Conflict between lorries and bicycles

To add to the distressing news of the death of the climatologist Kat Giles on Victoria Street two weeks ago, a young man on a Boris Bike was seriously injured in London last Friday in what seem to be very similar circumstances – crushed by a left-turning lorry, on Grays Inn Road. Both of these incidents are close to home for me – I frequently cycle down Victoria Street, and last Friday’s incident occurred directly opposite the Yorkshire Grey, the venue for Street Talks.

It wouldn’t be appropriate to speculate on the causes of these most recent incidents – just the latest in a very long line of deaths and serious injuries resulting from turning lorry conflicts. However, it is safe to say that human error is one of the most likely contributory factors; a failure of observation, too much haste, a lack of concentration, and so on.

While it is right that much more can and should be done to make lorries safer – better mirrors and warning devices, cabs with better visibility, higher industry standards – the fact remains that human beings are not infallible. They will make mistakes. And that does include cyclists, who will naturally incline to staying close to kerbs, and will consequently find themselves in dangerous positions at junctions (please do read Mark Ames’ excellent guide on this topic).

So for that reason I agree with Danny Williams, that the simplest and ultimately most effective solution is

to keep tipper trucks and people on bikes apart from each other.

One way of achieving this would be a lorry ban at peak hours, which has been mooted, but this doesn’t seem to me to be particularly likely, or workable. Even if it does come into force, there will (or at least should be) plenty of other cyclists about in London at other times of the day. Cycling should not just be about commuting; it should be about going to the shops, or visiting friends, or going out for the evening, or cycling to school – basic, everyday trips that will occur at all times of the day. We should try to protect all cyclists from interactions with lorries at all times, not just commuters. So a peak time ban would only amount to a stopgap intervention.

Danny quotes a Dutch Road Safety fact sheet, which states that

Truck drivers do not make the best possible use of the different mirrors [and] cyclists insufficiently take account of the fact that trucks have a limited visual field. The ultimate solution for the blind spot problem is a structural separation of trucks and cyclists.

The sad thing is that we should already know how to do this; we have examples of how to keep lorries and cyclists apart. The Netherlands has perfected road design that remove interactions between heavy goods vehicles and cyclists to an enormous extent.

Here’s an example, a busy junction in Amsterdam.


I’m cycling past a tipper truck, the type of vehicle involved in the latest collisions in London. But I will not interact with it at all.

If I’m turning right, I won’t need to go anywhere near the road; the cycle track merely continues around the corner, uninterrupted, and fully protected from the road. If I wish to go straight on, the truck will be held at a red signal while I have a separate green signal. Conversely, when the truck starts to move, I would then be held at a red signal.

Here’s another example of what this looks like, again from Amsterdam.

DSCN0119Here cyclists are progressing straight ahead across the junction (alongside pedestrians, who also have a green signal on their crossing). But note, crucially, that motor vehicles wishing to turn right are held at a red signal. 

So there is no turning conflict here, of the kind that has resulted in dozens of deaths and countless more serious injuries in London in just the last few years alone. Cyclists are separated in time and space from the movements of motor vehicles, just like pedestrians are.

What I find almost incredible is that we can potentially implement a very good approximation of this kind of Dutch design, right now. The elements are already in place.

We can put motor vehicles on different light signals for different turning movements. We already do this.

Different signals for motor traffic turning in different directions.

Different signals for motor traffic turning in different directions.

Likewise, we can put cycle tracks, separate from the road, around corners. They’re often called ‘shared use pavements’ (our disastrously bad version of off-carriageway provision) or they are poorly-implemented cycle tracks, with confused or dubious priority and separation. We can already do this; we just need to do it much better, with wider, better designed provision for cycling, and with clear priority for pedestrians, where appropriate.

A cycle track around the corner of a junction

A cycle track around the corner of a junction

And we can let cyclists and pedestrians cross the road simultaneously on green signals; we have toucan crossings. We even attempt to separate the movements of pedestrians and cyclists when they cross, although, again, we do it quite badly.

