All posts by Oliver O'Brien

Making Bikeshare Pay

Dockless bikeshare has been out in London for around 10 months now (not withstanding the brief but spectacular oBike launch and burnout in mid 2016). It looks like the operators are getting serious and trying to get dockless bikeshare to pay for itself – starting a transition from a “growth” to “profit” focus:

Mobike has introduced an out-of-operating-area fee of £10 for London (left). Interestingly, this is only £5 for Oxford (right), where arguably the competition is even more fierce:

It isn’t clear whether you can continue to start journeys from outwith the operating area, but you must finish within them to avoid the fee. However, you can also get the fee waived by taking any Mobike from outside the operating area, back inside it, within 12 hours. It could be the bike you took outside, or another one. This is interesting – let the user help the system work better. You either fix the problem (bikes outside areas the operator can manage effectively, or don’t have permission to be in) or pay the operator to do it for you.

Mobike is making a bigger deal of its hubs now, showing the numbers of bikes currently available at them. Bikes in a circular buffer around each hub are reported as at the point in some cities (Oxford – left) but individually in others (London – right). This makes them a little more like a hybrid system, with certain locations generally having a reliable pool of bikes available, but with out-of-hub parking still available to keep the system flexible. At this point, there is no user incentive for hub-based journey finishes or starts.

Mobike have also shrunk their operating areas – in Islington, it now doesn’t extend north of Holloway. Presumably Mobike are tired of moving their bikes back up the hill to Archway and Highgate, only to have everyone cycle them back down. However they have extended into a small part of Hackney – the De Beauvoir Town area. They have also shrunk their Southwark footprint, so that the bikes can only be used north of the South Circular, and their western footprint has also shrunk – no service in west Hounslow, Feltham, Southall or Greenford now. They has also completely pulled out of Newham. Looking at their app, it never looked like there were many bikes available for use there anyway. You can see the area of operation here (orange border).

In addition they have significantly increased their usage fee. Initially it was 50p for 30 minutes. Now, it is 50p for 20 minutes on the smaller-frame bikes with black baskets (show as orange on their map), or £1 for 20 minutes on the larger-frame bikes with orange baskets (shown as white on their map). [Updated 12 July – now £1 for 20 minutes for all their bikes].

Ofo has also increased their usage fee, it is now 70p for 30 minutes. So, in places where both systems operate, the small Mobikes are cheaper for journeys up to 20 minutes, the Ofos are cheaper for 20-30 minute journeys, the small Mobikes are cheaper for 30-40 minute journeys etc etc…

Single Journey

Ofo Mobike Small Mobike Big Santander Cycles
15 min 70p 50p £1 £2
25 min 70p £1 £2 £2
35 min £1.40 £1 £2 £4
45 min £1.40 £1.50 £3 £4
1h 15m £2.10 £2 £4 £6
1h 45m £2.80 £3 £6 £8

Subsequent Journeys in next 24h

Ofo Mobike Small Mobike Big Santander Cycles
15 min 70p 50p £1 Free
25 min 70p £1 £2 Free
35 min £1.40 £1 £2 £2
45 min £1.40 £1.50 £3 £2
1h 15m £2.10 £2 £4 £4
1h 45m £2.80 £3 £6 £6

Ofo and Mobike are also now showing specific forbidden areas on their app. This is interesting, because I assumed that all areas outside of their operating area were forbidden. Presumably these are extra-problematic areas for the operators. Ofo has marked various royal parks as forbidden areas, while Mobike has marked canal-sides, the London Bridge station complex, and large public parks, as forbidden. Mobike’s out-of-operating-area areas allow journeys to finish but the user (only) is then encouraged to move the bike back into an operating area within 12 hours, it is locked out of use for others.

