the UK.
Today’s network consists of 20 000 track miles, 2500 stations, 65 000 bridges and tunnels and thousands of miles of cabling. It is the responsibility of its owner, Network Rail.
The rolling stock consists of 1150 engines, of which 80% are diesel powered, together with 1000 diesel and 1700 electric passenger multiple units; 60% of the latter work on the third rail direct current network south of London.
In 1994 the majority of rolling stock was bought by three companies known as the ROSCOS (rolling stock companies), who lease the trains to 26 train operating companies, the TOCs. Of these, 23 are passenger carriers and three freight.
The TOCs’ ability to operate is by virtue of a licence and a franchise, bid for in competition and typically of seven years’ duration. New trains are specified by the TOCs but actually owned by the ROSCOS.
The last eight years has seen £4bn of orders for 1400 new trains, a scale of investment unlikely under state ownership. In the previous eight years, British Rail had placed orders for 582 trains to the value of £2.5bn. The introduction of new trains onto the network is a complex process of engineering and safety approvals. This process, criticised by many as being cumbersome, over-cautious and expensive, is one of many elements of today’s railway that we must improve.
How the system operates
The train operators gain access to the network through agreements negotiated with Network Rail and approved by the Rail Regulator. These permit access on given routes during particular periods of the day. With as many as ten companies having access to the same route sections, the potential for conflicting needs is high and the development of the six-monthly timetable with train operators, who bid for their preferred slots, is lengthy and complex.
Whilst the Strategic Rail Authority (SRA) lets the franchises to the TOCs, Network Rail’s activities are controlled by the Rail Regulator. As a monopoly supplier of track, Network Rail is regulated to ensure that the TOCs are fairly treated.
Cash flows in the system
The passenger or owner of goods pays a fare to the train operator, who in turn uses the fare to pay for the lease of the trains, his operating costs and to Network Rail for access to the network of track and stations. The balance is his profit. Network Rail in turn uses the track access charge to maintain, renew and enhance the network to meet the needs of the train operators.
The fares that are considered acceptable for passengers to pay are not enough to meet the payments the train operators must pay Network Rail which in turn are insufficient for Network Rail to meets its obligation of providing a safe, reliable and improved network. Hence there is a set of subsidy payments to the train operators, grants to Network Rail and an increasing debt burden for Network Rail as it borrows to meet its obligations. The contracts between Network Rail and the operators involve penalty payments for delays and poor performance. Disputes over these can be appealed to the Regulator. The sums of money involved run into hundreds of millions, so it is not surprising that a very commercial and sometimes adversarial environment has developed.
The Rail Regulator approves fares and access charges, sets penalties and decides how much money Network Rail should be allowed over a five-year period to operate the network.
The demise of Railtrack
Very simply, this financial and contractual cocktail, together with the impact of first the Ladbroke Grove and then Hatfield accidents on passenger revenues and costs arising, resulted in 2001 in Railtrack concluding that it had severe financial problems. The result of this was the Government’s decision to ask the courts to make a Railway Administration Order over Railtrack.
Today we have a new financial structure for Network Rail. This new purpose-built not-for-dividend company, has created a debt-driven financial structure with the SRA as guarantor of last resort, which has enabled them to buy the shares of Railtrack plc, the operating company from Railtrack Group.
A more efficient and reliable infrastructure
Most of the stations, structures, tunnels and cuttings in the system are over 100 years old. The elements that have been regularly replaced and upgraded are the ballast, the sleepers, still 25% of timber, the rails, 70% continuously welded but in parts of the network still joined by steel fish plates, and the signalling and communications systems. Even today there are 325 manually operated signal boxes utilising their Victorian operated iron interlock frames requiring the same maintenance and adjustment as the day they were built.
Nearly 5000 miles of track has overhead 25 kV power systems, much of it 30–40 years old. South of London, 2500 miles of track utilises the 750 V DC system. However, the improvement in modern turbo-diesel rolling stock, leaves a question mark over further major electrification in the UK.
Much has been made in the last couple of years of the fact that Railtrack did not have a full knowledge of its assets. The current reality is that we do have, on a zonal basis, a good understanding of the quantum, age and general condition of the assets but this knowledge is held in different forms and does not provide a detailed record. One of the consequences of privatisation
and the outsourcing of maintenance is that most records are now held on contractors’ own systems with no automatic rights of access for Network Rail. We are now putting that situation right with the introduction of our own network wide system called MIMS.
