Rail 697: safety is the hidden success story

Rail safety is a forgotten issue in the media. No longer does every minor incident get blown up into tabloid features on ‘danger on the tracks’ and there are debates on whether the fragmented railway is unsafe have thankfully disappeared from the Parliamentary agenda.

The reason, of course, is that we have just celebrated a decade in which only one passenger has been killed in a train accident, a remarkable and unprecedented record. Accidents to the workforce, too, are down and the terrible toll suicides and trespassers, which as highlighted in the last issue of Rail, is not seen as the responsibility of the railways.

This reduction in casualties has not come about by accident. Well, actually, to a great extent it has – or rather by accidents. There has been a long tradition in the rail industry stretching back to the 19th century of learning from mistakes that have caused accidents. Virtually every improvement from the introduction of interlocking and vacuum brakes after the Armagh disaster in 1889 to stronger couplings and safer coaches following the Clapham crash a century later has been introduced as a result of past failures. Most recently, the adoption of the Train Protection & Warning System (TPWS) that has greatly reduced the number of Signals Passed at Danger (SPADs) and vastly improved methods of detecting potential broken rails have been adopted largely as a result of the series of accidents in the aftermath of privatisation.

This trial and error method is no longer acceptable for two reasons. First, given the public outcry at any accident, it would be deeply damaging for the railways if it required a major accident to highlight any serious risks on the railway. Secondly, though, this method of improvement no longer works because all the obvious risks have largely been eliminated.

Let’s first of all stress one thing. The current record of the railways on safety should be a matter of pride for everyone who works in the industry. It is not only a high point historically in terms of safety, but also in comparison with other European countries Britain’s record on most measures is at or near the top on most measures. Managers from other railways come over to the UK to see how we have done it.

Despite the good news, everyone in the industry knows that with trains whizzing up and down the crowded network at speeds of up to 125mph, disaster is only one set of errors or bad luck away. And there is widespread concern within the industry that the background indicators, rather than the headline grabbing ones – are worryingly stable.  When Len Porter, the head of the RSSB – it likes to be called by its acronym, rather than the old Railway Safety and Standards Board – and Anson Jack, the policy director, gave me a briefing on the state of safety in the industry, they were absolutely clear about the need for a different approach in the past in order to remedy this. They produce a graph showing that the ‘precursor indicators’ have not reduced since 2006 and that is raising concern.

While SPADs have fallen dramatically thanks to TPWS, and trains are now much more likely to withstand a collision, the number of incidents caused by ‘public behaviour at level crossings’ has remained largely the same as have those from ‘irregular working’.

This suggests that a new way of reducing risk is required. The new approach must be holistic, which might sound like a rather meaningless business expression, but in fact has real resonance here, as Mr Porter explains: ‘The industry does not have a safety problem. It has a performance problem. There is no safety crisis in the industry.’ So the issue is, how do you improve performance without compromising safety.  He suggests, here that far from being contradictory aims, improving safety and performance go hand in hand – a win win situation and he says something which will be of great relief to those who worry that concerns about safety will lead to reductions in performance: ‘The next step change in safety will not be addressed by tightening standards but by doing all the things the railway companies need to do to improve performance’.

This is groundbreaking stuff but not necessarily easy to get across to the train companies and Network Rail. They cite an example relating to detectors used on the tracks to spot hot boxes (an overheating axle bearing). There is now a better, though more expensive, technology, which uses sensors that ‘listen’ to the axles going over and are able to spot which ones are likely to fail in the near future. This has the advantage of obviating the need for trains to be stopped on the main line when a hot box is spotted but Network Rail has just spent considerable sums on renewing the detectors.

Jack points out that it  is remarkable that while huge amounts of data are collected by 200 delay attribution clerks on delays to allocate responsibility and therefore payments, none of this is then used to create predictive models that could then lead to improving performance. Moreover, until trains are two minutes late, they are considered as on time, and yet minor delays are always, by definition, precursors to major ones. Analysing the causes of the short initial delays would undoubtedly lead to a better understanding of the causes of major ones.

Porter and Jack are quick to dismiss the notion that the RSSB is a kind of safety Taleban requiring ever more stringent safety standards. They tell an interesting tale around the issue of train horns. It used to be that trains sounded their horns whenever entering a long tunnel but the design of modern trains led to the sound being spread more widely across the neighbourhood. Porter says there ‘was a mountain of complaints on my desk’ relating the issue.  So the RSSB assessed the risk of doing away with horn use between 11pm and 7am and decided to allow the practice to be stopped, even though there was an added risk which was calculated as predicted to cause an extra death every five years.

The message of investing in performance to improve safety might sound a bit far-fetched. But the RSSB managers are clear that there is no alternative given the worrying flattening out of the improvement in the precursors to accidents. Let’s hope the lessons are learnt before that awful combination of bad luck and minor mistakes strike again. Ironically, that involves moving away from the safety agenda onto the performance one. The Office of Rail Regulation, for example, still focuses so much attention on safety, employing far more people in what used to be HMRI than on its regulatory function. Indeed, it still puts out press releases which suggest that there remains a big safety problem in the industry and still pursues transgressions mercilessly, resulting in those daft huge fines that go to the Exchequer. If Porter and Jack are right, then by focussing on performance it will save more lives than continuing to bang on about safety.


One way system needs two way thinking


The new concourse at Kings Cross is undoubtedly one of the modern wonders of the railway world, but the system being used to control passengers at the station following its opening is not proving universally popular. Hop off a bus or taxi in York Way, on the east side of the station, and you find yourself having to make a two minute detour down the overcrowded pavement in front of Euston Road – which will be relieved of the ghastly sixties additions early next year – to then turn back on yourself to reach the platforms. The East Coast staff who man the gateline on the platforms which are supposed to be ‘exit only’ are reasonably flexible and will let occasional passengers through who are about to miss their train but clearly the situation is not sustainable – there will be too many angry passengers with baggage forced to walk the long way round when in sight platforms.

Network Rail is opposed to any change, suggesting that it would create risk. It insists it wants one way traffic for both departing and arriving passengers, just like an airport. Yet, surprise surprise, Kings Cross is a station and very few stations in the world, however busy, have such a system. Indeed, the gates were used alternatively for both exit and entrance before the change, and there were no accidents. Keeping a few of the gates open for people to come into the station from the East would clearly help many passengers and seems a no-brainer – but perhaps the retailers lined up around the new concourse and eager for business to cover their high rents paid to Network Rail would not be so happy about it.


Scroll to Top