Smart motorways are (not) more dangerous

[Boustrophedon]

The US highways that do this do it by having a string of concrete barriers that a special equipped vehicle can move from one side of the shared lanes to the other in the middle of the day/night.

The other method is to have HOV/express lanes that are separated from the main lanes in both sides, and which are only open to toll-paying/vehicles with 2 or more passengers, and which can be closed for each direction as appropriate.

These existed on the M25 for a while but they sold off the barrier moving vehicle. Some were installed installed on the M20. https://www.highwaysindustry.com/moveab ... n-the-m20/

[bob sterman]

In the UK you also have a lot of major trunk A-roads with the same 70mph speed limit as motorways, no hard shoulder, bicycles and milk floats allowed (IIRC - maybe only sometimes?), and vehicles joining from stop signs coming out of little side at 90 degrees.

True. But on many A-road sections there is a grassy "soft shoulder" that you stand a good chance of being able to get most of your car onto. Whereas on new smart motorways with all lane running there is a crash barrier to prevent vehicles leaving the carriageway.

No idea whether lack of crash barriers and an accessible "soft shoulder" (perhaps with trees to crash into) is on balance a safer alternative - but it does give you the option of leaving the carriageway.

[Grumble]

Driving down a smart motorway this morning there was someone parked up on the grass verge at the side. Luckily there was a verge at that point.

[IvanV]

This country has a big problem that building infrastructure here is unduly expensive in comparison to our continental neighbours. As we are not richer than them, indeed mostly they are richer than us, this means we can't afford to build as much infrastructure as them. And since infrastructure is what lubricates our economy, it results in our economy being rickety.

The worst example is the railways. Railway construction seems to be about 3 times what it costs on the continent.

The countries that are more expensive than us for building roads in Europe are Norway, Switzerland and Austria. There is a reason building roads is very expensive precisely in throse three countries. They are very mountainous. And have a high cost labour too. So, adjusting for terrain and labour costs, I think we come out about 50% more expensive than the continental benchmark for road-building cost.

The present discussion is actually an illustration of one of the contributory reasons to this problem. Somehow the British public succeed in demanding higher standards of safety in the construction of our infrastructure than is normal on the continent. To the extent that the extra money spent on making each piece of infrastructure a bit safer would be better spent elsewhere in terms of saving lives. As you have noticed further up the thread, motorways are very safe roads in comparison to the average. 80% of traffic is on other roads. ALR smart motorways are similarly safe to unimproved older bits of motorway. And they will be a bit better when they get round to putting in the automated stopped vehicle detectors that they originally promised, and which the government has now told them they really have to install.

So ALR smart motorways are a relatively cheap way of increasing the capacity of a motorway, and produces a road safer than most other roads in the country. If that's not acceptable to you, you are in effect insisting on the goverment spending money saving a few lives, when money that would have saved many more lives if it had been spent somewhere else. (A trick that the railway operators have pulled off like almost no other.)

There are people who say they don't want more roads, because more roads just increases traffic. Yes, of course new roads increase traffic. It's kind of the point. If you have no roads, you have no traffic. Build a road and you get traffic. Build another road and you get more. Remains true as you make further improvements. he simple reason is that if you make it less annoying for people to make a journey, they are more likely to make it, or make it more often. And it is also obviously true of improvements to any other transport mode. It's not controversial, or shouldn't be. But it's not a relevant reason that means you shouldn't have more roads. 0 roads is clearly not the correct amount of roads.

So, the question I ask to people who don't want more roads, or road capacity expansions, is, "What is the right amount of roads?" An Ethiopian quantity of roads? When should we have stopped building/expanding them? In 1930? In 1950? In 1970? Were those the right amount of roads? Today's roads are just right but no more?

Perhaps we should have more railway lines. But that would be easier to support if we managed to build railways to continental cost levels. HS2 is costing something like twice the cost of the entire high speed railway program of Spain to date.

[shpalman]

some more stuff again

[nekomatic]

So, the question I ask to people who don't want more roads, or road capacity expansions, is, "What is the right amount of roads?" An Ethiopian quantity of roads? When should we have stopped building/expanding them? In 1930? In 1950? In 1970? Were those the right amount of roads? Today's roads are just right but no more?