Engineering activity on the network can be split into three types:
Each of these requires safe access to the network to carry out the work, which ideally means that trains are not allowed to run on the section of track being maintained. Renewal and enhancement activity is nearly all carried out in green zone working – which is when the trains are not running on the section being worked on. However, a significant proportion of maintenance is carried out on sections of track over which trains are running (red zone working). A recently introduced standard will reduce to a necessary minimum this type of work.
Green-zone working requires Network Rail to agree track possession with the train operators. The majority of these possessions are taken during mid-week nights between 11 pm and 6 am, or at weekends. This means most work is done under artificial light with low levels of efficiency. There are many opportunities for efficiency improvements.
Inspection of the network
Basic inspection is carried out by patrolmen walking the track looking for obvious faults and, if possible, making an immediate repair or adjustment. The public can relate to this approach, but it is far from perfect. Increasing use is being made of non-destructive testing, utilising rolling contact ‘hockey sticks’ which provide data on the internal structure of the rail and its welded connections. We are also building special track vehicles to do this vital work faster and with greater reliability.
A safe environment is provided by train-borne inspection. Detailed understanding of the geometry is achieved by the use of automatic track recording trains which can provide a trace record of horizontal and vertical alignment to an accuracy of 1 mm.
GPS technology is now enabling these same trains to physically relate the track to its surroundings so that monitoring of track clearances vital for safe running is available.
We will soon have digital cameras mounted on examination trains to provide a visual record of track and its surrounding area condition. We also plan to mount these cameras on passenger trains with automatic remote download to our systems.
A recently exposed rail defect, the cause of the Hatfield derailment, is rolling contact fatigue or gauge corner cracking. This is the propagation of short cracks across the surface of the rail which over time can develop at an angle down into the rail, join up and in the severe stages vertically migrate, with eventual failure and collapse of the rail. This is an increasing problem in other countries and the causes are not yet fully understood but it is thought that different types of train with different suspension and rolling contact characteristics can generate these defects at different rates. The solution is to grind off fatigued metal before cracks can develop.
Inspection and testing of points or switch and crossings sets with their many moving parts can only be done properly physically; it is time-consuming and ideally requires possession of the track.
Signalling and communication systems are in the main monitored remotely, although one of our largest problems today is wire degradation in systems 30–40 years old which also need to be physically monitored.
Repair and maintenance
Maintenance programmes can be developed and possessions planned as much as 60 weeks in advance. Immediate possessions are taken for emergency repairs. On an average day there are 82 system failures causing significant delay, the most common being points and cracked rails, track circuit failures, signalling faults and supply failures. Every night immediate and long-term planned maintenance takes place utilising approximately 10 000 engineers and operatives. In 2002 we are spending approximately £1.25bn on maintenance.
The sleepers are supported and held in place by the ballast. This crushed stone, typically granite or limestone, works less effectively as its voids become full of dirt, the stone moves or slowly breaks down. It can be automatically placed back into position under and around the sleepers or totally replaced by rail-borne machines. The track alignment can be improved by automatic machines known as tampers which move the track to its proper alignment and tamp the ballast. A British development is the stone blower, which achieves the same end without damaging the ballast through mechanical action.
One of the most effective ways to maintain rail life is to grind the rail surface. This removes surface defects and can reprofile the rail to its ideal shape, so giving a smooth ride.
Grinding can be done by hand-held machines or at the extreme by rail-mounted trains capable of grinding and reprofiling rail to an accuracy of microns at a speed of up to 19 km/h utilising 64 computer controlled carborundum grinding stones. We recently took delivery of a £10m plain line grinder and have two more plus five switch and crossing grinders on order. These are essential to keep on top of gauge corner cracking and rail quality generally. Our objective is to have them working 95% of available time.
Eventually, elements of the system have to be renewed. This year we will spend £2.5bn on system renewal. Automatic equipment can replace ballast, sleepers and rail separately or all at the same time. The average life cycles are 15 years for ballast, 25 years for rail and 35 years for sleepers. However, track life can range from as low as seven years in highly corrosive environments such as the Severn Tunnel, to 65 years in outlying parts of the network. Switch and crossing renewals are major activities often involving several hundred metres of track and heavy lifting equipment.
The last period of major track renewal was between 1965 and 1985, so it is clear that we now face another period of track renewal across the network.
Renewal decisions are based on inspection information and life-cycle factors. We are developing decision tools to enable us to optimise age, condition, possession costs and outputs to deliver the most effective solutions.