A toucan crossing at Hyde Park Corner. Technically cyclists should cross to the left, and pedestrians on the right, but the design is unclear and confusing.

A toucan crossing at Hyde Park Corner. Technically cyclists should cross to the left, and pedestrians on the right, but the design is unclear and confusing.

My point is that there is really nothing to stop us building a high-quality Dutch-style junction tomorrow. We don’t need to experiment; we know what works, because the Dutch have already done it. We just need to copy it, and do it well. Even better than that, take the things we can already do, and just implement them as well as the Dutch implement them.

When cycle tracks go round corners, make it clear that it is not a pavement, but also provide clear crossing points for pedestrians, where they have priority.

Continuous protected cycle track, with zebra priority for pedestrians

Continuous protected cycle track, with zebra priority for pedestrians

And where cyclists cross junctions with pedestrians, greater clarity is required.

Clear and distinct crossing routes for pedestrians and cyclists; not just a fudge that treats the two as equivalent users

Clear and distinct crossing routes for pedestrians and cyclists; not just a fudge that treats the two as equivalent users

In principle, there’s absolutely nothing stopping us from doing this right.

So it is more than a little disconcerting to read that Transport for London are

[hoping] to be able to test what interventions work

as the Alternative Department for Transport blog reports. He writes – and I’m inclined to agree –

TfL would rather figure it all out for themselves from scratch. This is madness – all the research is available from the Netherlands, which went through the learning process 35 years ago (and is still improving its cycling facilities). They made the mistakes so we don’t have to.

Yet TfL will “test what interventions work”? We already know what interventions work! They’re going to play around with our money, making it up as they go along because they can’t be arsed to go see what makes roads in the Netherlands work so well.

In a similar development, Transport Extra magazine reveals that

Transport for London is refining its traffic modelling to improve the representation of cycling and pedestrian behaviour. “Until recently, relatively little research had been undertaken worldwide to understand the behaviour of vulnerable road users at traffic signals and therefore be able to accurately represent them in traffic models,” TfL has told the London Assembly’s transport committee.

“We are leading on a world-first piece of research to understand cyclist behaviour as they discharge from signals and travel between signals. The research will also look at the impact cyclists have on general traffic discharge where they comprise a high proportion of road users.”

TfL says the work is being complemented by TRL research on pedestrian behaviour at traffic signals. “Once findings from the research are received [later this year], the new algorithms for cyclists and pedestrians will be available to update the capabilities of the modelling tools,” TfL explains.

Well, there are already  countries with high volumes of pedestrian and cycle flow at junctions, countries very near to us, so it’s hard to understand why Transport for London are spending time and effort coming up with fancy models when we have real world examples of large numbers of cyclists flowing through busy junctions. The impression being given is that TfL have no idea how cycling works in countries like the Netherlands and Denmark, and that cyclists ‘comprising a high proportion of road users’ is a unique, new problem, requiring new, unique modelling.

What we need to see is an end to bodging, and in its place, bold plans that would privilege cycling as a mode of transport, not just by making it convenient and comfortable, but most importantly of all by making it safe. We cannot go on with road designs which expect cyclists to fight for position with road vehicles turning across their path. We don’t send pedestrians across the road at the exact same time when motor vehicles will be bearing down on them; we shouldn’t do it with people on bikes either.

It is no exaggeration to say that people are dying because we are failing to take action.

This entry was posted in Go Dutch, Infrastructure, Junction Review, London, Road safety, Subjective safety, Transport for London. Bookmark the permalink.

