That stuff about Bristol illustrates that there will be a massive increase in capacity - but the demand that capacity will meet is not evenly distributed across all services, all day, so you do not need nine-car trains all day - same as it is now, but the only stock FGW has at the moment is full-length HSTs, so that's what they use.
Well they don't just have full length HSTs, but I will accept the argument so I can move on.
The 'massive increase in capacity' will likely be swamped within a few years, so there is nothing to expect that this capacity will even be
sufficient a few years from now.
The lessons of history show that reduction in journey times produce enormous increases in patronage, so DfT is probably conservative in the long run.
Turning back early? A meaningless expression. What is the point of running nine coaches all the way west of Oxford or Swindon, if there are fewer than 300 people on that leg of the journey, whereas demand east of those points is higher, so by splitting and joining sets you balance the capacity against the demand on each section of the journey?
My argument begins with the fact that a train that isn't splitting can have a far smaller dwell time than one that is splitting at that location.
Even with Sprinters it takes something like ten minutes with the appropriate padding to split a train.
I certainly wouldn't want to a joining in anything less than that with how troublesome modern trains are.
You also really have to have someone walk through the second unit and make sure everyone is off. So ten minutes each way.
The second point is that 'balancing capacity against demand' is often pursued as if it were an end in and of itself, the actual objective is ensuring that
farebox recovery is as high as it is possible for it to be.
Let us take the Oxford example, a train will cover the first leg of the journey in roughly an hour, and then the onward journey to Worcester (according to the Franchise Replacement spec for consulation this would be the normal terminating point - no idea if that has changed since) in roughly 1hr20 more.
This means that (with a customary 20 minute turnaround and cleaning time) the cycle time for each of those diagrams is something approaching 2:40 in the first place and 5:20 in the second case.
If we take the twin 5-car option then will require 6 sets to main 1tph on whole diagram and 3 more for the strengthening section.
Taking us to 45 vehicles.
If we take an alternative 9-car option which would offer comparable numbers of seats that is six sets or 54 vehicles.
So all running the through train has cost us is a set of 9 vehicles. Which requires a hire charge (all up) [assuming that figure of £180,000/month for 5 is still accurate] of roughly £324,000 a month.
Since the costs of running a train scale incredibly non linearly with length, the cost of running a train that you already have boils down to a marginal maintenence charge and the cost of fuel. This would be accounted for in the costs of ~£324,000/month alongside the capital costs.
Now we have to account for the operational savings of the unit train:
It is likely that each 5-car set would need a catering crew of roughly 3 staff members (one first class host and one in standard - indeed FHT seem to run with this settup), a quick search on First Group's website seems to put the starting salary of a catering team member at roughly ~£16,500; including Employer NI it probably puts it up in the region of ~£17,000.
So one three-person crew requires a cost of ~£51,000/yr, you will then need roughly three crews to be able to assume one sets diagrams are always covered, taking us to ~£153,000/yr - some £12,750 per month.
Having a split set will require three additional catering/CS crews, taking us to £38,250/month in costs.
That reduces the additional costs of through trains to roughly £285,750/month.
Next, we have additional staff at Oxford, it is likely you will require at least one shunter to handle movements with the second sets, since a set just detached from a Country-bound service will be on the wrong side of the station to be coupled to a City-bound one, and using a shunter reduces the number of conflicting movements from two (crossing ove the station twice) to one (the shunt move).
I have no idea what a shunter is paid, but the starting salary for a driver (and it might be a driver for ease of diagramming?) would cost roughly £45,000yr, which after multiplying by three to get 100% coverage gives us roughly £11,250 a month.
That reduces the costs of the through trains to roughly £274,500/month.
This sounds like a lot, but is is really?
There will be approximately 16 trains running between Worcester and London each day, taking us to ~£600 in additional costs per train.
That gets us 4 vehicles with a seating capacity of 312 seats.
So the marginal cost of providing the additional 312 seats is roughly £2 each
return.
Considering that the fare between Worcester and Oxford is currently £29.90 [return] that implies that if you could fill the trains with a fare of roughly £16 return you would be better off, and even if you can't you would not be that much worse off.
If you have unlimited capacity filling it almost certainly isn't hard since you can slash fares as required.
And let's have a coherent, rational answer for once please. Not yet another variation on the lines of 'they just should'.
How's that?
What drastically increased demands for infrastructure?
Well you now have crossovers at the splitting station which you didn't need before, you now have to have a train turning around at Oxford, blocking a platform, that you didn't have before and some other things that are rather hard to quantify.
Platform space has to be paid for.
What drastically increased demands for staff? Or do you also nurse this fantasy that every set will have a small army of staff, including their very own guard, on board all the time. They won't.
Well unless you are going to deprive all passengers going to Oxford of staffing at all by throwing them into the rear set, you will have to provide a catering crew and first class host.
They can't just pass through the noses of the sets to reach the rear one.
And if you aren't providing such space, why will all 5-car sets have kitchens aboard?
With that sort of attitude you will likely only be providing the most basic of fare to Worcester passengers which means you can put a FHT style buffet in and pack in more seats.
The fact that all the trains seem to be set up for Travelling Chefs in the DfT's projections sort of implies that all sets will require at least some form of catering staff.
(And three between 315 seats is hardly an 'army' - I even assumed they are all the lowest paid position possible).
If you want more capacity you, er, buy more coaches, which is what has happened with Pendolinos or 333s in West Yorkshire - unless those cases are in some way different to buying more IEP coaches.
The only way to buying more IEP coaches would be to convert some 5-car sets into 9/10 car sets, because you can't double up 6-car formations without requiring hideously expensive platform extensions.
This leaves you with a surplus of sets
with nowhere to put them.
ECML will be swimming in IEPs, GWML will be swimming in IEPs. MML and XC will be the only place to send them really and they will not be able to take as many as HS2 will release when it rips the heart out of the ECML IC market roughly 10-15 years into the life of the sets. [It now appears HS2 will be accelerated to score political points, so you don't even have until 2033].
333s were not pressing up against the platform length restrictions, IEPs really are.
You don't need platform extensions, the trains will have SDO and operating paired up will stop with part of each set on shorter platforms, whatever length the sets are, as was pointed out some months ago.
Put a pair of 6-car IEPs coupled into an 12-car formation into Paddington, tell me how that goes.
(That will put it 20m beyond the end of pretty much any of the platforms at Paddington, and rather further for most of them).
I do love all this stuff about coupling short trains. What is it that you and Rhydgaled know that train operators the world over have clearly failed to grasp, so they persist in doing it, day in, day out?
And yet Thameslink has abandoned coupling short trains in favour of unit trains?
It is nowhere near as clear cut as you seem to believe.
Noone in Britain has ever designed such IC trainsets to be operated in this manner, even Virgin only resorted to it out of desperation when Operation Princess fell to pieces.
