Level crossing stupidity...

[The Stationmaster]

The question of risk at level crossings and of them being the biggest identified risk on NR's infrastructure is an interesting one.  Comparative risk changes because of all sorts of factors and as some areas of risk are mitigated so others rise up the tree - level crossings in general represent a risk because part of their usage is unregulated and difficult to control and such control can only be achieved by eliminating them, at which time something else becomes 'the biggest risk'  (whatever that might means).

 

For example while tail end train collisions are, one hopes, virtually a thing of the past how many Section Signals are there on the network without full (i.e. 'Welwyn type) block controls and how much risk do they create?  It will probably be less than that of level crossings because there are fewer of them but the consequences are potentially far more serious in terms of physical damage and casualties and lack of such controls leaves an area 100% open to the frailties of human behaviour, it's remarkably easy to inadvertently restore a block instrument to 'Normal' if you attention has been diverted  (just as it is easy to open a level crossing before a train has passed).

 

I don't argue with the fact that level crossings, especially UWC crossings, involve considerable potential for collisions but the consequences are normally (fortunately) not particularly serious for the railway part of the incident apart from consequent line closures and delays.  And the cost of elimination compared with, say, 100% installation of Welwyn type controls is astronomic.  So I wonder if lack of block controls has been examined and assessed in the same way as the work on level crossings

[EddieB]

I thought that full barrier crossings always closed one side of the road to approaching traffic in each direction then after a delay close the other two barriers, thus allowing potentially trapped cars to escape.

 

Jamie

 

Maybe the case with gates, but not always with barriers.  There are many examples of barriers that extend the full width of the road.

[Grovenor]

Maybe the case with gates, but not always with barriers

IMHO Very rare, if ever, with gates but is or, certainly used to be, common with barriers when there are 4 provided.

Keith

[Edwin_m]

So I wonder if lack of block controls has been examined and assessed in the same way as the work on level crossings

Interesting thought Mike but I'm not sure I agree.  Level crossings have risen to the top of the list of risks because introduction of TPWS and various other measures have pushed SPADs well down from the top position.  I don't recall any accident for many years where lack of block controls was contributory, whereas fatal level crossing accidents are all too frequent.  Absolute block will be largely eliminated within the next couple of decades with the busiest sections where it is still used probably going first. 

 

Although the victim is sometimes to blame, Moreton-on-Lugg demonstrates that isn't always the case, and the fact remains that if the railway wasn't there there would be no accident so (donning a tin hat) I'd suggest there is a degree of moral responsbility to protect road users from their own stupidity.  Just as when driving on the road you will do your best to stop if someone runs out in front, and would share the blame if you didn't, even though any impact would be considered to be primarily the pedestrian's fault. 

 

Over the past few years we've had level crossing accidents where the train has been derailed, usually if a particularly heavy road vehicle is involved, but fortunately only at low train speeds so resulting in few casualties.  Hixon, Lockington and Great Heck demonstrate that a road vehicle collision (not necessarily on a level crossing) has the potential to produce the same sort of casualty numbers as a serious train collision.  Although things were done to address the causes of all three, there is still the scope for a series of unlucky coincidences to result in a similar event. 

[Glorious NSE]

Interesting thought Mike but I'm not sure I agree.  Level crossings have risen to the top of the list of risks because introduction of TPWS and various other measures have pushed SPADs well down from the top position.  I don't recall any accident for many years where lack of block controls was contributory, whereas fatal level crossing accidents are all too frequent.  Absolute block will be largely eliminated within the next couple of decades with the busiest sections where it is still used probably going first. 

 

Playing devils-advocate a little here - but the nature of statistics, especially for rare events like train crashes, is such that it's hard to tell. Just because there has been no accident like that for years doesn't mean it won't happen this morning. (Touch wood, etc)

 

Similarly if you look at the level crossing stats, ones serious enough to result in *rail* fatalities are very, very rare - in fact you can go many, many years with no *rail* crossing fatalities, then all of a sudden you'll wake up one morning to find the papers full of a Lockington...

