Today a friend from way back who is a structural engineer by profession dropped by. He is semi-retired now, but was not so long ago a pretty big cheese in the bridge designing trade. He still has quite a bit of influence on bridge designing, albeit rather less now than he used to have.
He told me of an engineering bee, concerning train safety, now buzzing about inside his head.
Trains are, on the whole, he said, very safe. But apparently level crossings are the big train exception. A trickle of deaths? A rather big trickle, he replied. “Scores every year” was the phrase I recall him using. I don’t know if that’s quite right, but level crossing deaths are certainly a big deal in Britain.
Perhaps partly because he is a bridge designer, my friend believes that where possible and where not too disruptive and expensive, level crossings should be replaced by … bridges.
Trouble is, there is an acronymic organisation (I think he was talking about this one) concerned with British train safety, which demands very large clearances, both upwards and sideways, for all new road bridges over railways. And it takes only a small increase in a demanded clearance size to require a greatly more elaborate and expensive bridge. Which means that a lot of bridges, that might be built, aren’t.
Partly this demand for big clearances is because at some future date the railway line in question might be electrified, and in the meantime, space must be left under all bridges for that.
My friend says: Fine. In the meantime build smaller but temporary bridges. If electrification ever happens, replace these small bridges with bigger bridges. (Or, I suppose, go back to having level crossings, although that possibility wasn’t mentioned.) Meanwhile, save many lives now lost at level crossings.
But partly, the reason is a safety consideration of another kind. Says the acronymic organisation: all imaginable train wobblings, including the most unlikely, must be allowed to occur under all future bridges without any train afflicted by such wobbling hitting the bridge, even though many existing bridges allow for no such wobblings, with no detectable effect on train fatalities.
Result? The rather big trickle of deaths at level crossings goes on, and on, and on. In the pursuit of even more perfect safety where perfect safety has pretty much already been achieved, a closely related and very unsafe circumstance is caused to persist.
The pursuit of safety, badly done, is resulting in the persistence of unsafety.
When people now speak, as they so often now do, of “health and safety gone mad”, this is one of the things they surely mean. It isn’t just that safety is pursued and damn everything else. Safety is also pursued in accordance with mindless rules, that have the effect of reducing safety itself.
Something tells me that this is not the only example of such perverse safety thinking.
“The perfect is the enemy of the good.”
A certain railway bridge in Kettering springs to mind.
Safety is dangerous.
There is (on average) one death each day in the US at a railroad grade crossing.
I have heard tell that the rate of deaths and injuries to pedestrian children in the US is going up, not down, and one of the reasons I have heard as being to blame for this is the design of roadaways in the sprawling exurb subdivisions which surround all major US cities. Apparently (I have heard tell), for the longest time, these roads were designed so that two fully-loaded fire trucks could pass each other – at emergency speeds. Safely. This led to roads that are wide, and straight, and completely free of any sort of traffic-calming design that might reduce speeds. The instances when fully-loaded fire trucks have to traverse these roads are vanishingly-rare. But they are driven on by the general public every day. And, funnily enough, roads that are designed to be driven at higher speeds, ARE driven at higher speeds, and so the rates of accidents between fast-moving cars and ball-chasing and bicycle-wobbling children are going up, not down.
The perfect is the enemy of the good. And the moment that you allow a complex system to be influenced without limit by a single parameter or user community is the moment that the Law of Unintended Consequences rears up to bite you. Governments never seem to be able to grasp this simple truth.
llater,
llamas
See also this:
http://www.samizdata.net/2013/01/16224/
Level crossings are bad for traffic congestion, too, as cars have to stop and wait every time a train comes along. This has additional safety consequences, but they are not easy to quantify. Also, for busy urban railway lines, the presence of level crossings can put restrictions on the number of trains that can use a railway, as the road has to be open a certain number of minutes per hour to allow traffic to pass. If rail capacity is limited in this way, there are going to be various knock-on effects in other places, one of which will be some (hard to tell how large, but non-zero) number of people getting into their (significantly less safe) cars instead of riding the train.
There was a proposal to introduce train service called “Heathrow Airtrack”, to run from Waterloo Station to Heathrow airport. This would have been highly useful to those of us who live south of the Thames, as public transport options to Heathrow from south of the river are not especially great compared to those from places north of the river. The proposal would have used the (presently unused) former Eurostar platforms at Waterloo, unused capacity on existing rail lines for much of the journey, reopening of a closed section of railway for one portion of the journey, and only a small section of new track right at the airport. As these things go, it would have been quite cheap to built.
