Single points of failure: are the UK's railways less resistant than they used to be?

[Western 52]

It seems these days that frequent delays and cancellations are caused by factors the railway coped better with years ago. Single incidents can bring the timetable down quickly - point failures, failed trains, crew going sick last minute, etc.

In recent decades we've seen reductions in infrastructure and staffing levels whilst the railway became busier. It seems that resilience to problems that arise regularly is now too low, making services less reliable.

Increased resilience costs money of course, but more reliable services will attract more custom.

What do others think?

 

[zwk500]

Increased resilience costs money of course, but more reliable services will attract more custom.

This is the eternal balance service industries attempt to strike.

There have been important moves towards standardisation of parts so that spare stocks are more cost-effective to maintain, but as ever the railway is a living beast and so there will always be odd places or old installations that haven't been captured yet. Staffing levels are a kettle of fish I'm not quite willing to look into yet.

In terms of a train sitting down bringing the job to a halt - there's only so much resilience you can provide on a high-spec (speed, frequency or tonnage) railway before it becomes more expensive than the cost of delays. There are certain areas where reversible working would be very much appreciated, but equally there are often incidents where a train just sits down at the wrong place and there's nowt that could have been done differently.

My big concern is how many embankments, cuttings, tunnels and bridges seem to be suffering in the more extreme weather we've been experiencing. NR has already shown it's records are far from perfect on this front in areas, and that it's engineers are stretched thinly trying to keep the trains moving instead of the formation. I think in the not too near future we will see some lines face very stern discussion about closure vs total rebuild.

 

[Brubulus]

My big concern is how many embankments, cuttings, tunnels and bridges seem to be suffering in the more extreme weather we've been experiencing. NR has already shown it's records are far from perfect on this front in areas, and that it's engineers are stretched thinly trying to keep the trains moving instead of the formation. I think in the not too near future we will see some lines face very stern discussion about closure vs total rebuild.

I have noticed that mechanically reinforced earth is little used in the rail industry, while it is a far more common site in the road sector and is a more stable, cost and space effective way to build embankments. I do think there should be a national move towards conducting larger track and signal renewals where an entire line is renewed infrastructurally at the same time, instead of trying to be smart and only renewing short sections at a time. Slab track should be considered on routes which require an exceptionally high level of maintainance, such as the Great Eastern.

 

[Gloster]

I wonder if the splitting of the railway into so many different companies, each solely interested in their own little area, is a factor. Back in the eighties plenty of station, yard and even office staff could do jobs like clipping points, winding points or manning level-crossings: I have mentioned how, as a Junior Railman with only a month or two’s service, I was sent out to man the barriers at Chilworth when the CCTV had failed (”Just do what the signalman tells you.”) How often nowadays do staff get taken off their normal job to do something that is the responsibility of another company in order to get things moving again rather than stopping the job while you wait for somebody from the ‘right’ company to turn up?

 

[The Planner]

I wonder if the splitting of the railway into so many different companies, each solely interested in their own little area, is a factor. Back in the eighties plenty of station, yard and even office staff could do jobs like clipping points, winding points or manning level-crossings: I have mentioned how, as a Junior Railman with only a month or two’s service, I was sent out to man the barriers at Chilworth when the CCTV had failed (”Just do what the signalman tells you.”) How often nowadays do staff get taken off their normal job to do something that is the responsibility of another company in order to get things moving again rather than stopping the job while you wait for somebody from the ‘right’ company to turn up?

I don't understand, its all Network Rail for an infrastructure failure, or are you on about fault teams not being responsible for the type of asset?

 

[Gloster]

I don't understand, its all Network Rail for an infrastructure failure, or are you on about fault teams not being responsible for the type of asset?

No, but I presume that if a failure, for example a points failure which would be (at least partially) solved by using a point clip, happens near to a station the staff there can’t go out and put a clip on: a MOM or whatever has to be found. I have suffered a couple of occasions where this sort of situation seems to have occurred.

