Absolute safety at any cost?

[Bletchleyite]

From the HS2 thread:

This is getting off-topic here, but I don't believe it at all acceptable to say that the level of safety and security can be reduced simply for the sake of cost-cutting. Human lives should not be measured in currency, and even if "100% safety" is not achievable, doesn't mean that we shouldn't be taking every measure in our power to eliminate all potential risks. Britain's railway network is statistically one of the safest in the world, after all.

But human lives are measured in currency, as there's only so much money available to spend. So if £1m will save more lives by being spent on roads rather than rail, or it's best not to spend it and so to cut fares to get people out of cars onto trains, shouldn't this be considered, rather than just considering rail as an island?

Note: this is not intended to be a DOO thread, while DOO is relevant to it as one method of cost cutting while potentially reducing safety, it's to discuss the issue more widely.

 

[eastdyke]

From the HS2 thread:
But human lives are measured in currency, as there's only so much money available to spend. So if £1m will save more lives by being spent on roads rather than rail, or it's best not to spend it and so to cut fares to get people out of cars onto trains, shouldn't this be considered, rather than just considering rail as an island?

Note: this is not intended to be a DOO thread, while DOO is relevant to it as one method of cost cutting while potentially reducing safety, it's to discuss the issue more widely.

But (simplistically) cutting rail fares to get people out of cars doesn't work on its own does it?

You need more rail capacity in the form of speed/trains/routes (eg HS2) to do that, which would take back your £1m.

 

[Deepgreen]

But (simplistically) cutting rail fares to get people out of cars doesn't work on its own does it?

You need more rail capacity in the form of speed/trains/routes (eg HS2) to do that, which would take back your £1m.

Or, does the simple process of transferring people from road to rail, without increasing rail capacity (i.e. cramming more people into trains which still have a vastly reduced safety risk than the road option) do the trick? This, of course, leaves out the huge range of social dissatisfaction and political ramifications, but is based on the pure safety element alone.

 

[DelW]

I think the last passenger fatality on British railways was at Grayrigg in 2005. Since then over 20,000 people have been killed on British roads (around 3,000 per year in the mid 2000s, dropping to around 1700 per year since 2012).

So it ought to be more effective to target safety improvement funding at roads rather than railways. However assessing the number of fatalities (or injuries) avoided per amount of money spent is not a straightforward calculation.

 

[Bletchleyite]

Or, does the simple process of transferring people from road to rail, without increasing rail capacity (i.e. cramming more people into trains which still have a vastly reduced safety risk than the road option) do the trick? This, of course, leaves out the huge range of social dissatisfaction and political ramifications, but is based on the pure safety element alone.

Or could spending that £1m on rail capacity rather than rail safety attract more people out of cars?

Possible example: allowing traditional manual SDO at all stations rather than requiring beacons, allowing train lengthening without platform lengthening. SDO itself is cheap, in a modern train it's software. Beacons etc are much less cheap.

 

[matacaster]

I think the last passenger fatality on British railways was at Grayrigg in 2005. Since then over 20,000 people have been killed on British roads (around 3,000 per year in the mid 2000s, dropping to around 1700 per year since 2012).

So it ought to be more effective to target safety improvement funding at roads rather than railways. However assessing the number of fatalities (or injuries) avoided per amount of money spent is not a straightforward calculation.

A day spent in Bradford or on the M62 would surely convince you that spending a large chunk of money on "Road safety" is often a complete waste of time when many accidents are caused by reckless morons in their souped up cars, taxi drivers doing U turns (actually saw one doing the same in the middle of some traffic lights) and HGV's which feature in many serious accidents (although not always the cause in fairness).

I am with Neil on this. Rail has priced itself out of certain markets because the cost of rail travel is being ramped up by the quest for ultimate safety - which is not achievable. Overall deaths would be reduced if less people travelled by road and more by rail. Yes, more capacity is needed on rail to cope, but if the safety goal is kept to a realistic level, then such costs may not be so bad.

An example is the amount of money spent on metal fencing (largely absent from foreign railways) to prevent people trespassing onto the tracks? Yet their ability to kill themselves running onto a road is largely unchecked and no one seems to give a damn.

 

[Bellbell]

The key question is would the money saved on x (in this thread, as per your example) rail, be spent on y (road) or not. Risk assessment and management does accept that the financial costs can be simply too high to mitigate every single risk and I understand that, but couldn't accept funds being limited under the implication that the money would be spent on safety elsewhere if that were not actually the case.

