Saturday 30 October 2021

Floating Bus Stops

One thing which seems to come up time and again in street design is as soon as a concept gains a name, it also gains controversy which ends up drowning out the voices of people who might have genuine concerns to work through.

Floating bus stops are one such "thing" and so this week, I'm trying going to try and get under the skin of the subject a little bit and along the way, pick up some of the concerns and also point you towards some research.

So, what is a "floating bus stop" (also known as a bus stop bypass)? It's simply a bus stop which has a carriageway or a cycle track running behind it so that people board or alight a bus from what is essentially either a large traffic island or perhaps something which looks like central reserve.

The photograph above shows a floating bus stop. Boarding is from a layby on a 40mph dual carriageway and the carriageway to the rear is a service road. As far as bus stop environments so, it's pretty poor with a narrow island between the two roads and no decent dropped kerbs to help people cross between the stop and the footway to the service road footway. This layout is decades old and apparently entirely uncontroversial, although if someone wanted to put parking restrictions into the service road to keep an area clear for crossing, then you'd hear about it.

The service road in the first photograph is probably not too busy and so here's another example. This time the island is essentially on the desire line along the street and so from the bus stop (behind the tree) the crossing of the service road will only be something people do for the shops or the side streets at each end of the parade, although this service road is going to be busier and it's parked wall to wall. Again, apparently entirely uncontroversial and probably because again, it is decades old and for better or worse, just part of the street fabric.

The photograph above isn't quite a floating bus stop because the designer decided to make the area behind shared-use, but to all intents and purposes, it is used as floating stop. It's on the far eastern reaches of London's Cycleway 3 which was opened in 2010. In fact, this cycleway was actually built around a decade earlier as part of the A13 being built to bypass its old alignment. If you go a little further east, you'll see the original green cycle tracks; the C3 scheme was merely a cosmetic job.

The point I am making here, is the concept isn't new and you'll be able to find service road and cycle track layouts going back decades. There's a gallery of 50 examples pulled together by Nick Kocharhook. Of course, just because something is widespread doesn't make it good or correct, but floating bus stops are simply a consequence of design philosophies adopted for particular sites and schemes over the years which have commonality in principles.

In more recent times (maybe the last decade), floating bus stops acquired a name and they became a "thing". On Lewes Road in Brighton (below), several were built as part of a bus priority scheme completed in September 2013 and which I visited in June 2014.

The old layout was a dual carriageway with two traffic lanes and a mandatory cycle lane in each direction. The bus stop was in a lay-by so buses would be swung across the cycle lane. The new layout provided a bus lane in each direction to prioritise buses (i.e. passengers) and although there was still a cycle lane, it was buffered from general traffic from the bus lane and bus traffic is obviously far lower than general traffic.

From a bus passenger point of view, the bus stops were made accessible with a high kerb to be more compatible with low floor buses and because there was no longer the need to pull into a layby, bus drivers can get very close to the kerb too. Having the bus lane means bus drivers don't have to pull back into the traffic flow and with the floating design, taking cycle traffic behind the stops also reduces the cognitive load on the bus drivers. This is sustainable safety in action.

The Lewes Road stop layout is become quite familiar as it repurposes old laybys where space was originally claimed to benefit drivers by taking buses out of the traffic flow and on 40mph and higher roads, laybys are still useful from a safety point of view. Having buses stopping in traffic (if not in a bus lane) is good from a bus priority point of view as this will help create a clear run ahead and is good for journey time reliability. A monitoring report can be read here.

While the Lewes Road scheme was being constructed, Transport for London was trialing floating bus stops with the Transport Research Laboratory as part of a much wider set of cycling innovation trials. This work helped to influence the Department for Transport in changing traffic rules to help cycling as well as introducing mini-zebra crossings for cycle tracks and low level cycle signals.

