Fault attribution: transport accidents

Discussions about serious topics, for serious people
Post Reply
IvanV
Stummy Beige
Posts: 2663
Joined: Mon May 17, 2021 11:12 am

Fault attribution: transport accidents

Post by IvanV » Thu Jul 22, 2021 6:46 pm

The jury in the Croydon Tram inquest has finished its deliberations, and the conclusion (we mustn't say verdict) is accidental death. A bereaved relative said, "It's not an accident, someone is to blame."

On the one hand, there is a tendency, as illustrated here, to want someone to blame, although often it's a bit more complicated than that. On the other hand, accidents such as these, where there is a big corporate body involved, it's easy to blame the workers and overlook the failings of management. See for example, Herald of Free Enterprise accident. Or, as in this case, blame no one, it falls between two stools as it were.

So the basic facts are these, to remind ourselves. A tram went around a tight bend 3 times too fast and derailed. 7 died, others were injured. There were no mechanical reasons for it, rather the driver was driving too fast of his own accord. The driver was seemingly disorientated, probably through tiredness from not getting enough sleep, and forgot to slow down for the bend coming up. You can't see it until the last second, you rely on track knowledge to drive it safely.

There were, however, some design flaws.
  • There was no advance signage of the up-coming speed restriction, which would have been a good idea. The driver had to know to slow down in time. This change was made very quickly. A cheap and obvious thing to do.
  • There is no automatic speed limiting system, comparable to the one that is a legal requirement on the mainline railway, in such locations. On the mainline, the TPWS system is present to slam on the brakes if the driver is approaching a speed reduction location too fast. This can be a red light or a speed-limited bend. There is a minimum speed reduction that applies. And temporary speed restrictions don't need it if they are sufficiently temporary.
  • There was a history of drivers going rather too fast around the bend and getting away with it. But this was only discovered afterwards. No one was routinely analysing the data to see if drivers obeyed the speed limits, and analyse possible reasons for it.
  • A disputed point is whether the shift patterns contributed to the driver's likely tiredness.
The questions I want to ask are:
  • Does the fact of the design faults absolve the driver of any blame for this?
  • Are the design faults sufficient that TfL should be blamed, either alternatively, or in addition?
As background, let me say a bit more about The Herald of Free Enterprise. A crew member were supposed to close the ferry door when a signal was given. But there was no system for ensuring it happened. Not a camera, not closed door detector, not a requirement to report it was done, or nor ask was done. The Captain took it on trust that it was always done when asked for, and set off shortly after. Occasionally it had been done a bit late in the past, but they got away with it. It was the particular weather conditions that resulted in it being a problem. On that occasion there was a conclusion of unlawful killing in the inquest. Various crew members were prosecuted. But it wasn't possible to pin any blame for the obviously inadequate system on either anyone senior, nor corporate manslaughter.

Also motoring background. Suppose someone failed to slow down for an unsigned concealed bend in their car, and crashed it and killed someone. Would/should they be found guilty of causing death by careless, or even dangerous, driving? I am aware of some cases like this where drivers did something quite like that, and were found not guilty of either offence. In one case a driver crashed their car into a ditch through taking a bend too fast on a mountainous single track road in the dark, killing some of the passengers. No other vehicles involved. The driver in effect admitted misjudging the bend. In the other case, a driver went around a blind bend 10mph over the speed limit, started to overtake some bicycles they encountered, and then a car came the other way. The driver cut in very close to the bicycles, one of them wobbled and came off, and was killed when the car struck them.
Last edited by El Pollo Diablo on Fri Jul 23, 2021 9:07 am, edited 2 times in total.
Reason: fixed bullets

IvanV
Stummy Beige
Posts: 2663
Joined: Mon May 17, 2021 11:12 am

Re: Fault attribution: transport accidents

Post by IvanV » Thu Jul 22, 2021 6:47 pm

...I keep hitting quote instead of edit.

plodder
Stummy Beige
Posts: 2981
Joined: Mon Nov 11, 2019 1:50 pm

Re: Fault attribution: transport accidents

Post by plodder » Thu Jul 22, 2021 7:53 pm

These are good questions, and risk assessments should show whether TfL or the designers thought about these things during scoping and design respectively.

