Great Western Coffee Shop

"Connectivity" has appeared as a buzz-word in HS2 documents, and in the new route studies, meaning something like serving other stations as well when you are discussing a main line. Network Rail have also undertaken a study (like a Network RUS) into a more concrete proposal they call "improving connectivity". This may be familiar as German-style clockface timetabling (though they identify it as Swiss, since the Swiss have taken it further, and it has been mentioned in this forum as being typical of the Netherlands).

In essence, it means this: when designing services linking between main lines, top priority in given to making them connect with the main line and each other, speed and direct trains are less important. Ideally, all connecting trains turn up simultaneously with the mainline trains. They also show how cross-platform changes can make three-ended routes work better (though I can't see how that would work with four ends).

Most of the study is an exercise in applying these principles to the Anglia Route, and they claim impressive reductions in travel time between almost all station pairs. Less surprisingly, it makes travel between more such station pairs by train a feasible option.

Unfortunately, they do not make any attempt at applying it elsewhere, nor say whether the Anglia Route is unusually suitable for it. So how well would it work for the Western Route?

The consultation is here (http://www.networkrail.co.uk/publications/long-term-planning-process/improving-connectivity/?cd=1), and the report itself is here (http://www.networkrail.co.uk/long-term-planning-process/improving-connectivity.pdf). Responses must be in by 31 January 2015.

(This is supposedly part of the Long Term Planning Process, though I have yet to find a route through the menus that leads to it there or anywhere else.)

It certainly seems a good idea to study this but I do wonder how many stations it would be possible to arrange cross platform interchange as in Germany.  Reading maybe when the Viaduct and the Festival line are fully operational. Cross Country to West Country Platorm 3 to 7

i still have my video of an ICE and IC arriving simultaneously either side of the platform in Mannheim.
No defensive driving in fact the IC overtakes the ICE runnig down the platform.

Westbury ... especially once we get an extra track along the unused platform  ;) .   

Salisbury ...

Parson Street ...  :P

I think it's fair to say this sort of thing already exists at Salisbury

Maybe ... or maybe it needs a bit of tuning there?

I'm an occasional traveller from West Wilts to lots of places including Farnborough (Hants), Basingstoke, and the South Bank in London for which Waterloo is a perfect arrival.

Arrival of the hourly service from Cardiff into Salisbury is at xx:29, which is just after the departure of the xx:21 Exeter to London Express.    Arrival of the hourly express from London is at xx:42, which is just after the departure at xx:40 of the Cardiff train.    We know there's a significant requirement for this flow, since there's a handful of trains from London that detach a Bristol portion at Salisbury.

Where it does work at Salisbury is for passengers arriving from Portsmouth and headed to Exeter and vice versa.

To enable cross platform Westbury London and vice versa connections at Salisbury the station would realy need to be two island platforms one Up one Down. Then the Cardiff to Portsmouth and Exeter London and and in the Down could be in the station at the same time.

Which was why I suggested that there are very few junction staions where this can happen

Agreed Parsons Street is an interesting one but like Westbury needs the 4th platform to be reinstated.

Then you could have a connection between Cross Country and or the IEP  and the Taunton Bristol stopper. It would mean two stops close together for the fast but but would keep changing passengers out of Temple Meads.  Although the passenger facilities might have to be improved
at Parsons Street.

I was thinking also of the connection with the new passenger services on the reopened branch line to Portishead.  ;) :D ;D

Then there's Avonmouth (for Henbury) and Stapleton Rd (for all sorts of places)...

Can't see a huge benefit here - not enough to stop express trains there - even after Portishead reopening.  A short trip into and out of Temple Meads would be more efficient and would give vastly more connecting opportunities.  

You will no doubt note that my suggestion of Parson Street was clearly marked  :P ...

CfN.  ;D

Interesting twist on cross platform changes is Newton Abbot.  At night, after the last Paignton train has gone, the down trains are usually crossed over to Platform 3 which offers no step access to the car park and taxis.

Yes, I see what you're saying - but I suspect that in many cases (Salisbury may be an example) that not all the connections are needed at the same.

Let's take a look:

Exchange from Waterloo to Cardiff AND from Portsmouth to Exeter (and vice versa) - useful
Exchange from Waterloo to Portsmouth AND from Cardiff to Exeter (and vice versa) - limited use
Exchange from any line back into a train returning the same way - pointless

Let me qualify "limited use".