I took part in some of the trials and in November 2013, I went back to the outdoor laboratory for a look around the floating bus stop layout (plus a few other things). It was from then, we saw the concept revitalised and used across the UK. Some designers did things slightly differently and the quality of the layouts varied, but the principle of taking cycle traffic behind the bus stop passenger area was rediscovered. The study report can be read here.

Of course, this approach has long been the norm across the North Sea and beyond. Floating bus stops are everywhere in the Netherlands as you might expect and have been operating for decades.

The photograph above is of stop in Utrecht with a two-way cycle track behind it. The passenger area is too small as you can see from the worn grass and the shelter arrangement blocks the intervisibility between pedestrians and cyclists at the formal crossing point. It took me a few minutes to work out that this bus stop no longer exists. I saw this in the summer of 2018, but the bus network had been changed by the time you see the site in 2021 - the Dutch are always adjusting things.

Above is a bus stop in Malmö, Sweden, which has a very large passenger area. This approach uses contrasting materials (color and texture) to pick out the footway/ passenger area and cycle track which seems to be quite common in Sweden and Denmark.

So, what is controversial about floating bus stops? Well I have to state that I can only speak generally about people and I cannot speak for them. People have different experiences and the risk of me generalising means I will miss personal nuance. But this has to be said.

The main area of concern comes from visually impaired people who may feel worried about trying to cross a cycle track, especially where they don't have enough vision to see oncoming cycle traffic. There is a perception that cyclists won't be cognisant of pedestrians more generally and coupled with cycle traffic being quiet, there is a genuine concern.

However, other people who might have issues are wheelchair users who may feel they are less visible to people cycling, people with young children who may not understand the layout, older people who find it harder to judge speed and neurologically diverse people who may struggle with an inconsistent street layout.

The way we can deal with these concerns in the round for my mind is about simplicity and consistency which I'll come to shortly, but it is important to pause and to try and get past some of the noise coming from those with genuine concern for people who need more consideration as well as those who like to use Disabled people and others to project their opposition to schemes which enable people to cycle.

One example was the objection of Guy's & St Thomas' Hospital in 2016 to floating bus stops outside their site on Westminster Bridge (above) which is covered in detail by Mark Treasure here. The objection was partly against the proposed layout because of impact on visually impaired patients using the hospital's specialist eye department and partly Transport for London's engagement approach. We often see cycling push back from well meaning groups and maybe less well meaning people who perceive cycling as threat to individual people's ability to travel independently or at it's most sinister, having other motives which see cycling as a general threat to society.

As a designer and someone interested in all of this, it is sometimes difficult to separate genuine concerns from the other, specially where people are using genuine concerns as a front for their culture war. All I would say is people should look very carefully at some of those who cluster around these causes because they do not have your interests at heart. It's a difficult and emotive subject, but with the Guys' & St Thomas' site, the scheme was built as you can see in the photograph above and as far as I am aware, the noise moved onto the next scheme that people wanted to derail. In fact, the hospital's charitable trust is now helping fund low traffic neighbourhoods in the local area.

The Westminster Bridge scheme uses mini-zebra crossings to give additional assistance and priority to people crossing the cycle track as in this case, Belisha beacons were added to increase the conspicuity of the crossing (a purely optional addition). Mini-zebra crossings have been used elsewhere in London and some were examined in detail looking at behaviour and user comfort. 

The photograph above is one of the trial sites on Blackfriars Road. Summary information on this and other cycle safety trials can be read here and in more detail here for floating bus stops.

In general, the research found mini-zebra crossings to be generally positive as they helped visually impaired people locate the crossing point. There was no real evidence that the Belisha beacons added anything to the layouts and there was little in terms of behaviour change of cyclists giving way or their speed, but priority was seen as being clearly established. This is a piece of cultural interest as zebra crossings are administrative controls and where people are moving under their own steam, they'll avoid stopping, but having a mini-zebra crossing ahead at least gets people prepared to adjust their speed or path - this is a blog post in its own right!

