To suggest that there's an added risk of fatigue-related accidents on Boxing Day rather questions the professionalism of rail staff, and I would be amazed if it were the case - indeed, if it were the case, then we would have other much mor serious problems. Where there
may be an issue is in finding staff ready, willing and able to provide the services.
On engineering works, I would classify them as
cause and
effect of the shutdown. Those which are causal may include (for example) a relay of the tracks between London Bridge and Borough Market Junction, where there's simply no easy alternative to a complete shutdown for a period. Those whihc are effectual are works that are done at this time of year because it's a convenient stoppage when actually they could be done at another time - for example, the lifting into place of a replacement bridge span over a two day period which could, I suspect, have equally been doen when the line was shut for 2 days for resignalling a month or so back.
At the current time, though, the idea of rescheduling 'effect works' that are already on the calendar for 26th December 2016 probaly wouldn't be a good one - yet another reason to be given for late deliveries and cost overruns on huge projects; pragmatically we should accept it for another handful of years, and look to having a Boxing Day service as a 'deliverable' as electrification and stock cascades come through?
I'm in little doubt in my mind that there would, in the current age, be good traffic levels on a 26th December service - though it might be variable year upon year. This year, it was a Saturday - next year, a Monday ... in 2017, a Tuesday and - although the days of the week are overwhelmed by the tidal wave of Christmas around this time, there is some effect and othere differing waves around the edges as the normal 7 day cycle changes to a few exceptional days and back to the cycle in the new year.
This modern expectation of traffic on 26th December is in contrast to what was observed at around the time of the removal of the previous service some 40 years ago. And one of the reasons given in those days was the cost of staffing all the signal boxes along the way for just a handful of trains. That excuse is largely gone; in our area, I think we still have signal boxes along the main line in Cornwall - for perhaps a couple of years. Crediton might need staffing to cover a small area? And there could be a few more.
What
would be the traffic? People visiting relatives. Sales shopping. Sporting events. I only have the knowledge to take a local(ish) look at this and how it would work out; I would see TransWilts as running with an 08:00 Westbury to Swindon, with the train then heading down to Salisbury, back to Westbury and cycling at 11:00, 14:00, 17:00 and 20:00, back into Westbury to close play arriving at 21:40 from Swindon.
As I write, noting this morning's restart issues:
07:11 Gatwick Airport to Reading due 08:35
This train will be started from Redhill.
This train will no longer call at Gatwick Airport.
This is due to a shortage of train drivers.
08:26 Weston-Super-Mare to London Waterloo due 11:54
This train will be started from Bristol Temple Meads.
This train will no longer call at Weston-super-Mare, Worle, Yatton and Nailsea & Backwell.
This is due to this train being late from the depot.
Somewhere, I noted a comment that we've done this subject to death over the years (paraphrasing). To some extend we have, because it will go away and be off the headlines come the New Year - just back for the short term next Christmas, when really it needs a long term view to be taken. But it's not exeactly the same subject each time - things are changing / melding and we may not see a sudden change, but yet nothing has turned at Chiltern into a London to Bicester serivce, to a London to Oxford Parkway service, and perhaps next year to a London to Oxford service. And as electrification goes live out from Paddington, perhaps the electic lines, and a few connections, will come along too.