Assuming the savings would go towards safety in another form, it's then a big discussion as to how to use that money, as the posts above show. Spending money on x, even if you've nominally cut spending on x might give safety improvements in y, or might not. One difficulty is assessing the effects of any scheme and if unsuccessful removing the funding - that in itself might cost money.

Interesting discussion though.

 

[tbtc]

"Safe" and "unsafe" get used a lot on here to represent some kind of absolutist black/white situation, where the reality is that there are several shades of grey.

Having two safety critical members of staff on board sounds "safer" than having only one safety critical member of staff on board - but then presumably having three would be safer still.

Where do you draw the line? On the rails, do we go back to "double manning" the front cab on 125mph services? That'd be "safer", but the status quo seems safe enough to most people.

Similarly, on the roads, should we insist that all cars have a yellow bonnet to improve visibility (in line with the way that the railway has been for generations)? That'd be "safer", but can you see any Government suggesting it?

I'd rather that we accepted that there are acceptable levels of risk and focussed resources on where the best safety improvements can be made.

Roads seem "safe" enough for me (given the number of deaths/ collisions per passenger mile), but there are some roads that are less safe than others - especially "local" roads near trunk routes, where motorists come off the Motorway (etc) still driving like they are on a fast segregated road and fail to adjust to the more complex situation on a single carriageway road that may see pedestrians crossing or have obstructed junctions.

Maybe focus more on the types of vehicle that seem to be involved in a disproportionate number of collisions (sports cars, HGVs)?

The NHS have NICE who have to rationally assess the benefits of one drug against another based on the cost to extend life (at a suitable standard of living - "quality of life") - maybe we should have the same rational approach to transport - based on hard statistics rather than an emotive "safe"/"unsafe" approach? Dunno.

 

[quantinghome]

A day spent in Bradford or on the M62 would surely convince you that spending a large chunk of money on "Road safety" is often a complete waste of time when many accidents are caused by reckless morons in their souped up cars, taxi drivers doing U turns (actually saw one doing the same in the middle of some traffic lights) and HGV's which feature in many serious accidents (although not always the cause in fairness).

I believe road safety traditionally followed the 'Three Es' - Engineering, Education and Enforcement. It's not just about redesigning road junctions or sticking in traffic lights (although that may help). Spending money on targeted enforcement would help reduce deliberately anti-social driving. HGVs can be redesigned so they have better visibility etc.

 

[Moonshot]

If anyone thinks rail safety is expensive,try having an accident.

 

[Bletchleyite]

Spending money on targeted enforcement would help reduce deliberately anti-social driving.

Indeed, though at present the trend is that outside London nothing seems to be enforced other than what can be enforced by camera. This makes little sense, as fines from enforcement could be used to fund more "manual" enforcement.

 

[Bletchleyite]

If anyone thinks rail safety is expensive,try having an accident.

There is that aspect. One example of it is anti-suicide fencing blocking fast line platforms. While this certainly saves lives, that isn't the main reason for it - its primary reason it to reduce the cost of disruption from suicides.

 

[xotGD]

Reducing speed limits on the roads would have the multiple benefits of reducing casualties, reducing CO2 emissions from better fuel efficiency and encouraging more people to switch to rail to save time. I suppose someone will claim that this would have a huge cost to the economy, but I see the upsides as being much greater than the impact of slowing down a few sales execs who can't do 90 in the fast lane any more.

 

[Bletchleyite]

Reducing speed limits on the roads would have the multiple benefits of reducing casualties, reducing CO2 emissions from better fuel efficiency and encouraging more people to switch to rail to save time. I suppose someone will claim that this would have a huge cost to the economy, but I see the upsides as being much greater than the impact of slowing down a few sales execs who can't do 90 in the fast lane any more.

That isn't reducing speed limits, it's increasing enforcement to ensure 70 means 70, rather than it meaning 90 as it does at present. The increased rollout of smart motorways will do that (making the cameras issue tickets at 71 is a simple matter of configuration), coupled with an increased prevalence of black box based insurance, whereby speeding[1] will cause hefty premium increases.

Though there are some reductions that would work. A blanket reduction to 50mph on single carriageways and 60mph on non-motorway dual carriageways (with the option to uprate some to 70 on a case by case basis, e.g. pseudo-motorway A roads like the North Wales Expressway and the A5D around MK) has a very strong case. The difference in attention required between 50 and 60 is noticeable on a typical rural single carriageway. If well-enforced it would also reduce risks from overtaking, as lorries and cars alike can do 50mph on single carriageways (this was recently upped from 40). Some Councils have a programme of doing this anyway, but it requires signage so is costly.