 

[Hadrian]

Two points about resilience today compared to the past:

a) When main lines were signalled from many (labour intensive) manual boxes almost all failures of signalling were highly localised. Keeping trains moving during failures was easier because there were the necessary competent staff (operating and permanent way) based fairly locally and able to put in single line working, clip points and/or provide flagmen at signals as necessary relatively quickly. The number of trailing crossovers between running lines has been significantly reduced as signalboxes closed.

b) When main lines were signalled from many (labour intensive) manual boxes almost all failures of trains could be dealt with relatively quickly because: competent staff were more readily available (as in (a) above), motive power was much more readily available to assist failed trains, and almost everything could be coupled to anything else.

In 1971 I was on a fast train from Birmingham to Derby which suffered a locomotive failure (I think Class 45) approaching Kingsbury Junction. The next locomotive hauled fast train was coupled up behind and the whole cavalcade (over 20 coaches and the 2 locomotives) proceeded to Derby without much further delay. At Derby the failed locomotive was replaced while the trains were split and the assisting train detached, set back, called at another platform and then proceeded northwards ahead of the train which had failed. As a then recent recruit to the railways, travelling as a passenger, I was very impressed with the speed with which the whole incident was sorted out.

My Swiss colleagues find it astonishing that main lines in the UK are still not fully bi-directionally signalled with relatively frequent facing and trailing crossovers to provide flexibility, not just for failures but also for permitting fast trains to overtake slower (usually freight) trains.

 

[SussexSeagull]

It is certainly the exception now at the weekend when I get a train anywhere and there is not some sort of diversion in place. To be fair I assume they aren't doing engineering works for the fun of it but this suggests some of the infrastructure is on it's last legs.

 

[Dr Hoo]

In recent decades we've seen reductions in infrastructure and staffing levels whilst the railway became busier. It seems that resilience to problems that arise regularly is now too low, making services less reliable.

What sort of reductions in infrastructure did you have in mind? Apart from obviously redundant facilities such as branch lines to closed collieries and power stations there seems to have been a modest stream of additions.

For example, on the ECML we have seen additional approach tracks to King's Cross, the link to take some services to Thameslink, proper six-tracking Finsbury Park-Alexandra Palace, independent Hertford Loop terminating platform at Stevenage, Hitchin Flyover, major re-modelling through Peterborough/Werrington, the Allington Chord in the Grantham area, various enhancements through the Doncaster area and quadrupling of Holgate Junction at York.

I get that the number of trains has increased, so the impact of incidents can be greater, but don't think that it's about cuts. And I'm certainly not sure that getting the booking clerk at Grantham to go and clip a set of points at Stoke Summit is the answer either.

 

[Western 52]

What sort of reductions in infrastructure did you have in mind? Apart from obviously redundant facilities such as branch lines to closed collieries and power stations there seems to have been a modest stream of additions.

For example, on the ECML we have seen additional approach tracks to King's Cross, the link to take some services to Thameslink, proper six-tracking Finsbury Park-Alexandra Palace, independent Hertford Loop terminating platform at Stevenage, Hitchin Flyover, major re-modelling through Peterborough/Werrington, the Allington Chord in the Grantham area, various enhancements through the Doncaster area and quadrupling of Holgate Junction at York.

I get that the number of trains has increased, so the impact of incidents can be greater, but don't think that it's about cuts. And I'm certainly not sure that getting the booking clerk at Grantham to go and clip a set of points at Stoke Summit is the answer either.

What I had in mind was the removal of various loops and crossovers, so trains are unable to get past blockages such as failed trains. Lack of staff to implement single line working past blockages too. Years ago there was more flexibility as there was more infrastructure.

 

[Starmill]

I have noticed that mechanically reinforced earth is little used in the rail industry, while it is a far more common site in the road sector and is a more stable, cost and space effective way to build embankments. I do think there should be a national move towards conducting larger track and signal renewals where an entire line is renewed infrastructurally at the same time, instead of trying to be smart and only renewing short sections at a time. Slab track should be considered on routes which require an exceptionally high level of maintainance, such as the Great Eastern.

As well as being far more expensive, installing slab track takes much longer. It also takes a very long time to renew. Just a platform length at Glasgow Central was a couple of weeks' worth of work.

