The future of smart motorways

[jnty]

I agree about our horrible media, but the reality is much more nuanced about the actual figures.

Yes, the average person driving between two points does appear to be safer on a smart motorway. BUT if they have the misfortune to break down, they are significantly less safe.

Is that extra risk a justifiable choice? Can you think of any other circumstance where something similar happens? Would the knowledge that everybody else is statistically safer make you feel any more comfortable in the event you broke down in a live lane on a SM?

Overall smart motorways are marginally safer than conventional motorways (as per the safety assessment previously linked, Table 1, page 44). They are much, much safer than conventional A roads. Yes, beyond this there's a lot of nuance, but this is the summary, and if the aim of smart motorways is to make additional capacity available without worsening overall safety compared to conventional motorways, then they've satisfied their aim and I for one am ok with that. Of course further improvements are always possible and I'm glad to see that many improvements have already been implemented.

In terms of further (cost effective) improvements I’ve never understood the priority given to standardisation. What are the arguments against discontinuous hard shoulders, or lots more refuges, where possible at low cost?

Smart motorways came about precisely because adjacent land take was difficult/impossible so presumably extra refuges and discontinuous hard shoulders are rarely possible at low cost. I think it's also hard to justify chucking in hard shoulder where possible from a risk/benefit point of view when the road is already generally safer and the extra signage and "churn" in road standard risks causing distraction and confusion.

[Bessie]

Overall smart motorways are marginally safer than conventional motorways (as per the safety assessment previously linked, Table 1, page 44). They are much, much safer than conventional A roads. Yes, beyond this there's a lot of nuance, but this is the summary, and if the aim of smart motorways is to make additional capacity available without worsening overall safety compared to conventional motorways, then they've satisfied their aim and I for one am ok with that. Of course further improvements are always possible and I'm glad to see that many improvements have already been implemented.

In terms of further (cost effective) improvements I’ve never understood the priority given to standardisation. What are the arguments against discontinuous hard shoulders, or lots more refuges, where possible at low cost?

Smart motorways came about precisely because adjacent land take was difficult/impossible so presumably extra refuges and discontinuous hard shoulders are rarely possible at low cost. I think it's also hard to justify chucking in hard shoulder where possible from a risk/benefit point of view when the road is already generally safer and the extra signage and "churn" in road standard risks causing distraction and confusion.

That all makes sense and you may well be right. But I drive a lot in France and my impression is that their approach to hard shoulders is much less standardised. I wonder what the accident records show for discontinuous shoulders where they have been used in the UK (e.g. M50, widened M5 near Brum, M4 Newport bypass). Difficult to adjust for confounding factors, I suppose.

[jackal]

The campaign against smart motorways, aided by our rotten and incompetent media, has cost lives. No-one cares.

I agree about our horrible media, but the reality is much more nuanced about the actual figures.

Yes, the average person driving between two points does appear to be safer on a smart motorway. BUT if they have the misfortune to break down, they are significantly less safe.

Is that extra risk a justifiable choice? Can you think of any other circumstance where something similar happens? Would the knowledge that everybody else is statistically safer make you feel any more comfortable in the event you broke down in a live lane on a SM?

It's absolutely routine, and the foundation of rational public policy, to accept large benefits for one group where they accompany smaller aggregate costs for another.

[Micro The Maniac]

Yes, the average person driving between two points does appear to be safer on a smart motorway. BUT if they have the misfortune to break down, they are significantly less safe.

But statistically, it is still in the noise, compared to the risks on ordinary dual-carriageways, and especially rural minor roads...

[Micro The Maniac]

It's a complete joke of system meta-design that the Highways Agency was allowed to change the fundamental driver-facing operational principles of the design, during roll out, meaning drivers could experience 3 different sets of rules on one journey (e.g. M6 > M42 > M5 to the South West).

Or just on the M1...