This means that in effect there are five types of dockless system geofences, with four being shown on the maps in the Mobike and Ofo apps:

Ofo Mobike
Hub area Shown in app as points overlaid on green-bordered rectangles. Parking here encouraged, occasionally incentivised with free next ride coupons. Shown as points, but a blue circular buffer is shown on clicking it. Parking here encouraged but not incentivised.
Operating area Show as green-bordered areas Shown as blue shaded/bordered areas
Out of operating area Shown as regular map. Parking locks bike for user-only redistribution back. Shown as regular map. Parking results in fine, unless it (or another) is taken back in in 12 hours by the user, but still available for use.
Forbidden area Shown as red shaded/bordered areas. Parking may result in credit drop or penalty. Shown as grey shaded/bordered areas. Parking may result in credit drop or penalty.
Out of region Not shown by either. But I assume there is a distance, out of the operating area, beyond which the operator would not seek to recover the bike into the operating area, but might sanction the user instead. For London, the GLA boundary might act as such an area

Meanwhile, the third dockless player in London, Urbo, has announced it is quitting the three London boroughs it is operating in – Enfield, Waltham Forest and Redbridge. Enfield will announce a replacement “soon”, Waltham Forest is getting Ofo in instead, and Redbridge already has Ofo. The numbers don’t sound great – only 6000 journeys in Waltham Forest in 5 months, on a fleet of 250 – so 20 journeys a day, or 0.16 journeys/bike/day, and 3000 miles clocked up in Enfield, again in 5 months, on a fleet of 100 – so (assuming average journey of 1 mile) 0.2 journeys/bike/day. This means these bikes are spending 99.8% of their time not being used (assuming an average journey of 15 minutes).

Barnet had just announced that Urbo was coming there, and Barking & Dagenham was due to sign-off Urbo coming there in July – I would speculate that these launches may be off. Urbo, an Irish company from the start, has just been bought by the operator of several bikeshare systems in towns in Ireland, and it has just launched Dublin, where it has access to the whole of the city. Just like in London, Dubliners likely want to cycle throughout the urban area and not within arbitrary political boundaries. So that would likely be where the London Urbo bikes are going.

Numbers of bikes and journeys are generally hard to come by for dockless bikeshare – the companies themselves have good commercial reasons for being coy with the numbers. In some cases, the operator will have fewer bikes out on the streets than they say – as they just don’t have that many (working) bikes – Santander Cycles have consistently overstated the numbers of bikes out there . The opposite can also be true – where dockless operators have agreed maximum numbers of bikes with boroughs, and then see these numbers be exceeded as users cycle over from neighbouring boroughs, or operations necessitate. There is little data, so we generally have to go on press releases:

Number of Ofos, Mobikes and Urbos in London

I am adjusting the following table as I get better information:

Ofo Mobike Urbo Santander Cycles
September 2017 200 750 250 9514
November 2017 400 (a + b) 1150 250 10038
December 2017 1000 1150 250 10211
February 2018 1000 (38km2) 2000 (130km2) 350 10513
March 2018 1350 Ann: 2200
Obs: 2616
475 10651
April 2018 1400 (a + b) Ann: 2200
Obs: 3315
475 10441
May 2018 2800 Ann: 3000
Obs: 3072
475 10279
June 2018 3000 (253km2) Ann: 4000 (68km2)
Obs: 1980
475 (177km2) 9716 (112km2)
July 2018
(provisional –
will be updated)
Obs (I): 1783 Obs: 1069 (68km2)
Obs (I): 889
CLOSED 9151 (112km2)

In better news, both Ofo and Mobike are both continuing to expand in London this summer, in areas where they do think dockless bikeshare will work. Mobike should be coming to Haringey soon, while Ofo have today announced their forthcoming expansions – Camden launches this week, with Waltham Forest following shortly (this week, says the borough at least), followed by Wandsworth and Hammersmith & Fulham. This means yellow bikes will soon be appearing in Bloomsbury, Fizrovia, Camden Town, Kentish Town, Highgate, Belsize Park, West Hampstead, Walthamstow, Chingford, Leyton, White City, Hammersmith, Fulham, Putney, Wandsworth, Battersea, Balham, west Clapham and Earlsfield.

Visit the new oobrien.com Shop
High quality lithographic prints of London data, designed by Oliver O'Brien

Where Might Bikeshare Succeed in Great Britain?

There’s lots of bikeshare systems in the UK now. As well as the third generation dock-based bikeshare systems, fourth generation dockless (and hybrid) systems are starting to appear on various streets around the country, led by Mobike, Ofo and Urbo, three dockless providers and operators.

I’ve put together this simple model to try and understand where systems are most likely to be successful, for which I’m defining as a lot of (legal) journeys made with each bike placed by the operator. To do this, I split the country into its local authorities, apply three scores, and then multiply them together to produce an overall “propensity for bikeshare”, or PFB score (the name is a nod to PCT.bike from Lovelace et al) which can then be ranked.

