Road Collision Investigation Branch

The study of British and Irish roads - their construction, numbering, history, mapping, past and future official roads proposals and general roads musings.

There is a separate forum for Street Furniture (traffic lights, street lights, road signs etc).

Registered users get access to other forums including discussions about other forms of transport, driving, fantasy roads and wishlists, and roads quizzes.

Moderator: Site Management Team

User avatar
Barkstar
Member
Posts: 2604
Joined: Sun Jul 14, 2013 16:32

Re: Road Collision Investigation Branch

Post by Barkstar »

In principal it sounds like a good idea but I wonder how it would work in practice? What level of incident will trigger a TCIB investigation? Collision investigation officers can investigate any RTC but in effect only look into fatals and the most major of non-fatal incidents. Would it be the same for such an organisation? I'm pretty sure a lot of seemingly minor incidents could tell us as much if not more than major RTCs, where, in a lot of cases, the driving has been clearly way worse than what is expected of a competent driver.
And in the wake of their findings would whoever was responsible for the highway be obligated to put such recommendations into effect immediately, regardless of cost?

Though I can see some benefits it might just be creating an additional layer of bureaucracy and likely to be opening a can of worms.
WHBM
Member
Posts: 9707
Joined: Thu Nov 30, 2006 18:01
Location: London

Re: Road Collision Investigation Branch

Post by WHBM »

For some of us who have been around for a good while, the area highways engineers used to be pretty good at this. A serious accident in their patch would likely see them going out the next day, generally without having to be "told" to go, look at the scene, read the police report (or actually go and speak to the officers who had attended it), take photographs, look at the sightlines and markings, drive through the junction, look up the history in the office, etc. Everything went in a file in the office, cross-referenced by location, accident type, etc, in a card index in pre-computer days which they or their predecessors has devised themselves. They probably wanted to go as well to note any damaged street furniture or other items to clear away. If they felt there was any engineering effort that would improve things it would go into the file of proposed improvements. Simple things would be written up on a DLO job card for attention. Sometimes they might even get called to a court case or, alas, an inquest.

That was then. Nowadays it's disappointing that this sort of initiative, and ability, has disappeared.

Police reports likewise have been reduced from written descriptions to Janet-and-John style tick boxes, along with desperation to find if certain items can be found to "get a prosecution" out of it, mobile phone in use being the current top investigation.
Micro The Maniac
Member
Posts: 1175
Joined: Thu Jun 25, 2015 13:14
Location: Gone

Re: Road Collision Investigation Branch

Post by Micro The Maniac »

WHBM wrote: Thu Nov 18, 2021 08:10 Government white paper launched on what seems an equivalent to the RAIB which investigates significant railway accidents
All in favour... as long as the subsequent recommendations it produces are constructive and not simply one of:
  • Lower the speed limit
  • Add traffic lights
  • Add a roundabout
  • Add traffic calming
which seems to the existing answer to every problem.

As many on here will agree, a thorough review of the existing design manuals would be a good start. Fixing the poor road layout never seems to be the answer.

I hope that its remit will also extend to the performance and operation of the cars themselves... in the same way that NHTSA is now looking at Tesla. As someone actively involved in the domain, I'm appalled at the slack regulatory regime for on-board software where standards such as ISO 26262 (functional safety) are purely optional, the type approval process doesn't even mention software controls, and guidance from (eg) MISRA is a nice to have - despite the big uncontrolled-acceleration legal case berating Toyota for not following it.
User avatar
Gav
Member
Posts: 1968
Joined: Sat Oct 22, 2005 17:44

Re: Road Collision Investigation Branch

Post by Gav »

they need this badly as we are getting more and more automonous systems in cars - they need to be able to identify common issues and make links to systems that have been installaed that could have bugs in them. a way to get such errors identified and ironed out.
User avatar
FosseWay
Assistant Site Manager
Posts: 19621
Joined: Thu Feb 23, 2006 22:26
Location: Gothenburg, Sweden

Re: Road Collision Investigation Branch

Post by FosseWay »

I'm broadly in favour in principle, providing (a) we get new information out of the process that we didn't have before (or we get it more quickly), and (b) the overall benefit from (a) is proportional to any increased delay in reopening roads.

