Why aren't the platforms at UK railway stations renumbered?

[zwk500]

Still fairly straightforward.

No it isn't, because there isn't just 1 big database that everything uses. It's 100s of systems all linked together by varying degrees, with a completely separate.document storage system that would need to be updated as well.

Also, general tip - If you've got no specific experience with an industry system, don't assume that nobody else on here has. FWIW, I've had experience of changing platform numbers in NR's planning system and it caused more than enough issues within just that one system, let alone any of the import/export functions.

 

[Falcon1200]

Glasgow Central low level platforms were not originally numbered following on from the station above, they were another example of low level platforms being designated A,B,C &D.

Thanks.

Glasgow Central did have a renumbering when the new platforms 12 and 13 were built on the former cab Road, the old 12 and 13 became 14 and 15.

12 and 13 were IIRC renumbered before the new platforms were completed; On a train one day I heard someone musing whether there wasn't (at that time) a platform 13 because it is an unlucky number!

 

[Railsigns]

Glasgow Central did have a renumbering when the new platforms 12 and 13 were built on the former cab Road, the old 12 and 13 became 14 and 15.

No. The high-numbered platforms at Glasgow Central were renumbered during the resignalling on 27 December 2008, meaning that no further renumbering of existing platforms was needed when the new Platforms 12 and 13 came into use on 10 May 2010.

 

[matchmaker]

No. The high-numbered platforms at Glasgow Central were renumbered during the resignalling on 27 December 2008, meaning that no further renumbering of existing platforms was needed when the new Platforms 12 and 13 came into use on 10 May 2010.

Glasgow Queen Street Low Level was renumbered during the reconstruction works. 10 & 11 became 8 & 9 (Low Level used to have 4 platforms. When 2 were removed during electrification the numbering wasn't changed)

 

[Grumbler]

No it isn't, because there isn't just 1 big database that everything uses. It's 100s of systems all linked together by varying degrees, with a completely separate.document storage system that would need to be updated as well.

Also, general tip - If you've got no specific experience with an industry system, don't assume that nobody else on here has. FWIW, I've had experience of changing platform numbers in NR's planning system and it caused more than enough issues within just that one system, let alone any of the import/export functions.

I was replying to the post about redesigning the systems. I am perfectly well aware of the difficulties when there a number of old systems, as I did work for many years on airline related systems dating back to the days before relational databases, where we experienced similar problems.

 

[The exile]

I was replying to the post about redesigning the systems. I am perfectly well aware of the difficulties when there a number of old systems, as I did work for many years on airline related systems dating back to the days before relational databases, where we experienced similar problems.

Even if you managed to link all the railway computer systems, there’s other systems to be taken into consideration (emergency services, utilities etc - not to mention good old Mk 1 human memories)

 

[algytaylor]

It'd end up being Platform 65,535.

Best reply!

I didn't realise that my daft question would end up being a computer programming discussion, though very pleased (in a way) at the turn of events.

My view/experience has that it's usually best giving 'old' code the benefit of the doubt. Hard-coding the platform numbers, I agree, seems like insanity to me, if you were writing a brand new piece of software. However, what's being dealt with isn't a brand new piece of software, and has almost certainly been written under very different conditions than you'd do it now. As others have pointed out, I can imagine interfacing complex mechanical infrastructure with a software system, which almost by definition is complex in itself, must've been an absolute nightmare. Especially one as safety-critical as the one we're talking about.

I really would be very generous when assessing it. To have written a piece of software that's still in use decades after it's original conception is an incredible feat, regardless of what it is.

 

[-Colly405-]

Surprised that nobody has mentioned Plymouth.
Platforms 3,4,5,6,7,8.
3 being the west-facing bay that throughout the 80's was always unnecessarily announced as "the west end of platform 3", since it had long been severed from being a through platform whereby the east end had been unnumbered to the public.
2 was also formerly a through platform, but was truncated into two halves by the same change as platform 3, but was completely unnumbered as used for stabling only.
1 was I think removed during the 1938-1962 rebuilding...

 

[Parallel]

I think Exeter St David’s has had its platforms renumbered in recent years to allow for a and b sections. These are also used on departure boards and via autoannouncements.