The risk assessments ought to also show why these measures were omitted, and give good reasons why (based on what they knew at the time).

The quality (or otherwise) of these risk assessments will tell you what you need to know. Don’t know if that’s the approach the coroner took, mind you.

User avatar
Martin_B
After Pie
Posts: 1614
Joined: Mon Nov 11, 2019 1:20 pm
Location: Perth, WA

Re: Fault attribution: transport accidents

Post by Martin_B » Fri Jul 23, 2021 12:38 am

IvanV wrote:
Thu Jul 22, 2021 6:46 pm
  • There was a history of drivers going rather too fast around the bend and getting away with it. But this was only discovered afterwards. No one was routinely analysing the data to see if drivers obeyed the speed limits, and analyse possible reasons for it.
  • A disputed point is whether the shift patterns contributed to the driver's likely tiredness.
I wonder if this was due to trying to make up lost time to stay on schedule.

During the early days of the Manchester trams (the current ones, so early 90s) I recall a driver saying that the schedules they were supposed to meet were based on light traffic, not winding their way through Manchester in the rush-hour and/or through the city centre with lots of commuters/shoppers around. And they were permitted 30 seconds stop at each stop, regardless of the number of people getting on and off, or seeing someone running to catch the tram if you just wait another 10 seconds. As being on schedule was their KPI for job performance, they had to speed up a bit more than they should have done in sections to make up the lost time, and then got berated by management for this.
Last edited by El Pollo Diablo on Fri Jul 23, 2021 9:07 am, edited 1 time in total.
Reason: fixed bullets in quote
"My interest is in the future, because I'm going to spend the rest of my life there"

User avatar
El Pollo Diablo
Stummy Beige
Posts: 3323
Joined: Wed Sep 25, 2019 4:41 pm
Location: FBPE

Re: Fault attribution: transport accidents

Post by El Pollo Diablo » Fri Jul 23, 2021 9:21 am

Probably of relevance to this is the RAIB Report into the incident, here*. The point of RAIB reports is to try to find the ultimate causes of the incident and make recommendations as to how to stop them happening again, rather than to blame specific people in perhaps the way sought by the families of those who died.

The report finds that the causal factors were
464 The tram did not slow down to a safe speed before entering Sandilands south curve because the driver did not apply sufficient braking (paragraphs 119 and 123). Although some doubt remains as to the reasons for the driver not applying sufficient braking, the RAIB has concluded that the most likely cause was a temporary loss of awareness of the driving task during a period of low workload, which possibly caused him to microsleep. It is also possible that, when regaining awareness, the driver became confused about his location and direction of travel.