Clearly 'every square metre of the country is road' is also undesirable, so there must exist an optimum greater than zero but less than that. What that is will obviously depend on how we weight the various costs and benefits of roads, which may change over time but is likely to stay controversial. People who oppose any given road project probably feel that the many externalities outweigh the benefits, and their position should be tested not with a silly strawman but by examining whether both sides understand all the costs and all the benefits and what the basis is for their disagreement on their weighting.

Applying the same good logic that you advocated for smart motorways, the question is not 'should we build roads yes or no', it is 'with the resources available, is any given road project the best use of them?' It certainly seems defensible to me that the answer will sometimes be 'no'.

[Fishnut]

I went to a talk not long before the first lockdown where one of the speakers was from the Welsh Future Generations Commission. They talked about how all their legislation has to be considered in terms of the impact on future generations. The planned M4 relief road was scrapped in part because of this - it was decided that spending £1.6 billion on 13 miles of motorway was not the best use of that money,

Not only did the project not align with Wales’ carbon reduction targets, [the Future Generations Commissioner] said, it also went against the Act’s well-being goals of supporting the resilience of ecosystems and a healthier Wales. It also did not take into account future trends such as increased homeworking – something which has been fast forwarded as a result of COVID-19 and research suggests is likely to become a new normal with the potential to reduce congestion on our roads.

[science_fox]

One problem here is that non-smart motorways are also stupidly dangerous.

Less dangerous than normal roads.

In the UK you also have a lot of major trunk A-roads with the same 70mph speed limit as motorways, no hard shoulder, bicycles and milk floats allowed (IIRC - maybe only sometimes?), and vehicles joining from stop signs coming out of little side at 90 degrees.

60mph!
Signage specified dual carriageways are 70 - and yes open to bicycles milk floats et al. Junctions are usually with feeder lanes. But all other roads including those that have a central reservation but no dual carriageway sign are 60.

I'm not sure it makes a big difference to your argument, but thought we should at least get the speed limit right.

[Gfamily]

Less dangerous than normal roads.

In the UK you also have a lot of major trunk A-roads with the same 70mph speed limit as motorways, no hard shoulder, bicycles and milk floats allowed (IIRC - maybe only sometimes?), and vehicles joining from stop signs coming out of little side at 90 degrees.

60mph!
Signage specified dual carriageways are 70 - and yes open to bicycles milk floats et al. Junctions are usually with feeder lanes. But all other roads including those that have a central reservation but no dual carriageway sign are 60.

I'm not sure it makes a big difference to your argument, but thought we should at least get the speed limit right.

A pedant writes:

A dual carriageway is a road which has a central reservation to separate the carriageways

So specific "dual carriageway" signage is not required for the 70mph limit to apply.

https://www.highwaycodeuk.co.uk/multi-l ... eways.html
https://www.gov.uk/speed-limits

[bob sterman]

A pedant writes:

A dual carriageway is a road which has a central reservation to separate the carriageways

So specific "dual carriageway" signage is not required for the 70mph limit to apply.

I hope traffic islands are excluded from the definition of "central reservation" - otherwise they could create a loophole for (very short) drag races that involve ridiculous g-forces.

[science_fox]

A pedant writes:
the Highway Code wrote:
A dual carriageway is a road which has a central reservation to separate the carriageways
So specific "dual carriageway" signage is not required for the 70mph limit to apply.

Good pedanting!

[Sciolus]

I've always wondered what the national speed limit is on roundabouts.

But IME most dual carriageways either have lower limits (60 or 50 in rural areas) or have been semi-motorwayised with slip road access, straightening and the like. Rarely, they have a cyclists prohibited sign. But I've never seen a 70 dual carriageway with a hard shoulder. But (again) real-world speeds tend to be lower than motorways (less expectation that 70 is the minimum and 80+ is the norm). So overall I dunno.

[IvanV]

So, the question I ask to people who don't want more roads, or road capacity expansions, is, "What is the right amount of roads?" An Ethiopian quantity of roads? When should we have stopped building/expanding them? In 1930? In 1950? In 1970? Were those the right amount of roads? Today's roads are just right but no more?

Clearly 'every square metre of the country is road' is also undesirable, so there must exist an optimum greater than zero but less than that. What that is will obviously depend on how we weight the various costs and benefits of roads, which may change over time but is likely to stay controversial. People who oppose any given road project probably feel that the many externalities outweigh the benefits, and their position should be tested not with a silly strawman but by examining whether both sides understand all the costs and all the benefits and what the basis is for their disagreement on their weighting.

