“No, I can’t sleep at night! I think of level crossings and level crossing safety all the time as we have so much to do, so much to change.”

Martin Gallagher is Network Rail’s head of level crossings. With a background in safety he deals with policy and strategy, assurance and national programmes for level crossings. And there are about 6,500 of them which might account for some of his temporary insomnia.

He has taken time to talk with the rail engineer to explain how Network Rail now manages what continues to be a very emotive subject.

Close the lot!

So, where does he start? With a list. Then all the level crossings are ranked in what is considered to be risk order.

The risk factors include obvious things like the speed of trains which impacts on recognition and decision times, the number of people using the crossing and the number of lines to be crossed. There can be local environmental issues such as a nearby school or pub. All of these, and many other sensitivities, are used to develop a sensible risk profile for each crossing.

So, why not close the lot and make all the problems go away? Well, the railways can’t just shut a level crossing. The majority of the network’s crossings have public rather than private rights of way associated with them and to close them needs planning and local authority approval to extinguish or divert a right of way. It is not within Network Rail’s gift to actually close a public level crossing.

Common misconception

Level crossings have been closed ever since they were invented, so what is the rate of closure nowadays? Martin is on home ground now, “We’ve increased the rate many fold. Back in 2008, I think we were closing crossings at a rate of about 20 a year nationally. Now we’re closing 200 a year.”

These have been mostly private, user-worked crossings because they’re easier to close than those with public rights of way. The rights to the crossing are owned by private individuals or companies and it’s within their gift whether they release the rights to those crossings or not.

There are 2,500 of this type of crossing and it is a common misconception that they all carry a low risk. The Sewage Works Lane accident on the Sudbury Branch in 2010 was a user-worked crossing. They present a considerable amount of danger, and closing them obviously eliminates any risk and also reduces operational costs.

The release of rights is fairly easy to achieve. There are situations where landowners have four or five user-worked crossings and Network Rail can enter into negotiations to suggest that use can be consolidated to just one or two.

“In the past a lot of things were done from the office, desktop studies and the like, but now we’ve got groups of staff based throughout the country who are experts in negotiating closures. We’ve closed over 500 since 2009 and we’re overlaying that with more public road crossings – but of course these are more contentious. Surprisingly, perhaps, it is not unusual for there to be a split in the community. Half love their crossing, the other half are quite happy for it to go.”

Obligations under Safety Law

Whilst there are benefits, there are, of course, associated costs. The balance of cost versus benefit can be extraordinarily difficult. After a long pause, Martin sighs and adds, “We’ve taken a decision internally that, as far as our obligations under Health & Safety Law is concerned, saying things don’t stack up financially is not always defendable. What we should be doing is making decisions based on expert judgement – based on whether it’s the right thing to do balanced against the cost benefit.”

Martin cited the case of open level crossings in Scotland. These crossings don’t have barriers – just a wigwag and a flashing light. Network Rail has begun a programme of enhancing these crossings by overlaying them with a half barrier.

This is a short term solution which has no real business case, but it’s an area that has been heavily scrutinised because open level crossings make up something like 2% of the total crossing population but account for 30% of accidents.

The conclusion by many is that the crossings are unsafe and so they need barriers – a classic case of “something has to be done”. But analysis in conjunction with the British Transport Police reveals that most of the people who have been prosecuted for violations lived within 12 miles of the crossing.

Drivers performing their own cost benefit analysis perhaps?


Martin admits that some of the damage from criticism is self-inflicted. Level crossing technology has to catch up with current developments. And there is a need to change the ways in which the railways introduce innovation. Traditionally it’s been clunky, slow and overly bureaucratic.

“I can understand why we need safety validations, why we need rigorous resilience and testing programmes, but consider this example: a set of miniature warning lights costs £¼ million. We have a problem at a number of rural crossings that could be mitigated by a set of miniature warning lights which themselves have a safety integrity level of 99.99995%. I think that’s the right number of nines….

