Can bad infrastructure be to blame for accidents?

[solocle]

Just was a passenger in an RTC here.



The light turned amber as we were crossing the stop line. The driver of our car stopped at the second stop line - the repeater signal was red.

The driver behind did not stop for amber, and hit us.

But the layout looks like a contributory factor to me. Red light and stop line.

"AMBER means ‘Stop’ at the stop line. You may go on only if the AMBER appears after you have crossed the stop line or are so close to it that to pull up might cause an accident!"

I would go 50/50 on that one.

We were already on top of the stop line when the light changed, and across it before a reasonable reaction time. Given that the driver behind us jumped said amber and rear ended us on the other side of the junction, I think pulling up might have caused an accident, don't you?

But at the 2nd stop line the light had just turned red. Although the driver was probably looking at the wrong light for the line, that's a major design issue.

The blame is between the idiot behind jumping the amber / driving too close, and whatever absolution for him the unexpected sudden stop is, that blame falls squarely on the council, IMO.

[tom66]

It is virtually always the rear-end car's fault in accidents like this. Usually caused by following too closely and not having sufficient space to stop; if they race through a junction to 'beat the light' then they could well crash into a stopped car in front. And the car in front can emergency stop for any reason, a kid could run out into the road, so you must always be able to stop if the car in front does an emergency stop. Sadly this seems to be a lesson that is forgotten pretty quickly by drivers.

[jnty]

It is virtually always the rear-end car's fault in accidents like this. Usually caused by following too closely and not having sufficient space to stop; if they race through a junction to 'beat the light' then they could well crash into a stopped car in front. And the car in front can emergency stop for any reason, a kid could run out into the road, so you must always be able to stop if the car in front does an emergency stop. Sadly this seems to be a lesson that is forgotten pretty quickly by drivers.

My mum was rear ended in a similar 'stopping at a green light' situation years ago - it was a blind junction and she heard sirens so paused to check before continuing. Slam.

[FosseWay]

It is virtually always the rear-end car's fault in accidents like this. Usually caused by following too closely and not having sufficient space to stop; if they race through a junction to 'beat the light' then they could well crash into a stopped car in front. And the car in front can emergency stop for any reason, a kid could run out into the road, so you must always be able to stop if the car in front does an emergency stop. Sadly this seems to be a lesson that is forgotten pretty quickly by drivers.

This exactly. The only sensible exception is really in situations where the vehicle in front deliberately brake-checks you.

If someone drops anchor for no good reason, their driving may be worthy of criticism, even prosecution, but that is a separate issue from whether the following vehicle left enough room to stop in the space that was clear.

[solocle]

It is virtually always the rear-end car's fault in accidents like this. Usually caused by following too closely and not having sufficient space to stop; if they race through a junction to 'beat the light' then they could well crash into a stopped car in front. And the car in front can emergency stop for any reason, a kid could run out into the road, so you must always be able to stop if the car in front does an emergency stop. Sadly this seems to be a lesson that is forgotten pretty quickly by drivers.

Yep - although I do think it's sometimes worth going beyond the simple blame game.

We don't know the cyclist in the first clip misinterpreted the left turn arrow for a straight ahead, they may be one of the many who blast through anyway.

We don't know for sure, no, but as I pointed out above and Ashley Neal mentions in his analysis, quite a lot in this particular case does point to it being a mistake - i.e. a lack of observation or misinterpretation of what was observed - rather than deliberate disobedience of the red light. People who run reds generally do this when they think they can make it. Quite often their risk assessment is some way off, but even in cases that result in accidents, there's clear evidence of last-minute attempts to avoid the collision. There wasn't in this case - I think the cyclist genuinely thought he had priority, and as you say, failed to check for people apparently proceeding against it.

Firstly, overall principle, the roads are As You Find Them. It is up to the road users to handle them at each point in an appropriate manner. This includes understanding what traffic signals mean. At the end of the day it is not "cars", nor "bicycles", that do the understanding of the signals, it is road users, their signals are the same, and it is up to all to understand them.