Applying the same good logic that you advocated for smart motorways, the question is not 'should we build roads yes or no', it is 'with the resources available, is any given road project the best use of them?' It certainly seems defensible to me that the answer will sometimes be 'no'.

Absolutely. And all public investment does have to be justified with quite elaborate analysis and cases. And Britain does this more carefully and consistently than most other countries.

Generally speaking, road projects that the government chooses to proceed with show quantified socio-economic returns that are high in comparison with most other public spending, and is high on the list for socio-economic return in comparison to alternative road investments. Sometimes they prioritise a scheme for another reason, and these will tend to be among the most controversial.

Stonehenge tunnel is one such particularly difficult case. It doesn't have a straightforward economic case - the road capacity benefit, and benefits of by-passing the small village of Winterbourne Stoke the road currently goes through, does not justify the cost of the tunnel by the usual criteria. The major additional benefit, that doesn't have a standard method for valuing it, is not having a busy traffic-jam-ridden road running visible/audible close by Stonehenge. The government chose to proceed with the scheme because it impliedly gave a large value to that benefit. Then there have been arguments about the damage caused by the actual selected scheme, which are again hard to value. Other schemes that reduce those damages would be much more expensive, and have other different local costs. It's all very difficult. Do nothing and you stil have a very busy traffic-jam-ridden road running close visible/audible from Stonehenge, which is not nice. You could just close the road, but the large volume of traffic would have to go somewhere.

[IvanV]

A pedant writes:
the Highway Code wrote:
A dual carriageway is a road which has a central reservation to separate the carriageways
So specific "dual carriageway" signage is not required for the 70mph limit to apply.

Good pedanting!

Not perfect pedanting. Speed limits are a matter of law and the Highway Code is not law. Although it may sometimes accurately state the law, it does not carefully distinguish between "advice" and "legal obligation". I'm pretty sure HC is right on this specific occasion, but for an authoritative pedanting on such matters, you should in general quote the law.

[Gfamily]

A pedant writes:
the Highway Code wrote:
A dual carriageway is a road which has a central reservation to separate the carriageways
So specific "dual carriageway" signage is not required for the 70mph limit to apply.

Good pedanting!

Not perfect pedanting. Speed limits are a matter of law and the Highway Code is not law. Although it may sometimes accurately state the law, it does not carefully distinguish between "advice" and "legal obligation". I'm pretty sure HC is right on this specific occasion, but for an authoritative pedanting on such matters, you should in general quote the law.

I understand its use of "You should" vs "You must" is indicative of whether a law is being mandated.

[IvanV]

(...Highway Code...)
I understand its use of "You should" vs "You must" is indicative of whether a law is being mandated.

Yes, you are right, it does indicate that, though a couple of the "musts" are questionable. What I should have said is that it tends to simplify, and if you need to know the details, or an authoritative statement, you have to go to legislation.

[TimW]

What we need (I think!) is
The 70 Miles Per Hour, 60 Miles Per Hour and 50 Miles Per Hour (Temporary Speed Limit) (Continuation) Order 1978

[headshot]

More or Less have a piece on this in their latest episode: https://www.bbc.co.uk/programmes/m000wsf0

[lpm]

Stonehenge tunnel is one such particularly difficult case. It doesn't have a straightforward economic case - the road capacity benefit, and benefits of by-passing the small village of Winterbourne Stoke the road currently goes through, does not justify the cost of the tunnel by the usual criteria. The major additional benefit, that doesn't have a standard method for valuing it, is not having a busy traffic-jam-ridden road running visible/audible close by Stonehenge. The government chose to proceed with the scheme because it impliedly gave a large value to that benefit. Then there have been arguments about the damage caused by the actual selected scheme, which are again hard to value. Other schemes that reduce those damages would be much more expensive, and have other different local costs. It's all very difficult. Do nothing and you stil have a very busy traffic-jam-ridden road running close visible/audible from Stonehenge, which is not nice. You could just close the road, but the large volume of traffic would have to go somewhere.

Government approval has been reversed by the High Court.

"The Transport Secretary had not properly assessed the risk of harm to each heritage asset within the World Heritage Site."

[El Pollo Diablo]

But I've never seen a 70 dual carriageway with a hard shoulder..

Here you are.

Edit: Or here

[shpalman]

National Highways adds emergency areas

It's almost as if being able to pull off the road is a good idea.

[Grumble]

National Highways adds emergency areas

It's almost as if being able to pull off the road is a good idea.

