• Our new ticketing site is now live! Using either this or the original site (both powered by TrainSplit) helps support the running of the forum with every ticket purchase! Find out more and ask any questions/give us feedback in this thread!

Cost Per Minute Of Delay

Status
Not open for further replies.

Taunton

Veteran Member
Joined
1 Aug 2013
Messages
12,313
It doesn’t cost that, but it’s charged at that
Fact is that these financial figures are just a management tool, to feature in business reports. They are not hard cash. If a train arrives 5 minutes late, even if it delays another one which arrives 5 minutes late as well, there is very little, maybe none, actual cash that changes hands or impacts on the financial Profit & Loss report of income and expenditure at the end of the year for the owners. Of course, sometimes there is. Sometimes you have to pay overtime, etc. But not always.

I've sometimes wondered where the costs of all the Delay Attribution staff are charged to :) . Are they just taken as an overhead, the cost of doing business, or are they also minutely analysed, salaries, office costs, etc, and then recharged to the delays?
 
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R

RailUK Forums

AlterEgo

Verified Rep - Wingin' It! Paul Lucas
Joined
30 Dec 2008
Messages
29,544
Location
LBK
I believe more operators should follow the lead of Caledonian Sleepers and put in a "performance allowance" to account for late running. For example 1S25 is due to arrive Inverness at 08:35 but is advertised at 08:45. Similarly 1M16 is due Euston 07:47 but the public time is 08:00. If it arrives 13 minutes late everybody's happy (well, the punters at least). It might be dismissed as "padding" by some, but it's good for the public perception of performance.
This happens anyway, the WTT and GBTT are different. See also "diamond time". https://wiki.openraildata.com/index.php/Allowances
 

The Planner

Veteran Member
Joined
15 Apr 2008
Messages
19,718
I believe more operators should follow the lead of Caledonian Sleepers and put in a "performance allowance" to account for late running. For example 1S25 is due to arrive Inverness at 08:35 but is advertised at 08:45. Similarly 1M16 is due Euston 07:47 but the public time is 08:00. If it arrives 13 minutes late everybody's happy (well, the punters at least). It might be dismissed as "padding" by some, but it's good for the public perception of performance.

Most operators have had WTT/GBTT differences for decades.
And plenty have removed it as it was considered as underhand fudging.
 

Pegpilot

Member
Joined
26 Mar 2020
Messages
120
Location
Northants
I spent a few years deep down in the detail of the London Underground PPP performance payment framework. The PPP contractors were penalised for each train delay on a sliding scale reflecting location and time of day. The measurement unit was the Lost Customer Hour and there was a set of calculation tables to calculate LCH's for any given incident. A key point was that for any disruption, each additional minute of delay incurred a higher rate of LCH accumulation as incrementally more trains were stuck behind the original failed train (or infrastructure fault). And when you're running a 2-minute service in the peak, a 10 minute initial delay will be affecting 5 loaded trains and the "fines" built up pretty quickly.
 

styles

Established Member
Joined
7 Dec 2014
Messages
4,930
Location
Gwynedd
I believe more operators should follow the lead of Caledonian Sleepers and put in a "performance allowance" to account for late running. For example 1S25 is due to arrive Inverness at 08:35 but is advertised at 08:45. Similarly 1M16 is due Euston 07:47 but the public time is 08:00. If it arrives 13 minutes late everybody's happy (well, the punters at least). It might be dismissed as "padding" by some, but it's good for the public perception of performance.
This is also easier to do when you run a service at nights when you aren't usually affecting other operators, and at low average speeds even on mainlines. The CS also benefits from aiming to just get people somewhere for the morning sometime, people don't generally rely on it for minute perfection.
 

Haywain

Veteran Member
Joined
3 Feb 2013
Messages
24,803
I'd imagine the 15th, 30th, 60th, and 120th minute of delay costs quite a bit more than the other minutes thanks to delay repay.
Not really - a delay of just a couple of minutes can be enough to push a connecting journey into Delay Repay territory.
I believe more operators should follow the lead of Caledonian Sleepers and put in a "performance allowance" to account for late running. For example 1S25 is due to arrive Inverness at 08:35 but is advertised at 08:45. Similarly 1M16 is due Euston 07:47 but the public time is 08:00. If it arrives 13 minutes late everybody's happy (well, the punters at least). It might be dismissed as "padding" by some, but it's good for the public perception of performance.
It happens all the time across the network, and frequently results in people travelling on a train half an hour later than intended being told they were only delayed by 29 minutes for the purposes of Delay Repay.
 

