Plymouth platform collision - RAIB report

[great central]

Permissive working is used many many times every day at Nottingham sometimes with three trains per platform, has been for years. As far as I know there have not been any incidents beyond the odd red face of a train too long for the space available.

[bescotbeast]

I think it's poor Report to be honest, typically RAIB in that it seems to miss some important issues while mentioning others of potentially lesser relevance.  According to the Report the Driver didn't to appear to know where he was going 'until he saw the points set towards Platform 6'  which seems very odd in view of the big illuminated figure 6 showing at P15 ( I bet he he knew exactly where he was going but simply didn't expect the HST to be standing where it was and couldn't see until too late that it was in Platform 6)

 

Hmm.

 

The driver was clearly told (by the signal aspect) that he didn't have a clear platform. There isn't any such thing as a 'nearly clear' or 'nearly full' platform in signalling, a platform is either occupied (approach it prepared to stop short of any obstruction), or not (so you can approach it normally).

 

The issue appears to be in this specific case that the driver didn't actually understand what the signal meant, and so didn't proceed appropriately.

One might speculate that a different driver that did understand what the signal meant could have also been lulled into a false sense of security or misled by the unusual circumstances, but that really would be speculation? 

 

Had the driver been approaching prepared to stop short of any obstruction as per the signal indication it would have ended up with a somewhat embarrassing operating cock-up (detraining folk from maybe one or two doors, and maybe wiping out any time gained by doing the platform share!) but not something that was inherently dangerous in terms of collision.

 

It doesn't matter what platform it was, if you are coming in on a sub you should be prepared to stop short of a train anywhere

 

 

As the esteemed Gentlemen above, have quite rightly pointed out, the error is purely and simply down to the driver.

A "calling on" signal conveys a simple message to a driver, the line ahead of you is occupied, drive accordingly. If the Driver had drove his Class 150 like a train and not his family car then we wouldn't be reading the accident report.

 

When I worked at Derby PSB, I would use permissive working on a daily basis and never once had to tell a driver that "the line ahead of you is occupied". I find it simply strange, that it's partly the Signalman's fault, for not having verbally reminded a driver what the signal is already telling him.

 

Simon

[bescotbeast]

Permissive working is used many many times every day at Nottingham sometimes with three trains per platform, has been for years. As far as I know there have not been any incidents beyond the odd red face of a train too long for the space available.

I've seen this happen at Brighton where "Bridges" has tried to put a 12 car on top of an 8 car. Normally sorted by send the offending train back to us and then re-offering the train up to them again.

[PaulRhB]

When I worked at Derby PSB, I would use permissive working on a daily basis and never once had to tell a driver that "the line ahead of you is occupied". I find it simply strange, that it's partly the Signalman's fault, for not having verbally reminded a driver what the signal is already telling him.

 

Simon

It's only partly the Signalmans fault if there's insufficient room and then it should only be a casual factor as it's only to avoid an irregular door incident not a collision, that's clearly the drivers responsibility in the rules.

I haven't had time to read it fully yet to ascertain if there was room.

[Titan]

It's only partly the Signalmans fault if there's insufficient room and then it should only be a casual factor as it's only to avoid an irregular door incident not a collision, that's clearly the drivers responsibility in the rules.

I haven't had time to read it fully yet to ascertain if there was room.

 