I'm afraid that that isn't a good enough response from "society" or "the authorities" or whatever you want to call the party that isn't the private individual. Yes, it is incumbent on all road users both to obey the law and to be aware of what's going on around them. But road users are human and humans make mistakes. If there's an issue with a road layout, surface, signage or whatever that either is clearly substandard or has been pointed out as being so by the fact of several recorded accidents or near-misses, and the authority responsible for it ignores those signals and does nothing about it, then that authority bears some of the responsibility for any loss or damage that results from further accidents there.

How much responsibility it's reasonable to allocate in that direction depends on what needs fixing, how problematic it is, and how realistic it is to fix it in a relevant timeframe. Whole junction layouts probably do need to be taken as found as a basic practical rule, because they can't be changed in the twinkling of an eye. But huge concrete monstrosities in the middle of the road that obscure the view for multiple road users at once can be moved (or not put there in the first place). Same with potholes, worn markings, missing signage. None of this is difficult to fix, so it is indefensible when it isn't fixed and contributes to an accident.

As individuals we have an absolute responsibility to keep our vehicles in an acceptable state of repair, such that obvious things like broken lights, bald tyres and bad brakes don't endanger us and others on the roads. We can't use the excuse of not having the money - it is a must. It's about time roads authorities took a similar responsibility for their part of ensuring road safety.

For an extreme example, this complete botch of signage fails on numerous counts.
1) - Not a TSRGD sign. Was special dispensation received?
2) - Display angle is variable, not fixed.
3) - Not retroreflective or illuminated.

OK, 2 doesn't look like an issue in the photo, it's pretty obvious is it not? As for point 1, who cares? Point 3, it gets in the way, does it not? Stop in the distance you can see to be clear.

4) - Human Error
Said gate is left propped open with a traffic cone.




Point 1 really remains irrelevant, but is symptomatic of the wider problem. 2 and 3 combine.

Now the combination of the sign being completely and utterly invisible in the dark to somebody turning left, and me mixing up my left turnings, three in the space of 300 metres, with the first not shown on ADS because it's a farm entrance...

The outcome was me cycling down the slip road onto the M40.

After revisiting the site to investigate, while I did confuse turnings, I apportion no blame to myself, and the entirety to Highways England (as they were then).

You cannot comply with a traffic sign if you cannot see the thing. And I could perfectly well see the proper retroreflective signs further along.


So my view is certainly not that road layout and signage cannot be to blame. There's a spectrum from "perfect", to "understandable error", to "completely and utterly criminal".

Even if the legal fault is 100% on one party, minimising the risk is a worthy endeavour for others involved.

[traffic-light-man]

Although the example in Ashley's video is a little unusual in the placement signal heads, I think this is a prime example of road users not considering a green arrow as a green arrow, or perhaps moreso not waiting to see the arrow before moving off.

Reading the comments, I was absolutely dumbfounded at the amount of people that seemed to be confused, at least to me, by a green arrow next to a full red, design decisions aside. I'd accept it if the general consensus was that on a starting amber, it's pretty much impossible to tell what the green aspect will display, but it seemed to be the green arrow and full red combination that was a confusion.

As I've said before, I'm a massive advocate for amber and red arrows. However, in addition to tackling the starting amber issues, the main reason I'm such an advocate of the red and amber arrows is in fact to give us the ability to introduce more split phase signalling like this, particularly around keeping things efficient for cyclists and pedestrians (as seen extensively in the Netherlands, for example) - and it's worrying to think that while I personally think red and amber arrows would aid understanding, maybe the understanding of an arrow by the majority of road users is flawed in the the first place, and that's a more pressing issue.

[jnty]

Reading the comments, I was absolutely dumbfounded at the amount of people that seemed to be confused, at least to me, by a green arrow next to a full red, design decisions aside.

Without wishing to overestimate the intelligence of the average motorist, it's worth remembering that Ashley Neal is a driving instructor and no doubt has a large following of people who are learning, or preparing for starting to learn, to drive.

[traffic-light-man]

Without wishing to overestimate the intelligence of the average motorist, it's worth remembering that Ashley Neal is a driving instructor and no doubt has a large following of people who are learning, or preparing for starting to learn, to drive.

That's a fair point, and I am perhaps being somewhat naive. I do know that a lot of Ashley's regular commentators are established drivers, though - and I'm not trying to say the ones commenting were those same folks - which is possibly why I discounted the 'in education' possibilities.