Yes, with the important caveat that a hard shoulder is not “off the road”.

[IvanV]

National Highways adds emergency areas
It's almost as if being able to pull off the road is a good idea.

Every road can be built to a higher standard and be much safer, but there's only so much money. Smart motorways have excellent safety levels, even among motorways. But there's a perception problem. People think, if I break down, something might crash into me. I can't get out of the way. What they overlook, and frankly is very hard for them to understand and evaluate intuitively, is the similarity with the very common situation of coming to a quick halt in traffic congestion, when someone can equally crash into you. One objective of smart motorways is to reduce the frequency of coming to a quick halt in congestion, and so overall, reduce the number of situations where you come to a halt and following vehicles might crash into them, even when the break-down situation is added in.

One of the amazing things we do in this country, and those have been reducing of late, is that we have Post Opening Project Evaluations of all major highway schemes. We even do it twice, at 1 year and 5 years, though currently National Highways is trying to argue it down to once at 3 years, because it is getting rather behind on its POPEs.

Here's a fascinating recent POPE. It's the 5-year POPE of M1 J10 to J13 smart motorway. It's a dynamic smart motorway, the most despised kind, with gantry signs opening and closing the hard shoulder to traffic when traffic levels demand.

The number of casualties per vehicle-km has materially reduced, even though the hard shoulder is unavailable when the road is busy. Ironically the average speed on the motorway has reduced, but journey times are more predictable, which it seems that road travellers on balance find that attractive. The road capacity, and the traffic it carries, has increased by more than the amount you'd expect given the increase in the amount of running lane capacity. Given that's more than traffic growth in general, there's a strong implication the motorway has improved service sufficiently to attract traffic off surrounding roads. And since many of those are not motorways, which are much the safest roads, it has transferred traffic from more dangerous roads to much safer roads, at the same time as reducing the casualty rate on the safer road. So probably it has also reduced accidents - and congestion - on the surrounding road network, although there is no attempt to investigate or quantify that effect. What an amazingly successful scheme, for both road capacity and road safety.

But National Highways failed to understand the perception problem, and is now being made going back in with the stopped vehicle detection it promised, emergency areas at the frequency it promised for those. To be fair, it made those promises and didn't deliver on them. But actually, as this shows, in this case the scheme was still great. Going back in and doing them will make schemes like these even better. But that money could have had been spent elsewhere, where it would have had higher casualty reduction effect. Spending them on smart motorways will have a very small casualty reduction effect, because the casualty level on motorways, including smart motorways, is already so low.

I'm no apologist for National Highways, and not all smart motorway schemes have been as successful as this. Some have been rather disappointing, at least in terms of the capacity they delivered. But I was rather amazed at what I discovered reading this. It challenged my preconceptions.

[bob sterman]

Here's a fascinating recent POPE. It's the 5-year POPE of M1 J10 to J13 smart motorway. It's a dynamic smart motorway, the most despised kind, with gantry signs opening and closing the hard shoulder to traffic when traffic levels demand.

Well it's a highly detailed report with all sorts of data about varying types of incident (of different levels of seriousness) - across various areas (e.g. the road itself and wider area).

It appears the "slight" collision rate has indeed decreased, but one very simple comparison that is not given a great deal of prominence in the report is this...

On the modified stretch of the M1 itself...

Fatal and serious collisions Dec 2006 - Nov 2009 (the 3 years prior to construction) = 22

Fatal and serious collisions Dec 2012 - Nov 2015 (the 3 years after construction) = 35

Change = +59.1%

[dyqik]

Here's a fascinating recent POPE. It's the 5-year POPE of M1 J10 to J13 smart motorway. It's a dynamic smart motorway, the most despised kind, with gantry signs opening and closing the hard shoulder to traffic when traffic levels demand.

Well it's a highly detailed report with all sorts of data about varying types of incident (of different levels of seriousness) - across various areas (e.g. the road itself and wider area).

It appears the "slight" collision rate has indeed decreased, but one very simple comparison that is not given a great deal of prominence in the report is this...

On the modified stretch of the M1 itself...

Fatal and serious collisions Dec 2006 - Nov 2009 (the 3 years prior to construction) = 22

Fatal and serious collisions Dec 2012 - Nov 2015 (the 3 years after construction) = 35

Change = +59.1%

What's the rate, though?

Small number statistics, plus no normalization, equals bad data.