Signalling problems Guildford/ Reading / Waterloo 18/07 lines

[ABDeltic]

What's the situation this morning? I have to travel from Portsmouth back to London

Seems fine so far.

 

[Islineclear3_1]

Seems fine so far.

Hope so, thanks

 

[norbitonflyer]

Seems to have been a residual problem on the Shepperton branch this morning. 2H05 (0640 Kingston to Shepperton, unit 701044) was terminated at Hampton Wick, ran ECS to Shepperton and back to Fulwell, then took up its diagram again at Fulwell as 2H14 at 0723 (should have been the 0711 ex-Shepperton). Signalling issues were cited as the cause. Not sure what the issue was, as this was not the first train of the day down the branch.

 

[Deepgreen]

T3 Possession unsurprisingly has been setup to get to the root of this issue. Hopefully the night team solve it. Best of luck to them.View attachment 184600

Where's "Earlesfield"...?

 

[NSEWonderer]

Where's "Earlesfield"...?

Haha must be a typo on siggy

 

[TheManBehind]

Also an ex-Controller, although never for a railway as busy as Waterloo (Scotland in my case, with different challenges).Contingency Plans are workable when the disruption occurs just after the service has shut down for the night, giving Control time to arrange and disseminate the amended train plan. If it happens once the service has started things are not so simple, and losing half the platforms at a station as busy as Waterloo must be close to unmanageable. The number of trains and Traincrew displaced as already mentioned will be immense.

Which is not to make excuses for the railway, signalling failures such as this should simply not happen - And especially should not happen twice. The design and reliability of the system needs forensic examination.

Yup - paper-perfect-plans! Although I'd rather be in control for the SWML than Scotland, at least the SW division was a pretty simple layout to play with!

There is likely an immediate contingency plan for partial or complete losses of platforms at Waterloo. And the service pattern would indeed have to be thinned, and the plan would probably tell you what service groups and what is cancelled / terminated short etc.

However, this is very different - as you know - to having an off-the-shelf train service plan for the various stock if it terminates at Clapham Junction / Wimbledon / Surbiton etc. Trains step up, don't form their booked workings, shunt, crew are displaced etc.

What the industry does need to be better on is not the actual contingency plans themselves - which prescribe the number of tph and the service patterns - but the practicalities of doing so with the stock and crew constraints. These need to be developed somewhat in advance and flexibility of crew arrangements required to pull them off diagrams sooner so long as they don't trip over the usual issues (late PNBs / finish at wrong destination).

Workload is of course another issue, but having the fundamental, 'off-the-shelf' "train service plan for the SX contingency plan 24A" already available to utilise and share, can surely help massively. But it can't be done on the fly with the train service controllers who are already firefighting the issue - it's another issue entirely to essentially make VSTPs for the rest of the day.

You're absolutely right - the plans themselves are generally sensible in principle, but rely on crew and stock all being in the new right place at a new right time - which is...challenging, to say the least.

 

[GordonT]

I suppose when Waterloo is in meltdown and the effects ripple out to the rest of SWR territory there is a slight crumb of comfort in that there is less impact on neighbouring TOCs than would be the case in many other TOC's locations.

 

[markle]

You're absolutely right - the plans themselves are generally sensible in principle, but rely on crew and stock all being in the new right place at a new right time - which is...challenging, to say the least.

A plan isn't worth the paper it's on if it's not implementable in practice.

I guess I'd made a grave error assuming that the people who drafted a contingency plan would have drafted something that could also be implemented.

 

[Horizon22]

A plan isn't worth the paper it's on if it's not implementable in practice.

I guess I'd made a grave error assuming that the people who drafted a contingency plan would have drafted something that could also be implemented.

Not really.

They (NR) look at the infrastructure available and look at how many trains can be reliability accommodated within that from a mostly planning, performance and signalling perspective. The operator(s) then negotiate (within reason) what those services may actually be.
For example, if it is a mix of stopping and express traffic now sharing the same line, there may be differences.

It is up to the operator to then deliver that plan, but the contingency plan will be theoretical maximum number of trains, not the minimum.

Full blocks and complete loss of infrastructure is much harder than partial blocks to accommodate as you get into the realm of trying to run shuttles as opposed to an overall reduction in booked trains. The former is much harder to manage from a crew, stock and customer information perspective.

 

[embers25]

I appreciate its's not simple, but surely even a 8-12 coach train every 20-30 mins Clapham/Wimbledon to Woking/Guildford and all other trains terminate at Woking/Guildfiord/Basingstoke wouldn't have been impossible.

Also, when so few trains are being allowed through, running the 1147 and 1247 from Salisbury through to Waterloo when both were only 3 coaches was a very odd decision.