15 Responses to Conflict between lorries and bicycles

  1. Rachel says:

    Another very interesting post making lots of good points. Thanks. Minor point, but I’m inclined to think that modelling cyclists’ behaviour more accurately at junctions is a good thing. I think the problem is more that in the UK, basing it on existing cyclists’ current behaviour risks limiting the modelling to ‘this is how the 2% will behave in hostile conditions’. I hope that they are aware of this limitation and are attempting to counteract it.
    Interestingly, I don’t think high cycling countries are massively ahead of us in terms of modelling cycling. I was speaking recently to a Danish transport consultant about this, and we concluded that because cycling is more politically accepted there, planners have not seen the need for the ‘extra’ evidence provided by models, cost-benefit analysis calculations, etc., to justify providing for cycling. However in the UK the models’ deficiencies matter more as we have an uphill political struggle so any apparent counter-evidence or lack of evidence counts more. And these deficiencies are substantial in relation to cycling, which is rarely modelled ‘in itself’ as opposed to being, like walking, what falls out when you model for the motorised modes. My other concern with modelling the ‘real behaviour’ of cyclists is that it may only be used to look at delays caused to motorists, not to also look at impacts on cyclists of different junction arrangements. I hope this won’t be the case. (Although we might get a more realistic estimate of delays caused to motorists by current inadequate junction arrangements, which might prove interesting).
    best wishes

    • That’s a fair point. I hadn’t realised that the Dutch and the Danes don’t do much modelling themselves – probably because they don’t need to! More research is needed, but as you say, it shouldn’t be based exclusively on the behaviour of “the existing 2%”.

  2. Jim Moore says:

    Great post again. I think though most of the Anglosphere does treat pedestrians and cyclists the same when it comes to left-turning vehicles being allowed to turn (filter) on green. Sure, in some places the left-turning vehicle may face a red left-turn arrow for a few seconds at the start of the pedestrian phase, but after that the arrow drops out and the motorist is free to cross through the pedestrian on-road walkway.

    • Thanks Jim. Just to be clear, the Dutch system doesn’t necessarily need to have that “conflicting green” situation; the phases can be set so there is no conflict at all. And importantly, left-turning cyclists aren’t strictly crossing the green phase, because the signalised bit of the crossing for pedestrians starts *beyond* the cycle track.

      • Jim Moore says:

        To clarify my comment, I was referring to *motor* vehicles (cars, trucks, buses etc) turning left across the cyclist and pedestrians who are going straight on, in the same signal phase, not cyclists who are turning left across pedestrians, which isn’t hazardous even outside of the Netherlands where they the paths are physicall separate. I must remember to watch my lexicon when commenting on bike blogs.

        And yes, it is possible to run these conflicting, hazardous movements in separate signal phases, as I understand is Dutch policy.

  3. Anne says:

    Hi there,
    I’m an avid Dutch reader of A view from a cycle path, and via this site I read other cycleblogs and thus I read your blog. And though usually I don’t comment, in light of yesterdays accident in The Hague (NL) where a cyclist had a similar accident as the one you blog about, I had to comment. I’m not saying your Tfl shouldn’t look to the Netherlands, wasting money when others have already done the research is just plain stupid (if indeed NL have done such research), but I would like to comment it isn’t yet perfect in the Netherlands as the accident shows (I do concur it’s a lot better than in the UK though, I wouldn’t want to cycle there). Deathly accidents accur less and less in the Netherlands, but we are still human beings and as such we make mistakes. With deathly outcome. I don’t think there will be a time when such accidents will never happen again, but perhaps with time we can bring it back to less then… say 10. If you hadn’t read the news, here’s a engllish link I found: and here’s also a dutch link with a video: (not gruesome or anything). And at googlemaps:

    • As I understand it, there is no separate green cycle phase at that particular junction; cyclists progressing straight on and vehicles turning right move at the same time. In other words, this is precisely the same kind of problem that occurs in the UK, and that could and should be avoided with the use of the better junction design available in the Netherlands.