There was not enough room. There was barely enough length for three cars, let alone the four that the train comprised of. I think people are being a bit unfair on the driver. He was expecting to be signalled in to a platform which had enough room for his train, a not unreasonable expectation. Every time he had done the move before only the far half of the platform had been occupied. All the examples in his training had shown the usual case of only the far half of the platform being occupied. Thus he had learnt that this is what the signal meant, and therefore is what he expected. His inexperience meant that he did not expect or realise that the HST he could see was in the platform in to which he had been signalled. On all other lines he could see far enough ahead to stop short of any obstruction that he could see.  Indeed in that respect the report mentions that his speed was commensurate with what would be expected for a normal stop under permissive working. It was only when he saw that the points were set directing him on to the HST that he realised that the HST was an obstruction rather than a train on an adjacent platform. As the points were within the platform the HST was standing only a little beyond the fouling point, so not only was the platform too short, In order to use any of the platform that was there his train would have been left fouling the other lines. This would have made it look like there was no room at all in front of the HST, not unreasonably reinforcing his belief that this was not an obstruction on his line, especially bearing in mind the training he had been given, and all his experience so far had matched what he had been trained to expect. It was only when he saw the points set in to the platform for the HST that he realised that it was an obstruction, and as the HST was so close to the points there was insufficient time to stop. It is also noted that it is not expected for drivers to check the points are set for the correct route, although it is good practice. In my experience some do, some don't. If this driver had not checked the points he would probably have had no time to brake at all, which could well have led to fatalities.

[caradoc]

Why should the Driver need to check points at all when he was clearly advised by the signal route indicator that he was signalled into Platform 6, and from his route knowledge he should have seen that the HST was in platform 6 ? The Signaller made a mistake, yes, in trying to make life easier for passengers, but the primary responsibility for the collision must rest with the Driver.

[Titan]

He mistakenly thought platform 7 was platform 6 due to inexperience and poor training, neither of which was his fault. Since he could see that what he thought was platform 6 was clear, he was driving at a speed appropriate for that platform.

 

He does not have to check the points at all, it is just good practice, which in this case reduced the speed of the collision. Or would you prefer that all the near misses due to this additional driver vigilance became accidents?

[jamesg]

I haven't re-read it since last night, but didn't the report say that the lifts on that platform were broken so the signaller was trying to make life easier for passengers connecting with the HST in platform 6?

The HST was timetabled to use platform 7 and the 150 multiple unit was timetabled to use platform 8, which would allow passengers to cross directly from one train to the other.  The HST was routed into platform 6 at the request of GWR staff because the buffet on the HST needed restocking, requiring the use of a lift.

 

There must have been a reason not to use platform 5, as using platform 5 would have avoided the permissive working while allowing passengers to transfer directly between trains.

[guzzler17]

If the driver had counted upwards from 3 then, as shown in Figure 12, he would've had a lot more time to realise that it was an HST in 6 rather than a short train at the east end of that platform.

 

Yes I appreciate that a clear signal of whatever sort should mean proceed assuming everything is fine but that's not always the case.  If you take this recent incident as an example

 

https://www.gov.uk/government/news/serious-signalling-irregularity-at-cardiff-east-junction

 

the signalling said OK but the driver realised in time that there was a problem with some pointwork and stopped.

[Kris]

There was not enough room. There was barely enough length for three cars, let alone the four that the train comprised of. I think people are being a bit unfair on the driver. He was expecting to be signalled in to a platform which had enough room for his train, a not unreasonable expectation. Every time he had done the move before only the far half of the platform had been occupied. All the examples in his training had shown the usual case of only the far half of the platform being occupied. Thus he had learnt that this is what the signal meant, and therefore is what he expected. His inexperience meant that he did not expect or realise that the HST he could see was in the platform in to which he had been signalled. On all other lines he could see far enough ahead to stop short of any obstruction that he could see.  Indeed in that respect the report mentions that his speed was commensurate with what would be expected for a normal stop under permissive working. It was only when he saw that the points were set directing him on to the HST that he realised that the HST was an obstruction rather than a train on an adjacent platform. As the points were within the platform the HST was standing only a little beyond the fouling point, so not only was the platform too short, In order to use any of the platform that was there his train would have been left fouling the other lines. This would have made it look like there was no room at all in front of the HST, not unreasonably reinforcing his belief that this was not an obstruction on his line, especially bearing in mind the training he had been given, and all his experience so far had matched what he had been trained to expect. It was only when he saw the points set in to the platform for the HST that he realised that it was an obstruction, and as the HST was so close to the points there was insufficient time to stop. It is also noted that it is not expected for drivers to check the points are set for the correct route, although it is good practice. In my experience some do, some don't. If this driver had not checked the points he would probably have had no time to brake at all, which could well have led to fatalities.