Passengers coming up from Portsmouth and Southampton to Basingstoke and London will go via Wincehester
Passengers from Westbury and north thereof, to Yeovil, Exeter and Westwards will go via Castle Cary
There would be some double-back flows with the extra connections, but none strikes me a large.
- Romsey to Andover
- Warminster to Axminster
- Grateley to Southampton
- Bath to Honiton
(please let me know if I've missed a big 'un).

I think everyone is rather missing the point. The proposition is that you could scrap the current timetable completely, adopt 'connectivity' in this sense as top priority, and then design a route network and timetable on that basis from scratch. The approach they describe would produce a pattern of 1 and 2 tph everywhere, with good connections (ideally all cross-platform). That ought to give better connectivity, obviously, but what would be the penalties?

The study claims that you can then overlay extra services: fast trains on the main lines, some of which also interconnect with slower ones; extra stopping trains for 4 tph 'metro' service, and peak commuter trains too. They imply that no capacity is lost, but that sound unikely (I imagine a lot of current timetablers will be sceptics). There will have to be some timetable padding, which we know happens in Switzerland and Germany. What about where the current timetable has to be carefully tweaked to mix fast and stopping trains on the same line can that capacity be maintained?

They do say that extra infrastructure is needed, most obviously those island platforms to change trains across, and line speed improvements where the clockface timings are just missed. However, they do say that some works can be omitted, such as doubling single track near stations (all trains arrive in a 'flight' and then leave likewise): they give the example of Trowse Swing Bridge at Norwich.

At Ely North, which is their main worked example, they want a new station, to make it easier for this to be a key interchange. While it's not made explicit, they would move the junction itself northwards, thus closing two level crossings. But this time they also ask for grade separation of the Peterborough line at this junction, which is not in the Anglia Route Study. I presume this is mainly for freight, but surely an avoiding line (maybe using the West Curve) and a choice of crossovers north or south of Ely North ought to be enough.

Very interesting stuving.

If they are really looking at a fairly radical change to service patterns, plus new stations and extra platforms to improve connectivity then it rather makes a nonsense of HS2 with it's 4 separate terminal stations. I suppose they could bring a train from the North into Curzon Street with a cross platform change into a London train. But that seems to defeat the objectives of HS2 of givng high speed journies to London from the North.

I suppose New Crewe is HS2's answer, I'm not sure Toton or Meadowhall meet the criteria.

New Reading with the flyunder from the Southern and the Viaduct and Festival lnes to the West with most lines bi-directional  could with clever timetabling/platforming give some cross platform interchanges.  Although like Salisbury you'd have to work out what flows to cater for. Heathrow (if West link is opened) and Gatwick from anywhere West. TV Crossrail stations to FGW and Cross Country.

(My italics!)

Bit of a sweeping generalisation!

I mentioned Avonmouth in the context of the Henbury Loop: There is currently a tension between those who wish to see a loop service, connecting Henbury to Avonmouth, and those who see that as impractical because the Severn Beach line is already running to capacity. When I read the consultation document, it immediately struck me that a cross-platform interchange at Avonmouth could allow something similar to the current level of service to be retained to Severn Beach, whilst providing a good compromise for those who'd like to see a loop service.

What we're talking about here is a slight move away from the hub-and-spoke railway that Beeching bequeathed us, to something more like a point-to-point service pattern. From my provincial perspective that can only be a good thing; too many journeys currently require time consuming (and often expensive) transits into and out of major urban centres.

Not at all. If you look at HS2 as all an overlay express service, it might not need to take part in this connectivity stuff at all. Certainly you could argue that it should, and connections with local trains are important, but that would only involve a few HS2 services.

This is something implicit in this connectivity concept, though not spelled out. If there are 2 tph (typically) on the main lines that do connect, and the rest are express overlay, can that be reflected in their stopping patterns and the way they are sold? That is, drop the idea that all trains can be expected to connect with others, and label them as something like "connecting trains" and "direct trains"? One of the inherent trade-offs of the concept is a (potential) loss of point-point running speed, plus needing to change trains, which get offset by shorter connection times.

Mind you, I've always thought this fixation on HS2's termini as missing the point too. If you think of it as trying to add a more express capacity as a new pair of faster-than-fast lines, and then conclude that it is impractical to put them next to the old line, you need a new route but could still could try to expand existing stations to take all the new trains. But that's going to be very difficult (read expensive), and a lot of this capacity is used by people who really do want to go between London/Birmingham/Manchester etc. and jump in a tube/tram/taxi. On those grounds it makes sense to run trains between new city stations for them. By all means complain about poor connectivity, but don't ask for all those direct trains to be added back into existing, hard-to-expand, stations.