The shortest life cycle elements of the system are signalling and telecommunications, but renewal of these invariably becomes enhancement of the system.
Enhancement is treated differently in financial terms by the Rail Regulator and, given that any system renewal by the latest model may well in some way improve and enhance, definitions become important.
Signalling is the key to control of the network. There is currently a shortage of signalling design engineers and any investment decision is always influenced by the availability of this resource. Given its safety-critical nature design is very carefully controlled, checked and carried through many, sometimes too many, iterations. The desire of engineers to use the latest and best, sometimes unproven, technology has led to time and cost overruns and we are currently reviewing our signalling strategy with the focus on what works and can be effectively delivered. However avoiding overruns has as much to do with good project management, particularly in the early stages, as it has with the choice of technical system.
Train protection systems
A major issue in recent years has been automatic train protection (ATP), which monitors train speed and automatically stops a train that is in danger of passing a red signal. ATP was rejected by British Rail on cost–benefit grounds in 1993. The cost of the system valued a life saved at £14m. After the Ladbroke Grove accident in 1999, the Cullen Report recommended the installation of ATP within ten years. However, after 1993 Railtrack had supported the development of an alternative, more cost-effective warning system known as the train protection and warning system (TPWS), which in its advanced form is capable of stopping trains travelling at up to 160 km/h. We are now on course to complete, in 2003 the installation across the network of TPWS at a cost of £550m.
This has not stopped the pressure for full ATP. The most recent report commissioned by the SRA has recommended that, through the introduction of the European Rail Traffic Management System (ERTMS), particularly on high speed lines and now mandated by a European Interoperability Directive, full ATP is delivered over a longer period. This report is now with the Department of Transport.
ERTMS is a signalling control system that enables the information to be displayed in the driver’s cab rather than on line-side posts and can take control from the driver if necessary. Trials are currently underway in Europe. UK network implementation is currently costed at £4–6bn. It will save one to two extra lives per year compared with TPWS, but the system will also enable better utilisation of network capacity through improved signalling control.
Another important planned improvement to the rail network is to the communication system between the driver and signal control. Today this is effected by a mixture of line-side telephones and in-cab radio systems. The in-cab systems are up to 18 years old and not all of them work in tunnels and certain other locations. We are currently introducing a £360m scheme known as GSM-R (global system for mobile communications – railways), which will provide a new in-cab system capable of operating anywhere on the network. Target completion is 2006. A new fixed telecommunications network is being designed and planned for completion in 2008, which will replace the current life-expired system. This has inevitably involved ‘Do you own or lease?’, ‘Do you utilise off-track systems?’ choices. Our decision has been to have our own track-side system for these vital safety controls.
Major upgrades, such as the Channel Tunnel Rail Link, West Coast Route modernisation, Thames-link 2000, the East Coast upgrade, and Cross Country Route modernisation represent billions of pounds of investment, with construction time frames of up to ten years and several years of prior engineering development.
In some cases improvements in rolling stock generate the need for infrastructure improvement. Thus the replacement in the southern region of old slam-door trains will require at least £500m of investment in improvements to power supplies to cope with the extra demand of high acceleration, heavier, air-conditioned new trains.
The costs and programme for major developments is significantly influenced by the nature of access to the railway. Naturally, train operators want maximum access for their services but unless long-term blockades are introduced construction work is done on a piecemeal basis. Recent research indicates that passengers may accept major closures or preferably diversions in return for more certainty once work is complete. The train operators, however, remain nervous about possible long-term loss of passengers.
Smaller enhancement schemes can often produce excellent operating improvements and there is a crucial balance to be struck between major, high-profile, long-term schemes and achieving improvements through smaller quickly delivered schemes.
Aspects of safety
When people travel on trains their expectation of a safe journey is greater than when driving, travelling by plane or even by ship. There may well be some simple explanations. For example, we recognise our own and others’ driving errors. Subconsciously many are still puzzled by the ability of hundreds of tonnes of aeroplane to be raised in the air, whilst whenever we board a ship we are reminded of lifeboat drill and often experience the full force and discomfort of the elements.
On trains, we jump on at the last minute, there is no security control, little if any safety instruction, we often stand throughout the journey and accidents are rare: in the last 50 years fewer than 1000 people have died travelling on our railways. In the same period 250 000 have died on the roads. Some 80% of deaths through rail accidents have been due to operational rather than technical causes.
The principal causes of accidents are:
signals passed at danger
level crossing misuse
derailment due to infrastructure defects.