If an accident investigation reveals something that was wrong with the road, signage or vehicles involved, or perhaps not "wrong" but capable of practical improvement, then that is very valuable. Also, if the investigations give us a better idea of what kinds of driver error are over-represented, the same applies. On the other hand, I'm not sure of the benefit of a complicated exercise whose conclusions are "that driver was on the phone/drunk/overtaking in a dangerous place" etc. if the same conclusions are reached through the current investigative process.

Broadly, I'd see the benefits of this as being to highlight risks and mitigations that we didn't know very much about before, not to reinforce our knowledge that driving like a dumbass causes accidents and costs lives. For the latter, I'd have thought we'd be better off spending the money on known mitigations of dumbass behaviour (more driver training, more police, better enforcement, take your pick).
Did you know there's more to SABRE than just the Forums?
Add your roads knowledge to the SABRE Wiki today!
Have you browsed SABRE Maps recently? Try getting involved!
WHBM
Member
Posts: 9707
Joined: Thu Nov 30, 2006 18:01
Location: London

Re: Road Collision Investigation Branch

Post by WHBM »

Micro The Maniac wrote: Fri Nov 19, 2021 10:12
I hope that its remit will also extend to the performance and operation of the cars themselves... in the same way that NHTSA is now looking at Tesla. As someone actively involved in the domain, I'm appalled at the slack regulatory regime for on-board software where standards such as ISO 26262 (functional safety) are purely optional
We have a Tesla at the office. I have to say I've never used any of the automated features in it, it just drives (very) nicely in conventional form. But there are those who are seemingly fascinated by it's automated features, show it off auto-parking round the car park, etc (which I would ban). Same people as are fixated by other new features of life - one of whom wasted 10 minutes of everyone's time on their phone in the City of London trying to get an Uber, while black cabs for hire were driving up and down past the door.
User avatar
the cheesecake man
Member
Posts: 2462
Joined: Mon Jan 07, 2019 13:21
Location: Sheffield

Re: Road Collision Investigation Branch

Post by the cheesecake man »

JohnnyMo wrote: Thu Nov 18, 2021 14:10 Cynic in me I expects reduce the speed limit and the junction will be much safer.
It depends whether excess speed was part of the cause and if anyone obeys the lower limit.
User avatar
Big L
Deputy Site Manager
Posts: 7517
Joined: Sun Nov 14, 2010 20:36
Location: B5012

Re: Road Collision Investigation Branch

Post by Big L »

Gav wrote: Fri Nov 19, 2021 10:24 they need this badly as we are getting more and more automonous systems in cars - they need to be able to identify common issues and make links to systems that have been installaed that could have bugs in them. a way to get such errors identified and ironed out.
I'm fairly sure it will be done badly !!
:D
Make poetry history.

Did you know there's more to SABRE than just the Forums?
Help with maps using the new online calibrator.
Add your roads knowledge to the SABRE Wiki.
Duple
Member
Posts: 161
Joined: Fri Apr 10, 2015 21:58

Re: Road Collision Investigation Branch

Post by Duple »

Totally agree with this,

Why it hasn't happened sooner is quite odd and this may actually release police time - if there is a "Branch" to investigate, they can get on doing other things.

Does make me wonder what happens with all the police accident investigations though, are these just used to apportion blame ?
User avatar
Ruperts Trooper
Member
Posts: 12031
Joined: Sun Jun 17, 2012 13:43
Location: Huntingdonshire originally, but now Staffordshire

Re: Road Collision Investigation Branch

Post by Ruperts Trooper »

Duple wrote: Fri Nov 19, 2021 17:57 Totally agree with this,

Why it hasn't happened sooner is quite odd and this may actually release police time - if there is a "Branch" to investigate, they can get on doing other things.