• J6a-10 - Don't use HS - Controlled
• J10-13 - Maybe use HS - DHS
• J13-19 - Do use HS - ALR
• J19-23a - Don't use HS - Traditional
• J23a-24 - Don't use HS - Controlled
• J24-25 - Do use HS - ALR
• J25-28 - Don't use HS - Controlled
• J28-31 - Do use HS - ALR
• J31-32 - Don't use HS - Controlled
• J32-35a - Do use HS - ALR
• J35a-39 - Don't use HS - Traditional
• J39-42 - Do use HS - ALR
• J42-end - Don't use HS - Traditional

What a confusing mess...

[Herned]

It's absolutely routine, and the foundation of rational public policy, to accept large benefits for one group where they accompany smaller aggregate costs for another.

Can you name another policy which directly increases risks for one group?

[jackal]

It's absolutely routine, and the foundation of rational public policy, to accept large benefits for one group where they accompany smaller aggregate costs for another.

Can you name another policy which directly increases risks for one group?

Expenditure on one policy area (e.g. healthcare) rather than others (e.g. transport).

Within that policy area, expenditure on one class of intervention (e.g. drugs) rather than others (e.g. staff).

Within that class of intervention, expenditure on one-sub class of intervention (e.g. cancer drugs) rather than others (e.g. cardiovascular drugs).

Within that sub-class of intervention, expenditure on one sub-sub-class of intervention (e.g. targeted therapy) rather than others (e.g. hormonal therapy).

Within that sub-sub-class of intervention, expenditure on one sub-sub-sub-class of intervention (e.g. breast cancer therapy) rather than others (e.g. lung cancer therapy).

These are just a few of countless thousands of policy decisions where one group's interests is, legitimately and unavoidably, weighed against another's.

[Patrick Harper]

The safety advantages of controlled D4 over controlled D3M are entirely the consequence of extra capacity, all else being equal. If peak demand for the road is not sufficient to cause average speeds and vehicle distances to drop significantly below optimal in a D3M formation (around 90k AADT?), then D3M is probably safer overall.

[Micro The Maniac]

The safety advantages of controlled D4 over controlled D3M are entirely the consequence of extra capacity, all else being equal. If peak demand for the road is not sufficient to cause average speeds and vehicle distances to drop significantly below optimal in a D3M formation (around 90k AADT?), then D3M is probably safer overall.

Indeed... there is also the fact that the various stages of SMART technology have been applied to the busiest motorways. Comparing the remaining traditional motorways with the M25 western section is like comparing apples with oranges.

I would be interested to see (for example) a comparison of the incident density on M3 J2-4a pre- and post-SMARTification - and, for example, pre-SMART J3 "Bagshot" was a daily mention on all travel bulletins

[DB617]

Alas, now we are getting into the complete nightmare of data science! Agree with all of the above though.

[jnty]

It's absolutely routine, and the foundation of rational public policy, to accept large benefits for one group where they accompany smaller aggregate costs for another.

Can you name another policy which directly increases risks for one group?

As jackal says this is true of almost any policy in any area of public life. An attempt at a close parallel might be an increase in the number of trains serving a station. The service level provided to the whole population of passengers increases significantly, but any passengers who fall on to the tracks are more likely to be killed. It's obvious that the benefit will massively outweigh this cost in almost all circumstances, but just like with smart motorways, it can be worrying to be standing on a busy eg. London Underground platform if you think about what might happen if you were to fall off the platform.

[Herned]

Expenditure on one policy area (e.g. healthcare) rather than others (e.g. transport).

Within that policy area, expenditure on one class of intervention (e.g. drugs) rather than others (e.g. staff).

Within that class of intervention, expenditure on one-sub class of intervention (e.g. cancer drugs) rather than others (e.g. cardiovascular drugs).

Within that sub-class of intervention, expenditure on one sub-sub-class of intervention (e.g. targeted therapy) rather than others (e.g. hormonal therapy).

Within that sub-sub-class of intervention, expenditure on one sub-sub-sub-class of intervention (e.g. breast cancer therapy) rather than others (e.g. lung cancer therapy).

These are just a few of countless thousands of policy decisions where one group's interests is, legitimately and unavoidably, weighed against another's.

Those are examples of selecting one group over another. Which is, as you say, entirely commonplace. However, in none of those examples will the losing-out group actually see their risk increase

[jnty]

Expenditure on one policy area (e.g. healthcare) rather than others (e.g. transport).

