Rail regulator to investigate rail work delay chaos

[whole diary]

[ChrisB]

Yep, just had a Journeycheck alert for its cancellation further towards the Thames Valley!

[bobm]

Ironically, although it is the first to leave Plymouth (05:09), it is not the first timetabled to arrive in London.  I was on the 05:30 yesterday and it was very sparsely populated.  The 05:09 goes via Bristol and the 05:30 via Melksham, putting it 20 minutes in front.

[TaplowGreen]

Ironically, although it is the first to leave Plymouth (05:09), it is not the first timetabled to arrive in London.  I was on the 05:30 yesterday and it was very sparsely populated.  The 05:09 goes via Bristol and the 05:30 via Melksham, putting it 20 minutes in front.

V busy on a regular Monday morning though!

[Jason]

There was a full page apology in The Times this morning for the events of the 27th

[Chris from Nailsea]

From the BBC^» (British Broadcasting Corporation - home page):

Rail disruption caused by bad planning, Network Rail says

Equipment failure and bad planning for delays caused the disruption to rail travellers over Christmas, a report by Network Rail has said.

Thousands of passengers had their journeys disrupted when work near King's Cross took longer than planned.

Trains were switched to Finsbury Park, a much smaller station.

"Not enough was done" in planning the switch, said the report, which included an apology from Mark Carne, Network Rail's chief executive.

Work near London's Paddington station finished on time, but a safety check took eight hours longer than planned.

"A number of things went wrong in these two instances. In addition it is clear that our project back-up plans and the train service plans should have done a much better job in protecting the travelling public from our engineering problems," Mr Carne said.

"Over Christmas, we undertook the biggest programme of engineering and investment work ever, on train lines across the country. Ninety-nine per cent went to plan but in the case of King's Cross and Paddington we let passengers down.

"I sincerely apologise for the disruption over the festive period and we are determined to learn the lessons so that we can continue to make the improvements the travelling public deserve."

Work should have been finished allowing King's Cross and Paddington to open on 27 December. But King's Cross had to close all day while Paddington only opened in the early afternoon.

Network Rail said that, by 10:30 on 27 December, Finsbury Park had become so crowded that incoming passengers could not get off their trains. Rail travellers had to queue for as much as three hours to get into the station.

The row over the delays led to Mr Carne renouncing his annual bonus, which could have been as much as ^135,000.

"On the basis of this report and the events on the day the industry has a long way to go to restore trust in how it handles these events," said Anthony Smith, chief executive of the rail customer watchdog Passenger Focus.

Network Rail says it will review how it plans for over-runs and will "provide better information to passengers".

The Network Rail report is no longer available at http://www.networkrailmediacentre.co.uk/Gallery/Christmas-2014-passenger-disruption-report-2208.aspx




Edit note: See posts below for explanation. CfN.

[Electric train]

This is worth reading, a well presented report with I feel sufficient detail to inform the layman and those who know how railway project work to understand where things went astray and where the lessons need to be learnt.

It is I feel quite frank and open.

[stuving]

I was interested to note this in Mark Carne's Foreword:

The timing of these events, over the Christmas holidays, has also made us question traditional
thinking. While our industry has historically seen the ^quieter times^ of railway use as the natural
time to carry out essential project works, I believe that it is appropriate to challenge some of this
thinking. Passengers who use the railway during holidays to connect with friends and family also
deserve reliable and predictable services. That is the thought behind a second review that we have
already announced, following discussions with the Secretary of State. This will be independently led
and will report through the industry^s Rail Delivery Group.

As I said earlier, it's the commuter lines that are less busy, so quite a lot needs to be done to take that spare capacity and run a suitable, robust, service for the holiday travellers - and suitable may mean more trains than normal, on some days.

[ChrisB]

I think he's referring to doing more work around the year, less at Christmas. Not necessarily persuading the TOCs^▸ (Train Operating Company) to run more services, which they can't really do

[Electric train]

Compressing all the work into a few major weekends (Christmas, Easter, August and to a lesser extent the May Bank Holidays) skews the market, resources are getting booked further and further out often over booked.   

The pain will be more weekends at other times of the year, perhaps with shutdowns commencing at 22:00 Friday through to 04:00 Monday to get some of these big jobs done, that's not to say the traditional holiday shutdowns will go. 

[stuving]

I think he's referring to doing more work around the year, less at Christmas. Not necessarily persuading the TOCs^▸ (Train Operating Company) to run more services, which they can't really do

I hope he (and the report's authors) don't prejudge the conclusions.