Does make me wonder what happens with all the police accident investigations though, are these just used to apportion blame ?
I presume they go to CPS to determine if charges should be brought and to the coroner in the case of a fatality.
Lifelong motorhead
Phil
Member
Posts: 2271
Joined: Sun Dec 15, 2002 18:03
Location: Burgess Hill,W Sussex, UK

Re: Road Collision Investigation Branch

Post by Phil »

Ruperts Trooper wrote: Fri Nov 19, 2021 18:10
Duple wrote: Fri Nov 19, 2021 17:57 Totally agree with this,

Why it hasn't happened sooner is quite odd and this may actually release police time - if there is a "Branch" to investigate, they can get on doing other things.

Does make me wonder what happens with all the police accident investigations though, are these just used to apportion blame ?
I presume they go to CPS to determine if charges should be brought and to the coroner in the case of a fatality.
Not so!

Please remember what was said up thread about investigative bodies!

Investigations performed by the RAIB / AAIB / MAIB do NOT apportion blame or liability.

Moreover any of the evidence they may gather is strictly inadmissible in a court of law when it comes to prosecution as per the legislation which set up said bodies!

If the ORR, Police, CPS, HSE etc want to prosecute then the MUST undertake THEIR OWN PARALLEL INVESTIGATION including re-interviewing folk, making their own notes, etc

This is deliberate - the whole point about the RAIB/ AAIB /MAIB is to learn lessons and try to prevent incidents from re-ocuring. Making the evidence they gather available for law enforcement hinders this as people may well chose to remain silent or not become 'whistleblowers' leaving valuable background information hidden and increasing the likelihood of a repeat incident occurring. Its also why individuals are not identified by name in the reports - just their job titles / roles.
Micro The Maniac
Member
Posts: 1175
Joined: Thu Jun 25, 2015 13:14
Location: Gone

Re: Road Collision Investigation Branch

Post by Micro The Maniac »

the cheesecake man wrote: Fri Nov 19, 2021 13:47 It depends whether excess speed was part of the cause and if anyone obeys the lower limit.
It never ceases to amaze me that the solution to (perceived or actual) excess speed always seems to involve reducing the limit further.

Someone who ignores the existing limit is unlikely to be impacted by a lower one, are they?
User avatar
Ruperts Trooper
Member
Posts: 12031
Joined: Sun Jun 17, 2012 13:43
Location: Huntingdonshire originally, but now Staffordshire

Re: Road Collision Investigation Branch

Post by Ruperts Trooper »

Phil wrote: Fri Nov 19, 2021 18:37
Ruperts Trooper wrote: Fri Nov 19, 2021 18:10
Duple wrote: Fri Nov 19, 2021 17:57 Totally agree with this,

Why it hasn't happened sooner is quite odd and this may actually release police time - if there is a "Branch" to investigate, they can get on doing other things.

Does make me wonder what happens with all the police accident investigations though, are these just used to apportion blame ?
I presume they go to CPS to determine if charges should be brought and to the coroner in the case of a fatality.
Not so!

Please remember what was said up thread about investigative bodies!

Investigations performed by the RAIB / AAIB / MAIB do NOT apportion blame or liability.

Moreover any of the evidence they may gather is strictly inadmissible in a court of law when it comes to prosecution as per the legislation which set up said bodies!

If the ORR, Police, CPS, HSE etc want to prosecute then the MUST undertake THEIR OWN PARALLEL INVESTIGATION including re-interviewing folk, making their own notes, etc

This is deliberate - the whole point about the RAIB/ AAIB /MAIB is to learn lessons and try to prevent incidents from re-ocuring. Making the evidence they gather available for law enforcement hinders this as people may well chose to remain silent or not become 'whistleblowers' leaving valuable background information hidden and increasing the likelihood of a repeat incident occurring. Its also why individuals are not identified by name in the reports - just their job titles / roles.
I was responding to Duple's question about where present police accident investigation reports go
Lifelong motorhead
ais523
Member
Posts: 1139
Joined: Tue Sep 08, 2015 19:52
Location: Birmingham

Re: Road Collision Investigation Branch

Post by ais523 »

cb a1 wrote: Thu Nov 18, 2021 18:15I wonder what proportion of rail and air accidents are due to 'operator error'?
There aren't supposed to be single points of failure on the railways, so anything going wrong is the fault of at least two people – either the two people who were both meant to be able to stop it, or else the person who made the mistake and the person who designed the process involved to have a single point of failure.