[1] I've got a Tracker Locate in my Landy, and I can view a console telling me of any speeding against their speed limit map, and lots of other telemetrics. My insurer doesn't get to see it[2], however - they would need to reward me with reduced premia for me to let them look

[2] Not that it'd be interesting, I tend not to speed anyway, not to mention that when driving a Land Rover Defender it is actually quite hard to speed in most of MK anyway - the limits are just too high for that to be feasible!

 

[eastdyke]

Or could spending that £1m on rail capacity rather than rail safety attract more people out of cars?

Possible example: allowing traditional manual SDO at all stations rather than requiring beacons, allowing train lengthening without platform lengthening. SDO itself is cheap, in a modern train it's software. Beacons etc are much less cheap.

Another possible example (that won't to happen):
Rail capacity enhancements (speed/frequency) without removal or upgrading of level crossings?

I do agree with (the principle of) looking to do stuff less expensively though.

 

[edwin_m]

The NHS have NICE who have to rationally assess the benefits of one drug against another based on the cost to extend life (at a suitable standard of living - "quality of life") - maybe we should have the same rational approach to transport - based on hard statistics rather than an emotive "safe"/"unsafe" approach? Dunno.

Something similar is done on the railway and the amount worth spending to avert a fatality is the same (or about the same) as on the roads. If I recall correctly it was initially set based on research asking people how much money they would be prepared to spend to save the life of someone they didn't know. It is essentially a method of establishing whether a risk is As Low As Reasonably Practicable.

One notable consequence of this policy was the decision to adopt TPWS instead of a nationwide ATP system in the 1990s. All accident reports since 1968 were reviewed closely to decide how the consequences would have been different, if at all, with either TPWS or ATP. This predicted that TPWS would have averted about 70% of the casualties of ATP, something treated with some disdain by Uff-Cullen but broadly borne out since. As originally conceived TPWS was within the value of preventing a fatality at the time, though as often happens it collected a few more bells and whistles during development which pushed it over. By contrast I don't know the extra cost of "crumple zones" at the ends of intermediate cars of units but I don't believe they have saved a single life in the 20 or so years since they first appeared (anyone who thinks of Lambrigg should go and read the report).

It gets more difficult when there is a trade-off between safety and some element that isn't safety related but affects how attractive the train service is and therefore how many people use it. One example might be the discussion on another thread where defensive driving can lead to some reduction in accidents, which can be monetised, but it's less easy to quantify the downside of slower services and less capacity.

 

[Chris M]

It's been a while since I've read the Grayrigg report, but I'm sure I've read of other accidents where crumple zones have likely (it's difficult to prove of course) saved lives and serious injuries. They reduce the amount of energy transmitted to the rest of the coach by absorbing it in a relatively controlled manner in the event of an end-on impact. There haven't been that many of those of course (which is a good thing).

I'm not in a position atm to search for anything to back up my recollections but will do later if nobody beats me to it.

 

[Bletchleyite]

TPWS is a good example - and it works well.

 

[philthetube]

It gets more difficult when there is a trade-off between safety and some element that isn't safety related but affects how attractive the train service is and therefore how many people use it. One example might be the discussion on another thread where defensive driving can lead to some reduction in accidents, which can be monetised, but it's less easy to quantify the downside of slower services and less capacity.

The main reason for defensive driving, in the main, isn't to avoid accidents but to prevent the consequences to the driver of a relatively minor mistake.

 

[3141]

Another angle to the issues raised by this thread is the larger clearances for OHLE, and the higher bridge parapet requirements, which have contributed to the rising cost of electrification. Most of these requirements do not address any actual problem. But they have helped to make further large-scale electrification unlikely.

As a result, many routes will continue to be operated by diesel-powered trains, and there's currently considerable attention being given to the pollution diesel engines cause. It will be with us on some routes for another 35 years at least. The cost of the safety requirements that have helped to make electrification unaffordable includes the extra weight of bimode trains that shouldn't have been required, and hence their higher running costs, in addition to their higher capital costs to start with.

 

[RichmondCommu]

An example is the amount of money spent on metal fencing (largely absent from foreign railways) to prevent people trespassing onto the tracks? Yet their ability to kill themselves running onto a road is largely unchecked and no one seems to give a damn.

I really don't understand why anyone has a problem with metal fencing on our rail network unless they themselves wish to trespass. Anything that keeps idiots off our railway can only be a good thing and our rail safety record is the envy of the rest of Europe.

If there was a better (and feasible) way of stopping poor souls from getting killed on our road network you can bet your bottom dollar that the Government would be all over it.