In seeking to prevent accidents we must identify risks and then the ways to minimise them. The minimisation of risk must be considered within a rational framework under the 1974 Health & Safety at Work Act. Risks should be reduced to ‘as low as reasonably practicable’ or ALARP, a concept initially developed for the nuclear industry.
Figure 4 shows a representation of levels of risk tolerability, applied to three exposed groups: employees, regular commuter passengers and other members of the public. These levels of safety performance are expressed as individual risk of fatality as a result of exposure to railway operations.
A key requirement of interpreting ALARP is applying an economic test. This test is defined as the cost being grossly disproportionate to the improvement gained. If our objective is to reduce the risk of death, then we must put a value on life or the level of investment appropriate to prevent a fatality. This in personal terms is very difficult but a rational view is important and for the railways we are currently using £1–3m. This compares with £1m for roads and £10m for the nuclear industry. Thus the 1993 decision not to invest in ATP at a cost of £14m can be put in context. These values are of course open to challenge and in the aftermath of a major accident what has been described as the outrage factor can push these values up exponentially.
TPWS represents a cost of £5m per life saved, above the normal level but less than that which could be called excessive or based on outrage.
Derailments and collisions
Derailments and train collisions can be catastrophic events.
Collisions can typically follow from unprotected signal failure. TPWS and ATP are designed to stop trains in the overrun length before a collision can occur but other aspects of signal design, such as sight lines, luminosity and separation of lights, can reduce the risk of a driver missing a signal. Every time a signal is passed at danger (SPAD) an investigation is carried out.
Derailments are caused by a variety of failures, typically broken rails, line obstructions, structural or earthwork failures. Broken rails can, as I described earlier, follow from the recently recognised phenomenon of gauge corner cracking or from impact loading due to wheel flats. We have recently increased our own detection of train wheel defects and the consequent action by train operators has coincided with a drop in the number of rail breaks.
Level crossing Incidents provide a good example of a line-side obstruction. Nearly 10 people a year die in road vehicles from these incidents, and the risk to passengers from derailment is considerable, so we have a risk assessment for every crossing on the network, many of which are user controlled.
When budgets are tight, maintenance of our structures is often the first area to suffer, while every day several bridges will be struck by lorries and buses whose drivers ignore warnings or are unaware of the height of their vehicles.
Earthworks failure has not been a major issue in the past but recent changing weather patterns have caused deterioration to embankments, particularly in the south and west where they are most likely to be of clay construction. We are currently increasing our expenditure forecasts to ensure stability of these structures which can be up to 150 years old.
The engineering process flowing through policy, standards, inspection, execution and verification is totally dependant on resources, particularly engineers. The combination of split responsibilities, increasing expenditure on a long-neglected infrastructure and the need to reduce risk and improve quality has increased demand for engineers in all areas. Engineers working on the railways are now recruited from around the world and at Railtrack we initiated conversion courses for engineers trained in other disciplines to become track engineers.
It is inevitable that the network is not always wholly compliant with standards. This is dealt with either by considered and permitted derogations or by placing restrictions typically on the speed of operating trains. Temporary speed restrictions of 20 mile/h were imposed across the network in the immediately aftermath of the Hatfield incident. Today temporary speed restriction (TSR) numbers are still in excess of pre-Hatfield levels. This is due to the continuation of gauge corner cracking, an increasing engineering workload, risk aversion, the stricter application of standards, inadequate maintenance or of course an increase in defects. The impact of TSRs on daily performance is high. Numerically they represent 20% of the delay minutes attributed to Railtrack but indirectly reduce recovery capacity in the event of other system failures and incidents. The identification of defects which will lead to new TSRs or the quick removal of applied TSRs is therefore a high priority.
The key drivers for our environmental policy are:
EU and UK transport policy
environmental regulation and directives
government taxation policy
government targets, such as those on CO2 emissions
directions and guidance we receive from the SRA.
The major environmental issues are:
surface water drainage
Contaminated land exists in many rail depots as the result of 100 years of activity. Today the management of diesel and oil storage and distribution in the depots is a key issue.
A new European noise directive will potentially lead to significant cost. If the primary source of noise, the wheel–rail interface, cannot be resolved at source then the alterative is a noise barrier. The first task is to understand current noise levels and we will shortly be starting a major noise-mapping exercise.
On the positive side, Network Rail recycles ballast, rail and sleepers. We manage 330 sites of special scientific interest, while the growing use of rail reflects either a real reduction, or at least reduced growth of road journeys and CO2 emissions.