 

Fact is, the number of crossing accidents is decreasing - that *may* be partly due to NRs concerted efforts to both reduce crossing numbers, raise awareness, and also make safety improvements, but it's also against a background of increasing road and rail traffic (when taken over the last couple of decades) 

[The Stationmaster]

Playing devils-advocate a little here - but the nature of statistics, especially for rare events like train crashes, is such that it's hard to tell. Just because there has been no accident like that for years doesn't mean it won't happen this morning. (Touch wood, etc)

 

Similarly if you look at the level crossing stats, ones serious enough to result in *rail* fatalities are very, very rare - in fact you can go many, many years with no *rail* crossing fatalities, then all of a sudden you'll wake up one morning to find the papers full of a Lockington...

 

Fact is, the number of crossing accidents is decreasing - that *may* be partly due to NRs concerted efforts to both reduce crossing numbers, raise awareness, and also make safety improvements, but it's also against a background of increasing road and rail traffic (when taken over the last couple of decades) 

Quite so.  And the whole matter of risk analysis can in itself be rather misleading.  While Edwin rightly makes the point above level crossing fatalities (usually of the Darwin Award variety as it happens so it is a different type of risk) we do tend to talk about rates of whatever incident so however many years.  As an example of this concern was expressed long before Eurostar operation commenced about the risks associated with the changeover from overhead to 3rd rail and an analysis of all known comparable, or vaguely comparable - such as Driver's reaction to misrouting, data and this produced an estimated risk of two pantograph collisions with fixed infrastructure in every 100 years. As it happens the figure was out by 50% (i.e the result was 3 instead of 2) the only difference being that when something is estimated against a timescale of 100 years it could all happen in one year or a couple - as it did in reality.

 

In other words lots of risk analysis (speaking as someone who has been quite nicely remunerated in the past for doing some) is often just a matter of how statistics are presented.  And coming back to level crossings it was a fact that for quite a few years after the introduction of automatic crossings there was a far higher incidence (and rate) of collisions at traditional gated crossings than at the new types.  What was different however - although not given such headline status - was the fact that the vast majority of collisions at traditional crossings were of road vehicles running into closed or closing gates, very few involved contact between a road vehicle and a train but statistically, in terms of both injuries and cost of repairs, gated crossings were not as safe as automatic crossings.

 

Equally at the present bridges are statistically safer than level crossings - until a few more vehicle drivers finish up on the railway (e.g Great Heck) or an underbridge is displaced and a train seriously derailed - then the 'greatest potential risk' might start to change.  Not an argument against closing level crossings but simply an illustration of the way potential risk rankings can always move - as Edwin has already said, the introduction of TPWS (and some other mitigations) has driven down SPAD numbers thus reducing that area of potential risk.

[Edwin_m]

Yes I agree that rail accident statistics are a minefield, particularly as the events in question are so rare as not to be statistically significant and the human brain is not good at dealing with very unlikely but very severe risks.  Better minds than I have worked to produce things like the RSSB risk model which attempts to get round this type of problem and presents the best estimate of the risk from a range of sources based on comprehensive data over quite a long period of time. 

[leopardml2341]

OK,maybe you seriously maybe need to read (re-read) the Highway Code...

 

'RED LIGHTS EQUAL STOP'  Even at a level crossing.

 

At one time IIRC the definition used to be:

 

Amber (traffic light - 'cause a railway signal is 'yellow') = Stop

Red = Wait

 

Subtle difference, but a highway code question that was asked of me when taking my PCV test (some time ago).

[caradoc]

Sorry to go against the flow, but considering the number of fatalities, maybe the "stupidity" is continuing to allow half barrier crossings.  We are forced to wear seat belts and crash helmets to protect us from ourselves, why not apply the same to level crossings?

 

Ed

 

As mentioned elsewhere, there are two major problems with converting half-barrier LCs to full barrier:

 

1. The crossing has to be monitored after the barriers are down, by a Signaller (on site or via CCTV) or more recently by an obstacle detector system, to ensure nothing is trapped on the crossing before signals are cleared. This greatly increases the installation, maintenance and operating costs of the crossing - Who pays ?