Many mid-sized cities in the UK have their own local authority, approximately covering the urban area, while Manchester and London are split into multiple LAs. Conversely, some LAs individually cover multiple smaller towns/cities and large rural areas too. Hence, this is a very simply model which is not going to be completely fair to every urban area (Stirling, in particular, gets pushed well down the rankings as its LA includes a huge rural area). Still, for most of Great Britain, it produces results I would anticipate and is a good start towards potentially developing a more sophisticated model. The local authority geography is also appropriate, as local authorities act as the gatekeepers for which access is negotiated (for systems which pay themselves) or as authors of bids for subsidised systems.

Model Inputs

For this first, simple model, my three compounding factors are:

  1. Residential and workplace population density – on the assumption that bikeshare systems need a critical mass of people passing by their bikes/hubs/docks, in order to be seen and used sufficiently frequently to justify the costs of equipment/maintenance, on the basis that a major source of income is per-use fees. Both residential and workplaces populations are used, as people have journey opportunities that can be facilited with bikeshare, both from their office (e.g. lunchtime errands, commute to evening socialising or back home) as well as their home location.
  2. Proportion of people who already commute by bicycle, again looking at both workplace location and residential location (bearing in mind that many commute journeys, particularly in London, cross local authority boundaries). While such people are less likely to convert to bikeshare, as they already have a means of cycling, their presence on the streets and associated culture and facilities (e.g. bike coffee shops, marked cycle routes or existing cycle-friendly infrastructure) help normalise the idea that cycling is a possible option for a journey need.
  3. Vandalism rate – theft and criminal damage to bikeshare bikes happens and it is an expensive one for the operators. It can be what causes systems to fail – particularly if they run out of bikes, or broken bikes litter the landscape and turn public opinion against the concept. While the dock-based systems are less vulnerable to this, as the bike is either securely locked to a clamp or in the hands of a paying user, dockless bikes are particularly vulnerable to vandalism, as they are not secured to immoveable objects, and their locks are, unfortunately, relatively easy to break by a determined offender looking for a free bike. The tendencies for vandalism of property that is not yours does vary significantly around Great Britain, while stereotypically it is likely to be more urban areas and areas with a younger and less educated population that is more likely to vandalise, using actual crime data on vandalism allows a more nuanced approach to be taken. Other crime classes (e.g. theft) were also considered for this model, but I think that vandalism rates act as a good proxy for how the local population around a bikeshare system will “care” for it or abuse it.

I have not included the absolute populations of local authorities in the calculation, as in general, with one exception (Isle of Scilly), LAs all have a significant night and/or day population, so they are all large enough to have a self-contained system. Another obvious factor, hilliness, is likely already correlated with proportion of cycling commuters and so is not included. N.I. is excluded from the model for now. Data sources include the latest police crime statistics (with populated-weighted averaging when across multiple LAs), and census data.

Results: Propensity for Bikeshare by Local Authority

Here are the results of the model run. Clicking an underlined title takes you to the main bikeshare for that area – forthcoming systems in brackets.

Some notes:

  • London boroughs score consistently highly, and even London as a whole, which is interesting as outer London is anecdotally not known as a particularly cycling-friendly place. Considering the size of the city, and the intimidating conditions cyclists often have to put up with in much of the capital, it is great to see it scoring so highly here.
  • Bikeshare operators are doing their homework and generally, the top end of the list is already well populated with bikeshare systems, in some cases multiple systems are competing.
  • The top local authorities without a bikeshare system (operating, announced or consulting) are Haringey, a north-London inner-city borough which has been surprisingly quiet until now, and Merton, an affluent outer London borough to the south. Oustide of London, the highest ranking areas without a bikeshare are Portsmouth (small system launching this summer), Gosport (neighbouring), Gloucester, Poole, Worcester and Hull. With the exception of Hull, these are all southern English urban areas, with generally affluent populations and some established cycling culture.

See also my London borough bikeshare scorecard.

Visit the new oobrien.com Shop
High quality lithographic prints of London data, designed by Oliver O'Brien

Bikeshare Docking Station Data Release

My research lab, the Consumer Data Research Centre, is making available much of the docking station empty/full observation data that I have been collecting, at frequent intervals (up to every 2 minutes) for generally the last 5 to 8 years, for over 50 cities around the world, to facilitate and encourage further quantitative research in the field. I already get numerous requests for this data – it is very interesting for all kinds of research projects, particularly because of it spans multiple years, so introducing this new mechanism is a good way to manage these requests. You can see the cities listed at the CDRC Data product page.

