Timetables

q8, there is one simple reason why such a system will never be implemented:

It's common sense stuff

Extended edition:

Splitting the branches of the Northern into High Barnet/Mill Hill East via Charing Cross to Kennington and Edgware via Bank to Mordern could also help services on there as well, considering how dire things are at present. Then Camden Town would be a lot more user friendly!

P.S. Has anyone in LU management actually considered this? I mean it really is logical. It helps the service get back to normal, it could even improve the normal service, and the passengers (err....'customers') will have a better understanding of the services available, especially touristy types.

Thing is, if passengers do want to ride from one end of the line to another, they can just haul themselves off the train at a branch station and get on another one. Most passengers will just want to head to somewhere in Central London, and surely it's them the schedules people should be targetting? (i.e. schedules designed for them, with little disruption as possible)

Basically, the idea behind scheduling most trips between each end of the line, is for the premise that there is greater recoverability from late running and delays. If a train running to the end of the line is running late, then it could potentially be reversed short of its original destination (remember that reversing point will be free as we are running end to end) to pick up its scheduled working in the opposite direction.

You can’t do that with every train, but timetabling every train to run end to end, you can reverse a fair amount short of the terminus in late running for as long as needed, while keeping a reasonable service to run to the terminus. So you progressively get more and more trains back to time.

If you have overlapping patterns of service – taking the District Line example mentioned, then because several reversing points are used at the same time, its complex to manage, and late running trains would be less able to recover. Hence you see end to end running, though having said that – on a typical District Line timetable – you have a series of Tower Hill, Barking and Upminster reversers anyway – which kinds of defeats my point. I reckon that’s more to do with number of available trains and platform space at Upminster (you can’t send everything upto Upminster!) – Plaistow reversers wouldn’t usually be scheduled as when the train reverses, its WB trip conflicts with the EB working. (I stress is scheduled, as opposed to it actually happening often to get trains back on its right time)

Also some terminuses are where the train drivers booking on/off points are, so it is advantageous to send as many trains to that terminus, to offer greater flexibility for drivers in terms of reliefs and more efficient use of drivers hours. I think Upminster is one such station, I know Morden is another station.

Well that’s all the theory anyway! But I’d say it’s generally the case most of the line managers want end to end running as described above wherever possible, through experience more than anything else. And of course that is sometimes not possible anyway. Due to factors like the run times, number of platforms available to reverse in terminus and so on…

Just to add to this Biolizard, schedules ARE designed as much as they can be with this in mind...on pretty much all timetables - you catch a train from the outer reachs of London, and it takes you into Central London anyway. So when you ask why schedules don't do this - well it happens already!

The debate is more about where it terminates on the other side of Central London and the reasoning behind various choices. And if you can kill two birds with one stone - passengers who want to travel to Central London on one train, and passengers who travel between both ends of the line without changing, remaining on the same train - well why not?

Most timetable patterns are arranged to allow nearly every train to pass a major depot oon route for changeover purposes.

To get them on time for the next trip.

Well there we have all sides of the picture don't we? All are valid and acceptable. However a lot of 'out-of-town' reversing point are under used in the off peak.. Take Dagenham East for instance. It is far easier [ and quicker] to reverse a train there than via Barking Sidings. I know there would be an extra 10 minutes running time each way to/from DE but that would be partially off set by the fact that a Driver just has to get out of a cab one end and walk along the platform to the other instead of walking through the train and having to open/close all the communicating doors while so doing. Dagenham East also allows a driver to take a PNR if needed instead of calling a out a spare man en route to do the same. Yes I know there are crafty gits who call out a spare for a PNR just to get out of work but they have existed since trains ran. The trouble with the District is that it is unbalanced. I.E. There are too many reversing points at the east end of the line and not enough in the west. My point is that all these reversing places were provided to increase the flexibility of railway working along with reformations and such. This seems to have been lost with modern day methods.

EDIT: Another thing that really b*ggers things up for Signal Controllers, Drivers and passengers is schedules office insistence on running services to differing headways then trying to fit them all together like a monkey puzzle. 10 minutes here, 8 minutes there, it does'nt make sense and results in trains standing still at various points waiting for a slot. A standing train is wasting money and time and it irks the customer no end. Lets get back to the 'common headway' principle of the past and make the thing more efficient and cost effective.

