I write this with a deep sigh. Not because it’s an unimportant subject, it’s just that it comes up every so often and I need to remind people that floating bus stops are not the starting position, they are a design response to policy and network decisions. So, here I am again writing about floating bus stops.

Campaigns against floating bus stops sometimes break out into the media, but there’s also the drip-drip output from some people and organisations (especially on social media) which far from galvanising people to seek improved designs, end up drawing out all sorts of oddballs to have a pop at people cycling, attitudes which all to sadly are replicated on the streets and ends up with views being entrenched and prejudices reinforced.

There are some people and organisations who advocate for a ban of floating bus stops and have got themselves into a position where they will absolutely not countenance discussion about that position, let alone compromise. At the moment I can’t see how they can come back from such a purist position, but neither do I see them presenting any credible alternative designs (or any designs at all). It is very easy to say no.

I covered the issue of floating bus stops back in 2021 which was going to be my go-to response on the subject and there isn’t much for me to add in terms of site design, but I there are a few things I’ve developed my thinking on since then, and I wanted to preserve a quick bit of research I did for a social media thread. I will come back so a couple of design issues at the end though.

On the thinking side, my 2021 post links to research and gives practical advice on how to improve designs which I think remain 100% sound, but I’ll cover some more research later. What my post didn’t cover is the network-level issues and the design of networks is something that has only really properly crystallised in my mind in the last year.

The fundamental question is why do we have floating bus stops? It’s as fundamental as why do we have anything on the streets and the answer is because of motor traffic, but it’s even more conceptual than that. Floating bus stops are a design response to a network-level decision to enable people to cycle as transport and that’s the starting position. If we do not use floating bus stops at all, we cannot possibly enable people to cycle in a way that meets their wants and needs.

In the UK, we are very used a two-mode transport solution of either walking and wheeling on footways or driving on carriageways. Where those modes cross is where design comes into play and it has all to often been at the expense of pedestrians with layouts that encourage drivers to take priority and pedestrians to cede it, as well as complex junctions and crossings that take an age to get from one side to the other.

Prior to the last decade or so, cycling has been seen as a mode that should be accommodated in the carriageway and treated like little motor vehicles. That was reflected in national guidance as well as actually what was done on the ground. If you have a look at the numbers of people cycling where that approach has been taken, it comes as no surprise that numbers are low and the approach failed.

When we weren’t trying to paint people onto the carriageway, we tried to bolt cycling onto footways which often took space away from walking and wheeling, made cycling slow and perhaps made the users of both modes defensive of their space; and all the while, we were actually maintaining the motor-centric status quo.

From a network-level perspective, a three-mode solution will have three-mode needs and each mode will have its own network-level needs. The key is how we design where the modes coincide or cross. I write about this more in one of my business blog posts which of course mentions floating bus stops.

There has been plenty of research to show that people won’t cycle with fast and heavy flows of traffic and so when we design a cycle network, we need to remember that people have the same wants and needs to travel to get to places as with any mode, and so they will be travelling along roads that have a high density of services, but which are also in use as busy motor traffic corridors. It is impossible to design cycling networks which purely rely on side streets because most services are not found there.

One might argue that buses (and trams) should be thought of as a fourth mode, but I would say that the difference here is that they are not door to door in the same way as cycling is and (conceptually) they follow fixed routes which should make the planning aspect a little easier. One solution which is sometimes advanced is that where space is constrained, then buses and cycles should be routed on different streets.

The problem here is the density of services issue and people using cycling as a door to door mode. I think this is a position which could be explored if we are willing to remove other motor traffic and where space is still constrained, but this is a rare. If we end up with a bus only street, but it has relatively low levels of bus traffic, then in fact most people will be happy to mix if it is all low speed. If there are high levels of bus traffic, then people will want to be protected and clearing out general traffic might be enough to build high quality floating bus stops.

Anywhere that cannot achieve this is going to be an edge-case needing bespoke network planning work in any case and will need a high density road network to be achievable. But, it has its place in the discussion and it needs the UK to move beyond designing routes as the starting point and really get under the network planning skin.

In England, we have the Local Walking Cycling & Walking Infrastructure Plans technical guidance which is meant to help us plan for walking and cycling networks over a geographic area and then prioritise delivery. I think the guidance is over-complex and some of the data sources it suggests we use are flaky. There is a basic premise that people want and need to travel between places and that is what we have mostly given motoring in the door to door sense. There is, however a key failure in the guidance as it completely forgets to mention the motoring network and by extension, bus networks. If those aren’t being looked at, how do we ever rationalise things? It’s more about the rural situation, but I have a look at network planning in my business blog post here.

It is worth highlighting at this point that the busiest cycle route on the planet is also on a bus only street, but because bus services are frequent, there are cycle tracks. At the network level, this is a main cycle route in the local city network and the streets either side are just low traffic city streets which are accessible for cycling, but nobody is using them unless they need to be there. I am talking about Vredenburg in Utrecht.

The photograph above is a view along one of the passenger islands of the floating bus stops on a street which are needed because if buses and cycles mixed, there would be chaos and people would get hurt. The live bus times at the stop in my photograph can be found here which make interesting watching. I am not suggesting that this design works for everyone, but it is very much the design response to network-level decisions and from a city perspective, Utrecht is really squeezing out private cars which gives more transport options for those who don’t or can’t drive.

It is also worth mentioning the London Cycling Campaign’s new report "After Dark" which takes a critical look at London’s designated Cycleways and suggests that a quarter of the network is socially unsafe at night. It also suggests there has been too much reliance on indirect routes which use or skirt parks, canals and other lonely places. This isn’t directly linked to the design of floating bus stops, but it is a network-level outcome where we haven’t tackled what is happening on the nearby and more direct routes along streets and in the event we keep cycle routes and bus routes separate, this type of thing will be one outcome.

