WTT138

As a pure guess, two less trains means greater stand time at termini.

Greater stand time gives more recovery time.

More recovery time gives Regulators (or whatever they're called these days) a bit more breathing space when it is needed to reform the service.

Just because there are more trains does not mean that the service levels are more reliable - less trains fighting for the same space means that the gaps between the service can be more even.

It's the new (and indeed correct) way of thinking.
Much better to have fewer trains and run more reliably.

To basically repeat what mrfs42 has said:

Have 15 trains, running as close together as they can, constantly losing time as they queue up at termini or queue up en-route to the termini. Then because of the late-running they then have to be short-tripped or reformed to recover - providing an un-reliable service during the reformation / short-tripping.

On the other hand have 10 trains, with lower frequencies, but more space to breath. Termini operate better. Much less likelihood of queuing. Much more likelihood of keeping to time, and less need to reform or short-trip. All-in-all a much more reliable service.

Better to have a 5-minute service that works, than a 4-minute service that doesn't.

One thing was missed - penalties ;D It seems to me that due to ever tightning on penalty margin, it is better to run a service late than to cancel it all together.... At least they then meet more targets and have fewer fines to pay!

Trip number: Any move in one direction on the line is called a trip, this can even be a train coming out of a siding to a platform. Each change of direction, the trip number goes up by 1.

Number of cars is simply how many cars (carriages) the train is formed of.

Snip turn - short duty (e.g. book on 07.00, book off 15.00 but with only a few hour's work - and certainly nothing like 8 hours!!).

On the bounce - a continuous stretch (e.g. "four days on the bounce").

;D ;D ;D

They are - aspect has already commented on this here.

The timing revisions for Picc 44 are indeed very interesting, but belong in another thread.

I was using terminal stand time as it is the easiest variable to calculate and to a certain extent is is also the most controllable variable in TT structure.

Intermediate stand time is a bit of a black art by comparison.

Surely not - less trains means a greater headway between them; would it not be at the expense of service density?

I might be wrong in this, so I'm open to correction, but 'the single platform at Barking or Plaistow' just doesn't really wash with me - by that token every individual platform is a 'single platform'.

Plaistow is a red herring as the platform occupancy is away from the main and successive trains; on book there isn't a consecutive stream of trains turning back in Plaistow; likewise platform 3 at Barking. Now, if you were talking of either platforms 6 or 1/2 at Barking I would concede that there would need to be a minimum time of 5 booked minutes between consecutive moves (if tipping out).

Likewise with the reversals I'd guess at 1 - 2 minutes clear space needed behind/around reversing trains, depending on track circuit layout, and looking at the booked working there's never anything arriving less than ½ min before a departure or 1½ min after a departure. Less trains means that the platforms at termini can be occupied for longer - ie new headway less 30 seconds, notwithstanding if you've got an odd service interval that doesn't fit cleanly the stand time might need to be tinkered with further.

If you're saying that the single platforms at Barking and Plaistow are a bottleneck, restricting the service: what about OB45/train waiting at OB37 or EE210/train waiting at EE225?

I apologise if I've got completely the wrong end of the stick, but I just can't see a straightforward (well, any; if I'm being honest) relationship between the stand time at Richmond/Wimbledon/Ealing and Hammersmith being a function of the bays at Plaistow and Barking. There are other factors en route that would need to be considered; not the least of which being WB services from Tower Hill, Manky House and the middle pair at Aldgate.

EDIT: You're not thinking of maintaining the same (WTT 137) headway with less trains are you? In that case I can understand completely the shortening of layover time - however I'm still flummoxed by the single platforms at Barking and Plaistow.

And snapshots, don't forget yer snapshots MetControl!! ;D ;D ;D

A quick explanation of the other way in which service control is measured against a target:

Starting at 6am (9am on Sundays), and then every 3 hours until midnight, service control has a snapshot target - the number of trains in service against the number specified at that hour in the WTT. It doesn't matter where they are, or whether or not they are on time, etc; as long as a train is not cancelled it is in service. If all scheduled trains are in service at the time specified, service control has met it's snapshot target. If a train is cancelled at 1 minute past the hour specified but re-enters service at xx59 in the hour before the next snapshot, both snapshots will not show the cancellation. Needless to say many things that happen are based around snapshot times ;D ;D

As for the cancellation verses service recovery benefits, lets try a couple of examples and see what people think:

Bakerloo line, weekday morning peak. There's a signal failure at Paddington on the northbound, and because there are points involved, there is an initial delay whilst they are scotched & clipped. We'll be kind and say it's taken 15 minutes to get the first train through, but of course as each subsequent train passes through the area, it too will be delayed as the driver applies the appropriate procedure. So, how long will it take for trains to start blocking back behind the failure? How long will it take to start affecting the Southbound? Don't forget that having trains sat in the tunnel between stations is a massive no no.