 

2. Because of 1, the road is blocked for far longer so that trains get clear signals.

 

 

The ultimate aim should be the elimination of as many level crossings as possible (again - who pays ?), but in the meantime Network Rail and the BT Police are doing a lot to highlight crossing misuse and catch offenders. 

 

When I cross the road there are no lights or barriers to make sure I don't step in front of a bus, yet how many pedestrians die on our roads each year compared to the number of people killed at level crossings ?

[The Stationmaster]

Yes I agree that rail accident statistics are a minefield, particularly as the events in question are so rare as not to be statistically significant and the human brain is not good at dealing with very unlikely but very severe risks.  Better minds than I have worked to produce things like the RSSB risk model which attempts to get round this type of problem and presents the best estimate of the risk from a range of sources based on comprehensive data over quite a long period of time. 

As an aside on this when i was out in Australia some years ago (on a risk assessment task as it happened - but a rather unusual one) I worked alongside the chap who had done a lot of the development work on the RSSB model and it was very interesting to look at how rating systems were arranged or set-up both that system and the one he was developing in Australia (until the client's money for that project ran out).  In fact a lot of it simply boils down to the way in which particular elements are weighted which can significantly alter the perceived risk or indeed the level of risk or comparative risk reflected by applying the system.

 

Another way of expressing this is that you can use (misuse?) even quite a  sophisticated risk assessment model or system to prove what you wanted it to prove before you started.  As it happened with my project I found that one of the biggest risks to the process I was examining was a particular individual suffering an accident which made him unable to work while close behind that came the Railway's General Manager and his habit of getting involved in the process.

[uax6]

There is a station foot crossing, not a million miles from where I stand, that everyday 3-5 people cross when the miniture warning lights are showing red. Why? I'm guessing they are in a rush and treat a train like a car, they can see it's coming, they assess it's speed, and then decide that they can make it. Either that or they are daft.

 

A couple of years ago (or was it last year, I can't remember now) the level crossing person for East Anglia arranged a meeting between the TOC, NR and the registered users for the crossings down here. We were expecting 30-40 of the users to turn up, guess how many did?  7. Yeap 7 and three of them where from the same farm, withanother 2 from another farm. The interesting thing that came out of the meeting is that the farmers that were present stated that UWC's were amongst the most dangerous things that the farm uses. They also stated that they avoid using them at all, even if it involves a long drive round.

But what of those that didn't turn up? Do they not care?

 

We have a crossing that has recently had phones installed, and still the main user of the crossing refuses to use the phone. Why? We don't know, he won't tell us, and he's been asked several times.

 

Attitudes are strange. Provide something that makes something safer, and we just don't use it. Hopefully the enhanced UWC (that is locked by a track circuit will help out, but sadly I doubt it.

 

Andy G

[phil-b259]

IMHO Very rare, if ever, with gates but is or, certainly used to be, common with barriers when there are 4 provided.

Keith

Correct. If the crossing has 4 individual booms (barriers to non railway people) then the facing booms must be proved electrically to be in the down position before the trailers descend. Obviously where only two booms are used (one each side) two close the road this does not apply.

(Also if the main and backup power supplies should fail then all 4 booms will descend together as the hydraulic valves in each pedestal release)

[PaulRhB]

The crossing has to be monitored after the barriers are down, by a Signaller (on site or via CCTV)

Good points apart from this minor detail at a CCTV crossing, the Signaller checks the cameras and presses crossing clear and the screen goes off, they don't continue to monitor it.

 

The new obstacle detector crossings do this job where a AHB is converted to full barriers, the difference is they continue to monitor the crossing all the time which is causing some 'issues' that they are working on.

[leopardml2341]

What really gets my goat about level crossing incidents is the way in which they're reported by the media.

 

"Train hits car............"

 

Almost always inferred to be the fault of the railway because there's an opportunity (at AHB crossings) for a car driver to weave around them.

 

Almost never inferred to be the fault of the driver for not obeying the rules of the road, or indeed self preservation for that matter.