I can think of examples of lonely routes which get you away from traffic (and the density of services) because there is no appetite to tackle the main roads. I can also think of examples where main roads have been tackled and where there are floating bus stops, but the alternative parallel routes that would have avoided floating bus stops are indirect, sometimes lonely and sometimes on local streets with too much rat-running traffic.

In terms of research undertaken since my original post, Living Streets has looked into floating bus stops (and continuous footways). The report and research backing it is dense, but at the very least it is worth reading the executive summary which essentially calls out the lack of design consistency and poor layouts seen out in the real world, but it also says:

*However, whilst there is some concern about bus stop bypasses, our observational **data suggested the level of discomfort or difficulty most people experience in using **these bus stops, when well designed, is very low. Where a cycle track is extremely **busy – a much wider group of pedestrians can be disadvantaged, unless reliable *support is provided to allow easier crossing.

This pulls at network-level and local contextual design responses, and in my opinion, is completely sensible and more effort is needed on the really busy places.

Guide Dogs commissioned University College London to undertake research on floating bus stops (and other design elements). Again, it is well worth a read and there are a set of recommendations which on the whole are sensible in my view. There is one recommendation I take issue with however:

*Investigate different ways of enabling cycle lanes and bus stops to interact which do not raise safety concerns amongst passengers and pedestrians using the stop. This investigation should include consideration of alternative positions, alignments and routes for the cycle lane to ensure that both cyclists and pedestrians are safe around bus stops. Until findings from this research are available, stop the further installation of Floating Island Bus Stops and Shared Bus Stop Boarders. *

I am not getting into shared boarders here as it’s another discussion, but as I have said above, floating bus stops are a design response to the decision to enable cycling through network design and calling for their use to be halted until some so far undiscussed and uncommissioned research is simply not reasonable.

The suggestion that there is somehow "alternative positions, alignments and routes for the cycle lane" is at best naïve because if there were a magic layout that had 100% support from everyone with an interest, people would be able to go and point at real-life examples. It is certain that there is no alternative where two-way cycle tracks are used. They don’t exist for a very good reason, and this is where I want to preserve a bit of a thread I did on social media.

I did a very quick literature review on international design guidance for floating bus stops (ignoring all of the network-level discussions for a minute). I am not going to critique each one, but there is very obviously a certain level of familiarity about all of them.

Brazil (2007)

British Columbia

Chile

England

India

Massachusetts

NACTO (USA)

New Zealand

Scotland

London (TfL)

Wales

Of course, I can be critical here and just say that all of these essentially copy the Dutch approach and given the discourse in the UK, we have folks unhappy with the approach. That’s fair to a point, but my counter is that nobody has proposed a reasonable alternative that doesn’t throw people cycling back into traffic or just ignores their needs (eg, bus lanes).

I did find this layout in Colombian guidance, but it doesn’t take an engineer to explain how awful it is.

Colombia

For completeness, people cite designs from Copenhagen where passengers board and alight from the cycle track and people cycling stop for them as a matter of it being of traffic rules. It is a solution, but one which those who don’t like floating bus stops will like even less. From the perspective of Copenhagen, there might, however, be an argument about consistency and everyone knowing how it works and that might take you so far, but they have plenty of floating bus stops and the reality of the compromise is actually space and in many cases, not taking it from general traffic.

Copenhagen

In rounding up other thinking, I need to link to Wheels for Wellbeing’s thoughts and perspectives which remind us the need for inclusive consultation but that people choosing cycling need continuous protection.

Transport for London has also undertaken a safety review of floating bus stops in London. It found the risk of injury to be very low, but more than a third of stops not meeting their guidance and that some people remain concerned, especially with confusing layouts. Personally, I dislike some of TfL’s guidance which has one layout which funnels people cycling into single file and requires them to negotiate what I would say quite tight geometry to the point where their attention might be on negotiating that and not people crossing, but again, I’m getting drawn into detail.

TfL does suggest there needs to be more education in how the infrastructure is intended to operate, that boroughs need to be more consistent with design and that guidance needs to be reviewed. They also want to look at innovation to help visually impaired people use [floating] bus stops and research the behavioural issues around cyclists not giving way (I assume at zebra crossings over cycle tracks).

On the technology side, I do think the idea of using detectors at busier cycle tracks which provides tactile information on large gaps in cycle traffic is helpful as it doesn’t rely on administrative controls to stop cyclists to allow people to cross (because zebra crossings and signals are motor traffic responses). I can’t find a link to a trial, but I will update this if I do.

Finally, on the zebra crossing issue my 2021 post did link to the research TRL did for Transport for London some time back, but which is is worth mentioning again here. The study was limited, but did suggest that zebra crossings on cycle tracks are a feature that most people understand in terms of who has priority, even if the compliance from people cycling was decidedly mixed. In legal terms, zebra crossings over cycle tracks are very much real zebra crossings and so people cycling should be giving way to people crossing.

On balance, I think the zebras are useful from a wayfinding and consistency perspective and I recommend them to be used wherever someone is being formally guided to cross a cycle track.

Floating bus stops seems to have got themselves stuck in a never ending debate about a design response which comes way down the line from policy and network planning decisions in a way I can’t think happens with anything else we find on the street.

To some extent they are a worry to folks who rely on buses and have fought for accessible services and floating bus stops perceptibly erode that hard won accessibility. I can’t argue against that, but equally, I can’t support positions that when taken back up the policy chains mean we have to maintain the status quo, which is pretty much motor traffic domination of most of the country. You can’t research your way out of it, you need to build you way out of it.