Disruption - running fast to make up time

[ComUtoR]

Is the "big picture", more often than not, the course of action/inaction that optimises the cost implications for the individual TOC?

My more cynical days are behind me and I'm in the camp where decisions are not being made specifically because of the cost.

More decisions are made 'to recover the service' at the expense of everything because that is the greater good. However, this can involve certain sacrifices on the ground. I'd also agree that some delays are made purely based on crew requirements. Again, sometimes at detriment of others.

Just the other day a decision was made to stop a fast service mid journey to pick up stranded crew. That delay would be at the detriment of those passengers but bigger picture is that the crew were needed to get other services out on time and recovered.

Nobody on the ground has the big picture. Some delays and recovery efforts look absolutely ridiculous but then chatting back in the crew room you discover that another service was picking up those you left behind or a decision was made because there was something happening elsewhere.

== Doublepost prevention - post automatically merged: 3 Jan 2023 ==

It may do, but if it is announced clearly before boarding starts that one will run fast to X and the other will be all stations, everyone for X and beyond will choose the fast one, and everyone for the intermediate stations will, perforce, use the other one. On a 4tph service few people go for a specific train anyway.

Humans are weird and wonderful creatures. Now they are armed and dangerous with apps.

Predicting what people will actually do is almost impossible.

You can announce changes till your blue in the face and I guarantee someone will ask "excuse me. Is this stopping at.." Or someone never listens for whatever reason and then ends up pulling a passcom as the train rolls through their stop.

I think passengers are so used to the service being unpredictable and they know the train is gonna get halfway down the line and then get caught in congestion or get stuck at a junction that their past experiences will inform their actions far more than what they are being told.

 

[Falcon1200]

Is the "big picture", more often than not, the course of action/inaction that optimises the cost implications for the individual TOC?

Yes, and not just for the TOCs, for Network Rail too if the delay cause is theirs; Because minimising cost to the railway also means that fewer trains have been cancelled or delayed and therefore fewer passengers have had their journey disrupted.

 

[Horizon22]

Is the "big picture", more often than not, the course of action/inaction that optimises the cost implications for the individual TOC?

Well yes but delays = costs so the “big picture” is ultimately good for both passengers and the TOC because overall delays are minimised impacting the fewest passengers possible.

For instance in peak you might run trains as booked leaving say London/Manchester/Birmingham/insert city but absolutely slash the stops on the return to make up the time.

 

[norbitonflyer]

More decisions are made 'to recover the service' at the expense of everything because that is the greater good.

Not necessarily. I have seen enough empty trains run past crowded peak hour platforms to realise that getting the train to the end of the line in time to take up its return working is not "the greater good", however many boxes it ticks. I would not expect to pay any delivery service just for turning up on time if the cargo it was supposed to be bringing was not delivered, and the same should apply on the railways, where the cargo is human. (This also applies where a connection is not honoured).

 

[Falcon1200]

Not necessarily. I have seen enough empty trains run past crowded peak hour platforms to realise that getting the train to the end of the line in time to take up its return working is not "the greater good", however many boxes it ticks.

It is for the 'greater good' if running a train fast means that a number of other trains, and therefore their passengers, are not disrupted. Doing nothing when trains are running out of course ensures that the service continues to deteriorate, with ever increasing disruption as late trains conflict with on time trains and incur more delay themselves, delay the on time trains, or both, leading eventually to cancellations; I have seen this happen and therefore took action to prevent it.

 

[JN114]

I’ve made this point on a number of occasions on threads like these -

As a controller who has to make these decisions as a fundamental part of my day job - I’m not assessed or judged by my management on the monetary cost of my decision making whatsoever. I have no awareness nor mechanism for finding out the actual monetary “cost” of the decisions I make; nor do I particularly want to.

The broad thought process is:-

1) Identify late running.
2) Assess impacts of late running for stock and crew.
3) Is there an “invisible” solution to those impacts (eg step-up stock or crew).
4) What is the impact if I do nothing.

Only once I’ve got to that step will I start going down the road of running fast to make up time. At that point I’m considering time of day, train formation and loadings, how much time I want to recover and whether that is actually achievable.

The other times it does happen is when NR implement reduced infrastructure contingency plans - so as an example if 2 of the 4 lines at XYZ are closed; there is a signed/agreed plan that all operators go into when instructed by NR - and that plan calls for these trains to run fast, these trains to start/terminate here and so on. We as an operator have very little choice in those scenarios; although we do get a lot of input into the production and maintenance of those plans.

