Across the world you’ll find countless ways of blocking the line for engineering work; some of them sophisticated, some of them not; from lockout switches and fixed TCODs to sleepers chained across the track. An American correspondent wrote to me recently, highlighting a method which seems only to involve recording your name and contact details on a register secured in a trackside cabinet. Canadian railways extensively use Track Occupancy Permits - a system that was trialled around Oxford a few years ago until resistance and paralysis successfully killed it off. And so, despite many documented high-level promises, we continue to rely on the ages-old approach of placing small explosive charges on the railhead. How come safety professionals exhibit a pathological loathing of lookouts - simple, cheap, safe and effective protection that’s equally rooted in the 19th Century - yet they seem eager to support something of the same vintage that could potentially blow your foot off?
However you do it, rules surrounding possessions can only be effective if they’re robustly implemented, and not always even then. Compliance relies on a range of factors such as whether the workforce has a one-track mind, focussed entirely on the job. Which takes us to Scotland during the autumn of last year and an event with fatal potential.
In the early hours of Sunday 28th October, after the last train had passed through, possession was taken of the 15-mile section of line between Stirling and Blackford. Six activities were planned within it which Network Rail’s computer-based Possession Planning System (PPS) had grouped together into three worksites. One of these incorporated S&T work around Bridge of Allan and Dunblane, undertaken by Babcock Rail, and drainage work at Park of Kier and Ashfield, managed by contractor QTS.
A light engine uses the Down line at Ashfield, close to the QTS drainage site.
Photo: Bob the Lomond (sourced from Flickr and used under Creative Commons licence)
The arrangements were discussed during a telephone conference on 17th October, printed in the Weekly Operating Notice and set out across the pages of the PICOP Briefing Pack. However the pack was incorrect in that the PPS had designated the S&T work to be this worksite’s “primary” activity and identified it as such on the paperwork. Normally, the Person in Charge (PiC) of the primary activity also acts as Engineering Supervisor, however it had been agreed that the drainage work PiC would take on that role. This error could have been corrected by the Possession Delivery Assistant but he was unaware that the software enabled him to do so.
Prolonged rainfall prevented the QTS team from accessing one of the drainage sites on the night of the 28th as it could only be reached over waterlogged fields. So their Engineering Supervisor asked the PICOP if his worksite could be shortened to just include their second site as this would be logistically easier. The PICOP agreed and recorded the revised limits on his copy of the WON. Habitually he did not enter details directly onto the Possession Arrangements Form (RT3198) as circumstances often changed at the last minute, requiring corrections to be made or new forms filled in. Instead he noted the information next to the relevant entry on the WON and then transferred it onto the form before the possession was given up. This approach is not contrary to the rules: they only state that the RT3198 must be filled in, not when.
One impact of this change was that the Babcock group - instead of their COSS signing-in with the QTS Engineering Supervisor - had to appoint their own ES and make arrangements directly with the PICOP. However, when recording details on the WON, the PICOP errantly drew an arrow suggesting that the QTS COSS would sign-in with the Babcock ES, overlooking the fact that he had already arranged a separate worksite for QTS. So there were now four worksites within the possession but only three were later transcribed onto the Possession Arrangements Form.
The PICOP’s personal life was in state of some turmoil. Possibly as a consequence, he had been involved in two irregularities over the preceding six weeks whilst acting as blockroadman. On both occasions he left single detonators on the line which were exploded by trains, causing their respective drivers to make emergency brake applications. The PICOP had worked in the industry for 34 years in assorted operational roles; he’d been taking possessions for 12 of them. But he couldn’t account for the lapses when his manager asked him about them; neither did he divulge the problems in his private life until the evening of 28th when the two met on site. Following a candid discussion, the manager asked if he was fit to continue with his duties. He said he was. It was at this point that he completed his Possession Arrangements Form using the notes he had made on the WON, but he failed to record the QTS drainage worksite.
At 07:04, the PICOP phoned the Blackford signaller to inform him that the work had been completed and the lines involved were now clear. Thus the railway was open to traffic again. But he had only spoken to the three Engineering Supervisors whose details were now on his RT3198. The QTS work at Ashfield was continuing, with seven staff and a roadrailer on track. At 07:58, almost an hour after the possession had been lifted, the ES tried in vain to contact the PICOP. Only when he called the signaller did it become clear that no protection was in place to prevent a train from being routed towards their worksite at 75mph. Fortunately none was scheduled. RAIB’s inquiry report does not confirm whether the ES had, as the rules require, set up marker boards exhibiting flashing red lights on the approaches to his worksite. We must assume that he had so there was at least one mitigation measure in place.
It’s clear that the PICOP’s state of mind, the error in the briefing pack and his use of the WON as a notepad could have contributed to this incident. But two other issues escape the gaze of RAIB’s visionaries, obscured by their blinkers. If planning is key to the safety and success of any on-track venture, who or what empowered those involved to jettison the agreed arrangements at the last moment, simply to make life a bit easier for the QTS Engineering Supervisor? With changes of this nature, due consideration should be given to all the potential consequences. What would have happened if the Babcock team did not have the resources to set up their own worksite: someone certificated to act as ES and a set of four marker boards? If a COSS wants to depart from a planned safe system of work, he has to discuss things with - and receive authority from - someone higher up the chain of command. What mechanism was in place here to ensure that the night’s activities were not compromised by an unnecessary late change?
The other question is what was it about the RT3198 that drove this PICOP to initially record details elsewhere, and is he alone in taking this approach? Is the form inherently deficient in some way? Indeed, in 2013, should we be relying on sheets of paper at all? This is an era of iGadgets - all connected to the internet - so why aren’t we using cloud-based forms, partly populated from the WON, which are easy to correct, can be infinitely expanded to record changing circumstances and are accessible for immediate reference by other interested parties? This approach would allow an element of “clerical interlocking” - for example, preventing a PICOP from inserting a time for giving up the possession until he has recorded times for the clearance of all worksites, providing on-screen prompts and alerts as to what is required next.
|
In 2013, should we be relying on sheets of paper at all? |
|
Beyond RAIB’s persistent failure to comply with statutory timescale obligations (though not in this case), the Branch’s investigators also seem to be locked in a box, unable to think outside it. Learning lessons from accidents should not just involve a compliance audit; it demands a willingness to challenge the status quo. Time after time, the Branch proves incapable of that. As a result, it makes no recommendations in relating to this incident; instead it simply asserts that “correct use of [the RT3198] would have prevented [it].” But this statement serves only to underline RAIB’s overarching position that paperwork is the panacea for all ills, ignoring human factors. If, as a result of his distracted mind, this PICOP made errors in transferring data from one piece of paper to another, he must have been equally capable of cocking-up when recording the information in the first place. Parroting “comply with the rules” exhibits a bury-your-head-in-the-sand mentality. Needed here were better safeguards. A shame then that RAIB couldn’t think of any.
Story added 1st June 2013
|