I've read a fairly large proportion of the RAIB reports, though, and my sense is that although human error is a major factor in many of the accidents investigated, human error by train drivers specifically is not one of the main causes (it definitely happens, and you'll find reports where it was a factor, but they're in a minority compared to the reports where the cause was something else).

For the example in this thread of "driving too fast", if it happened on the railways rather than the roads, there would be at least four points of failure to stop it causing an accident – the driver, who could have kept within the speed limit; overspeed sensors on the railway and train, which will automatically apply the emergency brakes if the train is speeding at certain points; the signaller, who can warn other trains away from the area if they realise that a driver is moving too quickly or running through red lights, or can contact drivers to remind/inform them of the speed limit if something has gone wrong with the normal mechanisms for telling them what it was; and routine monitoring of train data recorders to see if the trains had been speeding. There are reasons why you couldn't reasonably do all that on the roads, though – in particular, immediately forcing a speeding car to stop would likely cause an accident rather than prevent one.
User avatar
Vierwielen
Member
Posts: 5674
Joined: Sun Jun 08, 2008 21:21
Location: Hampshire

Re: Road Collision Investigation Branch

Post by Vierwielen »

ais523 wrote: Fri Nov 19, 2021 21:10
cb a1 wrote: Thu Nov 18, 2021 18:15I wonder what proportion of rail and air accidents are due to 'operator error'?
There aren't supposed to be single points of failure on the railways, so anything going wrong is the fault of at least two people – either the two people who were both meant to be able to stop it, or else the person who made the mistake and the person who designed the process involved to have a single point of failure.

I've read a fairly large proportion of the RAIB reports, though, and my sense is that although human error is a major factor in many of the accidents investigated, human error by train drivers specifically is not one of the main causes (it definitely happens, and you'll find reports where it was a factor, but they're in a minority compared to the reports where the cause was something else).

For the example in this thread of "driving too fast", if it happened on the railways rather than the roads, there would be at least four points of failure to stop it causing an accident – the driver, who could have kept within the speed limit; overspeed sensors on the railway and train, which will automatically apply the emergency brakes if the train is speeding at certain points; the signaller, who can warn other trains away from the area if they realise that a driver is moving too quickly or running through red lights, or can contact drivers to remind/inform them of the speed limit if something has gone wrong with the normal mechanisms for telling them what it was; and routine monitoring of train data recorders to see if the trains had been speeding. There are reasons why you couldn't reasonably do all that on the roads, though – in particular, immediately forcing a speeding car to stop would likely cause an accident rather than prevent one.
If I remeber correctly, one of the factors behind the Southall collision was poorly sited overhead signals which has led to a good deal more automation in the cab. It woudl appear that the underlying cause of the recent Salisbury collision was "leaves on the line". I recently saw a picture of that junction taken many years ago with steam engines pulling the trains. Not a tree is sight - you did not want sparks from the stream engine starting a line-side fire.
Phil
Member
Posts: 2271
Joined: Sun Dec 15, 2002 18:03
Location: Burgess Hill,W Sussex, UK

Re: Road Collision Investigation Branch

Post by Phil »

Vierwielen wrote: Fri Nov 19, 2021 22:23
ais523 wrote: Fri Nov 19, 2021 21:10
cb a1 wrote: Thu Nov 18, 2021 18:15I wonder what proportion of rail and air accidents are due to 'operator error'?
There aren't supposed to be single points of failure on the railways, so anything going wrong is the fault of at least two people – either the two people who were both meant to be able to stop it, or else the person who made the mistake and the person who designed the process involved to have a single point of failure.