For three cities (Cork, Limerick and Galway in Ireland) we have directly attached the data to the record page. It is the data that has been collected up to the point that the record was published a few days ago, and on request we will reload the data, capturing more recent observations.

For the other cities, the data record is “Safeguarded”. This is because the logistics and technical limitations of attaching the very large amount of data to the records. Namely, it takes quite a lot of time to prepare each dataset for download, and the platform we are using is not designed for hosting extremely large files – plus, it is likely that researchers would want the most up-to-date data, meaning that we would need to build a mechanism to update the record regularly. Using an application process also minimises spurious requests – we have invested time both in collecting the data and processing it, so we need to be confident that it will be used. Additionally, some systems (particularly those in the USA) come with their own data licence restrictions from the operating companies meaning that we cannot freely distribute the “raw” data. In Europe, most of the datasets are explicitly open, meaning use of the data is unrestricted (although normally requiring attribution). the Irish cities listed above have a slightly more restrictive licence, requiring us to distribute it on the same terms, which we have done.

The data is available on application to interested institutional researchers. In practice, this means academic, public sector and non-profit organisations who which to carry out public/publishable research with the data. Application details are attached to each record.

Above and below are simple graphs produced from the data for various cities. I have looked at the number of bikes available every day at around midnight and plotted it on a simple graph against time.

Access the data here.

Stay tuned because I am planning on releasing two further “data portals” of bikeshare data, soon. These are slightly more manageable in terms of file size and administration, so I am aiming for these to be directly downloadable.

Population Analysis of London Bikeshare Systems

Mobike, one of London’s four bikeshare operators (with Urbo, Ofo and Santander Cycles) as today expanded to Newham. The operators are being driven by different borough approaches and priorities, which is resulting in a patchwork quilt of operating areas, although the London Assembly is today pushing for a more London-wide approach to regulation of the field.

Over and above the map linked above, I’ve done some population analysis to look at how the four different operators compare in London. Demographic figures are from the 2011 Census so total population will have gone up a bit, and cycle-to-work population up a lot. Nonetheless, the figures still allow for a useful comparison. The populations here are working age (16-74) populations. The differences across the operators are dramatic:

Operator
 
# of
Boroughs
Area
/km2
# of
Bikes
Average
Day/Night
Pop
Bikeshare
Bikes per
Person
Pop
Density
pax/km2
Cycle
To Work
Pop %
Santander Cycles  5, + 6 (part) 112 10200 1,520,000 1:150 13500 4.8%
Mobike  6 196 1800 1,425,000 1:800 7300 3.5%
Ofo  5 123 1300 1,040,000 1:1000 8500 4.0%
Urbo  3 177 500 570,000 1:2800 3200 1.2%

I have calculated the populations by averaging the day-time workplace population and the night-time residential population, making a very rough assumption that people spend their waking hours split roughly between work and home. Santander Cycles, the dock-based system, has been around since 2010. The others are all dockless operators and launched in 2017.

The high population density where Santander Cycles works in its favour, as does its high bikes/population ratio, with one bike available for every 150 person who lives and/or works in the area. Urbo, on the other hand, is mainly targetting populations that both have a low population density, and a low cycle-to-work percentage – two factors that would work against it. Mobike and Ofo sit in the middle, with the former quite a bit larger than the latter at the moment, but the latter operating areas with a more established tradition of cycling (using here the Cycle to Work population as a proxy for cycling in general).

Changing Broadband Speeds in the UK

The Broadband Speed map has been one of the most popular maps that the Consumer Data Research Centre has ever published on our CDRC Maps platform. The map is based on data from Ofcom, the UK’s digital connectivity and broadcast media regulator, and I was invited to talk at their Innovation Workshop event, hosted by ODI Leeds, earlier this month. My brief was to demonstrate the Broadband map but also critique Ofcom’s open data offering (which provided the data for the map). The talk slides can be found below:

As part of the preparation for the event, I produced a new version of the the Broadband map, showing 2017 data from the Connected Nation report (the original was based on the 2016 data). This gave the opportunity to therefore prepare a third map, showing the change between 2016 and 2017. Note that this is showing the change in the average broadband download speed experienced across both business and residential premises conneections, averaged by postcode with each postcode averaged then averaged again across the local output area (which typically contains five postcodes for residential areas, but many more than this for business areas.) The metric population numbers displayed when you mouse across each area, therefore, is the number of business and residential connections – typically 50-150 for the latter.