 

[Class83]

I’ve made this point on a number of occasions on threads like these -

As a controller who has to make these decisions as a fundamental part of my day job - I’m not assessed or judged by my management on the monetary cost of my decision making whatsoever. I have no awareness nor mechanism for finding out the actual monetary “cost” of the decisions I make; nor do I particularly want to.

The broad thought process is:-

1) Identify late running.
2) Assess impacts of late running for stock and crew.
3) Is there an “invisible” solution to those impacts (eg step-up stock or crew).
4) What is the impact if I do nothing.

Only once I’ve got to that step will I start going down the road of running fast to make up time. At that point I’m considering time of day, train formation and loadings, how much time I want to recover and whether that is actually achievable.

The other times it does happen is when NR implement reduced infrastructure contingency plans - so as an example if 2 of the 4 lines at XYZ are closed; there is a signed/agreed plan that all operators go into when instructed by NR - and that plan calls for these trains to run fast, these trains to start/terminate here and so on. We as an operator have very little choice in those scenarios; although we do get a lot of input into the production and maintenance of those plans.

I had one trip where the train I went for was cancelled, the next train took us 1 station, then was made 'semi-fast' so 90% of the passengers got off (I think this might be a hint, if your semi fast stopping pattern removes more than half the passengers it's not helping), we eventually boarded the train behind which went 1 stop before they tried to kick everyone off. By this point about 90% of the passengers onboard were for one of the intermediate stops resulting in a lengthy discussion with the staff which took far longer than stopping at the station would have. I'm not sure control necessarily engaged their brain to work out that a train containing 3 trainloads of passengers to a station which is next to a site which employs over 1000 people between 8 and 9am should probably not miss that stop.

 

[Falcon1200]

I'm not sure control necessarily engaged their brain to work out that a train containing 3 trainloads of passengers to a station which is next to a site which employs over 1000 people between 8 and 9am should probably not miss that stop.

@JN114 has explained very well the process for altering trains to run fast, I was involved in such decisions too. Please bear in mind that Controls cover large areas, cannot be all-seeing and all-knowing, and are frequently dealing with multiple issues, and many trains, at the same time. It would be wrong to say that we do not make mistakes, or decisions which in hindsight could have been better, but we are only (mostly) human!

 

[Class83]

@JN114 has explained very well the process for altering trains to run fast, I was involved in such decisions too. Please bear in mind that Controls cover large areas, cannot be all-seeing and all-knowing, and are frequently dealing with multiple issues, and many trains, at the same time. It would be wrong to say that we do not make mistakes, or decisions which in hindsight could have been better, but we are only (mostly) human!

The Driver and Guard who could see the situation were communicating this via their radios, however, they kept being told to bypass all stations, rather than a compromise solution of missing some to make up time and stopping at 1 to empty the train. I'm sure there will be cases where the passengers are spread out so choosing where to skip is harder. But when the on train staff are clearly pointing out that a simple adjustment to the stopping pattern would be more efficient at getting paying passengers to their destination, control were perfectly aware of the situation and chose to ignore it.

 

[Jimini]

I've been on two northbound GA services from Tottenham Hale in the past, where the non-stop message hasn't been shared with the passengers about missing out the likes of Waltham X, Ponders End, Enfield Lock etc. until after departure from Tottenham. Lots of angry folk on the platforms at Cheshunt / Broxbourne as a result!

 

[miklcct]

But it wasn’t the evening rush hour - you were on the 1520 which is off peak territory.

The train it formed at Reading, the 1709, very much is the evening rush hour out of Reading (I don’t know the specifics of that train but I’m willing to bet it’s busy). This return service then forms the 1850 from Waterloo to Reading, which again is the evening rush hour. Just on the basis of these trains, more people had On time journeys than were delayed. And that’s before the impact of reactionary delay affecting other services as you try to force an out of path delayed train through the busy approaches to Waterloo.

Absolutely no doubt it was the right decision. No consolation for you, I know, and I’m sorry your journey was delayed.

If all trains run 30 minutes late on a line with 30-minute headway, no trains run late.

Of course running fast is the correct action if the trains before and after the late train are all on time, but if all trains on the line are late, it's better to continue to run them all late, by cancelling the missing departure and reform the trains to run one departure after the roster.

 

[Falcon1200]

but if all trains on the line are late, it's better to continue to run them all late, by cancelling the missing departure and reform the trains to run one departure after the roster.

Which might sound logical, but it would put every single member of Traincrew on the route 30 minutes off their booked diagram which would inevitably cause further disruption, sooner or later.