  4. Christine Jones says:

    I’ve been hit in London three times, twice by cement Mixers and once by a car, all turning left and taking me with them, I dove off my bike onto the pavement in all cases but I’m aware that there’s not always somewhere to dive so I was very very lucky. I think it helped that I didn’t bang the lorry or push back, I’d rather be alive than right.
    In all cases they didn’t stop and it all happened so fast, I couldn’t get number plates. Nowadays, I’d have reported it just to have the stats on record but back in 2000-2001 when it happened I put it down to experience.
    There is something seriously twisted with the way that the right of way, the road, the place where historically all vehicles have traveled has now become a place where lethal traffic go around oblivious of the harm they are doing. There could be road design, legislation, segregation, really a reality check, this place is for people! Not industrial machinery.
    Oh and I have had enough, I’m moving back to Utrecht, back to a place where they don’t always get it right but they at least think about road design. I’m fed up with Fen Roads, rubbish expensive trains, rubbish expensive buses and being looked at like I’m abusing my kids for riding a bike. At least if I work there I know my tax is going back to the people and not making the fat cats richer.

  5. Charlie says:

    I’d agree with Rachel’s comments here. While I agree that the “not invented here” attitude can be toxic – in general more research is good. Of course – it would be nice if it was open and could be critiqued. I certainly agree lessons drawn from studying commuters might not extend to families on school runs.

  6. It’s time for LCC to stop pussy footing around and start some serious direct action protests. Inaction will just lead to more broken families.

  7. fonant says:

    This is, to me, all so blindingly obvious (that we should copy what has been tried and tested in countries that have high cyclist numbers and low cyclist KSI figures) that I can’t understand why the UK is so reluctant to make our roads safer.

    Perhaps it’s “not invented here syndrome” in action? I remember talking to a transport officer in Chichester who didn’t want to implement 20’s Plenty in the city because he thought there were better ways to achieve road safety: although when pushed he couldn’t think of any.

    Thankfully well-researched blogs like this, and the improved communication between ordinary people that the internet gives us, is allowing people power to come to the fore. The more these issues are discussed in public, the better 🙂

  8. Good post as ever. 2 things;

    (i) I think Hyde Park Corner has now taken out the segregated arrangement and now peds and cycles mix – some “experts” have cited this as a nicer public realm experience, but of course, they don’t use the location – the guard rail could have come out, but the footway could have been built with a kerb upstand to the cycle track.

    (ii) If we do nothing else (I am not saying we should – I want decent infrastructure!) – we should take out guardrail as much as possible near junctions as at the the very least we have cyclists penned in between and trucks. Oh, if we take it out, we now have an extra 500mm of footway which might make the different in nicking a little space from the carriageway and the footway to provide at least a Copenhagen-style one-way track???

  9. Paul says:

    I’m not sure that interaction of peds and cyclists at Hyde Park Corner is all that much of a problem – you have to progress one toucan at a time, say from the Wellington Arch island to the Lanesborough Hotel side of Knightsbridge, then wait, then across to Hyde Park. You can hardly build up a speed differential in a few metres, can you?

    Either that, or you cut the corner in a gauntlet of white vans gunning their engines for a dragster start, to go straight to the park. I tend to do that these days, after I had my first (of five) “collisions” there with a car which moved away before its light was green, so I figure I might as well be squashed for a sheep as a lamb.

    There is absolutely no question that the lights around Hyde Park Corner are configured entirely for the benefit of motors. It is nigh impossible to cross a road in one phase, and the green man phases are ridiculously short. Unless Boris is prepared to modifiy his “network assurance” strategy to give more priority to unmotorised traffic, there is little prospect of improvement.

  10. Nico (@nfanget) says:

    I think you got TfL wrong. They do not want to to test what interventions work and improve the model to ameliorate cyclists’ lives, but to make sure it does not have an impact on motor traffic volumes. As a scientist I am very wary of computer models anyway, especially those that claim to describe human behaviour (and the TfL model is ridiculously pro-motor biased anyway). For proof, just look at all those model-tested junctions that completely ignore desire lines of all users.

  11. Pingback: Bicycling binary – still fighting over a comb | Buffalo Bill's Bicycle Blog

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.