Within that policy area, expenditure on one class of intervention (e.g. drugs) rather than others (e.g. staff).

Within that class of intervention, expenditure on one-sub class of intervention (e.g. cancer drugs) rather than others (e.g. cardiovascular drugs).

Within that sub-class of intervention, expenditure on one sub-sub-class of intervention (e.g. targeted therapy) rather than others (e.g. hormonal therapy).

Within that sub-sub-class of intervention, expenditure on one sub-sub-sub-class of intervention (e.g. breast cancer therapy) rather than others (e.g. lung cancer therapy).

These are just a few of countless thousands of policy decisions where one group's interests is, legitimately and unavoidably, weighed against another's.

Those are examples of selecting one group over another. Which is, as you say, entirely commonplace. However, in none of those examples will the losing-out group actually see their risk increase

To take one example there, the decision to relax the screening criteria for eg. breast cancer will, amongst those added to the screening population, benefit those who receive a 'true positive' result and receive treatment for cancer but seriously harm those who receive a 'false positive' result and undergo severely invasive unnecessary surgical treatment. This is actually a hugely challenging and controversial area of health policy. Of course, if the resources for extra screening and treatment are diverted from, say, lung cancer screening and treatment, then there will absolutely be an increase in lung cancer sufferers' risk of death. I'm not entirely sure what the difference between 'selecting one group over another' and 'causing the risk for one group to increase' is. You could just as easily say that smart motorways select the safety of the whole population of road users over those who catastrophically break down within limping distance of the nearside lane.

[jackal]

^ Quite.

Furthermore, the introduction of most medicines will have the effect of benefiting one group (for whom the medicine is effective) while increasing risks for another, smaller group (for whom side effects outweigh any benefits).

Should we ban effective medicines on the basis of statistically rare cases of severe side effects, just as anti-SMers propose banning SMs on the basis of statistically rare cases of severe collisions with stopped vehicles in running lanes?

[Bryn666]

I don't think it's hyperbole to suggest the implementation of smart motorways has been such a cock-up in getting fundamental basics right that the well has been truly poisoned with regard to doing further schemes.

As many of us have said many times, the incorrect decision to run with signal logic that behaves differently to the rest of of the world, inconsistent application, road signs that are not always user friendly, missteps from the control rooms... none of these have given people faith in the system.

And if you're going to run a system like this, you pretty much need faith in it from day 1.

[XC70]

I don't think it's hyperbole to suggest the implementation of smart motorways has been such a cock-up in getting fundamental basics right that the well has been truly poisoned with regard to doing further schemes.

As many of us have said many times, the incorrect decision to run with signal logic that behaves differently to the rest of of the world, inconsistent application, road signs that are not always user friendly, missteps from the control rooms... none of these have given people faith in the system.

And if you're going to run a system like this, you pretty much need faith in it from day 1.

Well said Bryn. HE only have themselves to blame. They claimed all sorts of nonsense from the start like when the CEO of HA (as it was at the time) went on the One Show to claim "100% CCTV coverage". This soon showed itself to an example of weasel words which people quickly saw through.

Every single day I see examples of signs which display issues which are not there and issues which are not signed. Then the moronic attitude of some in the call centres. I had one call where the operator confused M1 J24 with M25 J24, even when I twigged why he was talking about Potters Bar and tried to correct him, he seemed to not be able to compute. Then there was the operator who I rang to tell that there were lone 60mph signals displayed followed by blank signs at M1 J30. Instead of investigating, we got into a ridiculous conversation where she was trying to tell me that this meant that the 60mph limit was in force all the way up to J34 where the signs were displaying something again.

The idea of Smart motorways could perhaps work if done properly, but unfortunately it would appear that it has been implemented by - and is now run by - morons.

And don't get me started on the C5 documentary following the work of the traffic wombles. The exageration from the operatives and demonisation of all driver behaviour means I have had to stop watching....

[jnty]

I don't think it's hyperbole to suggest the implementation of smart motorways has been such a cock-up in getting fundamental basics right that the well has been truly poisoned with regard to doing further schemes.