Today's report does say that closing Euston and King's Cross at once was a plan with an obvious flaw in it, when the overrun happened. In addition, piling so many big projects into those two days meant using almost every driver there was, so there were none spare, and closing the railway in general hampered train movements. And they can't really say that wasn't predictable.

There is also mention of a recovery plan that was rejected - for want of drivers, mainly - pulling all the engineering trains out and using the two intact tracks to run the service by day, and coming back to finish the work by night. I hope that gets serious consideration as a way of using the full commuter lull (ten days), or maybe August (as was suggested in another thread), for major works. And of course just doing more in weekends, if nothing else proves workable.

Given the excess cost of trying to do everything at once in holiday weekends, or in tiny slices overnight, all sorts of other approaches might be viable when you (or rather they) look closer.

[ChrisB]

I hope he (and the report's authors) don't prejudge the conclusions.

Not sure what you mean by "prejudge"?

This is the report & has its conclusions, which the authors & Mark Carne are commenting on. So surely it's already "judged", not being "prejudged"?

Page 18.....wow.

You have to wonder just how NR^» (Network Rail - home page) got themselves into this mess.

They sourced a load of cranes.

They also sourced a load of *new* "log grabs", which appear to be bits that go on to the end of a crane's jib in order to allow it to pick up scrap rail.  They even had a spare one, and an on-site fitter.

But they didn't test in advance that these log grabs actually worked on these jibs, and when it came to it the hydraulic joints couldn't be made pressure-tight and all the cranes started leaking fluid and losing pressure.
And that's game over: they were unable to clear scrap rail, and it all goes down hill from there.

Given that these cranes are road-rail vehicles, and the business end of a small crane's jib is hardly new technology, sensible engineers would do, you would think, a quick acceptance test and make sure that there hadn't been a "whoops, the fittings are the wrong size" cockup on ordering brand new, untested grabs, and perhaps bolt on in the yard and spin it around and check it works.

NR, however, didn't do any testing until Christmas day, by which time it was all too late.  The onsite fitter couldn't bodge the connectors, and from then on they were b*ggered: they were running late, so the engineering trains ran late, so the drivers timed out, so they didn't have spare drivers available, etc, etc.

However, as noted above, the project had been cascading train drivers throughout the day and, shortly after midnight, the supply of new drivers  to support this cascade, ran out. There was one remaining driver and five engineering trains still on-site and this was the point at which the project started to rapidly lose time. Whilst all the drivers involved were cooperative  and committed to completing the project, they reached their maximum shift duration limits, which for safety reasons cannot be exceeded.

Ten minutes' testing would have found that the hoses didn't fit the jibs in advance. Then the railways ran out of railway staff, and had no plan to obtain more.  They didn't have enough drivers to handle the workload when recovering from the contingency they'd built in.  Contingency which isn't resourced isn't
contingency.

With only one driver and five trains spread out over the length of the site, progress on site slowed to a crawl as trains were in the wrong place to support the planned pace of work. This was compounded by the mechanical failure of one of the ballast wagons which failed in such a way that it could not be moved for a number of hours.

The lessons learned are shameful.  Point 2 is just a disgrace:

Contractors will be required to test any new equipment in an off-the-railway environment before it is used on live railway work.

How can you call yourself an engineer and not have some idea that the live workplace, with massive time pressures, at time on Christmas Eve, is not the time to test out whether previously untried equipment works?  Seriously?   

And similarly point 5:

Engineering train crew and contingency at times of peak work will be treated with the same level of nationwide cross-project scrutiny and planning as other resources in short supply, such as signal testers and overhead line engineers."

Is the whole point of contingency planning that you evaluate all the things that might fail, and decide to reduce, mitigate, transfer or accept the risk?  Does it really take a genius, or bitter experience,
to realise that having people to drive the engineering trains is quite important, and without them you're b*ggered?




Edit note: Quote marks fixed, for clarity. CfN.

[Electric train]

I hope he (and the report's authors) don't prejudge the conclusions.

Not sure what you mean by "prejudge"?

This is the report & has its conclusions, which the authors & Mark Carne are commenting on. So surely it's already "judged", not being "prejudged"?

Page 18.....wow.

You have to wonder just how NR^» (Network Rail - home page) got themselves into this mess.

They sourced a load of cranes.

They also sourced a load of *new* "log grabs", which appear to be bits that go on to the end of a crane's jib in order to allow it to pick up scrap rail.  They even had a spare one, and an on-site fitter.