As for Reading, I suspect there is too much going on there for the concept to work well. So far, proposals for new through services on the relief lines (e.g. Gatwick-Basingstoke) have been motivated by lack of capacity in the new platforms 12-15 (already!). I know the rubric for the study was "start with a blank sheet of paper, see what timetable it leads to, and then work out what new infrastructure that needs", but "expand Reading relief side" is probably the wrong answer for the foreseeable future. But it may still be possible to make some improvements within the existing capacity. You really do need to work through the process to see!

Note that I am not trying to summarise the whole report so you need not read it.

Just been re-reading the December Modern Railways article on Modernising the Midland mailine. Leicester would work well as connection point as you have two crossing flows East Anglia to the Midlands and Sheffield/Nottingham to London and the station has two island platforms.

However the plan for Derby seems to preclude any crossplatform interchange as the lines from Brimingham use one side and the lines from London use the other.

Just musing how difficult it is modernise 175+ year old infrastructure.

Yes the cross platform interchange at Derby (and probably elsewhere) creates conflicts and reduces capacity unless you provide grade separation. 

Surely the problem with any connectivity system is that it requires not only timetabling to eliminate long waits, but timekeeping. If your train is late in a no-change journey, you're just late by that much. If you have to change, it can be much worse.

I dare say this is also missing the point, but I don't know!

That's not a problem, surely? As long as you are Swiss, of course.

Another point is that you need some spare capacity to do it. If squeezing more trains or people through the same infrastructure is the priority, this stuff becomes very hard or impossible.

It depends.   Some through journeys get sliced, with a new train taking over from the old one at an intermediate station.   Example -  05:19 on Monday from Gloucester to Southampton, due there at 08:09.   Terminated at Swindon, restarted at Westbury.   Second example - Weymouth to Great Malvern services, sometimes terminates at Bristol Temple Meads when running late, with another train having left Bristol a bit earlier (i.e. right time).

In the first example, had you got on that 05:19 to travel to Southampton, I suspect you would have been 2 or 3 hours late getting to destination!

Indeed, if you are aiming to provide good connections by eliminating long waits, near-perfect timekeeping is essential. Personally, I don't think it is reasonable to expect the level of punctuality that would be required to make that work. The public performance measure for long-distance services is up to 10 minutes late at the final station. If that applied at every station where interchange is possible, not just the final stop, and that 10-minute target was met sufficiently often you could timetable connections at 15 minutes. That is probably much more acheivable than the 90-odd% right-time performance which would probably be needed to make five-minute connection times work across the board.

However, one problem with a 15 minute connection time that seems to be overlooked is that it is plenty of time to get uncomfortable while standing around waiting and not moving much. Five basic human survival needs:
    1. Oxygen
    2. Water
    3. Food
    4. Shelter
    5. Sleep
Most of these aren't particularly relevant to public transport connections, but shelter is, and the rail-industry-standard 'enlarged bus shelter' isn't much better than a bus shelter (which is utterly hopeless) at sheltering you from the wind. They also are not really big enough to fit a decent ammount of useable seating either.

Agreed the 'bus shelters' are pretty useless, but they tend to be found only on relatively minor stations, rather than ones where connections are more likely to be made. Though Parson St was mentioned upthread... (haven't been there myself but I know there's no 'proper building' at Bedminster.) Then again, I remember waiting at Pewsey for a train to Westbury a few Sundays ago. A 'proper station' with a good old Victorian waiting room. :) Locked.  :( Train was 34 minutes late  :((but I just made my connection :D).

Yes, it's not only being on time that's important, being there at all also counts!

'Connections' (or, at least, 'interchanges') can be made at quite a number of 'relatively minor' stations. Whitland for example is in a relatively rural area but is suituated at the junction to the Pembroke Dock line (so passengers from the 'main line' from Fishguard may want to interchange there) and is also the only station west of Llanelli at at which all Pembrokeshire trains call (meaning for example that passengers wanting to get to Fishguard from Carmarthen may have to change at Whitland). The 'bus shelter' type thing at Whitland is an old one that is, if anything, even worse the current rail industry standard ones. Also, it might not be a train you want to 'connect' with, rural stations with bus 'connections' rarely have a proper station building.

The limited opening hours of the facilities which do exist is a problem too. Aberystwyth, following the recent refurbishment, now has a waiting room but the building always used to be locked when the ticket office shuts at 17:30 (I assume that hasn't changed).