 

[RGM654]

Seems to have been a residual problem on the Shepperton branch this morning. 2H05 (0640 Kingston to Shepperton, unit 701044) was terminated at Hampton Wick, ran ECS to Shepperton and back to Fulwell, then took up its diagram again at Fulwell as 2H14 at 0723 (should have been the 0711 ex-Shepperton). Signalling issues were cited as the cause. Not sure what the issue was, as this was not the first train of the day down the branch.

What kind of signalling issue prevents a train from running in passenger service but allows it to run as empty stock?

 

[GW43125]

What kind of signalling issue prevents a train from running in passenger service but allows it to run as empty stock?

Guard caught up in the chaos the previous night needing their 12hrs rest?

 

[TheManBehind]

A plan isn't worth the paper it's on if it's not implementable in practice.

I guess I'd made a grave error assuming that the people who drafted a contingency plan would have drafted something that could also be implemented.

Trouble is, when do you plan for that plan to apply? To make a plan that's easily implemented in the current environment, you're talking a train-by-train, and even crew-by-crew plan, that has to account for the time of day as well, as you're operating these days with minimal full-crew/stock working and minimal spare crew coverage, and rationalised (read: reduced) route knowledge. With that in mind, 0900's contingency plan actions at 1000 (ie an hour into disruption) is likely to be completely different to 1000's contingency plan for an incident that started at 1000 (if that makes sense).

It's more expedient, and more sensible, to lay out a set of service principles that may be able to be adhered to in best case conditions, but with an awareness that in the event of something like what we've seen at Waterloo, that it may not be something that's possible, or indeed sensible, to do - several thousand people at Woking who've gone there because shuttle trains managed to get them that far and optimism said there might be a train to Waterloo afterward when the reality is there wasn't isn't much fun for them or the staff.

 

[norbitonflyer]

Guard caught up in the chaos the previous night needing their 12hrs rest?

Who knows? Butn it's odd it was only taken out of passenger service after it had got to Hampton Wick.

 

[Craig1122]

several thousand people at Woking who've gone there because shuttle trains managed to get them that far and optimism said there might be a train to Waterloo afterward when the reality is there wasn't isn't much fun for them or the staff.

This does seem to work on the assumption that Waterloo is the only destination people want to get to when in reality there are huge flows elsewhere on the network. It's poor when a signal failure in London often means no service for hours on end as far away as Portsmouth, Exeter or Weymouth and that local passengers are effectively abandoned.

There does seem to be either an inability or unwillingness to operate any kind of shuttle or ad hoc service. Or even to add extra stops whereas non stopping is common. It always seems to be a case of running as timetabled or nothing at all.

It certainly contrasts poorly with London Underground who will run shuttles either side of any infrastructure failure. I suspect both LU and the TOC's could learn a lot from each other if there was a willingness to be open minded!

 

[Bread & cheese]

This does seem to work on the assumption that Waterloo is the only destination people want to get to when in reality there are huge flows elsewhere on the network. It's poor when a signal failure in London often means no service for hours on end as far away as Portsmouth, Exeter or Weymouth and that local passengers are effectively abandoned.

This is a very important point.
I have long been of the opinion that the SWR network would be significantly more resilient, if as many stretches of line as possible had local shuttle services as part of the regular timetable, such as the Bournemouth to Winchester train.
I also doubt that the DfT would countenance such an idea.

 

[Falcon1200]

It certainly contrasts poorly with London Underground who will run shuttles either side of any infrastructure failure.

The SWR network is rather different from the Underground however - Weymouth is rather further from Waterloo that Epping is from West Ruislip, for example, making amending Traincrew workings very much simpler! However it does seem poor when multiple cancellations occur so far from the point of disruption, are SWR's crew diagrams so complex that arranging shuttles is impossible, or at least very difficult?

 

[Goldfish62]

The SWR network is rather different from the Underground however - Weymouth is rather further from Waterloo that Epping is from West Ruislip, for example, making amending Traincrew workings very much simpler! However it does seem poor when multiple cancellations occur so far from the point of disruption, are SWR's crew diagrams so complex that arranging shuttles is impossible, or at least very difficult?

Graham Eccles when recently interviewed on the Green Signals Podcast did consider that they are overly complex and need to be simplified if swifter recovery from disruption is ever going to be achieved.

 

[TheManBehind]

The SWR network is rather different from the Underground however - Weymouth is rather further from Waterloo that Epping is from West Ruislip, for example, making amending Traincrew workings very much simpler! However it does seem poor when multiple cancellations occur so far from the point of disruption, are SWR's crew diagrams so complex that arranging shuttles is impossible, or at least very difficult?

