Thameslink Logic

[Mutant Lemming]

Yesterday the 15:40 all stations St.Pancras to Luton was late so they decided to omit all the stops bar Luton Airport Parkway. The 15:54 all stations St.Pancras to St.Albans was also late so they decide to run that non stop to St.Albans (a train which departed with just one passenger on board). There had now been a bit of a gap for an all stations so they decide that the 16:18 fast to Bedford (on time) should call all stations.
Now while there may be "operational reasons" that they have done this for the average passenger they have :
- been turfed off a train that was supposed to take them where they going
- been stood on a crowded platform while they watch another train that was supposed to take them were they going leave virtually empty
- end up crammed on a train that should have been fast but is now slow.

Thameslink have improved in some areas (overcrowding isn't as bad since the new trains arrived) but they are continually poor in how they deal with service recovery and communicating with their long suffering customers.

 

[Mag_seven]

The 15:54 all stations St.Pancras to St.Albans was also late so they decide to run that non stop to St.Albans (a train which departed with just one passenger on board).

I can assure you that I was not the only passenger on board!

 

[bramling]

Yesterday the 15:40 all stations St.Pancras to Luton was late so they decided to omit all the stops bar Luton Airport Parkway. The 15:54 all stations St.Pancras to St.Albans was also late so they decide to run that non stop to St.Albans (a train which departed with just one passenger on board). There had now been a bit of a gap for an all stations so they decide that the 16:18 fast to Bedford (on time) should call all stations.
Now while there may be "operational reasons" that they have done this for the average passenger they have :
- been turfed off a train that was supposed to take them where they going
- been stood on a crowded platform while they watch another train that was supposed to take them were they going leave virtually empty
- end up crammed on a train that should have been fast but is now slow.

Thameslink have improved in some areas (overcrowding isn't as bad since the new trains arrived) but they are continually poor in how they deal with service recovery and communicating with their long suffering customers.

The problem is not so much recovery, but the fact that delays occur in the first place. Thameslink is highly prone to delays because it is fed from some of the most unreliable and congested parts of the railway network. The logic behind what method of recovery will be used is often dictated by factors unrelated to passenger convenience - getting the rolling stock, or often more importantly the driver, back on time for its next working. Unfortunately the best method of recovery is not to have the delay in the first place, followed by having plenty of slack in rolling stock and crew diagrams so delays can run out naturally. The latter is an area where I think improvements could be made, but this comes at a cost.

Having a terminus is a superb way of allowing delays to run out, unfortunately Thameslink reduces these by half - so it will always be noticeably less reliable than other networks. If one compares northbound services from St Pancras International to King's Cross, the vast majority of northbound Great Northern services at least start their journey on time. St Pancras Internaional Low Level is a very different matter entirely.

 

[yorkie]

Yesterday the 15:40 all stations St.Pancras to Luton was late so they decided to omit all the stops bar Luton Airport Parkway. ...QUOTE]
This makes sense; passengers only had to wait 8 minutes. It meant the next working was on time for the evening peak and didn't disrupt the rest of the service.

 

[John1975]

Seems logical to me. Recover the service decisively to prevent further delays and minimise the overall impact on the evening peak. The return workings of the two trains which ran fast achieved right time starts, as did the train which called additionally. I understand that it seems harsh and not customer friendly, but it was for the greater good.

As Bramling rightly says delay prevention is better, however in this case it's difficult to prevent someone trespassing on the track at Norbury forcing an emergency REC call which brought all trains to a stand in the area followed by a current isolation of all 4 lines which caused mayhem for the rest of the day. Compared to Southern, Thameslink came through relatively unscathed.

 

[bramling]

As Bramling rightly says delay prevention is better, however in this case it's difficult to prevent someone trespassing on the track at Norbury forcing an emergency REC call which brought all trains to a stand in the area followed by a current isolation of all 4 lines which caused mayhem for the rest of the day. Compared to Southern, Thameslink came through relatively unscathed.

Thus proving how despite the much-emphasised reliability features built into the core (*), performance can be affected by an incident which isn't even on the Thameslink route, let alone on the London-Bedford route. The greater good isn't much comfort if you happen to be intending to use one of those affected services, have somewhere to be by a particular time, and if it happens day-in-day-out. Still, not to worry, I'm sure a new regulating strategy will be the answer to everything, as some have seriously suggested.

(* for example bi-directional working which actually can't really be used in practice due to the density of the service)

London Underground tends to have a convention (I wouldn't call it a policy as such) that during a disruption they will try to avoid the first disrupted service being turned short, even if it causes a further problem down the line. It's felt unreasonable for the first lot of passengers already suffering as a result of the initial incident to have to suffer the further inconvenience of the train being curtailed. Another thing is to try to avoid altering two trains in succession. Some lessons to be learned here perhaps.

