How are delay minutes calculated?

[trainophile]

I understand that TOCs are fined when a service runs more than a couple of minutes late, and wondered how this is formulated.

If a train departs its original station on time but is subsequently delayed, having a knock-on effect that results in further delay at each station it calls at, presumably they don't calculate the fine compoundedly (is that a word?), to cover each station's delay?

If the train then manages to make up the lost time before its point of termination, is the fine no longer applicable?

Would be interested to learn some background to this, as it has been perplexing me since I started following my inbound trains on NRE.

 

[1D53]

Delay is typically done as a minimum of three minutes in a 'section', which is normally between major junctions or stations, delay is not double counted, so if 3 mins is lost beyween A and B and 3 between B and C the total 'fine' will be 6 minutes not 9.

Time made up is still applicable.

 

[trainophile]

I see, thanks for the information. I was also wondering how 'blame' and therefore 'charge' is allocated when the delay is out of the TOC's control, e.g. a trespass incident or fatality, or overhead wires problem - in fact there's probably far more reasons why delay isn't a TOC's fault, than those that can be attributed.

Must be a nightmare trying to sort out who pays what.

 

[Tomnick]

Anything that causes a delay (usually of 3 minutes or more, as above) will be detailed in TRUST, with an incident number associated. Any delay over a reporting section will be attributed either directly to a delay incident or, as the effects become widespread, to another train (for example, following another late-runner, or regulated to let something else go first), ultimately leading back to an incident number somewhere! Thus each incident has an ever-increasing number of delay minutes against it, both direct and reactionary, and an indication of who's responsible for the delay.

 

[Philip Elliott]

A system called Bugle is used.

 

[trainophile]

Phew . Makes you wonder whether the revenue gained from all the cross-charging covers the admin involved in calculating it all out.

Of course it should theoretically serve as a deterrent to encourage train operators to run on time, but in practice it's usually going to be down to unavoidable reasons anyway, as I'm sure they all want to run on time!

 

[Tomnick]

In some cases (particularly FOCs), I think it's seen as an important revenue stream rather than a deterrent! I don't know how the system works in detail, but the reactionary delays should all be attributed automatically once they're dobbed against the late runner that caused it. Still take a lot of manpower though - even at 0300 on a quiet morning .

 

[1D53]

A system called Bugle is used.

BUGLE is an internal not system which TOC's use for Performance, it manages nothing of the main delay process.

 

[wensley]

Take a read of this ... http://www.delayattributionboard.co.uk/documents/dag_pdac/DAG September 2012.pdf

It's a long and complicated process, the TOCs can dispute at various levels, with corresponding staff in NR who you have to deal with. The information is all retained by TRUST, TOCs then have their own systems for recording this data, which is only retained in TRUST for a few weeks.

3mins is the threshold when a delay becomes attributable. However, if a 2min delay (for example), caused a train in rear to lose 3mins this would become attributable also.

Obviously delay mins is one element of the performance regime, PPM is also seen to be important (probably the most important figure for a TOC's performance team), as well as CaSL etc.

The numbers can rapidly get eye-watering!

 

[306024]

A lot depends on the accuracy and enthusiasm of the delay attribution clerk. Easy to palm off a delay to an existing incident rather than have to do a detailed investigation which, when the job has gone belly up, there isn't time to do.

Recall a delay in Scotland once being blamed on a late running Sudbury to Marks Tey service. While that is theoretically possible at a push, a five minute replay of the service proved it was nothing of the sort.

A cynical view held by some is that there are too many people just measuring delays, rather than working pro-actively to avoid delays in the first place.

 

[wensley]

A lot depends on the accuracy and enthusiasm of the delay attribution clerk. Easy to palm off a delay to an existing incident rather than have to do a detailed investigation which, when the job has gone belly up, there isn't time to do.

Recall a delay in Scotland once being blamed on a late running Sudbury to Marks Tey service. While that is theoretically possible at a push, a five minute replay of the service proved it was nothing of the sort.

A cynical view held by some is that there are too many people just measuring delays, rather than working pro-actively to avoid delays in the first place.

And the enthusiasm of the TOC staff to dispute and trace it back to the true cause!!

Measuring delay does largely have no impact in improvement, other than tracing patterns which can be used going forward. A good example is identifying sub-standard sectional running times for timetable changes, or to tackle recurring issues at forums such as Right Time Railway.

 

[Bald Rick]

A cynical view held by some is that there are too many people just measuring delays, rather than working pro-actively to avoid delays in the first place.

Not cynical at all, it is a commonly held view, even amongst those who actually do the job!

Nevertheless, even if no money was changing hands, it would still be important to identify the root cause of delays, so attribution would still be necessary. However there wouldn't need to be the duplication of roles and 'constructive tension'.

How the delay minutes recorded and attributed turns into 'fines' is however a very very complex set of calculations and contractual machinations.
--- old post above --- --- new post below ---

And the enthusiasm of the TOC staff to dispute and trace it back to the true cause!!

It is well known that TOCs provide their delay management teams with special keyboards that only contain one button ISPUTE>

 

[wensley]

Not cynical at all, it is a commonly held view, even amongst those who actually do the job!

Nevertheless, even if no money was changing hands, it would still be important to identify the root cause of delays, so attribution would still be necessary. However there wouldn't need to be the duplication of roles and 'constructive tension'.

How the delay minutes recorded and attributed turns into 'fines' is however a very very complex set of calculations and contractual machinations.
--- old post above --- --- new post below ---


It is well known that TOCs provide their delay management teams with special keyboards that only contain one button ISPUTE>

That explains why I have to hit my 'A' key so hard when on the incident screen in TRUST

 

[tsr]

Recall a delay in Scotland once being blamed on a late running Sudbury to Marks Tey service. While that is theoretically possible at a push, a five minute replay of the service proved it was nothing of the sort.