Cardiff Central have recently labelled their platform ends on Platforms 1 & 2 with a and b sections too but it doesn’t look like the ‘back end’ has been updated as departure screens and autoannouncements never make any reference to 1a, 1b, 2a or 2b. (They do to 3a, 3b, 4a and 4b though but they have always done that on those platforms for as long as I can remember!)

 

[BrianW]

Surprised that nobody has mentioned Plymouth.
Platforms 3,4,5,6,7,8.
3 being the west-facing bay that throughout the 80's was always unnecessarily announced as "the west end of platform 3", since it had long been severed from being a through platform whereby the east end had been unnumbered to the public.
2 was also formerly a through platform, but was truncated into two halves by the same change as platform 3, but was completely unnumbered as used for stabling only.
1 was I think removed during the 1938-1962 rebuilding...

I misread that 1 as an I, and was imagining where you might have been removed to, and that that was some protest to make, for some 24 years, over a platform rearrangement removing not just 1 but 2 too!

 

[-Colly405-]

I think Exeter St David’s has had its platforms renumbered in recent years to allow for a and b sections. These are also used on departure boards and via autoannouncements.

From west to east, St Davids is 6-5 4-3 1-2 with 2 being the north-facing bay. I don't recollect any renumbering when a/b were added.

Of course, the numbers didn't change when the station was resignalled (90s?), with plat 1 changing use from Down Western to Up Southern, 3 from Up Southern to Down Southern, and 4 from Down Southern to Down Western, eliminating the Down Western conflicting with every Up or Down Southern...

I misread that 1 as an I, and was imagining where you might have been removed to, and that that was some protest to make, for some 24 years, over a platform rearrangement removing not just 1 but 2 too!

 

[Parallel]

From west to east, St Davids is 6-5 4-3 1-2 with 2 being the north-facing bay. I don't recollect any renumbering when a/b were added.

‘Renumbering’ may be not the right word, maybe I should’ve used the word ‘divided into’ instead, but looks like some sort of change has been made to the back end of the system, as you now occasionally hear trains being given a platform alteration from 6a to 6b etc.

 

[-Colly405-]

‘Renumbering’ may be not the right word, maybe I should’ve used the word ‘divided into’ instead, but looks like some sort of change has been made to the back end of the system, as you now occasionally hear trains being given a platform alteration from 6a to 6b etc.

Ah, got it!

 

[BrianW]

Surprised that nobody has mentioned Plymouth.
Platforms 3,4,5,6,7,8.
3 being the west-facing bay that throughout the 80's was always unnecessarily announced as "the west end of platform 3", since it had long been severed from being a through platform whereby the east end had been unnumbered to the public.
2 was also formerly a through platform, but was truncated into two halves by the same change as platform 3, but was completely unnumbered as used for stabling only.
1 was I think removed during the 1938-1962 rebuilding...

Plymouth Platform 3 in chocolate and cream as per Plymouth North Road and matrix ...

Tamar Valley Train, Platform 3, Plymouth... © N Chadwick cc-by-sa/2.0

SX4755 :: Tamar Valley Train, Platform 3, Plymouth Station, near to Plymouth, England by N Chadwick

www.geograph.org.uk

 

[-Colly405-]

Plymouth Platform 3 in chocolate and cream as per Plymouth North Road and matrix ...

Tamar Valley Train, Platform 3, Plymouth... © N Chadwick cc-by-sa/2.0

SX4755 :: Tamar Valley Train, Platform 3, Plymouth Station, near to Plymouth, England by N Chadwick

www.geograph.org.uk

Excellent!

 

[D6975]

It'd end up being Platform 65,535.

I think I'd prefer 1048575

 

[8A Rail]

Liverpool Lime Street was renumbered a few years back, when the old platform 1 disappeared and 2 new platforms were added between the old platforms 7 and 7. So now the station is still numbered continuously, from 1 to 10.

This coincided with major work to remodel the track work and install a new signalling system.

I have only just noticed this thread. To clarify the above, the platforms were renumbered in 2018 after the refurbishment was completed.

The old Platform One was disconnected and the line taken out. In consequence, old Platform Two became One, Three became Two, Four became Three, Five became Four and with Six becoming Five - all these platforms are under the North Train Shed.

In the South Train Shed, Platform Seven became Six with new Platforms built (on old taxi way) were they became Seven and Eight. The old Platform Eight then became Nine with Platform Nine becoming the new Ten.