465 The RAIB has identified a number of factors that may have influenced the driver’s actions:
a. low driver workload when approaching the accident site (paragraph 138 to 142, Recommendations 3 and 4);
b. although there is no evidence that the driver’s shift pattern carried an exceptional risk of causing fatigue, it is possible that the driver had become fatigued due to insufficient sleep when working very early turns of duty (paragraphs 143 to 152, Recommendations 3 and 4);
c. disorientation of the driver (paragraph 153 to 156); and
d. the infrastructure approaching the curve did not contain sufficiently distinctive features to alert drivers to their position relative to the curve at
Sandilands or to their direction of travel in the tunnel (paragraphs 157 to 164, Recommendation 5).
Underlying factors were
468 The underlying factors were:
a. LT and TOL did not recognise the actual level of risk associated with overspeeding on a curve (paragraph 195, Recommendation 2). This was for the following reasons:
i. route hazard assessments did not identify the need for additional mitigation due to the risk associated with overspeeding at Sandilands south curve (paragraphs 197 to 201, Recommendation 10).
ii. risk profiling for the Croydon network did not fully recognise the level of risk associated with a tram overturning (paragraphs 203 to 211, Recommendation 10);
iii. route hazard assessments and risk profiling relied on driver performance as the main means of mitigating the risk of overspeeding (paragraphs
202 and 212 to 214, Recommendation 10); and
iv. route hazard assessments and risk profiling did not take account of evidence from other tram, road and rail systems showing the level of risk associated with trams overturning (paragraphs 215 to 221, Recommendation 10).
b. Although senior managers recognised the importance of learning from experience, there were a number of factors which prevented TOL from gaining a full understanding of the extent of late braking (probably partly because of a lack of distinct cues) on the approach to Sandilands south curve. These factors included:
i. a reluctance of some drivers to report their own mistakes (paragraphs 224 to 243, Recommendation 12); and
ii. potential safety learning from customer complaints was not fully exploited (paragraph 248, Recommendation 13).
c. The risk associated with excessive speed around curves was neither fully understood by the safety regulator nor adequately addressed by UK tramway designers, owners and operators (paragraph 249, Recommendation 2).
Recommendations then follow, which are worth a read.



*As with all RAIB reports, one can append "Thomas and the" to the start of the report title.
If truth is many-sided, mendacity is many-tongued

IvanV
Stummy Beige
Posts: 2663
Joined: Mon May 17, 2021 11:12 am

Re: Fault attribution: transport accidents

Post by IvanV » Fri Jul 23, 2021 12:25 pm

El Pollo Diablo wrote:
Fri Jul 23, 2021 9:21 am
Probably of relevance to this is the RAIB Report into the incident, here*. The point of RAIB reports is to try to find the ultimate causes of the incident and make recommendations as to how to stop them happening again, rather than to blame specific people in perhaps the way sought by the families of those who died.
...
*As with all RAIB reports, one can append "Thomas and the" to the start of the report title.
I read the first version when it came out in 2017. I'd forgotten quite how clearly they made the points, within the constraints of what they can say, that there should have been better signage, and some automated control to reduce the risk of speeding round the corner. It's a pretty devastating report, looking back at it.

Looking back on it now, it does rather surprise me that the risk cases for Croydon tram got through the safety regulators with such apparently obvious deficiencies. I worked on a "is it being done efficiently" review of TPWS a long time ago, during its installation. It's budget had been increased by a factor of 3 or so after the initial design and decision to go ahead, and government wanted a bit more understanding and a view of where it might end up. We found a lot of he budget increase was down to HSE, who had been exceedingly hawkish forcing Network Rail to increase the functionality. That delivered a small safety benefit for large increase in cost, as it required more complicated connections between the trackside devices and the signal cabinets, which often suffered from fragile wiring. They also extended the number of situations it should be installed in, but that was fair enough. So I find it a bit surprising they weren't awake to similar, seemingly more obvious issues with the tram. Though I'm afraid it does rather confirm my prejudice that they have a tendency to fail to notice the wood for the trees.

And that raises a question, to the extent a designer puts up an inadequate safety case, and the safety regulator fails to spot it, where's the blame there?

I'm not sure you'd get many buyers for "Thomas and the overturning of a tram at Sandilands junction".

User avatar
Grumble
Light of Blast
Posts: 4746
Joined: Mon Nov 11, 2019 1:03 pm

Re: Fault attribution: transport accidents

Post by Grumble » Fri Jul 23, 2021 1:54 pm

Plenty of rail disasters in Thomas, not many trams though
where once I used to scintillate
now I sin till ten past three

User avatar
sTeamTraen
After Pie
Posts: 2553
Joined: Mon Nov 11, 2019 4:24 pm
Location: Palma de Mallorca, Spain

