Really interesting update to this.
Report doesn’t read as self-serving or arse covering at all. As someone who does incident investigations and writes similar reports I found it quite good – although could do with para numbering as did come across as “wall of text”. (yes, I get the irony of that!)
So, my rebuild of the key facts is below, as well as questions it raises (apologies if they are dealt with – don’t have time to read the entire thing).
2M48,
1737 attempting to leave station, eventually ceasing 1820 (43 minutes spent moving 100m??? including at least one evac whilst still next to the platform at 1747 after having already been at the station for 22 mins)
1830 Mobile Ops Manager (MOM, a team or a person?) arrives to start de-icing etc.
Has to stop due to self-evacuation of 2M50. Over next 2 hours various back and forward as people got off and so on. Staff accompanied them and BTP also evacuated some. Eventually got going 2153 (+4 hours).
PA system failed approx. 2015, driver had given a level of updating to pax on board.
Pax on board in pretty horrible conditions.
2M50,
1737 stopped waiting for 2M48 to clear station/signal. 10 metres from platform.
1810 (+43 mins!) told to move in which actually couldn’t do due to being in normal working rules.
1839 (+29 mins) told again but now explicitly using Emergency rules.
1841, pax began to detrain, 1 hour and 4 mins after coming to halt 10metres from station.
Cue track emergency isolation etc. and the snowballing of this incident as described in 2M48 above.
1945 (+60 mins from 1st evac) lots of handles and issues resetting plus pissed off pax. Driver and MOM began evac 2001 (+16 mins). Then large scale evac commenced with BTP/Fire Bde giving some control.
2052 train was empty.
2S54,
1740 stopped due to 2M50 ahead of it.
1900 (+80 mins) self evac started, and subsequent other evacs.
2R46,
1740 became trapped, 2106 (+3 1/2 hours) self evacs began (poss earlier?)
2H50,
1745 became trapped, circa 1845ish first self evac, 1952 a lot more, again 2051 and 2103.
2S56,
1752 became trapped. 1907 self evacs started, later many more with NR staff around and people shepherded off the railway.
2R48,
1750 became trapped. 1821 driver told to switch ends to go back to last station. Unable to do this due to first self evacs and power isolation. Lost heating and lighting.
Approx 1920 (+1h30) self evacs started. Persuaded to get back on. 2136 (+3h45) power on and went back to previous station.
2H00,
1745 became trapped. 1900 power isolated and pax could see people getting off train ahead. Lots of driver etc. walking through train but config made this difficult. No self evacs and eventually moved 2158 (+4 hours).
So what do I think from what the report identifies? Well overall what a horrible F****** night shift that must have been for a lot of people! And even worse for the passengers.
1) This thing snowballed fast.
2) The chance to nip it early died within 15 minutes. (2M48 self evac whilst still at platform, having already reported significant issues and taken nearly ½ an hour to fail to clear the platform).
3) Within 33 minutes the next opportunity to limit the damage was lost when 2M50 could have been brought into the platform under Emergency rules. This could have evacuated the train and cleared the points that were trapping most (all?) of the other trains. Enabling their progress and avoiding their evacuations.
4) First positive action (sending the MOM, which I think is an individual) was 45 minutes after 2M48 first reported issues.
5) Railway staff focussed entirely on getting trains running, BTP/Fire Bde appeared more human focused with evacuation assistance. Railway people convinced that keeping people on trains is the (only?) answer, and persisted with that even after people were self-evacuating. [note this is massively reinforced by the Railway posters on this thread who almost refuse to even consider an alternative]
6) Control Centre was suffering massive info lag with commensurate impact on the quality and relevance of their decisions, as and when any were made (and also impacting their sense of urgency). For anyone with a military history background – think how Blitzkreig overwhelmed the Allies in 1940 as by the time the Allied Command got some information and made a decision, the real situation was actually completely different and the decision was utterly wrong, if even possible for people to try to action. Hence operational failure through inaction giving massive strategic failure.
7) No effective Command system, BTP in separate chain, GOLD/SILVER/BRONZE although existed and part of the Railway Management system don't appear to have functioned as I'd expect - i.e. giving unity of command. Comms also poor.
8) There is actual substantial Railway policy that identifies the human, comms, command & control issues and identifies what should be done to alleviate them. This appears not to have actually been implemented in the breach.
So, as I posited months ago in my first post on this topic, huge sympathy for the people in the train cabs, on the trackside and in control – but ultimately, the system completely failed to manage this.
Basically, they (Railway staff/control) did exactly what I suspected. Focussed on doing what they wanted to do, and utterly failed to ever stand back and recognise that time was passing, that their plan wasn’t working and that they desperately needed to adapt. Yes pax on the track undoubtedly creates huge problems – but as this illustrated, persisting from the start and then over an extended period in trying to keep them on the trains, but then failing – creates even bigger problems. These are humans, not inanimate cargo.
As I also suggested – effective comms can work: where BTP were present and drivers/other staff were able to exert an effective presence in the train, self evacs were much reduced or avoided. Also notably despite the 3rd rail risks that the Railway Personnel are so preoccupied about (and which are so emotive on this topic), there were no significant injuries (partly due to BTP and other support).
Interesting its highlighted that the actual ATOC/NR Guidance Note has some good policy…
my italics
1) Take immediate action to prevent situations becoming worse.
Not done
2) How to meet pax needs – pax related considerations should take precedence over railway ones.
Not done
3) Demonstrate care, empathy, competence and confidence to retain control of the situation (numerous Railway staff posters on this topic take note!).
Sincere attempts but overall ineffective
4) Avoid relying on a single plan.
To misquote Highlander - the railway's view is that there can only be one plan.
5) Avoid temptation of deferring decisions in hope situation gets better/more info comes in. Initiate response plan in full early as easier to stand resources down and then ramp up as degrades into crisis. I.e. plan for worst, hope for best.
Not done
If I could sum up what we in the military have learn about managing situations it would be the above. Clearly the knowledge is there and being promulgated in the industry – just not adhered to. That suggests a lot of training is needed and some top down culture change.
Above all, my judgement is that given how the people approached the day, and how they responded to the issues – this wasn't a set of really bad luck -
it was 100% certain to happen the way it did. That's usually the case when such incidents are investigated - hardwired for failure even when the policy is actually good and the people genuinely trying hard.
Hopefully this incident will at least serve to drive some significant improvements…
ps. Don't take any personal "I'm amazeballs" from this, I've seen and learnt all this stuff personally with no lack of mistakes, and hence the sympathy for the people. Thing is, when you fk up, first thing is admit it, second is learn what you should have done and third is practice what you'll do next time.