However, I still think that, surely, most people have a basic understanding of the colours of traffic signals long before they learn to drive - especially in the UK where we're actually (still) quite basic with the rules. Applying an arrow to that, should, IMO, be self explanatory.

[Bristol]

On the A4 Bath Road in Bristol, there's been a temporary traffic light for a pedestrian crossing here for a lnog while now: https://www.google.com/maps/@51.4426168 ... 384!8i8192 because the building site on the left blocks the pavement behind where the camera is showing.

Some days, one finds the signal heads for the traffic turned 90 degrees to face away from traffic, all four of them, but the pedestrian ones still operating; one day when the road was fairly quiet, I pressed the button, got pedestrian green, and luckily I checked and saw that the approaching car didn't seem to be intent on stopping! Then I realised the cars weren't seeing any signals at all.

Of course, most pedestrians here ignore the lights and walk in the coned off cycle path past the building site instead.

[FosseWay]

Reading the comments, I was absolutely dumbfounded at the amount of people that seemed to be confused, at least to me, by a green arrow next to a full red, design decisions aside. I'd accept it if the general consensus was that on a starting amber, it's pretty much impossible to tell what the green aspect will display, but it seemed to be the green arrow and full red combination that was a confusion.

That's not quite how I interpreted Ashley Neal's theory.

AIUI, it's not the juxtaposition of the green arrow and red ball per se that's the problem, but rather the reversal of their relative positions. At the stop line, there is a green arrow for straight on and a red ball for right; the green arrow is to the left of the red ball. At the far side, there is a green arrow for straight on and a red ball for left, and the green arrow is to the right of the red ball. At the stop line the distant signal is quite a long way away and the niceties of ball vs. arrow may be missed, especially if the driver is already presuming that whatever's showing over the other side is just a repeater of what they've got next to them. In other words, they may get confused by being given a green at the stop followed by a red at the other side in the same relative position, although it's intended for someone else. This becomes more of a problem if they've got a red at the stop line that they can't see or aren't looking at, and an apparent green at the far side. A motorist has little excuse for not seeing the near signal if they're respecting the ASL, but plenty don't; but a cyclist *should* be in the box beyond the first stop line, and will naturally look for what they think is an ordinary repeater on the other side.

I agree that a competent driver or cyclist should be able to sort this out, and it's far from the most confusing situation you can find yourself in on the roads. Ashley Neal is of a similar opinion, in laying the blame for the accident on the cyclist. But no traffic regulation should ever be a case of testing people's observational skills and reaction times. Designers need to accept that there will always be people who misunderstand or don't notice things, and be open to changing their way of working if this happens a lot in a given circumstance. Accident "fault" should never be a zero sum game - the cyclist is to blame, so everyone else involved automatically has no responsibility at all for what happened.

As Ashley says, we can never know (unless the cyclist gets in touch with him) what led the cyclist to do what he did. But we can raise the possibility of confusion at the lights, which, being raised, should then be looked into by the roads authority concerned. Far more of an issue IMV is the socking great concrete planter obscuring everyone's vision. That should not be there, simple as. It is guarding against the possibility of a low-likelihood but high consequence risk (of a runaway tram hitting the workforce) by creating a reasonably foreseeable hazard to *everyone* using that junction - in other words, the hazard is there for everyone all the time, and is only mitigated by the traffic lights. As we know, for various reasons - sometimes possibly somewhat forgiveable, as here - some people sometimes jump lights. I still haven't grasped the point of the planter. If there is a risk of trams ploughing into the workforce on the other side of the junction, there's surely as much risk of the tram failing to stop under normal (no roadworks) operation and ploughing straight through the junction on red. There's no mitigation of the same kind as the planter for that, so I'm clearly missing something

[KeithW]

As Ashley says, we can never know (unless the cyclist gets in touch with him) what led the cyclist to do what he did. But we can raise the possibility of confusion at the lights, which, being raised, should then be looked into by the roads authority concerned. Far more of an issue IMV is the socking great concrete planter obscuring everyone's vision. That should not be there, simple as. It is guarding against the possibility of a low-likelihood but high consequence risk (of a runaway tram hitting the workforce) by creating a reasonably foreseeable hazard to *everyone* using that junction - in other words, the hazard is there for everyone all the time, and is only mitigated by the traffic lights. As we know, for various reasons - sometimes possibly somewhat forgiveable, as here - some people sometimes jump lights. I still haven't grasped the point of the planter. If there is a risk of trams ploughing into the workforce on the other side of the junction, there's surely as much risk of the tram failing to stop under normal (no roadworks) operation and ploughing straight through the junction on red. There's no mitigation of the same kind as the planter for that, so I'm clearly missing something

While some sort of barrier may well be required I see no reason for its form. This is a problem the railway/tram industry solved a long time ago.