Re: Fault attribution: transport accidents

Post by sTeamTraen » Fri Jul 23, 2021 2:16 pm

There is a case to be made that all accidents, other than perhaps those caused by a meteorite taking out a set of points a few seconds before the train arrives, are the fault of some person or process. I guess the question is the extent to which you can pick apart those issues from the overall complexity of the system, which in the case of something like a tramway necessarily has a non-zero minimum value.
Martin_B wrote:
Fri Jul 23, 2021 12:38 am
And they were permitted 30 seconds stop at each stop, regardless of the number of people getting on and off, or seeing someone running to catch the tram if you just wait another 10 seconds.
Just on this point: The other day I was running waddling for a bus and had got as far as the rear wheels when it pulled away. Part of me thought "Warwick Hunt" about the driver, but then I realised that there could potentially be a large number of people who might waddle up at 10-second intervals. Indeed, as I had just turned a corner, I didn't have any way to know whether the driver had, indeed, just accepted one or more other waddlers.
IvanV wrote:
Fri Jul 23, 2021 12:25 pm
I'm not sure you'd get many buyers for "Thomas and the overturning of a tram at Sandilands junction".
I'll see your unlikely Rev W. Awdry reference and raise you Enid Blyton.
Something something hammer something something nail

User avatar
Bird on a Fire
Princess POW
Posts: 10137
Joined: Fri Oct 11, 2019 5:05 pm
Location: Portugal

Re: Fault attribution: transport accidents

Post by Bird on a Fire » Fri Jul 23, 2021 3:59 pm

I always think it's a bit unfair for the media to dig up grieving relatives and quote them saying something silly.

Obviously something can be simultaneously accidental and somebody's fault.

The options available to a coroner aren't exactly limited, other than by convention:
There is no definitive list of conclusions available to a Coroner. The following are those most commonly used:

natural causes (including fatal medical conditions);
accident or misadventure;
industrial disease;
dependence on drugs/non-dependent abuse of drugs;
attempted/self-induced abortion;
disasters subject to public inquiry;
lawful killing (such as deaths caused during acts of war, or self-defence);
unlawful killing;
suicide;
open verdict (where there is insufficient evidence for any other verdict).

The commencement of the provisions in the Coroners and Justice Act 2009 have added some further possible conclusions to this list:

alcohol/drug related death, and
road traffic collision.

The conclusion of unlawful killing is restricted to the criminal offences of murder, manslaughter (including corporate manslaughter), and infanticide. Cases where driving causes death may, therefore, only be regarded as unlawful killing for inquest purposes if they satisfy the ingredients for manslaughter (gross negligence manslaughter) or where a vehicle is used as a weapon of assault and deliberately driven at a person who dies (murder or manslaughter depending on the intent).
The other conventional option available would have been, basically, an allegation of manslaughter by gross negligence, under the civil standard of the balance of probabilities. It seems like the blame may be sufficiently widely distributed, in a murky multiple-points-of-failure kind of way, which would make it tricky to pin the tail of guilt on the negligent donkey.

https://www.cps.gov.uk/legal-guidance/coroners

Presumably the "accidental" verdict precludes a criminal prosecution, but perhaps not a civil case?
We have the right to a clean, healthy, sustainable environment.

IvanV
Stummy Beige
Posts: 2663
Joined: Mon May 17, 2021 11:12 am

Re: Fault attribution: transport accidents

Post by IvanV » Fri Jul 23, 2021 5:06 pm

Bird on a Fire wrote:
Fri Jul 23, 2021 3:59 pm
Presumably the "accidental" verdict precludes a criminal prosecution, but perhaps not a civil case?
Certainly seems to preclude a criminal prosecution, given that it is not a road traffic accident. Negligence manslaughter would be an unlawful killing, although causing death by dangerous/careless driving isn't, in this country.

In fact TfL has already paid over £2.2m in compensation to 87 claimants, google tells me. You can be responsible to pay compensation for contributory negligence, falling short of manslaughter negligence.