The old sidings with in the station were completely taken out (i.e.Sidings A, B, D and E) and space created to widen the existing platforms especially under the North Train Shed.

Hopefully the above, makes it clearer, if not I tried.

 

[swt_passenger]

I have only just noticed this thread. To clarify the above, the platforms were renumbered in 2018 after the refurbishment was completed.

The old Platform One was disconnected and the line taken out. In consequence, old Platform Two became One, Three became Two, Four became Three, Five became Four and with Six becoming Five - all these platforms are under the North Train Shed.

In the South Train Shed, Platform Seven became Six with new Platforms built (on old taxi way) were they became Seven and Eight. The old Platform Eight then became Nine with Platform Nine becoming the new Ten.

The old sidings with in the station were completely taken out (i.e.Sidings A, B, D and E) and space created to widen the existing platforms especially under the North Train Shed.

Hopefully the above, makes it clearer, if not I tried.

Is it all set up now to squeeze in a short P0 then? Advance planning!

 

[Annetts key]

There were (still may be) some signal boxes that had a lever numbered zero…

The biggest cost of renumbering platforms, is where each line needs to be uniquely identified to a train driver as well as to the signaller, other railway staff as well as obviously passengers.

On the approach to the junction(s) where the line(s) diverge, the signal(s) must give route information to the driver. In practice where there are many lines, that means a one or two digit or letter code. For numbered platforms it’s logical for the platform number to be used.

The route information is displayed as an illuminated display (in railway terminology they are either SARI - Standard Alphanumeric Route Indicators OR MARI - Miniature Alphanumeric Route Indicators, or using the old terms, theatres or stencils) as part of the signal so that the driver can clearly see where the train has been routed to. Each SARI or MARI unit is custom built. Furthermore, all the interlocking equipment, including the documentation also would need changing if platforms were renumbered.

Additionally, some old stations have some or all of their structures as listed status. If this includes the signage, that further increases the complexity and cost.

Hence why platform renumbering ends up in the too costly or too difficult box.

It’s unlikely that negative numbers would ever be used. Far too much likelihood of confusion.

From a distance, could a driver clearly see the difference between a -1 and a 1 on the signal they are approaching?

And what about passengers?

I have to chuckle, as most of the perceived 'problems' that have arisen in more recent times (such as platform renumbering after a station rebuild etc) are driven by the computer age systems that tie everything together. Before the computer age things were so much simpler! No apologies for being an old fxxt

The costs of changing the signalling system existed long before anyone on the railway has even heard about computers… Mechanical signals had route indicators…

If there's a logical rule on the railway, chances are its been broken more than once!
Because of the cost of the disruption to operations platform renumbering would cause, you end up with just about every possible anomaly somewhere along the network.

Ahh, back to:
There’s the right way, the wrong way and the railway.
Also,
There’s always an exception to the rule/regulation/instruction/procedure…

 

[K.o.R]

Best reply!

I didn't realise that my daft question would end up being a computer programming discussion, though very pleased (in a way) at the turn of events.

My view/experience has that it's usually best giving 'old' code the benefit of the doubt. Hard-coding the platform numbers, I agree, seems like insanity to me, if you were writing a brand new piece of software. However, what's being dealt with isn't a brand new piece of software, and has almost certainly been written under very different conditions than you'd do it now. As others have pointed out, I can imagine interfacing complex mechanical infrastructure with a software system, which almost by definition is complex in itself, must've been an absolute nightmare. Especially one as safety-critical as the one we're talking about.

I really would be very generous when assessing it. To have written a piece of software that's still in use decades after it's original conception is an incredible feat, regardless of what it is.

Best practice would probably be to have every platform have an "ID" used by the signalling system and a "visible name" to allow renumbering without having to rewrite signal-platform relationships. Signal numbers are, after all, more-or-less arbitrary to the passenger on the platform, probably assigned in order of addition to the system.

 

[WAO]

If it's not been mentioned already, Reading (General) was renumbered on rebuilding. Sadly, it now has only one platform 4, instead of three, 4, 4a and 4b. (4a and 4b were the SR platforms built to replace Reading (South). They sent many an intending west-bound GWR passenger on a jolly ramble towards Waterloo.