The map shows a general light green gradient across the country, showing broadband connection speeds are gradually increasing, as more and more fibre to the cabinet (FTTC) is installed and people change organically contracts to providers with better service. The places where other colours appear are the interesting results. Large increases are seen in rural Lancashire, near Kendal in the Lake District, as a community-driven ultra-high-speed rural service there continues to roll out. More dramatic improvements are seen just to the east of Cheltenham, again a rural area with specialist high technology and defensive industries.

Cranham, for example, has seen a 11000% improvement, from 1.7mbit/s to 190mbit/s, as new business connections have come online:

Appleton, on the other hand, has seen a 99% decrease, from 540mbit/s to 2.3mbit/s:

In London, the drop around King’s Cross, the previous year’s fastest postcode, is almost certainly not due to a general decrease in available speed, but actually because residential connections have come online, and demonstrates the problem with aggregating by the residentially defined “Output Area” geography. The previous, ultrafast result was likely due to dedicated ultra-highspeed links into Google’s new UK office, and other high-technology businesses opening there. Since then, the residential blocks nearby have opened. These still have pretty nice connections, but not the business-level infrastructure needed. So, it shows as an average fall in London.

Rotherhithe is always an interesting area:

A traditionally very poorly connected area, both in transport but also digital connectivity, it has seen dramatic improvements in many areas. but also big falls in the newest area – again possibly due to an increased residential component in the mix.

Explore the broadband difference interactive map.

FOSS4G UK 2018 Meeting and OpenLayers 4

I attended and presented at the FOSS4G UK conference in central London, in early March. I was scheduled to present in the cartography track, near the end of the conference, and it ended up being an excellent session, the other speakers being Charley Glynn, digital cartographer extraordinaire from the Ordnance Survey, who talked on “The Importance of Design in Geo” and outlined the release of the GeoDataViz Toolkit, Tom Armitage on “Lightsaber Maps” who demonstrated lots of colour compositing variants and techniques (and who also took the photo at the top which I’ve stolen for this post):

…and finally Ross McDonald took visualising school catchment areas and flows to an impressive extreme, ending with Blender-rendered spider maps:

My talk was originally going to be titled “Advanced Digital Cartography with OpenLayers 4” but in the end I realised that my talk, while presenting what would be “advanced” techniques to most audiences, would be at a relatively simple level for the attendees at FOSS4G UK, after all it is a technology conference. So, I tweaked the tittle to “Better…”. The main focus was on a list of techniques that I had used with (mainly) OpenLayers 4, while building CDRC Maps, Bike Share Map, TubeCreature and other map-based websites. I’m not a code contributor to the OpenLayers project, but I have been consistently impressed recently with the level of development going on in the project, and the rate at which new features are being added, and was keen to highlight and demonstrate some of these to the audience. I also squeezed on a bonus section at the end about improving bike share operating area maps in London. Niche, yes, but I think the audience appreciated it.

My slides (converted to Google Slides):

Some notes:

  • My OpenLayers 2/Leaflet/OpenLayers 3+4 graphic near the beginning was to illustrate the direction of development – OpenLayers 2 being full-featured but hard to work with, Leaflet coming in as a more modern and clean replacement, and then OpenLayers 3 (and 4 – just a minor difference between the two) again being an almost complete rewrite of OpenLayers 2. Right now, there’s a huge amount of OpenLayers 4 development, it has momentum behind it, perhaps even exceeding that of Leaflet now.
  • Examples 1, 3, 4 and 5 are from CDRC Maps.
  • Example 2 is from SIMD – and there are other ways to achieve this in OpenLayers 4.
  • Examples 5, 6 and 9 are from TubeCreature, my web map mashup of various London tube (and GB rail) open datasets.
  • Regarding exmaple 6, someone commented shortly after my presentation that there is a better, more efficient way to apply OpenLayers styles to multiple elements, negating my technique of creating dedicated mini-maps to act as key elements.
  • Example 7 is from Bike Share Map, it’s a bit of a cheat as the clever bit is in JSTS (a JS port of the Java Topology Suite) which handily comes with an OpenLayers parser/formatter.
  • Example 8, which is my London’s New Political Colour, a map of the London local elections, is definitely a cheat as the code is not using the OpenLayers API, and in any case the map concerned is still on OpenLayers 2. However it would work fine on OpenLayers 4 too, particularly as colour values can be specified in OpenLayers as simply as rgba(0, 128, 255, 0.5).
  • Finally, I mention cleaning the “geofences” of the various London bikeshare operators. I chose Urbo, who run dockless bikeshare in North-East London, and demonstrated using Shapely (in Python) to tidy the geofence polygons, before showing the result on the (OpenLayers-powered) Bike Share Map. The all-system London map is also available.