Well this was one of the many scenarios I did as part of my controller training (something I will be repeating next year, but that's another thread on it's own!).

With a headway of 2 or 3 minutes, and only 2 platforms at Elephant & Castle (where stepping back is employed), it doesn't take a genius to realise that you must cancel trains if you are to have any hope of recovering the situation.

In another scenario, (which I can't remember right now), I had no option but to suspend the service Paddington to Piccadilly Circus both roads. I had trains reversing South to North at Paddington and North to South at Piccadilly Circus. Whilst the service North of Paddington wasn't such an issue, the time it takes to reverse there was and if I hadn't cancelled a few trains, they would have soon blocked back though Queens Park. The Elephant & Castle to Piccadilly Circus 'shuttle' was much more critical however - it only needed 5, maybe 6 trains maximum (again it takes time to reverse at Piccadilly Circus) yet I had something like 10 or 11 trains in that section. Cancelling trains was a must do. Sod the "penalties"!!

Remember I said you mustn't have trains in tunnels between stations? Well take the Victoria line - they have more trains in service than platforms - cancellations are one of the first actions a Victoria line controller will take.

On lines like the District, well all of SSR really, cancellations are not always an immediate concern/consideration - but it is a tool that is used quite often and can usually be more than justified when the bean counters start getting involved (usually long after the event and with plenty of time to analyse the ins & outs of whether or not a cancellation was justified :.

We seem to have wandered way off topic, but never mind, hopefully there's some food for thought within this post and I'm quite happy to continue the thread drift as it's already gone too far to recover anyway ;D ;D ;D ;D

Your the timetable expert MRFS, however my point was two fold. Because as soon as a H&C train leaves Barking bay or Plaistow bay, the next train is waiting to enter the bay. If you extend the platform dwell times at terminals (which is indeed excellent for service recovery) you hold the train outside and waiting to come in unless that train is not as frequent (ie. with a greater headway between). For the new timetable that is exactly the case. The longer dwell time at Plaistow and Barking is possible only because the H&C service headway is increased from a train every 8mins to a train every 10mins.

My contra reference in respect of the District Line (because this thread is WTT138) where we have multiple platform availability is that we are indeed maintaining the same headway as WTT 137 (x5 mins to Upminster and Wimbledon and x10 mins to Richmond and Ealing) and thus in order to give additional dwell time at Richmond (up from 12 mins to 18) and maintain the same headway an additional train is required when compared to WTT137, and that is set number T25. (I haven't yet had time to examine how reduced layover at Wimbledon is achieved with the same headway and same running time on what looked like the same number of trains).

Hence my conclusion that increased layover time can only be achieved at the expense of either, headway or running times or needing additional rolling stock. (thus expanding upon your thoughts that reduced rolling stock means additional layover time at termini)

Not in WTT138's case, no.

The extra trains are being provided to give additional layover time at terminus's - headways remain the same as before.

The additional running time on a peak rounder from Upminster to Wimbledon is in excess of 10 minutes when compared to WTT137, so this must certainly play a factor in the requirement of additional trains, allied to the terminus layovers increasing

That low humming noise you can hear?

It's not the Taos Hum it's JP Thomas spinning in his grave.

I spent an enjoyable hour this afternoon reading the District timetables 1908 - 1911 - such major alterations then in the schedule would have been picked up in the 'Black Book' long before they're necessary and introduced wholesale. Were the problems that chronic to require such an acute reaction - very, very interesting.

I suppose someone will say it - so I'll get a pre-emptive strike in: Could this extra 10 minutes be the price to pay for the Extended Circle? [1]
(I'll keep my thoughts under wraps, protempore - not 'cos I'm scared to say it).

aspect - TVM I'll revisit it when I'm not quite so pooped!

[1] I'm not saying it 'is', <Para Handy> 'chust ponderin' </Para Handy>

Regarding the difficulties inherent in turning trains at Gunnersbury, I wonder if a reversal there would be easier if the existing trailing crossover on the London side of the station were supplemented with a central reversing siding on the Richmond side of the station. The former four-track alignment provides plenty of space that could be used to get a siding of sufficient length for S7s and 378s, and with the station now under the TLC of LO, staff should be available to assist with reversals.

Would that make it easier to write a timetable like WTT138?