Road Safety News

Motorway hard shoulders shouldn't be used as permanent lanes – Transport Committee

Thursday 30th June 2016

The Government should not proceed with motorway 'all lane running' schemes while major safety concerns exist, the Transport Committee has warned.

In its report on the issue published today (30 June) the Transport Committee says that while Highways England may ‘consider the matter settled’, the Transport Committee believes that ‘the argument has not been won’.

Under 'all lane running' the motorway hard shoulder is used as a live lane of traffic. Previously, the hard shoulder has only been used at peak times or to deal with congestion.

The Transport Committee rejects the Government’s notion that this is an ‘incremental change’ and a ‘logical extension of previous schemes’.

Instead, it concludes that the permanent loss of the hard shoulder in all lane running schemes is a ‘radical change and an unacceptable price to pay for such improvements’.

Louise Ellman MP, chair of the Transport Committee, said: "The permanent removal of the hard shoulder is a dramatic change. All kinds of drivers, including the emergency services, are genuinely concerned about the risk this presents.

“It is undeniable that we need to find ways of dealing with traffic growth on the strategic network. But all lane running does not appear to us to be the safe, incremental change the Department wants us to think it is.

“While 'smart motorways' have existed for years, this is fundamentally different. Government needs to demonstrate that all lane running schemes do not make the road any less safe than the traditional motorway with a hard shoulder.

“The Government has a model which has worked. The scheme on the M42 has a track record of safety and performance but subsequent versions have gradually lowered the standard specification.

“The most recent incarnations of all lane running have less provision for safety measures than original pilot schemes.

“The Committee heard significant concerns about the scarcity, size and misuse of emergency refuge areas. We also heard about worryingly high levels of non-compliance with Red X signals.

“Levels of public awareness and confidence about using these motorway schemes are unacceptably low.

“Government needs to demonstrate considerable improvement in this area, including more emergency refuge areas, driver education and enforcement, before the Committee will endorse the extension of a scheme which risks putting motorists in harm's way."

In 2015, the DfT forecast that traffic on the strategic road network would increase by up to 60% by 2040. The Government sees smart motorways as a way of addressing this growth without incurring the costs of traditional motorway widening.

Plans are in place to permanently convert the hard shoulder into a running lane on around 300 miles of motorway. Highways England has a programme of 30 all lane running schemes to the value of around £6bn over the next nine years.

Stakeholder reaction

The RAC has welcomed the report and has used it to again express its concern over the use of the hard shoulder as a permanent running lane.

David Bizley, RAC chief engineer, said: “Whilst supporting smart motorways as a cost effective and relatively rapid way of increasing motorway capacity, the RAC has repeatedly expressed concerns about the latest design which turns the hard shoulder on motorways into a permanent running lane.

“We therefore welcome the Select Committee report and are pleased that this influential group of MPs has concluded that the decision to adopt ‘all lane running’ on all future smart motorways may be premature.

"The safety of motorists must come first and therefore new designs need to be trialled for sufficiently long to demonstrate their safety before they are introduced more widely."

However, the Transport Research Laboratory (TRL) says all-lane running motorways are no less safe than other motorways.

Alan Stevens, chief scientist, transportation, said: "“The volume of traffic on our motorways is increasing, so we need to take steps to increase capacity, improve traffic flow and ease congestion in a safe and pragmatic way. 

"Smart motorways allow this to be achieved usually within the highway boundary, limiting land use and disruption from road widening while ultimately providing drivers with shorter, more predictable journeys and less stressful driving. 
“Whilst the Transport Select Committee has raised valid concerns over the need for sufficient evaluation, TRL believes that the implementation of technology, such as smart motorways, is vital in keeping our networks flowing and can be achieved without increasing overall risk.  
"We have conducted several research projects using our driving simulator to see how people respond to both dynamic hard shoulder and all-lane running motorways and found both to be no less safe than other motorways. 
"Of course, like with all new transport innovations, implementation will need to be continuously monitored to ensure the predicted and desired outcomes are achieved with any safety implications immediately identified and addressed.
Photo: Highways England via Flickr used under Creative Commons


Comment on this story
Report a reader comment

What's your view - comment on this story:

I confirm that I have read and accept the moderation policy and house rules relating to comments posted on this website.
Your comment:
Your name and location:
Your email:

Charles, it's not as simple as you make out. The traffic modelling process has mostly failed to be linked effectively with land use policies, plus the effects of induced traffic are very rarely anticipated. Your "gems" are not necessarily examples of good practice. Sadly, I have first hand knowledge of this from many years as a modeller which is why I stopped doing it. Effective promotion of more environmentally friendly and healthy forms of travel are for many reasons the way forward rather than creating new roadspace that simply results in longer single-occupant car journeys (ie, more roadspace could offer less congestion meaning that more people can make longer journeys in a similar time period to the situation without the extra capacity...until of course journey times increase again as more people, developers etc "take advantage" of the new road space). Sadly (again), as I mentioned below, there will be a limit to the extent car sharing can help to make more efficient use of existing roadspace, unless it is promoted and incentivised effectively. More sensible land use policies could also help.
Andy, Birmingham

Agree (1) | Disagree (0)

I think Dave is right about progress being made in the introduction of evidence-led interventions and while RCTs would be the most robust, there are other evidence-led approaches that would suffice for smaller scale interventions.

I don't want to speak on behalf of others here, but from my dealings with a large number of people in the field of road safety there is a distinct desire from the majority to work in this way. However, this is not a simple problem to address. The issue is that most people in road safety work for political organisations (Local Authorities, DfT, Police and Crime Commissioners etc. etc.) and the political leaders in these organisations often need to be seen to be doing something to address public concerns that are brought to their attention. These people also hold the purse strings...and you start to get to the crux of the matter.

Sound, evidence-led interventions and evaluations take time and also, if using control groups, involve target groups not getting an intervention - try selling that to their local politician!

But despite these difficulties, as Dave has pointed out, progress IS being made and I would like to applaud and encourage all the practitioners that are doing their utmost, often in difficult financial circumstances, to make things better.

I'll get off my soap box now.
Matt Staton, Cambridgeshire

Agree (7) | Disagree (0)

1) I recognised that the quality of evidence in road safety was poor
2) I pinpointed the primary reason for the poor quality
3) I developed a method that could fully address this primary reason
4) I applied the method to real data
5) The method has been endorsed by the DfT and recommended by the RAC Foundation
6) The method has been applied to produce truly accurate evaluations

The pace of change is excruciatingly slow and what is holding road safety back is the reluctance by some to accept that a problem even exists. It may well be the case that only after we run RCT scientific trials, will we truly understand why it was necessary to run them in the first place. Progress is being made, though, and I believe that we will eventually start using an evidence-led approach.
Dave Finney, Slough

Agree (6) | Disagree (2)

Andy, traffic modelling basically relies on "guessing" future demand. It fails when the "guess" for that demand mis estimates the latent demand. As well as citing roads which are always congested we can all also cite roads (including motorways and trunk roads) which rarely, if ever, operate at full capacity. It is generally only the roads which have insufficient capacity which feature in the bad news. The others are hailed as the secret gems of our road network.

That some models do not correctly anticipate future demand accurately just means the assumptions were incorrect, and should not be interpreted as meaning the model would have failed if the assumptions were correct. The art is in knowing the market and correctly anticipating latent demand and success relies on that market not dramatically changing following project implementation.
Charles, England

Agree (6) | Disagree (1)

Dave Finney has for years called for RCT. I am not an expert on this but I do note that all the first All Lane Running schemes were selected on congestion characteristics and not existing safety performance, I may be wrong but to me the sites are indeed random. As the control data is all other motorways I think the statistical tests one year after are as good as they can be.
Pete, Liverpool

Agree (1) | Disagree (4)

"Alan Stevens, chief scientist, transportation, said: "“The volume of traffic on our motorways is increasing, so we need to take steps to increase capacity, improve traffic flow and ease congestion in a safe and pragmatic way...." As a scientist he may well be aware of Parkinson's Law...and therefore that traffic modelling/engineering is a failed science. Promotion/incentivisation of car sharing would be a better way forward, until all the spare seats fill up!
Andy, Birmingham

Agree (8) | Disagree (4)

Nick, I'm not convinced, based on the example you mentioned of a Think! Campaign, that you understand my point. What we need is a sound statistical analysis confirming (after the event) that casualties had actually been reduced (or otherwise) as a direct result of the intervention. The intervention you quote seems to have been based on various speculative assertions based on unspecified theoretical hypotheses of what the drink risk is and how it "could" (not "would") affect your chance of causing a casualty collision. Did it actually reduce fatal collisions? If so, by how many?