 

I have only ever come across a scenario once where there was obviously a problem with an AHB crosiing, one barrier up, one down red lights flashing big queue of traffic - so what did I do - picked up the phone to report it to the signaller and then watched as others simply drove across.

 

Even drivers with passengers didn't ask a passenger to 'get out and look' first - abrogation of their responsibility to the safety of the passenger(s) methinks.

[leopardml2341]

................, the only safe level crossing is a non-existent one, hence Network Rail's drive to close as many crossings as possible.....................

 

It shouldn't be necessary to close them, just to reduce tthe risk to an As Low As Reasonably Practical (ALARP) level, and if that includes transferring some of the responsibility to the user, by e.g. warning signs then so be it.

 

Just because there may not be any signage telling you not to do something - and that's definitely not the case with Level Crossings - doesn't mean it's ok to do it and it won't be your fault.

 

If a user of anything,  level crossing or otherwise, percieves a risk with which they are not happy, they, as humans, have the rational ability and more importantly responsibility, to weigh up that risk and take an alternative course of action if needed - even if that is initailly as unpalatable as waiting 2 minutes for a train to pass a crossing

 

What some of the human race really need to do is to start taking responsibility for their own actions and self preservation and stop assuming that there will always be some else to blame.

[beast66606]

Good points apart from this minor detail at a CCTV crossing, the Signaller checks the cameras and presses crossing clear and the screen goes off, they don't continue to monitor it.

 

The new obstacle detector crossings do this job where a AHB is converted to full barriers, the difference is they continue to monitor the crossing all the time which is causing some 'issues' that they are working on.

 

One of my local ones was cycling for a while earlier this week, yodels + lights on, barriers down, barriers up, barriers down, barriers up, barriers down and yodels off - presumably the radar and/or liedar weren't happy about something.

[Adrian Wintle]

What really gets my goat about level crossing incidents is the way in which they're reported by the media.

 

"Train hits car............"

 

But perhaps you are reading intent into a non-judgemental statement of facts... In the vast majority of level crossing accidents, the train does hit the car - there are very few where the car hits the train.

 

Adrian

[Edwin_m]

If the site is suitable then bridge replacement may be viable on a purely financial case leaving aside the safety benefits.  A bridge probably costs more than replacing/upgrading a level crossing but it lasts a lot longer and does not cause operational incidents.  The difficulty is often that there is not space for a bridge or local landowners/residents object to one. 

[The Stationmaster]

But perhaps you are reading intent into a non-judgemental statement of facts... In the vast majority of level crossing accidents, the train does hit the car - there are very few where the car hits the train.

 

Adrian

That is of course correct (there are occasional exceptions and in my experience they also tend to be reported as 'train hits car') but the impression given is that the train is in the wrong - because of the way we are used to our language and its usage.   Perhaps it wouldn't be so bad if the word 'collide' was used instead of 'hits' - and technically it is more accurate - but I suspect the average sub-editor is likely to say 'that won't sell 'papers'; and that, of course, is what it's all about.

 

As it is the existing language probably gives some many motorists the impression that they are in the right if a train hits them.  Clearly a percentage of motorists haven't got much idea about level crossings and have even less idea of what a train can do to a car and its occupants if they come into collision.  And equally many motorists regard 'beating the traffic lights' as some sort of macho sport.  Unfortunately overall there is a section of the road using community that has little or no concept of safety - for either themselves or others - but clearly it is impossible to provide bridges or close every level crossing in Britain.  All that can be done, in many cases, as Andy has already said is make sure that risks are reduced to ALARP, it is simply impossible to nurse-maid every motorist in the land but obviously we do need to do as much as possible to protect trains and passengers from the idiot element among road users.

[The Stationmaster]

If the site is suitable then bridge replacement may be viable on a purely financial case leaving aside the safety benefits.  A bridge probably costs more than replacing/upgrading a level crossing but it lasts a lot longer and does not cause operational incidents.  The difficulty is often that there is not space for a bridge or local landowners/residents object to one. 