WAO

 

[zwk500]

Best practice would probably be to have every platform have an "ID" used by the signalling system and a "visible name" to allow renumbering without having to rewrite signal-platform relationships. Signal numbers are, after all, more-or-less arbitrary to the passenger on the platform,

3 problems with this: 1. There are many occasions where signaller and platform staff need to be 100% clear which line is being talked about, using 1 name is the best way to achieve this. 2. What would you do at terminal or other large stations where Line Names are not obvious to assign? Your proposal still requires whichever system you end up with to being in the middle and be non-sequential, which creates potential issues of clarity for driver's route indications. 3. Numbers can usually be made slightly clearer than letters on a route indication.

probably assigned in order of addition to the system.

No, signal numbers are assigned according to careful numbering systems that (depending on exactly which system was in force when the scheme was numbered) can indicate direction, track, whether or not the signal is 'wrong-road', function and whether or not the signal has a red aspect. They're always numbered sequentially, usually from left-to-right as seen by the Signaller (whether that be on a Frame, Panel, or VDU)

E.g around Tamworth Trent Valley: Odd Numbers apply to Down Signals, Even Number to Up Signals. 3xxx Numbers are for the fast lines, 5xxx numbers are for the Slow lines. 9xxx for Bi-Di signals facing the 'wrong' direction. 1xxx are Shunt Signals. Numbers increase increase from Rugby to Colwich, as the VDU diagram is orientated with South to the left and north to the right.

 

[swt_passenger]

If it's not been mentioned already, Reading (General) was renumbered on rebuilding. Sadly, it now has only one platform 4, instead of three, 4, 4a and 4b. (4a and 4b were the SR platforms built to replace Reading (South). They sent many an intending west-bound GWR passenger on a jolly ramble towards Waterloo.

WAO

Post #32 did cover Reading…

 

[louis97]

‘Renumbering’ may be not the right word, maybe I should’ve used the word ‘divided into’ instead, but looks like some sort of change has been made to the back end of the system, as you now occasionally hear trains being given a platform alteration from 6a to 6b etc.

Arguably you are right to say renumbered, platform's 1, 3 and 6 actually had signage marking the western end of those platforms as A end's. I only ever saw it used once on the CIS, so it was rarely advertised. However the Western end is now the B end, with rather tachy alteration to the signage.

 

[Railsigns]

They're always numbered sequentially, usually from left-to-right as seen by the Signaller (whether that be on a Frame, Panel, or VDU)

The practice for some considerable time has been to identify signals (and points) with numbers that increase in the Down direction, irrespective of the orientation of the layout as it appears on the panel or VDU screen. Obviously, this doesn't apply to lever frames.

 

[Bikeman78]

There would appear to be at least some aversion to platforms being numbered out of sequence. No doubt there are a few exceptions to that, as there always is.

Edinburgh Waverley springs to mind.

 

[8A Rail]

Is it all set up now to squeeze in a short P0 then? Advance planning!

Yes indeed! To be honest, allowing for necessary gauge and signal clearances, it could handle at least a two car emu, may be three at a push but it would be limited to what train would be in Platform One for access. I think the latter is the reason why there is no track there as it would limit length of train on (new) Platform One.

 

[swt_passenger]

Yes indeed! To be honest, allowing for necessary gauge and signal clearances, it could handle at least a two car emu, may be three at a push but it would be limited to what train would be in Platform One for access. I think the latter is the reason why there is no track there as it would limit length of train on (new) Platform One.

Yes, points within a platform will cause problems like that, I know before the most recent alterations at Paddington there was a similar issue with the old P12 being reached through the country end of P11. In the reworked layout P12 was extended through what was P13.

But oddly enough in the context of this thread no one seems worried about the ensuing gap in the Paddington numbering sequence, presumably in future it will be explained as superstition.

 

[Annetts key]

But oddly enough in the context of this thread no one seems worried about the ensuing gap in the Paddington numbering sequence, presumably in future it will be explained as superstition.

And there I was thinking that the Paddington gap is the space between the train and the platform at the country end…

Will the Paddington gap overtake the Watford gap? Or the tube ‘mind the gap’?….

 

[The exile]

And there I was thinking that the Paddington gap is the space between the train and the platform at the country end…

Will the Paddington gap overtake the Watford gap? Or the tube ‘mind the gap’?….

This then raises the question of what the highest numbered “gap” is. Anything higher than Temple Meads’ missing 14? (Obviously the “next one after the highest numbered platform” doesn’t count - it’s not a gap!)