Dr Hoo

Established Member
Joined
10 Nov 2015
Messages
5,693
Location
Hope Valley
I've sometimes wondered where the costs of all the Delay Attribution staff are charged to :) . Are they just taken as an overhead, the cost of doing business, or are they also minutely analysed, salaries, office costs, etc, and then recharged to the delays?
This rather fails to recognise that Delay Attribution is about, err, delays and more particularly their root causes. This information enables Responsible Managers to know how their aspect of the rail system is affecting performance whether it’s asset related, planning related, staff related, etc.. Suitable plans to minimise original failures, speed up response, improve service management around problems and so on can then be developed and costed. That was behind BR’s decision to establish it and invest in the relevant data capture and software.

The Performance Regime is far less intertwined with the details of delay attribution than might be imagined. In crude terms it is interested in distinguishing between ‘operator on self’, ‘infrastructure manager’ and ‘operator on other operator’. It is also calculated on Average Lateness (at particular Monitoring Points) rather than delay minutes from individual incidents.

The Performance Regime is also detailed enough to recognise that lateness at (say) a London terminal on peak services is a bigger issue than lateness at Lowestoft or Aberystwyth in terms of the marginal revenue effect.

Both aspects of performance measurement have to also consider related matters such as cancelation and freight.

This remains a very brief and simplified overview.
 

NeakPed

Member
Joined
8 Nov 2025
Messages
93
Location
uk
This rather fails to recognise that Delay Attribution is about, err, delays and more particularly their root causes. This information enables Responsible Managers to know how their aspect of the rail system is affecting performance whether it’s asset related, planning related, staff related, etc.. Suitable plans to minimise original failures, speed up response, improve service management around problems and so on can then be developed and costed. That was behind BR’s decision to establish it and invest in the relevant data capture and software.

The Performance Regime is far less intertwined with the details of delay attribution than might be imagined. In crude terms it is interested in distinguishing between ‘operator on self’, ‘infrastructure manager’ and ‘operator on other operator’. It is also calculated on Average Lateness (at particular Monitoring Points) rather than delay minutes from individual incidents.

The Performance Regime is also detailed enough to recognise that lateness at (say) a London terminal on peak services is a bigger issue than lateness at Lowestoft or Aberystwyth in terms of the marginal revenue effect.

Both aspects of performance measurement have to also consider related matters such as cancelation and freight.

This remains a very brief and simplified overview.
(using an iPad on a train so too much to reduce the original quote - apologies).

There is no part of the financial/Schedule 8 regime that takes into account of ‘operator on other operator’ delays. The financial cost of this is met by Network Rail.
Section C1 of the DAPR on the Network Rail site covers this.
(April 2026 DAPR

For performance/delay purposes it is a useful tool/figure
 

Dr Hoo

Established Member
Joined
10 Nov 2015
Messages
5,693
Location
Hope Valley
(using an iPad on a train so too much to reduce the original quote - apologies).

There is no part of the financial/Schedule 8 regime that takes into account of ‘operator on other operator’ delays. The financial cost of this is met by Network Rail.
Section C1 of the DAPR on the Network Rail site covers this.
(April 2026 DAPR

For performance/delay purposes it is a useful tool/figure
That’s not really correct. Yes. there is one payment between Network Rail and the operator that rolls up Network Rail and inter-operator performance.

However, the operator’s own payment rate to Network Rail is fundamentally about the modelled likelihood of its own delays affecting another operator. So a largely self-contained operator such as Merseyrail will have a very low payment rate in Schedule 8 even if its own performance is atrocious because their interaction with other operators at Chester, Bidston and Hunts Cross is tiny. CrossCountry on the other hand, which mingles with multiple other passenger and freight operators on its routes, will have a much higher rate.
 

NeakPed

Member
Joined
8 Nov 2025
Messages
93
Location
uk
That’s not really correct. Yes. there is one payment between Network Rail and the operator that rolls up Network Rail and inter-operator performance.