FOSS4G UK was a good meeting of the “geostack” community in London and the UK/Europe, it had a nice balance of career technologists, geospatial professionals, a few academics, geo startups and people who just like hacking with spatial data, and it was a shame that it was over so quickly. Thanks to the organising team for putting together a great two days.

Bikeshare in London – Borough Update

While TfL ponders London-wide regulation and freezes Santander Cycles, borough-by-borough rollout of dockless in London continues. I’ve rated each borough based on its provision – current, announced and rumoured (so likely some errors) of this all-important last-mile mobility addition to our streets:

Key

  • A – 2+ operators in/confirmed, at least 1 of which covers/allows >75% of borough and the other covers/allows >25% of borough
  • B – At least one operator, in/confirmed, covering significant part (>25%) of borough.
  • C – Operator(s) only covering small area, and/or only rumours of new operator.
  • D – No operators or positive news.
  • + – Improvement expected soon.
  • * – Simple projected % for 2018, based on Census data – 2011% and 2001-2011 rates of change

I intend to keep this table up to date as things change, and make corrections where I discover things are wrong above. In case it doesn’t view correctly for you, you can see the full table on Google Sheets here.

A summary of recent news:

  • Wandsworth has a forthcoming pilot, and has published the responses to a questionnaire given to six potential operators (although, some blank lines suggest that some interesting rows, e.g. cost to the councils, may have been censored to spare commercial blushes.)
  • Nearby Richmond is also planning on trialing dockless “by April”.
  • Redbridge is bringing Ofo and Urbo in on 2 March. This will give Urbo a contiguous area across three north-east London boroughs, however all three are outer boroughs with generally lower existing cycling journeys or infrastructure than the inner city.
  • Ofo was in to the eastern half of Camden last week, but then abruptly withdrew back to the Islington border.
  • Ofo are also strongly rumoured to be arriving in Hammersmith & Fulham, soon.
  • Southwark borough is also bringing in two bikeshare operators, per their recently published strategy document – Mobike are hinting that they are one and I understand that Ofo is likely to be the other. For both Ofo and Mobike this makes a lot of sense – Southwark is a big inner city borough only partially served by Santander Cycles, and importantly for these two operators, they already are in place in the City immediately to the north. The launch date is unconfirmed but believed to be in March.
  • If you have any corrections, updates or tips please let me know via email (o.obrien (at) ucl.ac.uk) or Twitter DMs (@oobr).

I’ve also updated my previous map to show the latest state of affairs:

Previous (May 2018):

See also my thoughts on the future of bikeshare in London in general.

Future Transport Report and Bikeshare in London

The GLA published an interesting report last week: Future transport – How is London responding to technological innovation. It focused mainly on drones, driverless cars and app-based services (as an example, CityMapper’s experimenting on turning its huge desire-line dataset, created from the data of its millions of users and their journeys, into a group-based taxishare service), but also included some details on bikeshare, and in particular the dockless “revolution” that is currently underway in the capital.

Not Looking Good for Santander Cycles

Some tidbits from the report include some depressing news from Transport for London regarding future of the Santander Cycles dock-based bikeshare system – “the scheme is also limited geographically to central London, with TfL having no plans to extend it.” Ouch.