 

What you have described here sounds like complacency. 

[great central]

r

 

Yes I appreciate that a clear signal of whatever sort should mean proceed assuming everything is fine but that's not always the case.g

Old railway proverb. Assume makes an ass of u and me!!

[Titan]

What you have described here sounds like complacency. 

 

I would agree with you if the training he received had included the possibility that a signalman might accidentally direct him in to a platform with insufficient room. The fact that all other times in his experience there was sufficient room might have led to complacency. However from the report it would seem that this possibility was either not covered at all in his training, or had very little emphasis. In other words he was not being complacent, just driving according to his training and (limited) experience.

[Captain Kernow]

Report published today.

 

 https://assets.publishing.service.gov.uk/media/589d9072e5274a0410000000/R022017_170213_Plymouth.pdf

 

Martin.

For some reason, this link doesn't seem to be working.

[Free At Last]

Apparently been moved to https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/591413/R022017_170213_Plymouth.pdf

[martin_wynne]

The permissive rule to drive able to stop short of any obstruction clearly requires that you know which track to look for an obstruction on. If you can see several tracks clear ahead of you and believe you will be using one of them, it doesn't seem unreasonable to increase the speed a bit. Looking at the lie of the points from a distance is all very well in good daylight, but it's no good at night or in bad weather.

 

Do drivers now identify tracks by platform numbers? Are they required to count platforms to know which track the signal refers to? Didn't tracks at one time have names - Down Main, Up Bay, Down Loop, etc.? I remember such names on stencil indicators - as on the right here:

 

Linked from: http://www.roscalen.com/signals/Worcester/050323/SHNorth750.jpg

http://www.roscalen.com/signals/Worcester/ShrubHill.htm

 

Martin.

[Supaned]

that stencil indicator at Worcester is ONLY a route indication - it does not describe which line the train will be travelling along.

[The Stationmaster]

The permissive rule to drive able to stop short of any obstruction clearly requires that you know which track to look for an obstruction on. If you can see several tracks clear ahead of you and believe you will be using one of them, it doesn't seem unreasonable to increase the speed a bit. Looking at the lie of the points is all very well in daylight, but it's no good at night.

 

Do drivers now identify tracks by platform numbers? Are they required to count platforms to know which track the signal refers to? Didn't tracks at one time have names - Down Main, Up Bay, Down Loop, etc.? I remember such names on stencil indicators - as on the right here:

 

 

Linked from: http://www.roscalen.com/signals/Worcester/050323/SHNorth750.jpg

http://www.roscalen.com/signals/Worcester/ShrubHill.htm

 

Martin.

 

I'm not with you Martin.  The route indicators at Plymouth have shown platform numbers since 1960 and that makes wholly logical sense when the lines are named (e.g.) 'Platform 6' and so on.

 

The Worcester signal you have shown refers to routes away from the station - one of which goes towards Birmingham and the other towards Malvern and in the example you have shown that indicator doesn't use the name of a line because if it did it would say 'MAIN' instead of 'B'HAM' (although there is a more subtle reason for that I suspect).

 

As far as the Driver in the Plymouth incident is concerned - and this is very good reason for concern about both his training and competence/route knowledge.  He had a very clear indication of which line his train was signalled to and an equally clear indication of the condition under which the signal had been cleared.  No need to 'think' he might have to stop at whatever point because the signal indication was absolutely crystal clear that he should be prepared to stop short of any obstruction, and that 'obstruction' could be anywhere on that line.  And he should have known exactly where he was going otherwise he'd should never have signed his Route Card.