 

[jagardner1984]

It certainly feels fair to say we have less main line routes but far more intensive services on them - when the line is full in terms of train capacity; we lengthen platforms to mean more and more people can be carried, all of which means in periods of disruption many more people are disrupted. To have real resilience you surely need the ability to say at Preston “ah we have signal failure ahead on this Glasgow bound train, we’ll divert to Edinburgh via …. Another route and get onto Scotrail to get a special into Waverley to take this train full.”

There will now be numerous examples of whataboutery where this isn’t possible. But it should be. A few weeks ago LNER were cancelling dozens of services on a busy day because of the Colton Junction failure and resulting delays. But if Stansted runway is closed because of an at risk flight coming in, we divert to Luton and we sort onward connections. If the M6 is closed, we signpost diversions.

If the railway is meant to be a serious alternative to either - the numerous examples of “all lines blocked, get off here, maybe come back tomorrow” really need to stop.

 

[Starmill]

It is certainly the exception now at the weekend when I get a train anywhere and there is not some sort of diversion in place. To be fair I assume they aren't doing engineering works for the fun of it but this suggests some of the infrastructure is on it's last legs.

There's a lot of "technical debt" in the condition of the infrastructure. It's degrading on the whole more quickly than we're repairing and replacing it.

There are only two solutions, one is to put substantially more resources into maintenance and renewals (real terms increases of a few percentage points per year for a sustained period of at least a decade) or to rationalise and prioritise the infrastructure we have. The current approach makes the latter very difficult and of course the former is basically impossible, Network Rail can't magic up plant and engineers from thin air.

My big concern is how many embankments, cuttings, tunnels and bridges seem to be suffering in the more extreme weather we've been experiencing. NR has already shown it's records are far from perfect on this front in areas, and that it's engineers are stretched thinly trying to keep the trains moving instead of the formation. I think in the not too near future we will see some lines face very stern discussion about closure vs total rebuild.

This is a serious issue. Even without the threat of a full rebuilt vs closure on condition grounds, it remains my view that Network Rail is well underfunded to cover the cost of reactive maintenance to the kind of weather events which are now commonplace. Very little capital funding, proportionally speaking, is going into the areas which will be most affected by climate change.

My Swiss colleagues find it astonishing that main lines in the UK are still not fully bi-directionally signalled with relatively frequent facing and trailing crossovers to provide flexibility, not just for failures but also for permitting fast trains to overtake slower (usually freight) trains.

The timetable structure in Switzerland may allow overtaking on two-track routes, but in general that's not the case here.

 

[Dr Hoo]

What I had in mind was the removal of various loops and crossovers, so trains are unable to get past blockages such as failed trains. Lack of staff to implement single line working past blockages too. Years ago there was more flexibility as there was more infrastructure.

Fair enough, and your semaphore signal avatar clearly indicates a generally historical perspective. I had initially thought that the thread was intended to be about current infrastructure re-modelling.

To my mind the tipping point was the 1968 Transport Act with its Surplus track Capacity [hasty elimination thereof] Grants. These basically funded British Rail capital expenditure for rationalisation on a five-year-use-it-or-lose-it basis. Rapid de-quadrification, singling, ripping out of loops, abolition of wayside signal boxes, etc. followed.

Unfortunately the available budget from the Department of Transport to BRHQ had to be 'bid for' on a regional basis. So every region came up with loads of schemes in the hope that some would get through. The 'unsuccessful' schemes then had a tendency to become the baseline/default for the sort of layout that would be needed when any future re-signalling or track renewals came along even if not specially funded. Single lead junctions were an especially pernicious form of this type of planning because of course they facilitated easy subsequent singling of any connecting route.

All this coincided with the post-Beeching 'basic railway', especially conductor guard operation that enabled de-staffing of stations. Automatic half barrier level crossings enabled the abolition of many signal boxes and removal of crossing keepers. These were seen at the time as 'saving' the railway, along with fairly meagre year-to-year subsidy/grants in the early years but of course they set in train a fundamental constriction in resilience.