The main problem with theoretical hypotheses is that they don't generally take account of human nature, so do not predict or make allowance for the compensatory (possibly subconscious) actions of drivers to perceived variations in risk. For example, wasn't it shown that drivers with ABS took more risks than drivers without, thus cancelling out the advantage in braking effect it gives? Hasn't it been shown that drivers of large SUV vehicles take more risks because they feel better protected? What "cancelling out" effects were taken into account in the rationale behind the Think! 2015 campaign?
Charles, England

Agree (7) | Disagree (6)

It seems we have single lane roads and double carriageway roads and motorways or other arterial main roads without any hard shoulders and yet no one suggests that there should be one. Why should motorways be any different? We have had years of lane closures due to roadworks and is there a fuss made about that? No there isn't. So what's wrong in alleviating some driving pressure by allowing 4 or 3 lanes at certain busy times of the day? What's all the fuss about, its been happening for decades with no complaints or cencerns. My only wish is that HGVs be restricted to the inside two lanes and not to the third lane as that would create just what they are trying to alleviate, road crush.
R.Craven Blackpool

Agree (5) | Disagree (2)

I don't think it takes a lot of scientific study to conclude that emergency services will take longer to reach casualties in a crash on a SMART motorway in the rush-hour than they would have done had they been able to use a hard shoulder.

I may be a hard-nosed sceptic but I never had much faith in all the 'studies' that showed SMART motorways to be just as safe as those where use of the hard shoulder was prohibited. I always thought that things had been fudged to allow a cheapskate 'solution' to ever-increasing traffic volumes.
David, Suffolk

Agree (11) | Disagree (3)

Charles (and Dave)

Pat is correct - this is not the place for a detailed response as requested by Charles.

However, in brief, the THINK! 2015 festive drink drive campaign is an excellent example of an evidence-based initiative. The THINK! team used casualty stats to identify a specific problem (young males), then commissioned research to fully understand the problem (young men applying their own limit of two drinks and you're OK to drive), then developed a creative treatment to challenge this assumption (#ButAlive), and finally conducted post-campaign evaluation to see if the message resonated with the target audience and had changed perception and behaviour.

If Dave and Charles were saying that all road safety interventions should be evidence based I think that all, or virtually all, of our readers and road safety professionals would agree.

But to suggest or imply there is little or no evidence-based or scientific work going on is, to be honest, a bit ridiculous - and quite insulting to thousands of road safety professionals working in local authorities, private companies, the third sector and academic institutions.
Nick Rawlings, editor, Road Safety News

Agree (14) | Disagree (6)

That's an excellent point, Nick, and we need to ask what do we mean by "evidence"?

data: measurements that we have made
statistics: calculations made from the data
methods: the choice of what statistics to produce
evidence: conclusions that can be reached using the methods

We have tons of data and we can produce thousands of statistics but have we got the methods that can produce accurate evidence? The problem is that, to determine the effect of any intervention, we must separate it's effect from all of the other many factors. This has proved so difficult that, until recently, there was no known method that could fully exclude even the largest factor at intervention sites, let alone the smaller ones.

The first step in solving a problem is to recognize that the problem exists and the concerns raised by the Transport Committee are further steps towards recognizing the problem. In order to start an evidence-led approach we must raise the quality of evidence substantially. RCT scientific trials would do this by achieving the highest standards possible and, by adopting them, we could lead the world.
Dave Finney, Slough

Agree (7) | Disagree (9)

This is a news feed and I agree with Nick's timely reminder. Deeper broadbrush questions and that require detailed and comprehensive answers may be relevant and informative but belong elsewhere, not here.
Pat, Wales

Agree (12) | Disagree (8)

Nick, so that the sceptics here can understand where you are coming from, can you cite specific examples where scientifically and statistically sound analyses of the effect of the introduction of particular major road safety measures has concluded that the evidence unequivocally shows that the measures had delivered, in road safety terms, what they were forecast to deliver?
Charles, England

Agree (8) | Disagree (14)

I will repeat a comment I made earlier today in another thread.

For me, you generalise too much in your post. Much of the road safety work carried out by organisations such as TRL, Thatcham and some of our universities and road safety teams is scientifically sound and evidence based. I'm not saying all is perfect in the world of road safety, but it is certainly not correct to suggest that no one is using an 'evidence-led approach' - which is what your post implies.
Nick Rawlings, editor, Road Safety News

Agree (20) | Disagree (6)

The Transport Committee is right to be concerned about the poor quality of evidence for hard shoulder running. If all "smart" motorways had been installed within RCT scientific trials, we would now have "proof beyond a reasonable doubt" of what effect they are having - and not just on safety.

Surely it is time for us to learn the lessons of this and other road safety interventions? Let's make RCT scientific trials standard practice in road safety, let's start using an evidence-led approach,
Dave Finney, Slough

Agree (7) | Disagree (10)