However, the operator’s own payment rate to Network Rail is fundamentally about the modelled likelihood of its own delays affecting another operator. So a largely self-contained operator such as Merseyrail will have a very low payment rate in Schedule 8 even if its own performance is atrocious because their interaction with other operators at Chester, Bidston and Hunts Cross is tiny. CrossCountry on the other hand, which mingles with multiple other passenger and freight operators on its routes, will have a much higher rate.
It’s been a long time since i was involved in Schedule 8 (15 years +) so memory is a bit vague.
I recall that there was a ‘societal rate’ (i think), and busyness factors for monitoring points. A complex calculation that I was able to mirror in Access to provide a reasonable estimation of the Schedule 8 regime on a weekly/period basis for financial purposes.
At the TOC i worked, there were 2 service groups, that covered pretty much the same routes but with different traction and few traction specific origin/terminating points.
One group had lower financial values despite the fact that either could cause disruption to other TOC services and they both used (predominantly) the same stations.
I’ll have to have a look through my archives to see if I can find rekindle my hazy memory.
 

Bald Rick

Veteran Member
Joined
28 Sep 2010
Messages
35,909
What's the situation now where TOCs are government ran? Are the government now not attributing delay costs on themselves or is there some slight of hand where it still actually matters?

Delays will always be attributed, albeit in GBR I imagine there will be rather less arguing over individual minutes.

Some DfT owned TOCs are in the process of suspending the Schedule 8 regime though, ie money doesn’t change hands. You can probably work out which ones.


Fact is that these financial figures are just a management tool, to feature in business reports. They are not hard cash. If a train arrives 5 minutes late, even if it delays another one which arrives 5 minutes late as well, there is very little, maybe none, actual cash that changes hands or impacts on the financial Profit & Loss report of income and expenditure at the end of the year for the owners.

They very much are hard cash for the operators not in DfT ownership. Significant sums in fact. Historically several operators have run into big financial trouble because their assumptions about NR performance proved to be rather optimistic (or pessimistic, depending on your point of view).


Schedule 8 payments are there to compensate operators (or NR) for costs incurred and revenue lost through late running. A train that is even only 1 minute late can mean overtime on traincrew diagrams, extra fuel burnt, etc. Obviously above certain thresholds there is compensation to passengers, but there is also evidence that late / cancelled trains trigger demand suppression throughout the time bands, ie for travel right now as well as travel tomorrow and later.


At the TOC i worked, there were 2 service groups, that covered pretty much the same routes but with different traction and few traction specific origin/terminating points.
One group had lower financial values despite the fact that either could cause disruption to other TOC services and they both used (predominantly) the same stations.

When Thames Trains and First GW High Speed merged, they still had seperate track access agreements. If the (say) 1700 Thames stopping service to Reading failed across the throat of Paddington in the evening peak, NR would have been paying out potentially net six figures in Schedule 8 compensation to the combined FGW.
 
Last edited:

Solweytracker

Member
Joined
10 Oct 2020
Messages
51
Location
UK
They very much are hard cash for the operators not in DfT ownership.
So does this seriously affect the profitability of the FOCs also? For example, when the DRS locomotive on the 4S43 Daventry to Scotland trains failed on Shap and blocked the WCML operation for a few hours, with numerous delays and knock-on cancellations, who picks up the tab, and is it very expensive?
[Aside from the customer's own penalty clauses for late delivery].
 

Bald Rick

Veteran Member
Joined
28 Sep 2010
Messages
35,909
So does this seriously affect the profitability of the FOCs also? For example, when the DRS locomotive on the 4S43 Daventry to Scotland trains failed on Shap and blocked the WCML operation for a few hours, with numerous delays and knock-on cancellations, who picks up the tab, and is it very expensive?
[Aside from the customer's own penalty clauses for late delivery].

FOCs have a cap on their payments to protect them from high impact incidents. Otherwise they’d all have been out of business in fairly short order. It’s the same for open access, AIUI.
 

Dr Hoo

Established Member
Joined
10 Nov 2015
Messages
5,693
Location
Hope Valley
FOCs have a cap on their payments to protect them from high impact incidents. Otherwise they’d all have been out of business in fairly short order. It’s the same for open access, AIUI.
I’ve been retired too long to be certain but don’t the (optional) caps have to be ‘purchased’ up front? A bit like an insurance policy can ‘syndicate’ risks.

(Happy to be corrected.)
 

Bald Rick

Veteran Member
Joined
28 Sep 2010
Messages
35,909
I’ve been retired too long to be certain but don’t the (optional) caps have to be ‘purchased’ up front? A bit like an insurance policy can ‘syndicate’ risks.

(Happy to be corrected.)

Kind of. Not my area of expertise!
 