Imrael
Snowbonk
Posts: 505
Joined: Tue Nov 12, 2019 5:59 am

Re: Fault attribution: transport accidents

Post by Imrael » Sun Jul 25, 2021 8:50 pm

not at all my field, but the incidents etc that i do think about/read up on are most often multi factor. X happened, proedure Y should have prevented harm, but person Z had the wrong training because person J was on holiday that day.

My point, I guess, is that even if blaming an individual is appropriate foe liablity reasons, there always needs to be a root cause analysis aimed at not putting someone in that position again.

User avatar
jimbob
Light of Blast
Posts: 5276
Joined: Mon Nov 11, 2019 4:04 pm
Location: High Peak/Manchester

Re: Fault attribution: transport accidents

Post by jimbob » Mon Jul 26, 2021 6:37 am

Imrael wrote:
Sun Jul 25, 2021 8:50 pm
not at all my field, but the incidents etc that i do think about/read up on are most often multi factor. X happened, proedure Y should have prevented harm, but person Z had the wrong training because person J was on holiday that day.

My point, I guess, is that even if blaming an individual is appropriate foe liablity reasons, there always needs to be a root cause analysis aimed at not putting someone in that position again.
I'd say that is a failure of management. Person Z should not have been doing that part of the job until they had the correct training. It is not up to them to second-guess what parts of their training are missing if they're told to do something.
Have you considered stupidity as an explanation

User avatar
Martin_B
After Pie
Posts: 1614
Joined: Mon Nov 11, 2019 1:20 pm
Location: Perth, WA

Re: Fault attribution: transport accidents

Post by Martin_B » Mon Jul 26, 2021 9:07 am

jimbob wrote:
Mon Jul 26, 2021 6:37 am
Imrael wrote:
Sun Jul 25, 2021 8:50 pm
not at all my field, but the incidents etc that i do think about/read up on are most often multi factor. X happened, proedure Y should have prevented harm, but person Z had the wrong training because person J was on holiday that day.

My point, I guess, is that even if blaming an individual is appropriate foe liablity reasons, there always needs to be a root cause analysis aimed at not putting someone in that position again.
I'd say that is a failure of management. Person Z should not have been doing that part of the job until they had the correct training. It is not up to them to second-guess what parts of their training are missing if they're told to do something.
It's amazing just how often management try to blame Person Z, though, and ignore their own failings.
I was looking into the Texas City refinery fire a couple of years ago, and the management essentially tried to blame it all on one operator. Due to cut-backs and poor operating practices, this one operator was looking after two units (which should have had an operator each) plus starting up a third unit, without any supervisor assistance (the supervisor had left for a family emergency) and hadn't had a proper hand-over from the previous shift because that would have counted as over-time, and that was banned as a cost-cutting exercise.
"My interest is in the future, because I'm going to spend the rest of my life there"

Allo V Psycho
Catbabel
Posts: 734
Joined: Sat Nov 16, 2019 8:18 am

Re: Fault attribution: transport accidents

Post by Allo V Psycho » Mon Jul 26, 2021 9:44 am

Imrael wrote:
Sun Jul 25, 2021 8:50 pm
not at all my field, but the incidents etc that i do think about/read up on are most often multi factor. X happened, proedure Y should have prevented harm, but person Z had the wrong training because person J was on holiday that day.

My point, I guess, is that even if blaming an individual is appropriate foe liablity reasons, there always needs to be a root cause analysis aimed at not putting someone in that position again.
The Swiss Cheese model, of course https://en.wikipedia.org/wiki/Swiss_cheese_model

I watched a James Reason video lecture once which was really good, but I've lost the link. If anyone knows it (or a good equivalent), I would start showing it to medical students again.

As a manager (influenced by this) my response to the occurrence of an error was (1) how do we best alleviate the consequences? (2) how do we ensure that it doesn't happen again? Blame didn't figure, except as in trying to stop conscientious (i.e. most) people from blaming themselves too much.

Once, a bad thing did happen for which someone had to be sanctioned, but this was not about a mistake, but about a deliberate decision to do something which they knew was wrong, and had bragged about at the time.

Post Reply