 

[Horizon22]

If all trains run 30 minutes late on a line with 30-minute headway, no trains run late.

Of course running fast is the correct action if the trains before and after the late train are all on time, but if all trains on the line are late, it's better to continue to run them all late, by cancelling the missing departure and reform the trains to run one departure after the roster.

You are completely understimating the complexity of the timetable, the impact on crew and the turnaround times. The outbound trains will be late, they may miss their slot at the terminus / origin (which is more likely to be congested) and impact other services around it. If the crew are late on break, that might cause more impacts and they are off their booked diagram. If it's a fresh crew, then it's easier, but the impact from the original crew might spread to a later train. Acting - or not acting - at one stage might cause a worse, unintentional impact 1/2 hours down the line if you're not careful.

Everything is on a case-by-case basis and as others have worked in control environments have said, it is never an exact science and what average Controller A does might vary from what proactive Controller B does, to what reactive Controller C does although sometimes agreed contingency plans and infrastructure play their part too. Do mistakes get made? Of course, but people are only human and it comes with experience, an assessment of the circumstances (time of day, crew & stock availability, problem location, impact, staff support, the behaviours of the controller etc, signaller communications), the extent of the pressure and sometimes, luck. It is not for everyone which is why such roles tend to be relatively well paid. This is not to say there aren't people in roles who probably don't quite cut it!

 

[Bald Rick]

If all trains run 30 minutes late on a line with 30-minute headway, no trains run late.

not so. They all run 30 minutes late. Some passengers might not experience the lateness, but some will, not least those delayed on the train when it went from on time to 30 L.

 

[Horizon22]

not so. They all run 30 minutes late. Some passengers might not experience the lateness, but some will, not least those delayed on the train when it went from on time to 30 L.

Not to mention all eligible for Delay Repay (strictly they'll be eligible if they skip the stop too). But more services might be delayed too.

 

[riceuten]

I think the only problems I have encountered that merited comment (rather than the occasional train non-stopping Stevenage due to late running) is NOT being told the train is non-stopping, as I had once when the train nonstopped from Bigglewade to KX without informing passemgers who may have wished to have disembarked at Arlesey, Hitchin, Stevenage and Finsbury Park. Fortunately, the same train went out 20 minutes later stopping at just these stations, and the driver did at least apologise.

No, the absolute worst I had of this was a train I was on nonstopped Finsbury Park where I usually changed to catch the tube to Walthamstow. About 5 of us were met by revenue control staff who wanted to excess us/fine us for overriding our tickets/travelcards. "Pay the fine and appeal, sir" I was told. No, I stalked off and tailgated someone to be able to access the correct platform.

 

[Bikeman78]

Whilst passing through Brighton today, I noted that the 1258 to Bedford was delayed. It was stuck at Burgess Hill on the way down owing to an obstruction on the line. The ever helpful auto announcer kept going off, telling people that the next fastest direct train to Haywards Heath and Gatwick Airport will be the 1309 to London Victoria. Err, no it won't, because that is stuck behind the first train, and also the train that forms the 1314 to Cambridge, but never mind.

The human announcer was much more helpful, suggesting that people for Preston Park, Hassocks and Burgess Hill take the 1309 to Hove and then the Victoria train from there. Good on him. The 1258 departed at 1310 and only stopped at Haywards Heath en route to Gatwick instead of all stations. Given that the 1314 and 1309 were right up its backside, Haywards Heath was a strange choice. Why not Balcombe? It would have departed there more or less on time.

I subsequently discovered that the 1244 Cambridge didn't run either, owing to a train fault. I thought the passengers at Brighton seemed rather more cross than I would have expected.

 

[Bald Rick]

Given that the 1314 and 1309 were right up its backside, Haywards Heath was a strange choice. Why not Balcombe?

Because Hayward’s Heath typically has twenty times more passengers use it than Balcombe.

 

[Bikeman78]

Because Hayward’s Heath typically has twenty times more passengers use it than Balcombe.

As I stated, there were two other trains heading north right behind the late train. The poor people of Haywards Heath would have had to wait a whole extra four minutes. Anyone at Balcombe would be stuck for half an hour.

 

[Bald Rick]

As I stated, there were two other trains heading north right behind the late train. The poor people of Haywards Heath would have had to wait a whole extra four minutes. Anyone at Balcombe would be stuck for half an hour.

Balcombe has an average 2-3 people joining each train that calls there, and that includes the peak. It’s reasonable to assume that off peak it’s an average of 1 or 2.

Unquestionably the right decision.