 

I must admit, for a variety of reasons (now including this incident), to some misgivings about the quality and extent of road knowledge on the part of some Drivers.  This is not a new thing and there were some appalling examples of poor road knowledge in BR days but steps were being taken to deal with that in the late 1980s/early 1990s as part of the campaign to reduce SPADs. I do wonder if in the rush to get Drivers into revenue earning work some operators might have lost that emphasis while there has been clear evidence that at least two others (both minor operators as it happened) never had it in the first place.

[locoholic]

Inexperienced drivers, signallers and permissive working are a terrible combination.

[TheSignalEngineer]

Permissive working is used many many times every day at Nottingham sometimes with three trains per platform, has been for years. As far as I know there have not been any incidents beyond the odd red face of a train too long for the space available.

Likewise stations such as Birmingham New St, Leeds and Manchester Piccadilly. These stations just don't work without Permissive working in the peak periods.

 

Inexperienced drivers, signallers and permissive working are a terrible combination.

 

In my view the problems are not so much with permissive working as such but this case being an unusual move and the quality of 'Boil-in-the-bag' training nowadays . In the days of intensive freight working around Saltley every train on the Goods lines was on Permissive working. You just sat and waited for the train in front to move off then shuffled along a bit stopping just short of him. All of the drivers were used to it and they understood that when you were on the Goods lines you basically drove on sight until you came to a signal at Danger. Likewise at New Street permissive working has been used since Adam were a Lad. I watched it regularly in the 1950s and don't recall a platform incident put down to being due to a permissive move since that time.

[martin_wynne]

I'm not with you Martin.  The route indicators at Plymouth have shown platform numbers since 1960 and that makes wholly logical sense when the lines are named (e.g.) 'Platform 6' and so on.

 

That's what I was asking. If the track is called "Platform 6" and is shown as that on the signal box diagram, then clearly the driver should have known where he was going.

 

But if it is called "Up Main" on the signal box diagram and the length of concrete alongside it is simply labelled platform 6 for the convenience of passengers, it seems unreasonable to expect the driver to know that numbering for every station on the line. After all the word "platform" does mean a length of raised staging of some sort, not two rails running side by side.

 

To a railwayman "platform" might mean a length of track, but to the rest of the population it means the bit you stand on. That might include a new driver.

 

Martin.

[Oldddudders]

Inexperienced drivers, signallers and permissive working are a terrible combination.

It certainly reminds us that since the railway runs like clockwork and safely almost all the time, the comfortable everyday norm is accepted by staff, as with any job. But rules and regs have evolved over nearly 200 years to make it safe, and training in their application is critical on the day things are not quite normal. And it is the 'remote chance' stuff that we all tend to forget.

 

I have a bit of a parallel on the signalling side, where the training on a particular regulation must have seemed implausible and unnecessary to almost all the trainee signalmen (as they then were). This covered the eventuality of a signal being 'missing', which even my colleague Colin, who ran the courses, struggled to justify as a concept. After all, they can't phone in sick, can they? Until. In the small hours one morning, Colin was supervising changing of signal-heads, with a crane. A new head was almost in position when something went wrong, and it dropped thirty feet into a scrap metal yard, where it suddenly looked very much part of the scenery! So there was no signal for a couple of days. Had our unfortunate driver been convinced during his training that an occupied platform might mean an obstruction anywhere along it, rather than the distant train he was expecting and had always found before, we wouldn't be discussing this now.

[Glorious NSE]

 