I joined BR as a teenager in the early 1970s and as a junior clerk was able to read some of the files in idle moments. It was patently obvious even to my youthful eyes that the steam-hauled train to school as an 11-year-old, running on a double-track line between closely-spaced staffed stations, each with their own signalbox(es), etc. had already largely disappeared.

These ships sailed over 50 years ago.

 

[The exile]

"Thanks to" the internet and Social Media, we actually know about so many more incidents in (near) real time - if it wasn't enough to make the broadcast news (which would usually mean fatalities) the first most 1970s enthusiasts would hear about it would have been a couple of months later in the Railway Press.
On top of that, many practices that were condoned (unofficially or even officially) years ago no longer are - whether rightly or wrongly is here not the issue - some of those may well have sped up a return to what at least for the passenger would have appeared to be "normal service".

That said - in an entirely unscientific look at may own travel - given the vast increase in my frequency of travel compared with 40 - 50 years ago - the incidence of severe delays doesn't seem to have increased greatly - their cause, however is different. Most of the recent ones (last 10 years or so) have been due to "person hit by a train" - whereas journeys in the 1970s seemed to be a lottery of whether you'd end the journey with the same loco you'd started with. I can remember some very long waits at various out of the way points of the ECML in particular (if only I knew all the details of the traction - particularly the "white knights"!)

 

[swt_passenger]

I have noticed that mechanically reinforced earth is little used in the rail industry, while it is a far more common site in the road sector and is a more stable, cost and space effective way to build embankments...

But how often are NR building brand new railway embankments? Presumably HS2 is using more up to date methods?

The evidence at the landslip site at Hook seems to be that repairs are of a much better quality than the original, but it seems all they’ll ever be funded to do is repairs, no one is proposing pre-emptive reconstruction of embankments over long distances (yet).

 

[Horizon22]

What I had in mind was the removal of various loops and crossovers, so trains are unable to get past blockages such as failed trains. Lack of staff to implement single line working past blockages too. Years ago there was more flexibility as there was more infrastructure.

In the “last couple of decades” includes anything from 2000 onwards, and in that timeframe there’s been almost no infrastructure removal on key routes - indeed it’s increased. Lots of evidence suggests that rationalising and simplifying pointwork around major terminals actually reduces delays.

 

[zwk500]

no one is proposing pre-emptive reconstruction of embankments over long distances (yet).

And this is quite likely to cost a lot more in the long run...

 

[HSTEd]

Fundamentally the railway network is pushed harder than it has been before, in a stricter safety environment, and under pressure over spiraling costs.

If we were starting from a blank sheet of paper I don't think our railway system would look anything like the one we have now. We have ancient infrastructure with unclear engineering limitations (often constructed using methods that would be deemed amateurish by modern standards) that has major legacy maintenance and operational issues and we are trying to patch it up as it continues to age out.

 

[Adrian1980uk]

Part of the issue is that even the post COVID timetable is pushing the infrastructure harder than ever, just take the GEML, 2 tph from Norwich when there used to be 1, then from Ipswich numerous more tph. The example is that if the lines are blocked at diss for more than an hour it's potentially 4 plus trains affected when it may only have been 2 before.

 

[Taunton]

No, but I presume that if a failure, for example a points failure which would be (at least partially) solved by using a point clip, happens near to a station the staff there can’t go out and put a clip on: a MOM or whatever has to be found. I have suffered a couple of occasions where this sort of situation seems to have occurred.

As the Carmont accident showed, the points at the Carmont signalbox crossover could be reversed by the signaller, but had no facing point locks (on a line used almost wholly by passenger trains - what else is going to need to be reversed there), and had no point clips in the signalbox, so an MOM had to drive through the storm the best part of two hours, delaying everything on the way, just to fit two clips.

Talk about not being joined up. What a waste of time and cost. And what a hazardous journey by road, but presumably if the MOM had a road accident in the flooding it would not be reportable to RAIB as it would happen outside railway premises. So that's alright then.

 

[Pigeon]

no one is proposing pre-emptive reconstruction of embankments over long distances (yet).

And this is quite likely to cost a lot more in the long run...