However, the proposal was eventually cancelled due to the number of level crossings on the existing railways on the route. Running the additional trains to Heathrow on the lines would have increased the number of minutes per hour the roads were closed to a sufficient extent that the consequences with respect to increased traffic congestion were considered too severe, so the project was cancelled. Building bridges across the level crossings using current regulations was sufficiently expensive as to make the project uneconomic, so no trains from Waterloo to Heathrow, which is a shame.
Looking on the bright side, ‘thinning the herd’ is supposed to improve the species….
Indeed, aren’t most such accidents the results of people ignoring the quite generous padding between the gates & lights and the train actually arising.
Actually, that’s possibly a point in support of the original premise. If the time (three minutes if I recall correctly) was reduced, people might be less tempted to “cheat”.
A British author, Ken McLeod, has written a darkly satirical dystopian novel about Britain in the future, where, for instance, women are being relegated back to the home because of an expansion of health and safety laws- the Labour government justifies this as an enlightened example of the perfect market, because individuals can’t know enough to really know what is best for them, but the government can, and therefore should make these decisions for them! I think it is called ‘Intervention’.
It’s ironic that the banning of smoking on planes caused the air less to be less fresh.
The ratio of accidental strandings of vehicles on crossings to people deciding to drive into the path of a train would make interesting reading. Haven’t these luck-pushers seen what a moving train does to a stationary car?
Health and safety rules with respect to railways also generally only apply to new infrastructure. A safety rule comes in – much stricter than old safety rules, as these things always are – and grandfather clauses often exclude all the tracks, bridges, stations etc that already exist from the new rule, as to close half the network due to the new rule would be ridiculous. This can lead to money being spent in the wrong places – upgrading the safety aspects of old infrastructure can often lead to much greater safety improvements per pound spent than the additional money spent on new infrastructure with ultra-strict safety rules compared to that spent on new infrastructure simply with current good practice. (Modern infrastructure is built with stronger and otherwise better materials and design than old infrastructure, just due to the advance of technology and without explicitly thinking about safety). In addition, there are perverse operational incentives. For instance, if a railway line remains open, the grandfathering in the regulations continues to apply. If the railway is closed, it cannot reopen without the new safety rules being in place. This can prevent reopening of infrastructure in cases when it might be useful, and this can be one (but not the only) reason why railway lines will sometimes be kept running with one service a week in one direction at 5am on a Wednesday rather than closed entirely.
Regional
January 16, 2013 at 4:42 am
It’s ironic that the banning of smoking on planes caused the air less to be less fresh.
Not only that, but the incidence of toilet fires dramatically went up when the ban came in.
While I was working for the airlines, one instructor informed us that ‘complacency’ was the biggest threat to safety, the ‘It’ll never happen to me’ mentality. I cant think how you would counteract that unless you allow small scale hazards to occur occasionally.
Ironic, too that near to my house the authorities introduced a one way road system to improve traffic flow and then introduced road humps to slow it down!!!
“The perfect is the enemy of the good.”
And the good is the enemy of the perfect.
There are times when a very useful upgrade doesn’t happen because the existing system is adequate, and there is insufficient discontent to overcome inertia.
It happens a lot in politics: parties and officeholders learn to be good enough not to alienate a motivated majority. Those who don’t get the chop. (Sometimes literally, e.g. Charles I and Louis XVI.) It’s evolution in action.
Brian:
Rich Rostrom:
Moi:
(Of course, what I want to write is, “No blanket rule is always right,” but– 🙁 )
Michael Jennings makes some very good points and it’s refreshing to see people talking sense on this instead of the usual nonsense. I’m coming at this from an insider’s perspective as until retirement last year I had spent most of my railway career working signalboxes with level crossings and in all that time I never witnessed one accident, that’s an indication of how safe controlled level crossings actually are. It’s actually minor crossings which are the dangerous ones, many of which have no protection at all apart from user worked gates, not even that in some cases. Misuse of crossings is rife but those that are controlled or monitored are largely accident free, those that aren’t can be death traps for the unwary or stupid. I’m in two minds about the solution to this, on the one hand I’d agree that getting rid of these places is a good idea but there are hundreds of them and even if the replacement procedure was improved and costs reduced it would still be hideously expensive. I’m inclined to the view, which is common amongst rail workers, although it’s not considered the done thing to say so by Network Rail, that stupidity shouldn’t be reinforced, if people are daft enough to do dangerous things on level crossings then no one should be too bothered if they end up dead. However there’s also the cost of that to consider and especially the effect on train drivers and other staff who have to deal with the results which are usually pretty horrific.
