There's been intermittent talk of timetable padding on this forum and elsewhere. Personally I don't like the term 'padding' as I think it covers too many variables. Britain benefits from relatively open industry data so how the working timetable is put together can be relatively easily understood. The principal components of a GB WTT are
Sectional running times (SRTs) between timing points. These are modelled using rolling stock technical performance data and adjusted to reflect human behaviour and cover generalities. From my slightly limited understanding, things that might be perceived as 'slack' include:
- Generalisation to allow trains within the same class but different lengths to use a common set of SRTs - a longer train will take longer to clear a low-speed section, so if a shorter formation is used on the day then there'll be have some slack in the running time calibrated to the longest reasonable formation
- Generalisation to reflect the longer routing into/out of a timing point - e.g. the furthest platform / slowest routed platform
- A percentage uplift to reflect driver behaviour and adhesion. Drivers generally do not accelerate or decelerate to the full rolling stock technical capability
(where differences are too large to be generalised separate adjustments are applied)
My feeling is British SRTs don't tend to be exactly generous. I've heard anecdotal accounts that drivers used to the defensive / cautious driving styles don't achieve SRTs.
Station dwells. Highly dependent on dispatch procedures and passenger behaviour. I have strong views on dispatch methods but if there is no DOO, the time taken between wheel stop and door release and door locking and and wheel start can't be considered as slack. Passenger behaviour can be unpredictable. Again, British dwells don't tend to be overly generous - losing dwell at intermediate stations on intercity services isn't uncommon.
Pathing time. Unavoidable time in the schedule if you have to wait for a slower train in front or have to wait for your slot to cross a junction, and it's not possible to depart/pass the previous timing point later. This isn't slack. A schedule with a lot of pathing time isn't one with a lot of slack in it - it's simply a bad schedule.
Performance time. This is discretionary time and is generally genuine slack. There are instances of mandatory performance time like at Welwyn but this is rare. Sometimes performance is inserted to standardise the timetable - say if the xx00 needs a minute of pathing time because of a conflict but the xx30 doesn't have the same conflict, but to keep the two train exactly 30 minutes apart at the next station call you put a performance minute in the xx30.
Engineering time. This I'm told is a bit of a dark art. Its purpose is to allow for temporary speed restrictions (TSRs) from unplanned infrastructure degradation. Engineering time is mandatory and is specified in the Train Planning Rules. There can be large variations in how engineering times are applied. The Chat Moss route between Liverpool and Manchester needs 2 minutes of engineering time. The much longer Peterborough - King's Cross stretch of the ECML only requires one minute. This is slack if infrastructure is in a good state and there are no TSRs. A more recent approach is to have engineering allowance built into the SRT as a % uplift.
There is another form of slack found in prolonged transitional arrangements. If a fleet replacement is particularly drawn out you have a long period of a new faster train running to old SRTs. You can't adopt new SRTs until full fleet replacement is complete.
Do other European railway systems have similar components in their working timetables?
Sectional running times (SRTs) between timing points. These are modelled using rolling stock technical performance data and adjusted to reflect human behaviour and cover generalities. From my slightly limited understanding, things that might be perceived as 'slack' include:
- Generalisation to allow trains within the same class but different lengths to use a common set of SRTs - a longer train will take longer to clear a low-speed section, so if a shorter formation is used on the day then there'll be have some slack in the running time calibrated to the longest reasonable formation
- Generalisation to reflect the longer routing into/out of a timing point - e.g. the furthest platform / slowest routed platform
- A percentage uplift to reflect driver behaviour and adhesion. Drivers generally do not accelerate or decelerate to the full rolling stock technical capability
(where differences are too large to be generalised separate adjustments are applied)
My feeling is British SRTs don't tend to be exactly generous. I've heard anecdotal accounts that drivers used to the defensive / cautious driving styles don't achieve SRTs.
Station dwells. Highly dependent on dispatch procedures and passenger behaviour. I have strong views on dispatch methods but if there is no DOO, the time taken between wheel stop and door release and door locking and and wheel start can't be considered as slack. Passenger behaviour can be unpredictable. Again, British dwells don't tend to be overly generous - losing dwell at intermediate stations on intercity services isn't uncommon.
Pathing time. Unavoidable time in the schedule if you have to wait for a slower train in front or have to wait for your slot to cross a junction, and it's not possible to depart/pass the previous timing point later. This isn't slack. A schedule with a lot of pathing time isn't one with a lot of slack in it - it's simply a bad schedule.
Performance time. This is discretionary time and is generally genuine slack. There are instances of mandatory performance time like at Welwyn but this is rare. Sometimes performance is inserted to standardise the timetable - say if the xx00 needs a minute of pathing time because of a conflict but the xx30 doesn't have the same conflict, but to keep the two train exactly 30 minutes apart at the next station call you put a performance minute in the xx30.
Engineering time. This I'm told is a bit of a dark art. Its purpose is to allow for temporary speed restrictions (TSRs) from unplanned infrastructure degradation. Engineering time is mandatory and is specified in the Train Planning Rules. There can be large variations in how engineering times are applied. The Chat Moss route between Liverpool and Manchester needs 2 minutes of engineering time. The much longer Peterborough - King's Cross stretch of the ECML only requires one minute. This is slack if infrastructure is in a good state and there are no TSRs. A more recent approach is to have engineering allowance built into the SRT as a % uplift.
There is another form of slack found in prolonged transitional arrangements. If a fleet replacement is particularly drawn out you have a long period of a new faster train running to old SRTs. You can't adopt new SRTs until full fleet replacement is complete.
Do other European railway systems have similar components in their working timetables?