There was a detailed part of this research where Disabled people were accompanied by researchers to get a more detailed appreciation of lived experiences and this can be read here. This is only recording those who took part and everyone might have different experiences, but it is at least some input from those who perhaps need the most help from street layouts. I would recommend Table 6 to designers as this gives some suggestion on what would improve layouts such as;
  • Ensuring the island is large enough for wheelchair users to manoeuvre and for the numbers of people using it,
  • Making sure the correct tactile paving is used, included stems reaching the back of footway to guide people to the crossing point,
  • Provide consistently laid out dropped kerbs,
  • Training for cyclists and visually impaired people, including guide dog users they understand how the layouts work,
  • Audible announcements on buses to people know they are leaving at a floating bus stop.
  • Good bus stop layouts with consistent layouts so bus drivers always stop in the right place. This allows rear doors to align with the crossing point so users have an easy route.
  • Making sure there is no clutter for people to walk into on the desirte line (such as bins).

There are other things discussed and I would add that we need to make sure bus shelters don't block the views between passengers and cyclists which is done well at the stops in Cambridge (above). The floating bus stops built in Cambridge have also been subject to analysis, this time by Sustrans. Their report from December 2015 is here. It doesn't stop the claims that the layouts are dangerous. In this report, someone with political ambition suggesting that "lots of old people" were getting hurt without any actual data. The person wanted to explore alternative cycle routes and I would suggest that this is yet again projection by someone who didn't want the cycling scheme at all.

Transport for London has incorporated floating bus stops into its accessible bus stop design guide (here). Although it does rather push the idea of having the change in direction and narrow cycle track in an effort to slow cyclists down. Unfortunately, this type of layout (as above) often means high kerbs on both sides and changes in direction which create collision risk and which are more difficult to pass using a non-standard or cargocycle. There is also extensive advice in LTN1/20 - Cycle Infrastructure Design.

In general, two-way layouts make this idea of slowing people down redundant anyway and in my view, trying to do this won't affect the reckless too much, but risks other people having a collision. Other ideas which have been tried are rumble strips which just make cycling uncomfortable for many and painful for some and more exotic ideas which using traffic signals or lights embedded into the cycle track to try and manage interactions at crossing points. Frankly, if cyclists aren't stopping at mini-zebra crossings, they won't stop at signals, again, a function of how administrative controls are less effective where people are moving under their own power.

I have seen other suggestions such as barriers which close the cycle track when a bus arrives, but failing to understand that people will just jump onto the footway to go around or people who say they are happy with cycle tracks but that cyclists should reintegrate with traffic at bus stops. Floating bus stops are designed to protect people cycling and simply throwing them back into mixing with buses is just downright dangerous - cycling with buses is not attractive to most people (above). There's also people with the to provide some sort of detection and messaging to visually impaired users that there is a clear gap in cycle traffic. This will be interesting to follow, although by implication a technological "fix" will add costs and complexity. It might be helpful in very specific circumstances such as the Westminster Bridge example above. I remain to be convinced.

So, are floating bus stops safe? Well nothing is ever 100% safe and perfection is an impossibility. There are always risks, but from a designer point of view they are low risk and the residual risks can be managed through good layout design and potentially some local education for users (which we seem to have missing more generally) - they are safe enough. At a population level, more people cycling is a good thing for so many reasons, but we have to remain mindful of people who need help. 

I think the skill here is to concentrate on site specific issues with the actual people who might need additional help or at least additional engagement. This requires us to filter out the projectionists and may I respectfully suggest that some campaign groups need to work harder to support individuals rather than issuing blanket objections to schemes. There has been lots of research on floating bus stops now, and quite frankly, I have had enough of having to justify a decades-old concept every time someone doesn't want change. In future, I'm going to point at this post and say "there, that's my view".

Update 31/10/2021
Thanks to Twitter, I have been alerted to another piece of research, so thanks to @davelostdave for the reference. This time, it's an evaluation report by Transport for Greater Manchester for a floating bus stop design proposed for use on Oxford Road, Manchester. A design was developed through stakeholder engagement with a trial built on Oxford Road near the junction with Hathersage Road (similar to the one also built a little further north - below and which is outside Manchester University Hospital).