Thornavis’ comment reminds me of the town in–was it Denmark or the Netherlands? The latter I think–which had an intersection, with a stoplight, that was so accident-prone that the city council (or whatever it’s properly called) finally threw up its hands and removed the stoplights, leaving the intersection completely unmarked. I suppose you all know this story, but anyway, the punchline is that this had the effect of reducing the accidents there to zero. (I believe I’ve subsequently read that they removed all the stop signs and lights in the town. Won’t swear to it though.)
The theory is that once people knew they had to look out for themselves, the lights failing to do it for them, they became interested in being on the alert for crossing traffic.
I can think of other explanations, but anyway, there it is.
Julie
The problem with that, in relation to level crossings, is that trains and traffic don’t mix very well, certainly not trains of the sort of speeds and weights we find in the UK and even more so the US – there’s some really spectacular Youtube images of grade crossing accidents in the US. Even light rail is vulnerable to this despite lower speeds and people being, theoretically at least, more aware of the presence of trams.
I ought to clarify somewhat my comment about never having witnessed an accident on a crossing. I saw plenty of ‘incidents’ as we term them but nothing involving serious injury, still less death. The only train/vehicle accident we experienced at the box I worked in for the last twenty four years of my railway career was at an unmonitored user worked crossing where a van driver drove his vehicle across the line and was struck by a train, derailing it and wrecking the van, which sounds like carnage but amazingly nobody received more than minor injuries. This box has ten CCTV monitored crossings all of them very busy both for rail and road traffic and yet there has never been any major incident at any of them. If they were left unprotected I doubt that would be the case, in fact I know it wouldn’t be.
Thornavis, thanks for the reply and information. As I said, I can think of other possible explanations for the success in the road-intersection case.
Yes, we have many RR-Xing accidents here.
What’s wrong with gates (those wooden barricades, if they are still wood) that swing down from BOTH sides and overlap for a ways in the center, with virtually no clearance between them, on both sides of the track? (I’m sure there’s a problem — I just don’t see what it is.)
You have “user-operated” gates? You mean the gates are always closed unless a driver (or pedestrian) leaves his vehicle to push a button to raise them?
Julie
The old fashioned wooden gates some of which still exist, in Britain they always operated in the horizontal plane, were actually more prone to accidents than modern lifting barriers. There are a number of technical reasons for this which I won’t bore you with but even with a keeper on site to work them accidents were very common.
As for user operated gates, yes that’s correct they have to be opened by the user, which in the case of a vehicle driver means getting out opening the far gate first, then the nearer and then driving across, repeating the process on arrival, all the while checking constantly to make sure there is no train approaching. That’s the correct way of doing it and as you can see even that is inherently risky. When you consider that many users don’t follow the correct procedure you can see why there are so many accidents. Also a lot of crossings of this kind have no gates or barriers at all, some have road traffic type lights but many are completely unprotected. There are far too many instances of people just driving walking or cycling straight across without looking and getting killed as a result. As a railwayman I find this inexplicable because I know how dangerous trains are but most people just don’t appreciate that and disregard them in a way they never would with road vehicles. Another new hazard is earphones from mobiles or music pods, cyclists in particular seem to think it perfectly OK to wear these when riding and there have been a number of railway fatalities as a result. As I said before, what can you do about stupidity ?
Ah, I see. Ours swing up and down in the vertical plane. Unfortunately some are far enough apart that you can get to the tracks by weaving through them, and the rest don’t even have a barrier across the other lane.
I will admit there’s sometimes a reason to defeat the one-barrier type, and that’s when a train has stopped for some reason after setting the lower-the-gate switch. Typically I’d say 1/4-1/2 mile away. Then, after you watch a long time and it’s obviously not moving, you go ahead. But I don’t think those are the circumstances of any crash I’ve heard of.