As for the junction its a horrible cluttered mess.

[FosseWay]

While some sort of barrier may well be required...

It's this that I'm unclear about. Why is a barrier needed at all?

In normal operation, I presume the tram runs along the tracks through the junction, including stopping at both the signal-controlled junction and the tram stop. Just now while the work is going on, there are no trams running through the junction, but evidently they stop fairly close to where the planter is. In any tram system there is always a small risk of mechanical failure, driver illness or deliberate mischief by a driver or passenger that could lead to the tram failing to stop where it should and putting people at risk.

During normal operation (without the planter or roadworks), there is a small chance of something going wrong that makes the tram disobey a stop signal and hit vehicles or pedestrians crossing the junction on green. If it's going fast enough, it may derail and hit the stop shelter or other road users. During the closure, without the planter there, there's a similar small risk of the same thing happening. So my questions are these:

1. Why is the small risk associated with a tram hitting the workforce at the stop considered to be more worthy of mitigation than the same risk of the tram running a stop light and hitting traffic?

2. Why it is considered acceptable to mitigate this small risk - for small it must be not to warrant any mitigation under normal operation - by restricting visibility to road-borne users of the junction (not just cars and cyclists, but pedestrians as well)?

As I say, I must be missing something here, because these questions really shouldn't be beyond someone whose job is to ensure safety on a tram system.

[jnty]

While some sort of barrier may well be required...

It's this that I'm unclear about. Why is a barrier needed at all?

In normal operation, I presume the tram runs along the tracks through the junction, including stopping at both the signal-controlled junction and the tram stop. Just now while the work is going on, there are no trams running through the junction, but evidently they stop fairly close to where the planter is. In any tram system there is always a small risk of mechanical failure, driver illness or deliberate mischief by a driver or passenger that could lead to the tram failing to stop where it should and putting people at risk.

During normal operation (without the planter or roadworks), there is a small chance of something going wrong that makes the tram disobey a stop signal and hit vehicles or pedestrians crossing the junction on green. If it's going fast enough, it may derail and hit the stop shelter or other road users. During the closure, without the planter there, there's a similar small risk of the same thing happening. So my questions are these:

1. Why is the small risk associated with a tram hitting the workforce at the stop considered to be more worthy of mitigation than the same risk of the tram running a stop light and hitting traffic?

2. Why it is considered acceptable to mitigate this small risk - for small it must be not to warrant any mitigation under normal operation - by restricting visibility to road-borne users of the junction (not just cars and cyclists, but pedestrians as well)?

As I say, I must be missing something here, because these questions really shouldn't be beyond someone whose job is to ensure safety on a tram system.

The Trams to Newhaven project seems to have an active and friendly PR team - I'm sure they would give you an informative reply if you asked: https://www.edinburgh.gov.uk/tramstonewhaven/contact-us

[Big L]

While some sort of barrier may well be required...

It's this that I'm unclear about. Why is a barrier needed at all?

In normal operation, I presume the tram runs along the tracks through the junction, including stopping at both the signal-controlled junction and the tram stop. Just now while the work is going on, there are no trams running through the junction, but evidently they stop fairly close to where the planter is…

If I’m in the right place, the tram line is being extended through the junction and off towards Leith. No (modern) trams have run through the junction.

This is how the line ended before.

[FosseWay]

If I’m in the right place, the tram line is being extended through the junction and off towards Leith. No (modern) trams have run through the junction.

This is how the line ended before.

Ah, OK, thanks - I hadn't realised the works were to extend the track.

But the question still stands - when it's completed, it will be theoretically possible for a tram to ignore a stop signal and hit something, and there won't be a big planter in the middle of the track.

[Glen]

But the question still stands - when it's completed, it will be theoretically possible for a tram to ignore a stop signal and hit something, and there won't be a big planter in the middle of the track.

