Level crossings and dead train drivers

Submitted by Janine on 27 February, 2005 - 12:22

The driver was amongst the seven people killed at the level crossing crash at Ufton Nervet on 6 November (pictured).

The rail industry and the government were relieved that there was an individual they could blame, as the crash was caused by a suicidal man deliberately parking his car on the crossing. And the scale of the disaster would not have been so bad if it were not for the highly-unusual circumstance of the train dragging the car onto the next set of points, causing the derailment.

This gave them the ideal excuse to deflect awkward questions about the industry's failure to protect its passengers and staff. A freak accident, they said. Nothing could have prevented it. We don't need to spend any money. But do the facts bear this out?

Level crossings: facts

A few months before the Ufton Nervet crash, the Health & Safety Executive's annual report into railway safety highlighted level crossings. The HMRI’s top cheese, Dr Alan Sefton, said that level crossings are the "greatest potential risk" to rail safety.

Of the 18 deaths in 2003/04, 17 were members of the public and one was a train driver working on the Romney Hythe and Dymchurch railway, who died when his train hit a car. The number killed at crossings that year was the highest since 1991/2, when 20 people died. And this was before the Ufton Nervet crash even happened.

Recent accidents at level crossings have brought their safety into question. Various measures have been suggested to prevent them happening again, some with more sense than others. To understand the options, we need to know what types of crossings are in use and their problems.

The first one is probably the safest: the staffed level crossing - a worker stands at the crossing and closes the gates when warned that a train is approaching.

The next type is a remotely worked set, where the operator is sited away from the crossing and a set of barriers is lowered to prevent public passage. Often, the signaller in their box will check that the level crossing is clear before the barriers are lowered.

A variation on this are remote controlled barriers which have closed-circuit television (CCTV) where the signaller can be miles away, but still checks that the crossing is clear before lowering the barriers.

'Open Crossings' have no barriers, but have a set of lights which warn of the approach of trains. These are contentious, as the opportunity to 'make a dash' is too obvious, and there is a history of people doing just that.

On rural lines there are many unstaffed crossings, which has always been a problem as it relies on the public to check that there are no trains before crossing. Some have telephones, but by no means all.

'User-operated' level crossings have been the scene of serious accidents. In July 2003, three farm workers died, and six minibus passengers were injured, at a user-operated crossing between Pershore and Evesham, Worcestershire. The minibus driver failed to get permission to operate the crossing.

The above list is not exhaustive and there are probably dozens of variations.

Because of the need to keep the trains moving and the public's delay to a minimum, most level crossings are not interlocked with the signalling system. So the ability to stop a train as it approaches a crossing is severely limited as was seen at Ufton Nervet.

The upshot of all this is that the drive for profit has made a bad situation worse. The cheapest system with least staff takes priority over a safer system which is more labour intensive.


The obvious answer would be bridges or underpasses at all intersections between the railway and other rights of way. RMT suggested this after the Ufton Nervet crash; the industry replied that it would be prohibitively expensive and take a huge amount of time. It may be impractical for all 7,937 crossings in the UK, but it should certainly be an option for some of them.

Another alternative would be to staff all crossings, but this is not infallible - as seen recently in Lincolnshire, when gates were left across the rails in the path of a train.

The most effective method would be to combine more staffing with more use of CCTV, alongside the interlocking of signals with barriers in the same way that track and junctions are interlocked. Short of a suicide attempt, trains could not approach a level crossing without the barriers being proven to be down and the crossing clear.

It should be possible to replace all open crossings and user-operated crossings with safer crossings with barriers, staff and/or surveillance systems.

Other measures which could reduce level crossing deaths include: plastic bollards to discourage car drivers zig-zagging their vehicles through half-gated crossings; sensors under the track which could detect a stationary vehicle above it and alert signallers; lasers or radars which sweep for obstructions; improving communication between drivers and signallers.

Profit before safety?

All these ideas would need further consideration. But one thing we can be sure of: the capability exists to seriously reduce the danger of level crossings. If we had a railway industry which was both publicly owned and democratically controlled, then workers, passengers and communities could work out the most effective way to cut and eliminate the risks.

But we do not. We have a profit-driven railway industry, in which that obscene phrase 'cost per life saved' will appear again and safety improvements will be ruled out.

When a crash like Ufton Nervet is still in the news, industry bosses always say that they are 'looking at' possible solutions. But when the TV cameras stop looking at them, will they stop looking at the solutions that could save lives?

Add new comment

This website uses cookies, you can find out more and set your preferences here.
By continuing to use this website, you agree to our Privacy Policy and Terms & Conditions.