From http://www.geplus.co.uk/technical-p...lay-embankments-in-the-south-east-15-01-2018/: "The costing model found that preventative engineering would be less cost-effective than maintenance and delay costs over both 20 year and 50 year lifespans."

Translation: Rather than take the trouble to fix dodgy railway embankments that might fall down, properly, and so make sure they don't, they're just going to bodge them up in patches as and when bits do start to fall down, and happily pay the fines when it makes the trains late. For 50 years.

This sort of attitude needs to stop.

 

[al78]

A couple of things spring to my mind:

1. The UK lost around a third of its network during the Beeching years. This has removed duplication which may reduce inefficiency and costs, but means there is often only one practical route to get somewhere, and if it goes belly-up, getting around it can be very difficult and time consuming.

2. On the busiest routes there is very little slack in the timetable to absorb minor niggles, so a small delay can have a ripple effect that propagates through the network locally and takes considerable time to get back to normality if crews and trains are out of place.

3. There seem to be frequent delays (in my part of the SE anyway) caused by a fault with the signalling. Could this be primarily a result of copper wire theft and if so, is it more of a problem not than in the past?

 

[12LDA28C]

Also worth remembering that drivers generally have less route knowledge than they used to and many no longer sign any diversionary routes - this has been discussed on this forum in the past. It's cheaper for a TOC for drivers to just sign core routes and not maintain knowledge of any diversionary routes for the rare occasions when those routes might be used due to infrastructure failure, failed train etc.

 

[zwk500]

A couple of things spring to my mind:

1. The UK lost around a third of its network during the Beeching years. This has removed duplication which may reduce inefficiency and costs, but means there is often only one practical route to get somewhere, and if it goes belly-up, getting around it can be very difficult and time consuming.

Is this true? A lot of the old routes crossed other lines but often only connected via goods-only or very operationally painful spurs. The grouping had led to a lot of connecting lines being built but even then BR had to build a lot of it's own. And of course a lot of dead end branches were close by Beeching as well

3. There seem to be frequent delays (in my part of the SE anyway) caused by a fault with the signalling. Could this be primarily a result of copper wire theft and if so, is it more of a problem not than in the past?

Wire theft tends to follow the price of metals. At the moment copper prices are very high, hence the risk is worth it. The move towards Fibre-optics isn't making much of an impact yet as thieves don't know which is which so cut all of them before finding out they're useless. Signalling delays can also occur for a lot of other reasons as well.

 

[Western 52]

Has the move towards highly centralised signalling systems made signalling less reliable? There seem to be more failures than I remember years ago, and a major issue at one of these signalling centres could presumably affect a wide area. Do these systems have built in redundancy to cope with failures?

 

[HSTEd]

Has the move towards highly centralised signalling systems made signalling less reliable? There seem to be more failures than I remember years ago, and a major issue at one of these signalling centres could presumably affect a wide area. Do these systems have built in redundancy to cope with failures?

The ROCs are built with reserve power supplies and other such things.

Whilst failures affect a larger area, the reality is that failures in signalling in key areas can knock out half the network anyway.

If you have failures at Clapham Junction it doesn't matter if all the rest of the signalling on the SWML or other lines through it works perfectly, the service will fall over.

Whilst some faults might be worse, you bank the savings from centralisation every hour of every day.

 

[zwk500]

Has the move towards highly centralised signalling systems made signalling less reliable? There seem to be more failures than I remember years ago, and a major issue at one of these signalling centres could presumably affect a wide area. Do these systems have built in redundancy to cope with failures?

Yes, they have redundancy, although things like a track circuit failure can't really be mitigated against. ROCs also have data logging which helps speed up fault finding but also gives lots of good data for future designers to iron out issues in later installations (among other uses).

 

[The Planner]

Why is it any different to the move to PSBs in the 60s? Lose one of those and it would have all gone down the pan.

 

[matacaster]

Why is it any different to the move to PSBs in the 60s? Lose one of those and it would have all gone down the pan.

True. Worth noting that Mr Putin would only need to target the major signalling centres to cripple the rail network. Electrification also means that getting the railway working again would take much longer.
Perhaps these issues should be examined by ministry of defence!!