But they didn't test in advance that these log grabs actually worked on these jibs, and when it came to it the hydraulic joints couldn't be made pressure-tight and all the cranes started leaking fluid and losing pressure.
And that's game over: they were unable to clear scrap rail, and it all goes down hill from there.

Given that these cranes are road-rail vehicles, and the business end of a small crane's jib is hardly new technology, sensible engineers would do, you would think, a quick acceptance test and make sure that there hadn't been a "whoops, the fittings are the wrong size" cockup on ordering brand new, untested grabs, and perhaps bolt on in the yard and spin it around and check it works.

NR, however, didn't do any testing until Christmas day, by which time it was all too late.  The onsite fitter couldn't bodge the connectors, and from then on they were b*ggered: they were running late, so the engineering trains ran late, so the drivers timed out, so they didn't have spare drivers available, etc, etc.

However, as noted above, the project had been cascading train drivers throughout the day and, shortly after midnight, the supply of new drivers  to support this cascade, ran out. There was one remaining driver and five engineering trains still on-site and this was the point at which the project started to rapidly lose time. Whilst all the drivers involved were cooperative  and committed to completing the project, they reached their maximum shift duration limits, which for safety reasons cannot be exceeded.

Ten minutes' testing would have found that the hoses didn't fit the jibs in advance. Then the railways ran out of railway staff, and had no plan to obtain more.  They didn't have enough drivers to handle the workload when recovering from the contingency they'd built in.  Contingency which isn't resourced isn't
contingency.

With only one driver and five trains spread out over the length of the site, progress on site slowed to a crawl as trains were in the wrong place to support the planned pace of work. This was compounded
by the mechanical failure of one of the ballast wagons which failed in such a way that it could not be moved for a number of hours.

The lessons learned are shameful.  Point 2 is just a disgrace:

Contractors will be required to test any new equipment in an off-the-railway environment before it is used on live railway work.

How can you call yourself an engineer and not have some idea that the live workplace, with massive time pressures, at time on Christmas Eve, is not the time to test out whether previously untried equipment works?  Seriously?   

And similarly point 5:

Engineering train crew and contingency at times of peak work will be treated with the same level of nationwide cross-project scrutiny and planning as other resources in short supply, such as signal testers and overhead line engineers."

Is the whole point of contingency planning that you evaluate all the things that might fail, and decide to reduce, mitigate, transfer or accept the risk?  Does it really take a genius, or bitter experience,
to realise that having people to drive the engineering trains is quite important, and without them you're b*ggered?

I cannot understand how they got to use these grabs with out product acceptance, I have been through pain worse than wisdom teeth recently trying to get a manipulator altered to put OLE^▸ (Overhead Line Equipment, more often "OHLE") stovepipes up.

There are some very well know "critical" resources OLE linesmen, S & T Tester in Charge, HV Distribution nominated persons, and yes train drivers.  We (the railways) are in state where a driver can turn up to an engineering train as say sorry mate not passed out on that traction or route or both!!!




Edit note: Quote marks fixed, for clarity. CfN.

[SandTEngineer]

NR^» (Network Rail - home page) has quietly changed the report since it was first issued (and apparently not told anybody: I wonder if the National Press have noticed?).  A number of people have pointed out various errors and discrepancies in the original, the KX possession Map Fig 5 being one of them.  This showed lines under possession that were closed over 50 years ago 

Personally I think its a very poor response.  It would have been better if NR had put its hands up and admit that it (not its contractors) c****d it up big time and honestly stated that it wanted to learn from the mistakes rather than keep repeating the mantra about 'building a better railway' which is wearing a bit thin

....oh and the bit about the 'best practice' planning software that showed all of the possessions coming in on time at 95% probability.  I have been on several projects like that where the work plan has been 'manipulated to fit' so that the project manager can show that the work should go ahead as to defer it would cost ^s.

[Chris from Nailsea]

Thanks for your comments, SandTEngineer. 

The Network Rail report is available at http://www.networkrailmediacentre.co.uk/Gallery/Christmas-2014-passenger-disruption-report-2208.aspx

The link to the Network Rail report which I quoted previously no longer works: their (apparently revised) report is now available at http://www.networkrailmediacentre.co.uk/News-Releases/Network-Rail-publishes-Christmas-passenger-disruption-investigation-report-2231.aspx 

[SandTEngineer]

Apologies CfN I meant to add the new link to my post but got overcome by frustration with the NR^» (Network Rail - home page) approach in the report