There was not enough room. There was barely enough length for three cars, let alone the four that the train comprised of. I think people are being a bit unfair on the driver. He was expecting to be signalled in to a platform which had enough room for his train, a not unreasonable expectation. Every time he had done the move before only the far half of the platform had been occupied. All the examples in his training had shown the usual case of only the far half of the platform being occupied. Thus he had learnt that this is what the signal meant, and therefore is what he expected. His inexperience meant that he did not expect or realise that the HST he could see was in the platform in to which he had been signalled. On all other lines he could see far enough ahead to stop short of any obstruction that he could see.  Indeed in that respect the report mentions that his speed was commensurate with what would be expected for a normal stop under permissive working. It was only when he saw that the points were set directing him on to the HST that he realised that the HST was an obstruction rather than a train on an adjacent platform. As the points were within the platform the HST was standing only a little beyond the fouling point, so not only was the platform too short, In order to use any of the platform that was there his train would have been left fouling the other lines. This would have made it look like there was no room at all in front of the HST, not unreasonably reinforcing his belief that this was not an obstruction on his line, especially bearing in mind the training he had been given, and all his experience so far had matched what he had been trained to expect. It was only when he saw the points set in to the platform for the HST that he realised that it was an obstruction, and as the HST was so close to the points there was insufficient time to stop. It is also noted that it is not expected for drivers to check the points are set for the correct route, although it is good practice. In my experience some do, some don't. If this driver had not checked the points he would probably have had no time to brake at all, which could well have led to fatalities.

 

I would agree with you if the training he received had included the possibility that a signalman might accidentally direct him in to a platform with insufficient room. The fact that all other times in his experience there was sufficient room might have led to complacency. However from the report it would seem that this possibility was either not covered at all in his training, or had very little emphasis. In other words he was not being complacent, just driving according to his training and (limited) experience.

 

 

I don't have an intention of being unduly harsh on the driver, specifically, as a person, but it is pretty undeniably that it's the drivers error that turned this from being a cock-up into a collision. The issue seems pretty clearly to be to do with either his training, or in verifying he understands his training, rather than his intentions.

 

Let me put it differently - I agree - if I'd understood the rules/signal to mean that I had a clear road into that platform then I'm sure I'd have acted similarly. But that doesn't change the fact that the actions I'd have then taken were fundamentally wrong ones, as the understanding they were based on was fundamentally wrong.

 

A subsidiary signal means you drive prepared to stop in the distance you can see to be clear, that hasn't changed in....if not forever, in a very long time so far as I can see - nobody should be being taught to expect it to always mean you are clear for 'half a platform' or 'up to X car stop', even if that is the usual situation. Whilst he might not have had a huge amount of varied route knowledge, getting a 'calling on' signal ought to have been reasonably feasible in at least two other places he'd have worked to (Par and Penzance) - if not more so it shouldn't have been some kind of abstract concept that could never come up.

 

Whilst in this specific case the signaller did create the issue by his misjudgement, in another scenario the signaller may not be able to see the platform and be basing his decision on what other staff have told them - or the signaller could have been misinformed of what the inbound train formation is (believing it to be a 2 instead of 4, or 4 instead of 8 or whatever) and so be confidently expecting the train to fit a space he knows is available - or where a join is occurring the inbound crew of the first train could have misunderstood their instructions and left their train at the wrong end of the platform - those kinds of things should be very rare, and it's right to say they shouldn't happen, but the protection against them causing a collision is the driver controlling his train prepared to stop short of any obstruction. It's pretty fundamental.

[russ p]

That's what I was asking. If the track is called "Platform 6" and is shown as that on the signal box diagram, then clearly the driver should have known where he was going.

 

But if it is called "Up Main" on the signal box diagram and the length of concrete alongside it is simply labelled platform 6 for the convenience of passengers, it seems unreasonable to expect the driver to know that numbering for every station on the line. After all the word "platform" does mean a length of raised staging of some sort, not two rails running side by side.

 

To a railwayman "platform" might mean a length of track, but to the rest of the population it means the bit you stand on. That might include a new driver.

 

Martin.

That's what route knowledge is all about,

[martin_wynne]

A subsidiary signal means you drive prepared to stop in the distance you can see to be clear

 

But this makes sense ONLY if you know which track to look along for any obstruction.

 

In this case the driver could have stopped in the distance he could see to be clear.

 

The problem is that he was looking along the wrong track. Which is surely the problem here, and not his speed or driving or understanding of the permissive rules?

 

Martin.

[russ p]

If a driver doesn't know what to look at or where to look he shouldn't be driving, sorry!