Harpo

Established Member
Joined
21 Aug 2024
Messages
3,905
Location
Newport
Historically several operators have run into big financial trouble because their assumptions about NR performance proved to be rather optimistic (or pessimistic, depending on your point of view).
I remember an account team being puzzled when a coastal branch that repeatedly flooded in winter didn’t become an issue with its TOC. Analysis showed the S8 payment vastly outstripped the out-of-season farebox revenue, hence the silence. Leaf fall could also be a ‘dog that didn’t bark’.

However, when performance later hit a rare golden patch, that team didn’t know how to invoice the TOC as they hadn’t done it before.
 

Krokodil

Established Member
Joined
23 Jan 2023
Messages
6,597
Location
Wales
I believe more operators should follow the lead of Caledonian Sleepers and put in a "performance allowance" to account for late running. For example 1S25 is due to arrive Inverness at 08:35 but is advertised at 08:45. Similarly 1M16 is due Euston 07:47 but the public time is 08:00. If it arrives 13 minutes late everybody's happy (well, the punters at least). It might be dismissed as "padding" by some, but it's good for the public perception of performance.
Night trains need to have considerable amounts of slack built into their schedules to allow for a wide variety of possible diversions without the need to alter the public timetable.

Adding more than the odd minute here and there to a normal train slows down the advertised schedule which makes it less attractive. It's also not very practical for anywhere besides the terminus.
 

Clarence Yard

Established Member
Joined
18 Dec 2014
Messages
3,272
FOCs have a cap on their payments to protect them from high impact incidents. Otherwise they’d all have been out of business in fairly short order. It’s the same for open access, AIUI.

No, Open Access is the same as the other passenger TOCs. Both have caps for trains/monitoring points which they do not have to pay extra for. It’s charter operators that pay a per train mile charge for their cap because they are more directly exposed to the costs of individual delays, especially the effects on other operators.

The 28 day average nature of schedule 8 combined with the different thresholds, monitoring points weightings and payment rates for each TOC service group makes it almost impossible to say this delay cost x pounds. It’s a system of averages, not absolutes.

Projecting schedule 8 (and schedule 4) is a significant job for any TOC as it is effectively an (automated) liquidated damages regime that has to be budgeted for. The sums involved (and periodic swings) can be substantial.

The DfT has always regarded any detailed TOC schedule 8 figures as commercially confidential so you won’t find anything in the public domain and there are always suitable redactions to any TAA on the ORR website so you can’t work it out in theory.
 

Dr Hoo

Established Member
Joined
10 Nov 2015
Messages
5,693
Location
Hope Valley
A typically thorough and contextualised post by @coppercapped .

Something that might be added is that if the performance regimes are just ‘turned off’ or ignored the financial signals go blank at the same time. All too easy for people to lose awareness of how much unreliable infrastructure, unreliable trains, poor measures to address leaf fall, cable theft, trespass, etc. is actually ‘costing’ as the revenue line sags.

When I was at ScotRail in BR days it took a vast effort to explain to signalling and track colleagues that repeated points failures at Greenhill Upper Junction were wrecking the Edinburgh-Glasgow route’s reputation as a dependable service while coach competition on this short route grew. Some major work followed and the revenue line turned up immediately.
 

Harpo

Established Member
Joined
21 Aug 2024
Messages
3,905
Location
Newport
When I was at ScotRail in BR days it took a vast effort to explain to signalling and track colleagues that repeated points failures at Greenhill Upper Junction were wrecking the Edinburgh-Glasgow route’s reputation as a dependable service while coach competition on this short route grew.
Same in Railtrack/NR days. Asset performance came after safety, compliance and budget and was a consequence, not an intention. So frustrating!
 

The Planner

Veteran Member
Joined
15 Apr 2008
Messages
19,718
Yet other bits of NR are fiercely trying to get a load put in, including intermediate dwells to make their current measures look better...
Can you prove they aren't needed? Dwells are a well known issue.
 

87015

Established Member
Joined
3 Mar 2006
Messages
5,078
Location
GEML/WCML/SR
Can you prove they aren't needed? Dwells are a well known issue.
PB doesnt affect the dwell time though? WTT 1000-1002 being public 1002-1002 is what is being pushed, which i'd very much put as 'underhand fudging'!
 
