Great Western Coffee Shop

February's Today's Railway Europe has Februaries Today?s Railways Europe has a very long article (10 pages) on the German Governments  plans for Deutschland Takt 2030, a clock face timetable for all  Germany y mostly based on 30 minute intervals for particular services and regular connections between services.  The article outlines the steps necessary to achieve this goal.  This consists a long list of infrastructure improvements from New High Speed lines some electrification extra-long freight loops etc.
However,  to my mind the most telling principle behind the whole plan is ?first the timetable then the infrastructure planning.
This seems to me a way of planning for rail enhancements in an integrated way rather than looking at individual enhancements in isolation.
Taking Melksham how frequent a service is required from where to where how many freight paths are required etc. Possible services could include.

Shuttle Westbury to Chippenham/Swindon.  30 minute service
So a possible loop/s on line, improvements to Thingley and Bradford Jns and, fourth platform at Westbury,  third platform at Chippenham

From Southampton to Oxford? Hourly  takes one of the shuttle paths.
All of the above but capacity enhancements in Southampton area  Swindon to Oxford.plus  any where else.

Bristol to London via Bradford on Avon stopping service takes 2nd shuttle path.
All of the above with possible exception of Westbury 4th platform. Reinstatement of Bradford North Curve, new station on outskirts of Trowbridge.

This approach needs to be carried for all the stations served or potentially served from Melksham this could reinforce enhancements already planned and suggest further   enhancements. The plan will require great deal of iteration

Having worked out the future timetable and enhancements needed to implement it you can add the clincher to this approach. Electrification
 
At a minimum electrification from Chippenham to Bristol, Didcot to Oxford,  Newbury to Westbury and Bath plus Chippenham to Trowvbridge.  The electrification to take into account the other enhancements so that they can be incorporated in the planning of the electrification.
Implementing Takt is a complicated process and would take many years but the integrated approach of timetable first would I am sure provide a better railway

That sounds like the same concept as Network Rail's "Improving Connectivity" - as examined in their report in 2014 and consultation. I can't see that report available any more, and I don't think it ever led to much at all. It was discussed in this thread (http://www.firstgreatwestern.info/coffeeshop/index.php?topic=15136.0).

Working out the services you want, then working out where you don't have the capacity to run them and looking to add that capacity make sense and - err - has been done in practice for a while, and is the way a lot of the way future requirements are looked at.

Want a half hourly service Exeter to Yeovil Junction, with alternate services onward to Salisbury and onward (when they're a GWR diversion) to Castle Cary?   Take a look at where they'll be passing each other and that's where you need your loops.    But, yet, not as simple as that.  There are places that a loop will be expensive, impossible, delay trains by adding a stop away from stations, and you'll slide the trains up and down, perhaps change stopping patterns, etc, to make best use of existing loops, platforms, etc - there's no point in spending for the sake of a train running 5 minute later throughout provided that doesn't mean adding a train to the cycle or missing a connection.

Another benefit ... clock face = same each hour = you will make really good of the facilities and not have lots of them each used only occasionally.  And another benefit - where there's a good connection it will be a good connection all day and (please!) if you can get it to work as well in the other direction you have a very marketable product on the flow.

Now ... I was actually looking at "Here's the service how do we do it" for the Wiltshire, Somerset and Weymouth.

Hourly trains - six in the cycle:
Weymouth passes trains at ...
* Maiden Newton (existing loop)
* Around Spakford (new dynamic loop, perhaps from Pen Mill - see also services via Yeovil Junction)
* Between Frome and Westbury (on main double line)
... time at Westbury to connect into Bath and Bristol and from Southampton area train (add 4th platform?)
* Between Melksham and Thingley (new dynamic loop most of the way)
* Shrivenham/Grove (on main double line)
... on way to Oxford
With the train having a turn-around of just over 12 minutes at each end

Hourly trains - South Wales or Bristol to Solent, for those Westbury connections (broadly current stops in our area plus Dilton Marsh and Wilton)

Every 30 minutes Bristol Temple Meads to Bradford-on-Avon, alternating on to Chippenham or Swindon and Westbury or Warminster (this is where the "slide it" exercise comes in and perhaps we're looking far, far ahead) or Radstock.

But is starts ... not even from the service but from:
* the passenger journey which gives
* the service pattern which gives
* infrastructure needs
or as Network Rail are saying - "putting the passenger first".

Lets hope the money follows.

There was a discussion of why NR's attempt at doing this ground to a halt at such an early stage here, at Railfuture (https://www.railfuture.org.uk/article1601-Improving-Connectivity). This refers back to the original Swiss Taktfarhplan.