Delay Repay - GTR says delay was much lower

Envy123 · 7 Aug 2019

My mother intended to take the 08:25 train on 22nd July from New Southgate to Hadley Wood. This was cancelled and she ended up taking the 08:55 train.

I claimed to GTR for Delay Repay, but they said that the delay was 15-29 minutes instead of 30 minutes.

Can I appeal this, somehow? As GTR already confirmed that 80p would be paid by BACS to the nominated account.

RailUK Forums

paddington · 7 Aug 2019

The 0824 was scheduled to arrive at 0833 and the 0854 was scheduled to arrive at 0903, but actually arrived at 0902, so strictly speaking she was delayed by 29 minutes.

I am of two minds as to whether this sort of delay should just be treated as a 30 minute delay, but it might be difficult to automate. I was recently delayed by 10 minutes on a Northern service connecting to Merseyrail so had to take the next 4tph train. The Merseyrail service was slightly early so my actual delay was 14¾ minutes. I have filed a claim and will see what they say.

221129 · 7 Aug 2019

Envy123 said:
My mother intended to take the 08:25 train on 22nd July from New Southgate to Hadley Wood. This was cancelled and she ended up taking the 08:55 train.

I claimed to GTR for Delay Repay, but they said that the delay was 15-29 minutes instead of 30 minutes.

Can I appeal this, somehow? As GTR already confirmed that 80p would be paid by BACS to the nominated account.

Your mother was only delayed by 29 mins so therefore GTR are correct.

furlong · 7 Aug 2019

paddington said:
The 0824 was scheduled to arrive at 0833 and the 0854 was scheduled to arrive at 0903, but actually arrived at 0902, so strictly speaking she was delayed by 29 minutes.

A company shouldn't profit from mistakes, deliberate or otherwise, in its timetabled arrival times - in the event of an arrival earlier than the published time, then it should still be using the published time - or else it should have corrected the timetable. The awkward question for the company is how it was possible for the train to arrive early and if the journey can be done in less than the published time wouldn't the original service also have been able to arrive a similar amount of time early and shouldn't that also have been taken into account when calculating the delay? You can't adjust one arrival time without also adjusting the other.

najaB · 7 Aug 2019

furlong said:
The awkward question for the company is how it was possible for the train to arrive early and if the journey can be done in less than the published time wouldn't the original service also have been able to arrive a similar amount of time early...

Pathing.

Envy123 · 7 Aug 2019

Thanks, guys. I didn't know TOCs can make the trains arrive as early as possible, to try to reduce the delays.

You learn something new everyday.

AlterEgo · 7 Aug 2019

furlong said:
A company shouldn't profit from mistakes, deliberate or otherwise, in its timetabled arrival times - in the event of an arrival earlier than the published time, then it should still be using the published time - or else it should have corrected the timetable.

There is no mistake - this is just diamond time.

Hadders · 7 Aug 2019

furlong said:
A company shouldn't profit from mistakes, deliberate or otherwise, in its timetabled arrival times - in the event of an arrival earlier than the published time, then it should still be using the published time - or else it should have corrected the timetable. The awkward question for the company is how it was possible for the train to arrive early and if the journey can be done in less than the published time wouldn't the original service also have been able to arrive a similar amount of time early and shouldn't that also have been taken into account when calculating the delay? You can't adjust one arrival time without also adjusting the other.

Good point. I've got an issue with GTR denying me a claim due to a cancellation because the train I eventually caught arrived 1 minute early meaning the overall delay to my journey was only 14 minutes.

JBuchananGB · 8 Aug 2019

Most customers would see it as a positive thing that the railway does its best to help passenger delays to be as few as and as short as possible.

In very few cases is an arrival time published in the time table, usually it is the departure time, so I cannot see the argument that the published departure time should be treated as if it was the actual arrival time for delay repay purposes. It is not a mistake for a train to arrive early. It is a bonus.

Spurs · 8 Aug 2019

JBuchananGB said:
Most customers would see it as a positive thing that the railway does its best to help passenger delays to be as few as and as short as possible.

In very few cases is an arrival time published in the time table, usually it is the departure time, so I cannot see the argument that the published departure time should be treated as if it was the actual arrival time for delay repay purposes. It is not a mistake for a train to arrive early. It is a bonus.

Depends how it's regulated. You wouldn't want a situation where trains were always scheduled to take twice as long as they actually do so that they were never late but always early - it's just blatantly confusing to passengers and only serves to let the railway get away with never having to compensate anyone. Of course a minute's difference isn't this severe, but the same principle applies. If the journey takes 13 minutes when things run well, that's what should be in the timetable and what delay repay claims are based on.

island · 9 Aug 2019

Ultimately there has to be a cutoff somewhere and some people will wind up unluckily delayed by just under the requisite time.