I grew up in the country back in the 40’s and 50’s, and very few country crossings were protected. Busy lines often did have RR-Xing “towers”–I’m sure you know, a large black-and-white-striped “X” made of two timbers, attached to a tall wooden post, sometimes with two lights next to each other that would flash red alternately when a train approached. No sound. But a lot of crossings just had a standard yellow sign the size of a regular stop sign, that said “Caution–RR Crossing.” And I still see those sometimes, out in the country.
We had our share of near misses. I have no idea what the casualty rate was. But one of the things stressed HARD by our driving teacher (Grandpa, in my case) was STOP, LOOK, LISTEN. In fact they taught us that in grade school, for pedestrians and the kids who rode bicycles (that would be everybody, almost).
When RR traffic began to decrease significantly on some of the rural lines we started having more problems–carelessness because the likelihood of an oncoming train was less.
You would think that with all the movies over the last umpty years that feature murder-by-train (bad guy shoves good guy’s car onto tracks) people would be well aware that trains are nothing to mess with.
Julie, i think that the town was Boehme, or some such name, and i think it might be in the Netherlands. It was some time ago, but it made a lot of headlines at the time!
Thanks, Nick. I thought it might be the Netherlands. –Can we just say “Holland”? I’d swear there was a followup story to the effect that a few years later, a lot of towns had done it.
Julie: yes, I remember that too.
I think the effect in play at “unguarded” road intersections is that right of way cannot be assumed by any party and a sort of visual negotiation is undertaken by motorists meeting at the intersection. Although some drivers may abuse the priveledge, sanity will prevail over most others and they will yield to the one asserting a right of way. With Trains and road traffic the train has an assumed and absolutely necessary right of way which leaves no room for any road vehicle to assert a difference of opinion (albeit often inadvertantly or unknowingly as often the trains are out of sight and unheard even when close).
Another observation is that the German Autobahns where no speed limit is in place are safer and cause less road rage because the slower users are always aware that faster traffic could be coming up rapidly behind them. Slower traffic generally keep to the slow lanes and rarely deliberately hinder faster traffic as many self appointed “policemen” do when speed limits are in place. Again here there is no assumed obligation upon the other driver to yield or to obey some regulation to stay under an arbitrary speed so sanity and courtesy prevails. The exception here i have often observed is what I assume to be (probably american) tourists driving underpowered hire cars staying solidly in the overtaking lanes at 100 km/h or less and not moving over to allow fast traffic (often coming up at 200 km/h) to pass unhindered.
I have also noticed, from spending some time in Italy, that Italian drivers are notorious for lack of “road discipline” and adherence to the traffic rules or lane markings, etc. And yet the accident rate in Italy is often better than that in Germany or Britain where obeying the road rules is often regarded as almost a sacred duty and assumed right of way is often defended to the point of causing accidents. In Italy most drivers (except the tourists from countries where road rules are regared as holy writs again) expect the unexpected from other drivers and allow themselves margin to correct.
Of course that logic cannot be taken to extreme, because then you get Africa.
An observation from commuting many years by bicycle through city centres is that pedestrians do not look out for traffic. I can only assume that they rely on their hearing but bicycles and electric (or hybrid ) cars are silent and give no aural warning of their approach. I have collided with pedestrians, at high (for a bicycle) speed on at least 3 occasions and in every case they simply walked into my path without looking. I forsee a lot more pedestrian accidents involving electric cars in the future. I suspect this effect is what happens at railway crossings, where drivers do not look but implicitly rely on there being some other undefined cue to alert them of the presence of a train.
This is not really an issue.
We are only talking about a couple of feet maximum. Considering that wholescale electrification is now on the agenda. Plus slowly there is a programme to raise clearances so larger freight trains can run. There is not point building new infrastructure to old standards, especially as the cost difference between them is not that high.
So it would be more expensive rebuild these bridges in the future.
As for the Airtrack scheme. It was ill thought out. The idea was good, but if they wanted to plough a viaduct through the town centre of Staines and double the length the barriers were down on the other side of the river then it should not have cancelled the new town centre station it promised. it was this that allowed people to use extra services from the West and South of town. Once they cancelled that. what was the point of routeing it through the town centre (other than it being cheaper),they should have just stuck it down the side the Motorway avoiding the town centre and the level crossings.
It could be revived though with just a link to Central London via Staines. That way it avoids the viaduct and the level crossings and would still serve the majority of the traffic. others could just change at Staines or Clapham Junction.
Bridging the level crossings in this area was a non starter as it would mean dozens of house no longer having driveways or even curbside parking.