Unfortunately, yes. Simplest diagrams when I was crew were often Farnham-Waterloo-Basingstoke with driver and unit, then Basingstoke-Waterloo-Alton-Farnham with driver and a new unit. The worst would have been Farnham-Waterloo with driver and unit, Waterloo-Staines via Hounslow (new unit), Staines-Waterloo via Hounslow (new unit), Waterloo-Reading (new driver and unit), Reading-Farnham empty (or by taxi if you outstabled).

ETA: Outer depots too! Waterloo-Weymouth services frequently had a crew change which could have been any combination of Bournemouth, Southampton, Basingstoke or Woking.

And we had broadly simple diagrams with a high percentage of full-crew working - suburban depots I suspect had a much bigger challenge. This was all before we had a reduction in full-crew working in the name of the God of Net Subsidy Reduction!

 

[Annetts key]

The SWR network is rather different from the Underground however - Weymouth is rather further from Waterloo that Epping is from West Ruislip, for example, making amending Traincrew workings very much simpler! However it does seem poor when multiple cancellations occur so far from the point of disruption, are SWR's crew diagrams so complex that arranging shuttles is impossible, or at least very difficult?

It's not just the SWR network. A large part of the railway has tied itself up in knots making everything overly complex in the name of flexibility, under the illusion that this is a good idea, because it apparently saves money.

Yes, it is nice to have plenty of through trains rather than making the passengers change from one "shuttle service" to another. But, there has to be a better way of using your train crew (and other staff, because it's not only a train crew issue).

The overall system should be designed to be able to cope with at least some major disruption without the whole service collapsing or being cancelled.

 

[Farnborough]

I have long been of the opinion that the SWR network would be significantly more resilient, if as many stretches of line as possible had local shuttle services as part of the regular timetable, such as the Bournemouth to Winchester train.

I echo this... it would be more resilient (IMHO) to have:

• Shuttles running all stations between Waterloo/Woking, Woking/Basingstoke, Basingstoke/Southampton, Southampton/Bournemouth, Bournemouth/Weymouth
• A semi-fast service only stopping at Woking, Farnborough , Basingstoke, Southampton, Bournemouth and Weymouth - plus maybe Eastleigh and Poole

This sort of thing works well in Switzerland...

 

[Craig1122]

I echo this... it would be more resilient (IMHO) to have:

• Shuttles running all stations between Waterloo/Woking, Woking/Basingstoke, Basingstoke/Southampton, Southampton/Bournemouth, Bournemouth/Weymouth
• A semi-fast service only stopping at Woking, Farnborough , Basingstoke, Southampton, Bournemouth and Weymouth - plus maybe Eastleigh and Poole

This sort of thing works well in Switzerland...

It's an interesting idea in theory, trouble is that "Station x loses direct service to London" is a guaranteed headline generator that any politically savvy manager will shy away from. And of course you also have to separate the crew/stock diagrams as that appears to be half the problem anyway.

 

[GordonT]

I echo this... it would be more resilient (IMHO) to have:

• Shuttles running all stations between Waterloo/Woking, Woking/Basingstoke, Basingstoke/Southampton, Southampton/Bournemouth, Bournemouth/Weymouth
• A semi-fast service only stopping at Woking, Farnborough , Basingstoke, Southampton, Bournemouth and Weymouth - plus maybe Eastleigh and Poole

This sort of thing works well in Switzerland...

Switzerland tends to have invested in generously resourced infrastructure, fleet and manpower.

 

[bahnause]

Switzerland tends to have invested in generously resourced infrastructure, fleet and manpower.

Not really. The infrastructure is built based on the desired timetable and does not support much more than that. Everything is highly automated and controlled from a few signaling centers. There is not a lot of manpower available, but a lot of flexibility (drivers sign quite a lot of routes and traction) and less fragmentation. Only very few restrictions regarding vehicles and routes, almost everything can go everywhere.

There are numerous contingency plans in place for line closures, diversions, and individual measures for delayed trains. These have been agreed upon with all parties involved, e.g., bus operators and TOC. It's all about planning and flexibility.

 

[Farnborough]

trouble is that "Station x loses direct service to London" is a guaranteed headline generator that any politically savvy manager will shy away from.

True... but that's mostly a symptom of our typical 1 or 2tph service patterns where, if a connection is poorly scheduled or missed, you have a 30 minute wait.

With 4tph, it's always less than a 15 minute connection - which isn't that long, and you will probably make up on the fast section of a journey.

 

[Craig1122]

True... but that's mostly a symptom of our typical 1 or 2tph service patterns where, if a connection is poorly scheduled or missed, you have a 30 minute wait.

With 4tph, it's always less than a 15 minute connection - which isn't that long, and you will probably make up on the fast section of a journey.

I do think a lot of station pairs outside London are actually quite poorly served with the current timetable on the SWML due to services skip stopping on different patterns. It's certainly interesting to contemplate where a radically different approach would lead, and if it could also be more robust in disruption. Probably one for a new thread!