“But there is no business case to spend £¼ million at every rural crossing, so nothing happens. Then someone comes up with a safety enhancement that’s a different type of technology but that has a safety integrity level of 99.95% – slightly fewer nines. It costs just £15,000 to install in a day, but the railway shuns it. The Rail Regulator has been really supportive and has the view that if something enhances safety, then it should be introduced. It may not be as good as the gold-plated version but if it’s better than what you had previously then go ahead.”

“In fact we have developed something that is GPS based to track train position. It was designed initially for trains on the Sudbury line and it’s on trial at the moment. It gives train location information to the signaller in Liverpool Street IECC, so hopefully the problem of signallers and users not knowing the accurate position of the train in long signal sections should be a thing of the past. Shortly the crossing user too will be able to see this information on a screen at the crossing. It is a really good system, installed in a day, at a fraction of the cost of a set of miniature warning lights.”

Overseas experience

There is much to learn from overseas experience. Martin’s team has spent a lot of time in dialogue with contacts across the world looking at some of the things they do. A good example is Israel, where they improved safety hugely by taking some very low-cost pragmatic steps. They’ve developed luminous paint that’s equally visible by day or night and they paint the approaches to their crossings with a distinctive blue colour.

They’ve put a lot of cameras in at their crossings as well as obstacle detection systems. Japan has been using obstacle detection systems for years, as has Germany and many other countries. Other low-cost measures are available, such as rumble strips and measures that slow vehicles down on the approaches. But in this country this needs local authority cooperation or draft approval.

Many European countries have taken policy decisions which state that they will not have crossings on anything wider than double track or with line speeds >120 km/h. If you go to Norway you won’t find any level crossings on high speed lines. Portugal has gone through a similar process. In this country a lot of funding would be required for such a policy decision.

Changing people’s perceptions

Public opinion is often informed by professional commentators. Martin asserts that, “The way that we need to change people’s perceptions around level crossing safety is by demonstrating to these commentators that things have changed, and things are changing, through delivery; through all the promises we’ve made, the investments, the improvements to asset condition and to risk management. If their opinions can be changed then this may affect the commentary that they give to the public on the way that Network Rail is dealing with level crossing safety.

“After a fatality at a crossing there are often remarks like ‘an accident waiting to happen…; they’ve had so many warnings….; hardly surprising…’

“Hopefully the slant may change to an acceptance that we’re never able to completely eliminate accidents, but there have been massive efforts to improve the way this risk is managed. So far as some of the local authorities and local politicians are concerned, there’s a huge amount of lobbying to be done on safety responsibilities around level crossings. The perception in general is that level crossing safety is a rail-only responsibility.”


Martin has a team of 20 specialists based at Network Rail’s National Headquarters working on forty different projects to improve safety.

Half of these projects are around safety enhancements, with some big capital investment schemes looking at introducing enforcement cameras at crossings, and also asset condition monitoring cameras.

These would allow real-time monitoring of rural crossings, checking sighting, and detecting whether gates are left open.

“Camera technology is great these days and these projects will change the way we manage crossings on the ground.”

At the moment, assessment and inspection processes are fragmented – they’re not carried out by the same person. A mobile operations manager goes out to collect data on usage then a risk coordinator enters that information into the national database and comes up with a score that shows which options are available to improve safety.

An asset inspector from Off-track goes out and inspects the crossing. A recurring theme in accident investigations has been a criticism of the quality of assessment, the quality of inspection, and the fragmentation of communication. Martin’s aim is to create dedicated level crossing managers who would be responsible for all those activities.

They’ll be professionally trained and would be risk-qualified. “I don’t imagine the oil or nuclear industries allow anyone to risk assess a safety critical asset who isn’t professionally qualified.”

Firm commitment

“Nobody would be able to claim that they are content with level crossings at the moment, but I am content with the plans we have in place and in our progress.

“The Board has set us some really good challenges and supported us. We have been given £100 million, which is a lot of money in these austere times. They’ve given a firm commitment that they want level crossing safety to improve. I couldn’t ask for more support than that.

“At the moment it’s still work in progress, but perhaps, reasonably soon, I will be able to sleep better at night.”