LAX54 · 10 Aug 2019

Envy123 said:
Thanks, guys. I didn't know TOCs can make the trains arrive as early as possible, to try to reduce the delays.

You learn something new everyday.

Time 'allowed' for ESRs / TSRs, there is also a bit of 'make up' time, in case it's further delayed for any reason, this ensures, or tries to ensure it gets to it's destination to time, no point having a timetable that gets to it's destination only if everything lines up with the Moon, imagine the complaints when connection after connection is lost !

JamesRowden · 10 Aug 2019

island said:
Ultimately there has to be a cutoff somewhere and some people will wind up unluckily delayed by just under the requisite time.

There doesn't. The function for compensation against delay doesn't need to have step signals in it (the compensation suddenly changing for an infinitely small change in delay). Smoothing the function will make a minute insignificant in terms of compensation in all cases. We just need the DfT to be a bit smarter.

AlterEgo · 10 Aug 2019

JamesRowden said:
There doesn't. The function for compensation against delay doesn't need to have step signals in it (the compensation suddenly changing for an infinitely small change in delay). Smoothing the function will make a minute insignificant in terms of compensation in all cases. We just need the DfT to be a bit smarter.

It is far easier to administer the current way though.

JamesRowden · 10 Aug 2019

AlterEgo said:
It is far easier to administer the current way though.

An extra few lines of computer code is simple.

Adsy125 · 10 Aug 2019

JamesRowden said:
An extra few lines of computer code is simple.

But not simple for the passenger to understand, and a bit unnecessary. The current delay repay scheme is far better than anything that has existed before it (to my knowledge anyway).

najaB · 10 Aug 2019

JamesRowden said:
The function for compensation against delay doesn't need to have step signals in it (the compensation suddenly changing for an infinitely small change in delay).

Zeno's compensation?

JamesRowden · 10 Aug 2019

Adsy125 said:
But not simple for the passenger to understand, and a bit unnecessary. The current delay repay scheme is far better than anything that has existed before it (to my knowledge anyway).

Additionally have a plot of compensation against delay for the passenger to read rather than just using words.

AlterEgo · 10 Aug 2019

JamesRowden said:
An extra few lines of computer code is simple.

I think you have a big misunderstanding in how Delay Repay is actually administered.

JamesRowden · 10 Aug 2019

AlterEgo said:
I think you have a big misunderstanding in how Delay Repay is actually administered.

The staff (or automated process) don't have access to a computer capable of calculating a simple mathematical function?

AlterEgo · 10 Aug 2019

JamesRowden said:
The staff (or automated process) don't have access to a computer capable of calculating a simple mathematical function?

This isn't the issue - can you imagine how many disputes there would be, where people could argue the toss over a single minute or how the algorithm calculated their compensation? This doesn't even begin to account for the delays on missed connections where the subsequent train/s also didn't run to time?

Dibbo4025 · 10 Aug 2019

JamesRowden said:
The staff (or automated process) don't have access to a computer capable of calculating a simple mathematical function?

Shouldn't even require a computer - thr function
R=P x t/60 for 0<=t<60
R=P otherwise
for a single or
R=P x t/(2x60) for 0<=t<120
R=P otherwise
where R is the refund, P the ticket cost and t the size of the delay
gives the linear interpolation of the current delay repay scheme and isn't much harder to work with, to the point anyone with a piece if paper and pen let alone a phone or calculator wouldn't have to think, let alone a computer

In practice though I can see this would be best used with an oyster like payg system or other digital ticketing with automatic DR to make it anymore useful than the current system. Also since it would give DR for any delay at all we would probably need to be using right time railway as when combined with ppm (which is itself very much suboptimal) I could forsee it making dekay figures definitely not more transparent if not actually more opaque

Bletchleyite · 10 Aug 2019

paddington said:
The 0824 was scheduled to arrive at 0833 and the 0854 was scheduled to arrive at 0903, but actually arrived at 0902, so strictly speaking she was delayed by 29 minutes.

I am of two minds as to whether this sort of delay should just be treated as a 30 minute delay, but it might be difficult to automate. I was recently delayed by 10 minutes on a Northern service connecting to Merseyrail so had to take the next 4tph train. The Merseyrail service was slightly early so my actual delay was 14¾ minutes. I have filed a claim and will see what they say.

I had a similar refusal from LNR and appealed it because I felt it was over-picky on their part (as in practice I was delayed precisely 30 minutes, as the one before also arrived 1 minute early) and they paid it on appeal.

JamesRowden · 10 Aug 2019

AlterEgo said:
This isn't the issue - can you imagine how many disputes there would be, where people could argue the toss over a single minute or how the algorithm calculated their compensation? This doesn't even begin to account for the delays on missed connections where the subsequent train/s also didn't run to time?