Well, that’s a pity, but not entirely surprising. The reason are both political and financial:

  • It was a system launched and promoted by a previous mayor, of the “other party”, who was more focused on cycling and also less afraid to challenge the status quo. The existence of the system (despite its problems), and the associated disruptive – but brilliant – segregated cycle lanes – have been revolutionary. But they will be ever associated with a previous mayor.
  • Also despite the system being almost eight years old, TfL is still subsidising the system by over £3m a year. Too much unnecessary redistribution going on – using vans – presumably because they always have done it or other schemes have. Vans and drivers are expensive! And they clog up the streets. Incentivise users to do their own distribution – by using the bike itself – for rewards – much cheaper! The New York City system has “Bike Angels” (pic below) which actually has pretty decent rewards for riders who rebalance. There has also been an, I think, frankly misguided focus on keeping docking station density unnecessarily high throughout the area, and a focus on journey numbers over user utility, means the main beneficiaries are city workers getting from Waterloo to the Bank during weekdays and tourists circling Hyde Park at the weekends. Both groups have numerous other options, so the bikeshare here isn’t really solving any “last mile” mobility problems.

A flat rate of say 50p/20 min or £1/30 min, similar to the dockless systems (generally 50p/30 min), and moving the 20% least used docking stations so they were outside every tube/train station and inside every park and larger open spaces in Zone 2 to provide a ready source of obvious rides, would probably end up being more profitable, as there would be a massive surge in usage (right now it’s cheaper to get the bus!), particularly if it was integrated into Oyster card/contactless and was covered by daily/weekly capping. Why does it need to be profitably anyway? The buses (in London) aren’t, and the bikes encourage healthier outcomes, both to the user and surrounding pedestrians, than sitting on a bus emitting pollution.

London-Wide Regulation of Dockless

One important point from the report is suggesting that TfL should take the lead in regulating and so controlling the provision of dockless bikeshare systems on London streets, instead of the individual boroughs as present. This is an “obvious” thing to do – it is time consuming and inefficient for each operator to, assuming that they want a pan-London rollout, have 33 sets of negotiations with the 32 separate borough councils and the City of London, when much of London’s other transport operations are specified, operated or regulated by one entity – Transport for London. It also slows the rollout, makes the systems very expensive for operators to maintain if they are fragmented, particularly if each borough has slight differences in their specification (e.g. on maintenance response times). More importantly, it makes it less useful and more confusing for the user – both the start and the end of an intended journey need to be within an operating zone. Borough boundaries are somewhat arbitrary to most people living in London, and not easily noticed when crossing them. “London” is a more familiar area. Most commute journeys in London are radial – going towards or away from the centre (see pic below). The reluctance of many of the operators to even publish their operating zone except on their individual apps, also makes discovering the allowable area of a system – let alone using it – a challenge.

Curiously, a lot of the press leading from this report seemed to suggest that the recommendation of TfL taking on regulation of dockless bikeshare in London was in some way a clampdown on their operations, or a tightening of the environment, and would be therefore bad for the operators and presumably therefore good for those wanting to see these bikes off the streets. Rather an odd, negative spin. I don’t think it’s not that at all. Having TfL specifying and setting operational parameters for dockless systems, rather than 33 geographically limited entities doing so, can only be good for everyone concerned.

Smarter City? It’s All About the Data

My own take is that TfL regulation would be welcomed, but it should be linked with an incorporation of data into TfL’s own open data platform. A smart city is an open city, one where availability data on useable assets in the public realm is accessible to all, not to those tied to a particular platform (even if the transaction part needs operator control) or hidden to hide poor public value for money or failing/missing assets, as is happening for a system in a certain city up north. There is a need to democratise the information, in order to digitise the city. To this mind, dockless operators should be mandated to open up data on the location of bikes available to hire, hubs (preferred parking areas) and operating zones. This kind of information has generally always been made available by dock-based systems, dockless systems have so far not been compelled to release this as open data yet – at least in the UK.

To this mind, I am contemplating launching an operator data openness scoreboard for the UK – similar to one that was unveiled at a recent bikeshare hackathon event in Washington DC. I’ve also put together an interactive map showing the current coverage in London of the four active systems (Urbo, Ofo, Mobike and the dock-based Santander Cycles), plus known “hubs” where you are most likely to find bikes available for hire – N.B. not all the operators are using a hub-based approach across their whole areas. Because of the lack of open API data from most of the operators, much of this map is manually put together and so may be a little out of date – I will do my best to keep it current. The crucial data missing here of course is the current locations of bikes available to hire. For now, you’ll need to use each individual app.