I've read a fairly large proportion of the RAIB reports, though, and my sense is that although human error is a major factor in many of the accidents investigated, human error by train drivers specifically is not one of the main causes (it definitely happens, and you'll find reports where it was a factor, but they're in a minority compared to the reports where the cause was something else).

For the example in this thread of "driving too fast", if it happened on the railways rather than the roads, there would be at least four points of failure to stop it causing an accident – the driver, who could have kept within the speed limit; overspeed sensors on the railway and train, which will automatically apply the emergency brakes if the train is speeding at certain points; the signaller, who can warn other trains away from the area if they realise that a driver is moving too quickly or running through red lights, or can contact drivers to remind/inform them of the speed limit if something has gone wrong with the normal mechanisms for telling them what it was; and routine monitoring of train data recorders to see if the trains had been speeding. There are reasons why you couldn't reasonably do all that on the roads, though – in particular, immediately forcing a speeding car to stop would likely cause an accident rather than prevent one.
If I remeber correctly, one of the factors behind the Southall collision was poorly sited overhead signals which has led to a good deal more automation in the cab. It woudl appear that the underlying cause of the recent Salisbury collision was "leaves on the line". I recently saw a picture of that junction taken many years ago with steam engines pulling the trains. Not a tree is sight - you did not want sparks from the stream engine starting a line-side fire.
Yup.

Unfortunately the decision by British Railways to stop clearing the lineside of small saplings / bushes (driven in later years by repeated Treasury cuts to funding and thus preserve cash for the business of operating trains) before they had a chance to mature has meant that many lineside residents and green groups have become accustomed to having lots of trees screening their property / acting as a lineside nature reserve.

Attempts by NR to undertake wholesale clearance a few years ago created a mighty stink with said residents / green groups who got MPs involved and forced NR into a rethink of the practice.

It should also be noted that vegetation control is an ongoing costs - and with the DfT demanding cuts, it represents an easy thing to put on the back burner for a while. The fact that its detrimental to long term thinking / safety rarely bothers a Whitehall that seems incapable of looking beyond the next election....
Last edited by Phil on Mon Nov 22, 2021 02:17, edited 1 time in total.
Phil
Member
Posts: 2271
Joined: Sun Dec 15, 2002 18:03
Location: Burgess Hill,W Sussex, UK

Re: Road Collision Investigation Branch

Post by Phil »

Vierwielen wrote: Fri Nov 19, 2021 22:23
ais523 wrote: Fri Nov 19, 2021 21:10
cb a1 wrote: Thu Nov 18, 2021 18:15I wonder what proportion of rail and air accidents are due to 'operator error'?
There aren't supposed to be single points of failure on the railways, so anything going wrong is the fault of at least two people – either the two people who were both meant to be able to stop it, or else the person who made the mistake and the person who designed the process involved to have a single point of failure.

I've read a fairly large proportion of the RAIB reports, though, and my sense is that although human error is a major factor in many of the accidents investigated, human error by train drivers specifically is not one of the main causes (it definitely happens, and you'll find reports where it was a factor, but they're in a minority compared to the reports where the cause was something else).

For the example in this thread of "driving too fast", if it happened on the railways rather than the roads, there would be at least four points of failure to stop it causing an accident – the driver, who could have kept within the speed limit; overspeed sensors on the railway and train, which will automatically apply the emergency brakes if the train is speeding at certain points; the signaller, who can warn other trains away from the area if they realise that a driver is moving too quickly or running through red lights, or can contact drivers to remind/inform them of the speed limit if something has gone wrong with the normal mechanisms for telling them what it was; and routine monitoring of train data recorders to see if the trains had been speeding. There are reasons why you couldn't reasonably do all that on the roads, though – in particular, immediately forcing a speeding car to stop would likely cause an accident rather than prevent one.
If I remeber correctly, one of the factors behind the Southall collision was poorly sited overhead signals which has led to a good deal more automation in the cab. It woudl appear that the underlying cause of the recent Salisbury collision was "leaves on the line". I recently saw a picture of that junction taken many years ago with steam engines pulling the trains. Not a tree is sight - you did not want sparks from the stream engine starting a line-side fire.
Not automation, rather the introduction of TPWS (and expansion of the BR designed ATP) which should prevent trains passing a red light before they end up foul of other train movements.