 

[Failed Unit]

Glad I don't live at Welwyn North. I suspect that will be skipped lots. (The GTR want to add more trains to block the biggest bottleneck on the southern ECML)

 

[bramling]

Glad I don't live at Welwyn North. I suspect that will be skipped lots. (The GTR want to add more trains to block the biggest bottleneck on the southern ECML)

Anywhere with a half-hourly service will suffer - this will be most of the GN side. One of the features the Thameslink apologists point to is the extra choice and the fact that you won't have to wait long til the next service if something goes wrong. Apart from the fact that this doesn't help you if there's nothing coming north for whatever reason, this false promise completely ignores the fact that most of the GN services will be 2tph as opposed to much of the current Thameslink being 4tph, and the fact that a lot of stations won't be seeing any extra off-peak frequency at all - Hitchin to Peterborough for a start.

This all ties in with my past experience of using Thameslink for 7 years. In those days London to St Albans had 4tph fast and 4tph stopping, so in theory I should never have had to wait on a core platform for more than 15 minutes for a train going north. Reality was rather different, with waits of up to 40 minutes being far from uncommon, particularly at the end of the morning peak when things were being got back to book. I simply dread to think how it will be with 2tph services.

 

[Mutant Lemming]

Yesterday the 15:40 all stations St.Pancras to Luton was late so they decided to omit all the stops bar Luton Airport Parkway. ...QUOTE]
This makes sense; passengers only had to wait 8 minutes. It meant the next working was on time for the evening peak and didn't disrupt the rest of the service.

All stations passengers had to wait nearly 30 minutes not 8 - and watch two all stations trains be reformed to fast and leave very lightly loaded.

 

[ComUtoR]

What would you have done and would you accept full responsibility and cost of all the impact minutes of that decision ?

 

[tsr]

As Bramling rightly says delay prevention is better, however in this case it's difficult to prevent someone trespassing on the track at Norbury forcing an emergency REC call which brought all trains to a stand in the area followed by a current isolation of all 4 lines which caused mayhem for the rest of the day. Compared to Southern, Thameslink came through relatively unscathed.

That was not the only issue in place. A Thameslink service was immobilised at East Croydon due to an operating incident, Southern lost track of its trains and crews and ended up blocking out Victoria with stock for a very long time (a worrying theme over the last 18 months or so), and Network Rail did not have enough control staff on duty. It was a catastrophe waiting to happen, and frankly the trespassing (and the subsequent issues recharging power to several important areas) was just the trigger.

Thameslink got off relatively lightly, indeed, but it could have gone a lot more badly for them in slightly different circumstances. I wonder if a split service north & south of the Core would have been better as a precaution. I also wonder if it would be better to have more defined ways of terminating severely delayed northbound Thameslink services earlier, but making the same stops, and doing [cross-platform, if possible] interchanges onto relief stock doing special extra stopping services from that point onwards.

 

[hairyhandedfool]

The 1540 left St Pancras 18 late, but arrived in Luton 3 late (1632), plenty of time for it to form the 1644 to Sutton. An 18 minute delay would have seen the return one run late.

The 1558 left St Pancras 12 late, but arrived into St Albans 2 minutes late, allowing it time to get in and out of the centre siding and depart as the 1644 to Sutton. a 12 minute delay would also have seen that return one run late.

The 1618 left St Pancras on time, but dropped 14 minutes by St Albans due to it's extra stops. It made up time after that, arriving in Bedford 8 minutes late, 3 minutes before it's return working, the 1730 to Three Bridges, was due to leave.

Passengers who couldn't use the 1540 or the 1554 (or the 1610), and didn't want to stand on the 1618, could have waited for the 1624 St Albans service instead, though they may still have had to stand.

 

[bramling]

The 1540 left St Pancras 18 late, but arrived in Luton 3 late (1632), plenty of time for it to form the 1644 to Sutton. An 18 minute delay would have seen the return one run late.

The 1558 left St Pancras 12 late, but arrived into St Albans 2 minutes late, allowing it time to get in and out of the centre siding and depart as the 1644 to Sutton. a 12 minute delay would also have seen that return one run late.

The 1618 left St Pancras on time, but dropped 14 minutes by St Albans due to it's extra stops. It made up time after that, arriving in Bedford 8 minutes late, 3 minutes before it's return working, the 1730 to Three Bridges, was due to leave.