The trial site was also evaluated using video surveys and additional engagement (including user interviews) which led to adjustments being made to the design for the wider scheme. What is notable is that Oxford road was (and still is) a very busy corridor for buses and cycle traffic and so there was lots of data on "interactions" between passengers and cyclists.

A couple of interesting points for me is that cyclists were travelling at 13mph on average (so some faster and some slower) and over the 7 day trial, nearly 52,000 passengers crossed the cycle track (309 an hour on average) with no actual collisions recorded in the video surveys and very few interactions which (my interpretation) were dangerous.

User surveys showed perhaps more caution from segments of the population who may require more consideration or assistance and indeed, there was disagreement between different groups on what adjustments should be made. The following statement from the report for me is pretty helpful;

There was clear consensus amongst the workshop attendees that the segregation between the footway, cycle bypass lane and bus stop platform was positive and effective. The recessed bypass lane improve awareness of the different areas for pedestrians, cyclists and bus users.

Personally, I think the rumble strips used on the final scheme are uncomfortable and the cycle track is a bit narrow, but the layouts on Oxford Road and clear and legible. Again, lots of work put into testing a layout which only became necessary because we gave it a name. More on Oxford Road here.


  1. Thank you, a good read. My thoughts, for what they are worth, as a cyclist and former (now occasional) bus user.
    I think a lot depends on individual cases. I do have some reservations where cycle lanes and bus stops or pavements are very busy, or there isn't much space etc.
    There are certainly some bad examples around. C20 seems to deposit passengers down directly into the (shared) cycle lane.
    Much as I would love to see everyone cycling, I do think great caution has to be taken not to deter or hinder bus-use in order to entice relatively few additional cyclists (compared to numbers of bus users, many of whom are especially vulnerable or have limited transport options).
    Perceived danger is a big thing I feel for pedestrians and bus-users - this has to be taken seriously (they shouldn't have to feel anxious and at risk), but also it's difficult to know how much it will lessen as changes (like floating bus stops) are normalised.
    It can also be frustrating and worrying for a cyclist to have to carefully pick their way through bus stop areas, often I would prefer to just be in the main carriageway but leaving a physically segregated lane is difficult. I wonder if one option (on low speed roads) would be returning cyclists to the main carriageway but giving them the whole lane as priority cycle lane for the length of the bus stop, with drivers required to give way with dashed white lane right across the lane, obviously not physically segregated but with ramps/bumps/coloured tarmac to enforce driver caution). No idea if this is feasible.
    Lastly, part of the reason cyclists are reluctant to mix with buses (and indeed to sit behind heavy vehicles at junctions when it may be safer than nipping up the inside) is the exhaust fumes. It's possible that electric buses would make sharing lanes a bit less unwelcome and perhaps obviate some of the need for separate infrastructure.

    1. C20 certainly has issues, although there are no floating bus stops as such; most are essentially bits of shared paths and some look kind of floating, but only by virtue of materials - they are at the same level and there isn't much contrast - I prefer to see the cycle track at a different level.

      In terms of busy sites, the extra case study I have added is very busy for passengers and cyclists, so if properly designed, it's fine.

      The problem with integrating people back into the carriageway at bus stops is apart from be predictably dangerous on high frequency bus routes it wouldn't feel very pleasant for people we are trying to get cycling (electric buses or otherwise) and in any case, there is no existing law, sign or marking which gives cycle traffic priority.

      My view is well designed floating stops are fine and given the number of trials since we have named them, there really isn't a reason not to deliver them.

  2. Yee Olde Bus Stop Bypass,-0.1723857,3a,75y,208.02h,68.13t/data=!3m6!1e1!3m4!1sERFkOp3-HW041sMmQ-WJaQ!2e0!7i13312!8i6656

    Shame the cycle track ends into nothingness just to the left.