But as another example, I'm fairly sure the late running of a train via Gatwick last summer was blamed on a cow on the line at Nailsea & Backwell, which caused congestion due to the subsequent late running of an FGW service via Barnham. And a slight delay to an early-evening stopping service on approach to Heyford the other day was almost certainly caused by the landslip at Unstone. So these sorts of things do happen, if not on quite such a "silly"-sounding scale!

Occasionally, it's possible to trace the delay fairly readily and then inform any customer who enquires in depth, which can result in some amusing comments...!

 

[455driver]

Recall a delay in Scotland once being blamed on a late running Sudbury to Marks Tey service. While that is theoretically possible at a push, a five minute replay of the service proved it was nothing of the sort.

I have seen one where a delay on a SWT service (which knocked a XC service for 4 minutes) had a lot of delays from Scotrail added to it. It was only when I pointed out that the XC service had actually left Birmingham New Street (BHM) on time that the Clerks queried it and it disappeared up somebody elses exhaust.

DA was set up to try and find out all the problems with the timetable so they could be dealt with but it has turned into a money go round with people trying to palm off delays anywhere they will stick regardless of actual blame.

Dont get me started on signallers O codes!

 

[1D53]

3mins is the threshold when a delay becomes attributable. However, if a 2min delay (for example), caused a train in rear to lose 3mins this would become attributable also.

2 minutes is only attributed on LNE for the moment.

 

[The Planner]

Also helps if you are on good terms with a the PDQS bods, a quick phone call explaining it often gets it moved quicker!

 

[D1009]

Not cynical at all, it is a commonly held view, even amongst those who actually do the job!

Nevertheless, even if no money was changing hands, it would still be important to identify the root cause of delays, so attribution would still be necessary. However there wouldn't need to be the duplication of roles and 'constructive tension'.

How the delay minutes recorded and attributed turns into 'fines' is however a very very complex set of calculations and contractual machinations.

It is well known that TOCs provide their delay management teams with special keyboards that only contain one button ISPUTE>

I agree with that, but what the process has done over the years is making managers personally responsible for the delay their various departments cause on the railway, and it is always discussed at performance reviews. Without it I don't think our railway would be as punctual as it is.

 

[GB]

In some cases (particularly FOCs), I think it's seen as an important revenue stream rather than a deterrent!

This is true. We have it drummed into us periodically that all delays must be advised and attributed correctly. When your looking at tens of thousands of pounds across the year its easy to see why.

Of course Network Rail have the keen ability to implement their MFSDD policy (Management of Freight Services During Disruption) for freight services over 120 minutes late to help massage the figures.

 

[lemonic]

I agree with that, but what the process has done over the years is making managers personally responsible for the delay their various departments cause on the railway, and it is always discussed at performance reviews. Without it I don't think our railway would be as punctual as it is.

Some might argue that the railway is too punctual as a result of delay attribution. By that, I mean many trains are not held for 1 or 2 minutes when it would benefit passengers transferring from a delayed connecting service, simply so that the train is 'on time'. I am aware that in some circumstances holding a train could lead to a lot of knock-on delays, but there are also situations where holding a train will not lead to (m)any knock-on delays and the train is still not held.

 

[Bald Rick]

Also helps if you are on good terms with a the PDQS bods, a quick phone call explaining it often gets it moved quicker!

Especially if you make him a regular cup of tea...

 

[The Planner]

and it is always discussed at performance reviews.

It isn't used on our performance reviews, which amuses me.

 

[trainophile]

Presumably there are some occasions where delays cannot be attributed to either the TOC or NR - I'm thinking suicides for one example?

Same with trespass I suppose, although they could argue that it is NR's responsibility to make the railway inaccessible - however that would be impossible on stretches that have level crossings.

What about animals on the line? Is any action taken to charge the farmer concerned?

 

[The Planner]

Suicide is a NR delay, XC delay code I think? animals is dependant on the state of line side fencing

 

[GB]

Its my understanding that suicides originating from a station the TOC will pick up the minutes, all others are picked up by Network Rail.

 

[trainophile]

Blimey, that seems a bit harsh.

 

[Bald Rick]

Its my understanding that suicides originating from a station the TOC will pick up the minutes, all others are picked up by Network Rail.

Not quite right - this was tested at a very early delay attribution board. It is due to the track access contract and station access contract being separate contracts.

The only fatalities (or attempts) that are not given to NR are those that either originate on a train (ie jumping out of a train) or those that originate from a station but do not touch the rail infrastructure, ie jump from platform and bounce back on to platform.

For any fatality, once the line is open, any delays due to the station remaining closed go to the TOC.

There are some other cases where delays are split between NR and TOC, but the above covers most eventualities.

Almost all 'external' delay goes to NR, eg animals on line, planes on line, weather, unexploded bombs, exclusion zones causing the line to be stopped etc etc.

 

[Tomnick]

One amusing exception, as I understand it, being a bird 'strike': smaller than a pheasant, the TOC gets it, otherwise it goes to NR.

 

[trainophile]

Considering NR don't receive any direct fares revenue, it must make a huge impact on their balance sheet having to fund all these eventualities that arguably aren't caused by them.

Just had another really silly thought - who does the fine money involved get paid to? I'd only thought about who pays it, not who receives it!

Sorry to harp on the morbid, but if someone jumps off a bridge in front of a train, then that's NR's liability? :?

Also, in a station fatality, it would be charged to the station operator, rather than the TOC whose train was involved, assuming they are not the same?

 

[455driver]

the delay minutes are thrown around the system until they stick somewhere whether it had anything to do with the actual cause or not!

I think that Delay Attribution adds nothing to the modern railway and it certainly isnt being used for the purpose it was intended for when it all started!