 

[miami]

I really don't understand why anyone has a problem with metal fencing on our rail network unless they themselves wish to trespass. Anything that keeps idiots off our railway can only be a good thing and our rail safety record is the envy of the rest of Europe.

If there was a better (and feasible) way of stopping poor souls from getting killed on our road network you can bet your bottom dollar that the Government would be all over it.

If the fencing was free, that would be fine. If it takes money away from more useful schemes (like electrification) then that's where the problem comes in

 

[Chris M]

Fencing saves more money than it costs when you tot up the cost of injured and killed trespassers (intentional or otherwise):
Delays to trains (delay repay, train crew overtime, delay attribution staff at TOC and NWR, delay attribution payments to/from TOC)
Station staff costs (overtime)
Cleanup of p-way and train (staff time, materials)
Emotional recovery time off for staff involved (possibly leading to overtime payments for others)
Staff time for investigations, inquests, coroner's court, etc.

Then the there is money that it costs the exchequer not directly via the railways:
Emergency services (police, ambulance, coroner, fire brigade in some circumstances)
Hospitals
Sick leave/compassionate leave for victim and/or relatives
etc.

All this is money that could otherwise be spent on electrification.

 

[RichmondCommu]

If the fencing was free, that would be fine. If it takes money away from more useful schemes (like electrification) then that's where the problem comes in

Can you put a price on preventing someone from getting on to the tracks and ultimately getting themselves killed? I'd say that was very useful.

 

[edwin_m]

It's been a while since I've read the Grayrigg report, but I'm sure I've read of other accidents where crumple zones have likely (it's difficult to prove of course) saved lives and serious injuries. They reduce the amount of energy transmitted to the rest of the coach by absorbing it in a relatively controlled manner in the event of an end-on impact. There haven't been that many of those of course (which is a good thing).

I'm not in a position atm to search for anything to back up my recollections but will do later if nobody beats me to it.

Any you can think of are most welcome. As I posted I can think of none.

The type of accident mitigated by crumple zones tends to be medium-speed end-on collisions, which are also mitigated by TPWS so are much less common than they were a couple of decades back. Crumple zones in cab ends may be of value in mitigating low speed collisions and collisions with HGVs on crossings.

 

[TimG]

Can you put a price on preventing someone from getting on to the tracks and ultimately getting themselves killed? I'd say that was very useful.

Yes you can. That's the whole point of the thread!

 

[lejog]

The NHS have NICE who have to rationally assess the benefits of one drug against another based on the cost to extend life (at a suitable standard of living - "quality of life") - maybe we should have the same rational approach to transport - based on hard statistics rather than an emotive "safe"/"unsafe" approach? Dunno.

And the acceptable cost is:

The health benefits are expressed as quality-adjusted life years (QALYs).Generally, we consider that interventions costing the NHS less than £20,000 per QALY gained are cost effective. Those costing between £20,000 and £30,000 per QALY gained may also be deemed cost effective, if certain conditions are satisfied.

https://www.nice.org.uk/advice/lgb10/chapter/judging-the-cost-effectiveness-of-public-health-activities

If you multiply this by 40 (the UK average remaining life expectancy), you get around £800k-£1m, which I think is also around the value of a life used by the courts when assessing damages.

 

[lejog]

I should add the large sums of money spent on safety in the transport sector (air and car as well as rail) appears to be a result of avoiding the headline news caused by major accidents, which has the potential for a large effect on revenue and even affect the viability of the industry.

But deaths from suicides say or perhaps the occasional fatality on a platform which don't make the news are a different matter.

 

[furnessvale]

Yes you can. That's the whole point of the thread!

Everything has a price including safety.

That price translates into the cost of a ticket. Have we reached, or even passed, the point where the safety add on cost of a ticket means people travel by a cheaper, but less safe mode i.e. road?

 

[miami]

Fencing saves more money than it costs when you tot up the cost of injured and killed trespassers (intentional or otherwise)

And if that's the case, then it's a no-brainer.

Presumably the cost and perhaps impracticality of installing and maintaining platform edge doors would not be value for money though.

While some platforms have to serve a variety of stock, many stations don't (say Acton Bridge, which I believe only serves LM 350s). The question is: is the money better spent elsewhere.

Another option that would improve safety even more than metal fences would be to encase the railway in corregated iron, perhaps with landscaping over the top. Basically a cut-and-cover. This would completely prevent people scaling the fences, prevent jumpers at bridges no matter how prepared they are, or how remote the bridge, and have added benefits of preventing storms from bringing down wires and reducing noise pollution too.

The downside as ever is the cost - there are better places to spend money to save lives.