As many of us have said many times, the incorrect decision to run with signal logic that behaves differently to the rest of of the world, inconsistent application, road signs that are not always user friendly, missteps from the control rooms... none of these have given people faith in the system.

And if you're going to run a system like this, you pretty much need faith in it from day 1.

I agree entirely - implementation does seem to have been poor. You could, however, argue that the fact that the safety stats look good overall despite this means that SMs, in principle, are actually extremely safe. A 'properly implemented' one might even be able to close the gap on breakdown safety.

However, I do feel like the extra 'gold plating' that has SMs have ended up acquiring along with the operational limitations added by not having a hard shoulder probably means they're not cost effective in the long run.

It's a shame that the government that banned them clearly has no intention of dealing with the institutional malaise they're symptomatic of - that ceding an inch of ground to new infrastructure is to be avoided at almost any cost.

[XC70]

I don't think it's hyperbole to suggest the implementation of smart motorways has been such a cock-up in getting fundamental basics right that the well has been truly poisoned with regard to doing further schemes.

As many of us have said many times, the incorrect decision to run with signal logic that behaves differently to the rest of of the world, inconsistent application, road signs that are not always user friendly, missteps from the control rooms... none of these have given people faith in the system.

And if you're going to run a system like this, you pretty much need faith in it from day 1.

I agree entirely - implementation does seem to have been poor. You could, however, argue that the fact that the safety stats look good overall despite this means that SMs, in principle, are actually extremely safe. A 'properly implemented' one might even be able to close the gap on breakdown safety.

However, I do feel like the extra 'gold plating' that has SMs have ended up acquiring along with the operational limitations added by not having a hard shoulder probably means they're not cost effective in the long run.

It's a shame that the government that banned them clearly has no intention of dealing with the institutional malaise they're symptomatic of - that ceding an inch of ground to new infrastructure is to be avoided at almost any cost.

They are taking an age to build a dozen refuges between 32 and 35 on the M1 at the moment, so of course we are down to 3 lanes for the next 18 months or so I guess (another disadvantage of no HS), so we have traffic chaos and I imagine the cost is going to be well into 8 figures. They have also just installed SPECS between 30 and 31 so I am braced for more works, but slightly confused as they built an extra refuge n/b by Woodall srvcs about 2 years ago, so not sure what they are going to be doing. Whatever, I guess I will need to get up at 5.50am, rather than 6am for a long time if I still want to get to work on time.

[Bryn666]

M6 J21A-26 has just had an additional 18 months added to its contract time too - apparently "poor ground conditions" and the need for extra ERAs is the reason cited.

1. It's the South Lancashire coalfield. A GCSE geography student could've told you there'd be numerous issues with ground settlement, poor quality soils, drainage trouble, etc. That's before you factor in any of the land crossing former peat bogs around the flashes of Wigan. I don't do structural engineering for a reason, I know my limitations, but I am aware of the basic geology of the region here at least.

2. Even on here we said the original proposed number of ERAs seemed ridiculously low for the traffic volumes and extent of the scheme.

It's almost become a meme that I criticise NH for virtually everything, but is it any wonder? These are people with degrees and expertise supposedly dribbling out of every orifice imaginable and this is what we get lumbered with.

[domcoop]

Unless I have misunderstood (quite possible!), my point stands: more people have died as a result of the anti-smart motorway campaign. Why does nobody think this is an issue?

I heard an interview with the Transport Secretary Mark Harper on the radio (think it was with Nick Ferrari on LBC). The interviewer was obviously an anti-smart motorway person talking about "death trap motorways" and "when are you going to remove the existing schemes if you think it's too unsafe to build more". This was a few days before the report came out.

I was actually surprisingly pleased. Harper said right away that Smart Motorways are safer and removing the existing ones would increase casualties and also drive congested traffic onto even more dangerous A roads. The way he put it was that they should actually build more Smart Motorways, but that they recognise the general public don't feel safer on them and you can't roll out a programme like this without the public being on board. Certainly sounded like it would be left open for the programme to resume in the future.