Dagenham East isn't a good design for reversing trains because a train departing from the bay platform has to cross the path of EB trains. This is likely to either delay the EB service, or delay the departure of the reversed train.

Oh piffle. With a 10 minute headway you have plenty of time to get in and out of the bay at DE. In my time you were going in just after a westbound on the main left. While in bay you got time to go to the toilet and change ends and then when the Bay reverser behind you was due you were signalled out. However I do acknowledge that in those days the headway to Upminster was 15 minutes so too was the Dagenham service. If they went back to the 2.5 minute headway off peak with 7.5 min Circle/H&C and 15 min branch service some of the delays now present would be eliminated. It would also remove the ridiculous amount of time circles get to go round. 52 minutes is far, far too long an allownce.

Can you provide actual examples of where this does happen on the network? I’d like to know which sections of the underground you are referring to. Like for example where do we have a 10 minute headway here and 8 minute headway there, as you point to?

You portray the schedules office as INSISTING on running services to different headways that are not compatible with each other. But its not as simple or casual as that. It would be much less difficult work for us if they just compiled train timetables with compatible headways – and sometimes they can do just that.

But there are other more complexities and factors that need to be considered when compiling timetables. The line managers would have specific requests and ideas, and S&SD would have their own ideas about what level of service should be provided on certain sections of the line according to their demand data. And then there’s the number of trains available that influence what levels of service, and hence headways, that can be run. And there are many more members of staff that need to be consulted. Its like a no win job!

The biggest challenge which I think should be mentioned is where you have different of lines using the same sections of track.

The common headway principle you mention – well if that was applied, then the Picc, Met, Circle, H+C, District lines would all need to work to a standardised headway – which during peak hours, is impossible as each line has different remits for providing rush hour services on these shared sections of lines. And don’t forget that Network Rail is thrown into the mix as well – meaning that its impossible to schedule a uniform headway throughout the whole of some lines.

Regarding the original question -
There is nothing wrong with the current off peak District service.

We have [(XX)= usaul late running reverse point]
Upminster - Wimbledon (Parson's Green)
Upminster - Richmond (Dagenham East)
Tower Hill - Ealing (Mansion House)
Edgware Road - Wimbledon (Putney Bridge)
High Street Kensington - Olympia (usually waits time at HST)

This gives a balance of early turning points at different points on the line, and allows some sort of service to all destinations.

The other possible reversing points are Barking bay, Plaistow & Whitechapel - which are all used by the Hammersmith & City line. There are other places such as East Ham, for example, where a train can be reversed - but not in the context of the conversation we're having here.

Here's something Pete UXB wrote a while ago, which fit's in nicely here I think:

It is true to say that each line has differing frequencies, but because of the 17 minute thing - at least on paper - it works.

17 minutes? Where's the logic in 17 minutes? It does not fit into an hour, 2 hours or multiples of hours. Are schedules running to a different timescale than human beings? I have an old book which gives the unalterable criteria for good timetabling and it's not 17 minutes. 12 minute headways may be more acceptable

And of course a Met (sorry H&C) every 8 mins to Whitechapel with every other extended to Barking off peak, meaning an 8 or 16 minute headway into a District headway of 10 mins. That's always puzzled me too.

The reason that happens is to strengthen layovers to increase robustness from late running. If all the H+Cs terminated at Whitechapel, then you’d have to reverse in advance of the 8 min frequency. Sending alternate trains to Barking, means that the Whitechapel has plenty of time to reverse before the next Whitechapel reverser arrives.

The alternative is to reduce the H+C service to every 10 minutes, which would certainly ease pressure on C stock. But the line remit based on passenger data I guess, demand that 8 minute H+Cs are scheduled.

Try and think of it in terms of Trains Per Hour (TPH) – what Colin has posted up is how 28 TPH is run on the sections. That is several trains from a range of destinations – something that is demanded by passengers. And all of those trains are crowded to capacity. You advise the example of repeating the pattern of service every 12 minutes rather than every 17 minutes. This means that you would need to change the level of services from those range of destinations, to fit into that 12 minute cycle.

But there are only a certain number of trains available to make more frequent trips from say one branch section into the City – and you can’t reduce a branch section frequency to that which fits in nicely with trunk headways – coz that’s reducing the number of TPHs, which are needed for the high numbers of people.