It would be a simple function (like the existing system) rather than an algorithm. An example would be a straight line between 0% at delay A, and 100% at delay B. Nothing below delay A, and 100% above delay B.

This whole point is that 1 minute extra delay would be insignificant with regards to the compensation paid.

All of those other issues exist with the existing system (where 1 second can make a £50 difference).

35B · 10 Aug 2019

JamesRowden said:
It would be a simple function (like the existing system) rather than an algorithm. An example would be a straight line between 0% at delay A, and 100% at delay B. Nothing below delay A, and 100% above delay B.

This whole point is that 1 minute extra delay would be insignificant with regards to the compensation paid.

All of those other issues exist with the existing system (where 1 second can make a £50 difference).

Delay Repay is simple and easy to understand, even if some operators make a pigs ear of how they administer it. Changing the scale to avoid a cliff edge deals with one minor problem - those disgruntled at missing a cut off - at the cost of making it more complex and harder to explain to the majority of people who just want to know a rule, and may not realise that they can claim in the first place.

Far better to fix some of the bigger issues, like differences between working and public timetables, or basic punctuality, than tinkering around the edges of a decent system.

JamesRowden · 10 Aug 2019

35B said:
Delay Repay is simple and easy to understand, even if some operators make a pigs ear of how they administer it. Changing the scale to avoid a cliff edge deals with one minor problem - those disgruntled at missing a cut off - at the cost of making it more complex and harder to explain to the majority of people who just want to know a rule, and may not realise that they can claim in the first place.

Far better to fix some of the bigger issues, like differences between working and public timetables, or basic punctuality, than tinkering around the edges of a decent system.

I think you underestimate the ability of the public to understand a plot with a single straight line (plus a marker to show where their claim is on the plot with the output compensation shown) displayed to them on the website when they would make their claim (or didn't think about how the claim process could be made as transparent/comprehensible as possible).

AlterEgo · 10 Aug 2019

JamesRowden said:
I think you underestimate the ability of the public to understand a plot with a single straight line (plus a marker to show where their claim is on the plot with the output compensation shown) displayed to them on the website when they would make their claim (or didn't think about how the claim process could be made as transparent/comprehensible as possible).

So, basically under the current system the vast majority of disputes happen where a passenger is experiencing a delay of 26-30 minutes or 56-60 minutes and is caught by the cliff edge. This is a small minority of claimants.

Under the new regime, you’ll have no cliff edge and instead a sliding scale, with far greater opportunities for disputes. Passengers who are 36 minutes delayed but think it was 42 will dispute.

None of this even begins to account for the problems you have when validating missed connections.

I have worked administering Delay Repay and my experience tells me that what you propose is a total and utter nightmare.

furlong · 10 Aug 2019

The scheme should advertise round numbers of minutes - 15, 30, 60 etc, but implement it as a minute or two lower i.e. always rounding times in the passenger's favour, to allow for small inaccuracies in recording, small delays opening doors etc. The railway is already familiar with this concept in the way it advertises railcards that offer a headline "1/3 off rail fares" but implements it in a way that normally provides very slightly more than 1/3 off.

JamesRowden · 10 Aug 2019

AlterEgo said:
So, basically under the current system the vast majority of disputes happen where a passenger is experiencing a delay of 26-30 minutes or 56-60 minutes and is caught by the cliff edge. This is a small minority of claimants.

Under the new regime, you’ll have no cliff edge and instead a sliding scale, with far greater opportunities for disputes. Passengers who are 36 minutes delayed but think it was 42 will dispute.

None of this even begins to account for the problems you have when validating missed connections.

I have worked administering Delay Repay and my experience tells me that what you propose is a total and utter nightmare.

My interests are fairness and encouraging TOCs to produce a good service for passengers (by making them want to avoid a 59 minute delay to a passenger's journey almost as much as a 60 minute delay).

AlterEgo · 10 Aug 2019

JamesRowden said:
My interests are fairness and encouraging TOCs to produce a good service for passengers (by making them want to avoid a 59 minute delay to a passenger's journey almost as much as a 60 minute delay).

The cost of Delay Repay is such a tiny drop in the ocean that it doesn’t even come into the thought process when improving performance. I worked in a control room and know lots of controllers and I can confidently say they know pretty much nothing about Delay Repay. That’s one of the biggest misconceptions on this forum. Schedule 8 payments are far, far greater.

Delay Repay - GTR says delay was much lower

Member

RailUK Forums

Member

Established Member

Established Member

Veteran Member

Member

Veteran Member

Veteran Member

Member

Member

Veteran Member

Established Member

Established Member

Veteran Member

Established Member

Member

Veteran Member

Established Member

Veteran Member

Established Member

Veteran Member

Member

Veteran Member

Established Member

Established Member

Established Member

Veteran Member

Established Member

Established Member

Veteran Member