Automation as regards mainline train movements has only recently been put in place in the Thameslink core - elsewhere the emphasis from technology is on stopping the driver making mistakes rather than removing the task of driving from them.
User avatar
FosseWay
Assistant Site Manager
Posts: 19621
Joined: Thu Feb 23, 2006 22:26
Location: Gothenburg, Sweden

Re: Road Collision Investigation Branch

Post by FosseWay »

Vierwielen wrote: Fri Nov 19, 2021 22:23 If I remeber correctly, one of the factors behind the Southall collision was poorly sited overhead signals which has led to a good deal more automation in the cab. It woudl appear that the underlying cause of the recent Salisbury collision was "leaves on the line". I recently saw a picture of that junction taken many years ago with steam engines pulling the trains. Not a tree is sight - you did not want sparks from the stream engine starting a line-side fire.
Swedish railways are gradually making more and more lines "tree-safe" by ensuring a given width of treelessness either side of railway lines. When it gets very windy (into weather warning territory) you can tell which lines have been dealt with and which not, since the safe ones remain open and the others are shut and provided with rail-replacement buses because the risk of a tree blocking the line is considered too high. The risks involved here are not just the obvious (train hits fallen tree, or train stops in time but must then be evacuated) but also, especially in the parts of Sweden most covered in trees, that a stationary train without power must be evacuable within the time it takes for the residual heat in the carriages to dissipate, otherwise passengers will start dying of hypothermia. They don't run passenger trains at all on some routes if the temperature goes below -40C for this reason - they just can't get to the relevant bit of bugger-all by road or helicopter in time if something goes wrong with the train.

There are other reasons for making lines tree-safe - wildlife is less likely to run straight across the tracks, but will instead enter the cleared area and see/hear the train and stop or run back into the forest. Fire is also an issue, even with modern trains - a spark can ignite a forest fire, and these can be very large and difficult to control in places where you can't get fire engines to.
Did you know there's more to SABRE than just the Forums?
Add your roads knowledge to the SABRE Wiki today!
Have you browsed SABRE Maps recently? Try getting involved!
Graham
Member
Posts: 287
Joined: Thu Sep 19, 2002 12:37

Re: Road Collision Investigation Branch

Post by Graham »

cb a1 wrote: Thu Nov 18, 2021 18:15
My recollection from when I did the Road Safety module at Uni was that for road collisions, driver error was 97%
This may well be true, but the danger with this statistic is that it encourages a "prosecute the driver at fault and all will be well" attitude, when a simple change to the road design may be all that is needed to prevent other drivers from making the same mistake in future.
WHBM
Member
Posts: 9707
Joined: Thu Nov 30, 2006 18:01
Location: London

Re: Road Collision Investigation Branch

Post by WHBM »

Graham wrote: Sat Nov 20, 2021 15:37
cb a1 wrote: Thu Nov 18, 2021 18:15
My recollection from when I did the Road Safety module at Uni was that for road collisions, driver error was 97%
This may well be true, but the danger with this statistic is that it encourages a "prosecute the driver at fault and all will be well" attitude, when a simple change to the road design may be all that is needed to prevent other drivers from making the same mistake in future.
In the time I've been a road user (ie when I could first balance a bike) fatalities have been reduced from 8,000pa in 1966 to 1,800pa now. One aspect is the vast majority of potential engineering ameliorations have now actually been done, and the remaining hard core are those more difficult to crack, like deliberate stupidity. There are also certain classes, such as motorcycle accidents, which form an ever-increasing proportion.

I regret that police forces have reduced accident analysis to a simple ticklist, which just makes it easy to tick driver error without much thought.
Post Reply