But a tram running away isn't going to run off the end of the track and derail once the line is complete. The planter is essentially a temporary buffer stop at the end of the track, as the permanent one was removed to allow construction of the new track.

[jnty]

But the question still stands - when it's completed, it will be theoretically possible for a tram to ignore a stop signal and hit something, and there won't be a big planter in the middle of the track.

But a tram running away isn't going to run off the end of the track and derail once the line is complete. The planter is essentially a temporary buffer stop at the end of the track, as the permanent one was removed to allow construction of the new track.

My take is that the issue is that, before, the tram would immediately derail and stop, whereas now it could keep running all the way down the hill into the worksites and through pedestrian crossings etc. I think this is perhaps considered more of a risk as trams will routinely be left unattended there as the drivers change end (and they may even stable there overnight) whereas in normal running they won't be doing that routinely any more, and there may be further safety systems in place.

[FosseWay]

But the question still stands - when it's completed, it will be theoretically possible for a tram to ignore a stop signal and hit something, and there won't be a big planter in the middle of the track.

But a tram running away isn't going to run off the end of the track and derail once the line is complete. The planter is essentially a temporary buffer stop at the end of the track, as the permanent one was removed to allow construction of the new track.

My take is that the issue is that, before, the tram would immediately derail and stop, whereas now it could keep running all the way down the hill into the worksites and through pedestrian crossings etc. I think this is perhaps considered more of a risk as trams will routinely be left unattended there as the drivers change end (and they may even stable there overnight) whereas in normal running they won't be doing that routinely any more, and there may be further safety systems in place.

Thanks, the bit I've bolded makes this whole thing start to make a bit more sense

Re derailing - if you've got a runaway tram, it is preferable for all concerned if it *does* derail, as the friction of the wheels on tarmac or whatever is far greater than on the rails. It might fall over and/or crash into something and cause deaths, injuries and damage, but the potential for these outcomes is far greater if it stays on the track downhill, accelerating. In 1992 13 people died in a tram accident in Gothenburg when a tram did precisely this, reaching up to 60 mph down a hill before derailing and crashing into a bus stop and vehicles. (Swedish Wikipedia)

I'm obviously coming at this from a situation of considerable ignorance. But it strikes me that the likelihood of an out of control tram is basically tiny, while the fact of obscured visibility for users of the junction is there 100% of the time for all users coming from the direction of either the car or the cyclist in the video. There must surely be better ways of dealing with this.

[AndyB]

Put slightly differently, could a smaller planter have done the same job equally effectively without the same line of sight issues for other road users?

[jnty]

But a tram running away isn't going to run off the end of the track and derail once the line is complete. The planter is essentially a temporary buffer stop at the end of the track, as the permanent one was removed to allow construction of the new track.

My take is that the issue is that, before, the tram would immediately derail and stop, whereas now it could keep running all the way down the hill into the worksites and through pedestrian crossings etc. I think this is perhaps considered more of a risk as trams will routinely be left unattended there as the drivers change end (and they may even stable there overnight) whereas in normal running they won't be doing that routinely any more, and there may be further safety systems in place.

Thanks, the bit I've bolded makes this whole thing start to make a bit more sense

Re derailing - if you've got a runaway tram, it is preferable for all concerned if it *does* derail, as the friction of the wheels on tarmac or whatever is far greater than on the rails. It might fall over and/or crash into something and cause deaths, injuries and damage, but the potential for these outcomes is far greater if it stays on the track downhill, accelerating. In 1992 13 people died in a tram accident in Gothenburg when a tram did precisely this, reaching up to 60 mph down a hill before derailing and crashing into a bus stop and vehicles. (Swedish Wikipedia)

I'm obviously coming at this from a situation of considerable ignorance. But it strikes me that the likelihood of an out of control tram is basically tiny, while the fact of obscured visibility for users of the junction is there 100% of the time for all users coming from the direction of either the car or the cyclist in the video. There must surely be better ways of dealing with this.

Can't help but wonder if this is a symptom of differing safety standards for rail and road transport (even if trams are a bit of a grey area.) The rail people say that there's a fractional risk of a very serious runaway tram incident, let's eliminate it simply but effectively with a big concrete thing. The roads people say fine, it's a signal controlled junction and it's not like restricted visibility signal controlled junctions are unusual.