Last edited:

citycat

Member
Joined
18 Dec 2013
Messages
334
Location
Woerden, The Netherlands
What if the late running passengers had arrived on a late running connection and that was the reason for them boarding late? It seems extremely poor to publicly shame them.
I used to be a train guard on the ECML and I wouldn't dream of making such an announcement. It seems today's train managers are quite happy to go off track (excuse the pun) from the official spiel book and make their own judgemental announcements. I wouldn't blame the late running passengers for making an official complaint if they had a justifiable reason for running for the train. And the service would normally have some slack built into the timetable for it to be a non issue delay by the late runners.

I once delayed departure of the train I was working from Bradford Forster Square by two minutes because I was running back from the fish and chip shop. The driver wasn't bothered as I had his fish and chips too, which I delivered to him during the Shipley stop. We never got a 'please explain' from control.
 
Joined
7 Jan 2009
Messages
1,031
A couple of minutes will normally be regarded as a 'sub-threshold delay' (normally 3 mins is the margin) and so is not recorded as a delay at all for the purposes of Schedule 8.

The rate the LNER conductor quoted seems the right kind of level, given the high revenue nature of the route.

It is interesting to note that some of the comments here recognise that the much of the Sched 8 Delay Attribution machinery is likely to be kept post-GBR. Indeed, a fair amount of it predates privatisation! When senior managers say they wish to 'abolish Sched 8' they are normally of course only thinking about payments, not the measurement machinery behind it...which, although cumbersome, is vital management information for improvement.

NR has always been vocal about 'abolishing Sched 8'. Unfortunately for all of us, their routes often end up having up to pay up to TOCs as performance is, on a long term basis, below the benchmarks set by ORR. So they have always seen it as an unecessary cash outflow from a company that is insufficiently funded. When the regime was created (thinks: do I admit to being around then?) the idea was that the benchmarks should normally be more balanced, ie they could just as easily be exceeded by NR as they are not reached. Somewhere along the line, this got garbled (perhaps because the whole industry's average performance in absolute terms has been very slowly declining for about a decade although there are some signs that it is now flattening out?) And, was always the case, some routes/TOCs do much better on performance improvement than others.
 

Dr Hoo

Established Member
Joined
10 Nov 2015
Messages
5,693
Location
Hope Valley
A ‘spot on’ post from @Gerald Fiennes .
I was quite happy to retire after a mere 20 years of being involved with Schedule 8 (and Schedule 4) in various ways but ‘yes’ the regimes are intended to be, on average, financially neutral around the benchmarks (which have been re-calibrated at each Periodic Review).
Sadly for many years performance has tended to lag behind aspirations and in the process undoubtedly restricted the industry’s ability to attract passengers and increase farebox revenue. The amounts changing hands under Schedule 8 have been modelled to broadly reflect this.
Nobody should be in any doubt that ‘flaky’ service delivery has cost the industry literally billions of £,£££,£££,£££s over the decades; just as it did in BR days, of course, at commensurate pricing levels - it’s just that the data capture, modelling and analysis of the issue didn’t exist back then.
 

dcsprior

Member
Joined
28 Aug 2012
Messages
862
Location
Edinburgh (Fri-Mon) & London (Tue-Thu)
I believe more operators should follow the lead of Caledonian Sleepers and put in a "performance allowance" to account for late running. For example 1S25 is due to arrive Inverness at 08:35 but is advertised at 08:45. Similarly 1M16 is due Euston 07:47 but the public time is 08:00. If it arrives 13 minutes late everybody's happy (well, the punters at least). It might be dismissed as "padding" by some, but it's good for the public perception of performance.

A TOC only really has the option of advertising a later-than-expected time at stops they aren't also picking up passengers from - because if they are picking up, they'd need to sit and wait until their advertised time anyway.

CS is unusual in that most (perhaps all?) of its Scottish stops are pick-up-only northbound, and most (all?) of its English stops are pick-up-only southbound.

That isn't the case for most TOCs!
 

Travelmonkey

Member
Joined
16 Aug 2023
Messages
773
Location
The Midlands
A couple of minutes will normally be regarded as a 'sub-threshold delay' (normally 3 mins is the margin) and so is not recorded as a delay at all for the purposes of Schedule 8.

The rate the LNER conductor quoted seems the right kind of level, given the high revenue nature of the route.
they are eagarlly watched from above, I know control were on the phone to one of my gurads before he even had time to get the ramp down for me due to the dewll at a intercity stop.

does make one wonder would the guard OP is refreencing been as so blunt to a wheelchair user. I dread to think the dressing down they'd get if they did it to someone like Doug Pauley,
 
Status
Not open for further replies.

Top