Passengers who couldn't use the 1540 or the 1554 (or the 1610), and didn't want to stand on the 1618, could have waited for the 1624 St Albans service instead, though they may still have had to stand.

All that is operationally lovely, however if I've understood correctly your suggestion amounts to a 44 minute delay to passengers, or presumably an hour's gap in the service, on what is supposed to be a 15-minute headway to St Albans at least. I don't doubt it was the best operational solution, but it doesn't alter the fact that this situation was utterly arse for passengers. It is simply yet another example of what a shambolic unreliable and undependable operation Thameslink is, and a golden example of why the railway needs *less* Thameslink, and certainly not more. 2018 is going to be an unmitigated disaster, and it's farcical people keep trying to defend this inherently awful service.

Thameslink will also have to in future pay some thought to the fact long gaps on the service will cause core platforms to become increasingly crowded. It will be very hard indeed to implement effective measures when there are five separate northbound flows all trying to use the same platform space. The traditional British solution of some kind of informal queue just won't work. There's a saying wouldn't the railway work beautifully if it wasn't for those pesky passengers... What appears to work nicely in the comfort of 'control' may well not be so appreciated by the station staff who have to deal with the chaos.

By contrast Crossrail has some mitigations in place for reducing imported delays, and they are joining together two not three networks, which themselves are much more self-contained than Thameslink will ever be. One scheme will be a success and one will be a shambles, and I hope those responsible for designing Thameslink never get let near a rail project ever again.

 

[Bedpan]

The 1540 left St Pancras 18 late, but arrived in Luton 3 late (1632), plenty of time for it to form the 1644 to Sutton. An 18 minute delay would have seen the return one run late.

The 1558 left St Pancras 12 late, but arrived into St Albans 2 minutes late, allowing it time to get in and out of the centre siding and depart as the 1644 to Sutton. a 12 minute delay would also have seen that return one run late.

The 1618 left St Pancras on time, but dropped 14 minutes by St Albans due to it's extra stops. It made up time after that, arriving in Bedford 8 minutes late, 3 minutes before it's return working, the 1730 to Three Bridges, was due to leave.

Passengers who couldn't use the 1540 or the 1554 (or the 1610), and didn't want to stand on the 1618, could have waited for the 1624 St Albans service instead, though they may still have had to stand.

So, in hindsight at least, there would have been time for the 1540 to have stopped at two or three stations on the way and still started back from Luton at 1644, assuming a 6 minute turn round is possible. The 1558 could have stopped at a couple of different stations on the way to St Albans and still started back on time, and the 1618 could have missed out the stations that the previous two trains had stopped at and so been less late arriving at Bedford. And the people waiting at St Pancras would have been less delayed.

 

[ComUtoR]

So, in hindsight at least, there would have been time for the 1540 to have stopped at two or three stations on the way and still started back from Luton at 1644, assuming a 6 minute turn round is possible.

Nope. You need to allow for speed affected by the stops and then allow for dwell times. You also need to then decide which stations to stop at and which line to send the train up. Typically they send the stoppers across to the fast and then you can hit 100 and make the time back. If you stick it on the slow you lose speed, ergo more time. The assumption about the turn around is indeed an assumption. Turnaround times are set out by the TOC (8/RLU is longer than 6 minutes) and it will depend on the unit length and if the Driver is relieved or not.

You also need to check the Drivers diagram too. With connections so tight the impact minutes can stack up again. Not only impact minutes but sometimes we have a minimal connection between units and a single minute may mean you miss your pass and ergo the next service typically gets caped.

In my experience. My preference is to be fast to St Albans and sent over to the fast at Carlton Road so I can maximise the fast line and get the service back on time at St Albans. That way delays are limited as you get back on time very efficiently. What I hate is running late then getting run semi fast and then because I'm still running late I typically get a 'stop order' on the return trip, thus affecting another service.

A lot of the time they run you fast to help clear the line for the service behind. Nothing worse than being run fast only to hit the back of a stopper.

 

[Mutant Lemming]

A lot of the time they run you fast to help clear the line for the service behind. Nothing worse than being run fast only to hit the back of a stopper.

Not much chance of that when they keep making the stoppers fast.... unless of course you happen to want one of those stops.

Incidentally - more and more people are becoming delayed prior to travel by closed ticket offices and operational gatelines preventing platform access. I thought there were rules about this even if it is only 'for a few minutes' (32 on the first occasion and 17 on a subsequent occasion). There have also been issues with people unable to leave stations during these periods. Surely an unstaffed gateline needs to be left open.... or do Thameslink have rules of their own on this?