Organisational structure and contractual arrangements
One element of privatisation was the sale of the maintenance and engineering organisations within British Rail, typically to UK construction companies. British Rail’s design engineering capability infrastructure was also sold, leaving Railtrack with a core resource. In numerical terms 20 000 people had been transferred out of BR’s engineering capability and Railtrack retained 500.
The result was that Railtrack, to meet its obligations to provide the infrastructure to the train operators had to buy in engineering and construction services. Railtrack, however, retained the actual operation of the network. Today, of the 13 000 people in the company, 6500 are responsible for the signalling operation and control, electrical control centres and communication control of the network. The majority of the stations are manned by the train operators, but Network Rail manages the largest 16.
Network Rail itself consists of a centre based at Euston and six regional units called zones. The centre sets engineering policy standards and operating procedures, and oversees procurement of all services.
The infrastructure is maintained, renewed or enhanced through a series of contracts between the company and external consultants and contractors managed directly by the zones.
The maintenance contracts are typically let for five years and the contractor, having been set standards to be achieved by Network Rail, inspects the railway, decides on an annual programme of work which is agreed with Network Rail, plans and executes the work and ensures it is carried out to the required standard.
Hundreds of separate work packages are carried out every day by 20 major contractors, with most of the materials and construction plant being delivered by rail. To bring some order to this major logistical exercise, Network Rail has a national control centre which ensures that materials and plant are in the right place at the right time.
Following the Hatfield accident, concern was expressed both by Railtrack and its various stakeholders as to whether Railtrack had sufficient control of the engineering works on the network. The result was a restatement of the company’s engineering policy, set out in these principles:
to set detailed engineering policies, specifications and key work instructions for maintenance, renewal and enhancements
to specify the technical competence required to work on the infrastructure
to control directly examination of the network and automate it where possible
to take charge of asset information including current condition data and infrastructure records
to make or control work decisions and prioritisation.
In support of these principles, we are reviewing the relationship with our maintenance contractors. In future Network Rail will set out in more detail the inspection regime and specify the competences required of the individuals carrying out inspection. Network Rail will interpret the results and determine the work to be carried out by the contractors and it will take a more intrusive role in verification.
This is a significant shift of risk and responsibility from the contractors to Network Rail. There have been calls for all maintenance to be carried out directly by Network Rail. This would be a massive change and one we do not intend to pursue. The majority of Network Rail expenditure is on maintenance and construction, and while we believe we should make the key engineering decisions that ensure the safety and sustainability of the infrastructure, the physical work should be organised and managed by contractors who have the necessary skills and resources. This is no different from what happens in, say, the oil industry and with other utility suppliers. Looking forward, working with our contractors, we have to improve efficiency and productivity in all aspects of the maintenance process. We have started by reviewing our approach to possessions management and control. Possessions are booked months in advance but last-minute changes, cancellations of work and too much time spent setting up and demobilising work all reduce planned access.
I believe that in the next three to five years we could achieve a stable sustainable network. Future expansion and enhancement of the network will be strategic decisions for the SRA. The last five years have seen a 30% increase in passenger miles, with a growth in freight of nearly 50%. The Government has set a target of a 50% increase from the year 2000 by 2010, together with an 80% increase in freight tonnes carried. The further pressure on the infrastructure is obvious. With increasing expenditure on maintenance and renewal, funding of major enhancements will be difficult and the intention is to use privately funded special-purpose vehicles. SPVs will entail considerable risk which I believe will ultimately involve Government. The railway, past and present, has always required tough decisions; they will continue to need to be made but with the new sense of direction and control at the SRA I am confident that with flexibility and pragmatism by everyone in the industry, we can engineer and deliver a safe, reliable railway for the next 100 years.
John Armitt CBE FREng
Chief Executive, Network Rail
John Armitt was appointed Chief Executive of Railtrack PLC on 14 December 2001. He has extensive experience in the building, civil engineering and industrial construction markets. From 1987–93 he was Chairman (International and Civil Engineering Divisions) for John Laing plc. From 1993 to 1997 he was Chief Executive of Union Railways where he was responsible for developing the Channel Tunnel Rail Link. In 1997 he joined the Costain Group plc, also as Chief Executive, and is credited with rescuing the company. He was awarded the CBE in 1996 for his contribution to the rail industry. He is a Fellow of The Royal Academy of Engineering and of the Institution of Civil Engineers.