And someone has to pay for it of course.  Interestingly I lived only a few miles from an AHB crossing which HMRI had recommended for closure, which Railtrack, and then NR, wished to close, but - spurred on by a handful of locals - the District Council objected to closure - notwithstanding a couple of previous incidents at the crossing and despite the fact that there are bridges not far away which literally add no more than a handful of miles to road journeys for those few objecting locals.  So the crossing remained (and still remains) open although it was announced late last that it is intended to replace it with a bridge 'subject to purchase of the necessary land'.

 

It's name?  Ufton.  Death toll 2004 - 6 plus 12 seriously injured (plus the suicidal motorist who had parked on the crossing to end it all.  Plus one road misuser killed in 2011, plus a near miss - which finally prompted yet another (this time successful) attempt to think about a bridge).   And still there are more than reasonable diversionary routes for most regular users, the crossing has good approach visibility for road users yet the solution is now seen not as closure but spending millions on a bridge that only needs to be used by a handful of folk everyday.  It would probably be cheaper to build a link road to one of the existing bridges than to build an overbridge across the railway - it seems that option has not even been considered.

Edited April 12, 2013 by The Stationmaster

[uax6]

Well having witnessed one of the 'numptys' that NR employ to close crossings, I can well belive it. The sheer crassness of the bloke I met was un-nerving. Very good on the legal side, but IMO shouldn't be allowed to talk to anyone outside his own family!

 

Then again, what records do the railway hold on there crossings? I've just had a call from someone in Brum, asking about our one that recently got phones (18months ago!) asking if they had been fitted, commissioned and the board here ammended! And it's not the first time I've had a call about the same things. Shocking.

 

Andy g

[Adrian Wintle]

That is of course correct (there are occasional exceptions and in my experience they also tend to be reported as 'train hits car') but the impression given is that the train is in the wrong - because of the way we are used to our language and its usage.   Perhaps it wouldn't be so bad if the word 'collide' was used instead of 'hits' - and technically it is more accurate - but I suspect the average sub-editor is likely to say 'that won't sell 'papers'; and that, of course, is what it's all about.

 

As it is the existing language probably gives some many motorists the impression that they are in the right if a train hits them.  Clearly a percentage of motorists haven't got much idea about level crossings and have even less idea of what a train can do to a car and its occupants if they come into collision.  And equally many motorists regard 'beating the traffic lights' as some sort of macho sport.  Unfortunately overall there is a section of the road using community that has little or no concept of safety - for either themselves or others - but clearly it is impossible to provide bridges or close every level crossing in Britain.  All that can be done, in many cases, as Andy has already said is make sure that risks are reduced to ALARP, it is simply impossible to nurse-maid every motorist in the land but obviously we do need to do as much as possible to protect trains and passengers from the idiot element among road users.

 

The problem is that any other wording (that would work as a headline) would probably be more judgemental "Train hits idiot... Idiot drives under front of train... Crossing jumper hit by train...", especially if it were then proven that there were extenuating circumstances (crossing failure, brake failure...). Other options like "Level crossing collision" don't necessarily provide all the info (it could be two cars).

 

"Train hits car" is at least succinct and factual and won't come bact to bite...

 

Adrian

[lmsforever]

I used to drive over this crossing regularily when I used to deliver fish to a school in the vicinity and traffic was always light ,two cars = rush hour but locals  very bolshy always complaining about vehicles delivering goods.

[leopardml2341]

But perhaps you are reading intent into a non-judgemental statement of facts... In the vast majority of level crossing accidents, the train does hit the car - there are very few where the car hits the train.

 

Adrian

Not at all, quite simply that when the train hits the car.................

 

...........the car is where the train has the right of way. Unless of course it's as a result of a SPAD incident.

 

Edit for 'car' read road user

Edited April 12, 2013 by leopardml2341

[Adrian Wintle]

Not at all, quite simply that when the train hits the car.................

 

...........the car is where the train has the right of way. Unless of course it's as a result of a SPAD incident.

 

Or a crossing failure (in which case it is not clear who has right of way), or in the case of an unforseen mechanical failure (where the train does have right of way, but the road user is not necessarily guilty), etc. 

 

By saying 'train hits car' you are describing the facts, without allocating blame. How would you word it?

 

Adrian