How Dockless Might Work in London

Personally, I think that hub locations are one of the most important features of the new dockless systems. They can act as a point that is regularly stocked with bikes, meaning that regular users might not need to keep doing different journeys on foot to hunt down a bike. They also help the system keep order – they may help to minimise the amount of inconsiderate parking. Currently, the reward for finishing a journey at such a hub is minimal. Operators will have to be more ambitious if the hub-inducement model work. Something like halving the fee to finish a ride in a hub, and doubling it to finish a ride outside of an operating zone altogether. It should also be possible to have hubs outside of operating areas, for example, on specific pieces of private land where access has been granted, but the local authority has not granted permission.

More generally, I think, for large western cities like London, where pedestrian space will always be at a premium, the dockless model, as it stands, will not work. Either there won’t be enough bikes, so that no one uses the system, or the labour costs will be too high for the operators, or there will be too many bikes causing clutter. The system for London that I envisage, will work, is that Zone 1 gets around 1000 “mandatory hubs”, boxes painted on the ground, each around the size of a car parking space but mainly on the pavement, where any cycles from any of the dockless cycles may be left. TfL or boroughs get tough with the operators and users and impound bikes left outside of these zones by users, with operators getting fined for each impound and the operators passing on these fines (should they choose) to the delinquent user. Outside of Zone 1, the system could work as present, with recommended hubs. However boroughs/TfL need to stop operators creating hubs branded with only their own logos and signs, and instead create generic hubs. They can still just be painted lines on the ground – the bikes themselves should be all the branding needed.

See also my borough by borough scores, or the interactive map.

Dockless Continues to Expand – Here’s the Data

It looks like Brent will be the next London borough to get dockless bikeshare, with Mobike likely to be setting up in the borough within the next few weeks. The operator is not confirmed, although strongly hinted at, including by the council themselves. The choice of Mobike makes sense, as the operator already has a fleet of bikes in the two boroughs to its south, allowing journeys across quite a wide area of west London – although if you want to journey to the centre of the capital on a Mobike, you will still need to finish your journey in Islington or the City, or return the bike back to the three West London boroughs.

Above is my latest map of London boroughs, showing bikeshare provision. The red circles show Santander Cycles docking stations, while the other colours show the spread of dockless bikeshare – yellow (Ofo), orange (Mobike) and green (Urbo). Darker grey shades show areas of the capital with a high proportion of the resident population cycling to work (2011 census).

As well as these maps, I have now published interactive maps of the Ofo, Mobike and Urbo current operating areas in London (for all three) and other UK cities (for Ofo). Ofo were kind enough to supply me with their boundary files, for the others, which define their areas by borough boundaries in London, I have simply dissolved these boundaries. Simple spatial analysis shows the current area, in square kilometres, of these systems. See the “Numbers” tab for the latest numbers.

Currently, there are:

There are also individual dock maps for other operators in London (Santander Cycles, 110km2) and Oxford (Oxonbike, 8km2).

Operators do not supply consistent numbers of bikes operating in their areas, so I am not currently listing those. Various press quotes suggest 230 Urbos, 1000 Mobikes and 400 Ofos in London, 150 Ofos in Cambridge, 100 Ofos in Oxford and 200 Ofos in Norwich. This numbers likely include significant numbers that are not currently on the streets and available for use.

In parts of Islington, there are noticeably fewer Ofo bikes than a few weeks back, where they were previously very visible on on the main roads, while Mobikes are very visible but lined up in neat, operator-placed rows, rather than in the more organic placement you would expect from bona-fide users. There have been no London usage numbers released by the operators so far, despite having been operating for nearly half a year, which suggests they are perhaps not great. To be fair – it is wintertime and it is still early days.

Railway Station Numbers

You can click on all the images in this blogpost to go explore each view further on the interactive map.

The ORR publishes station entry/exit numbers on an annual basis, on a “best guess” basis, using ticket sales, gate information and modelling. The data is split by ticket type – full fare, reduced fare (off-peak tickets, tickets bought with railcards, advance tickets, child tickets etc) and season tickets. They make this data available as an Excel spreadsheet, so I’ve crunched it and have produced a couple of maps based on this data. I have also consolidated the total counts and ticket type counts data on CDRC Data.

The first shows the total numbers of entries/exits across the last year that the data is available for (2016-7), with a blended colour, with different red/green/blue strengths proportional to the % numbers for season tickets (red), full fare (blue) and reduced fare (green) entering/exiting National Rail services at that station. The area of the circle is proportional to the total numbers, combined across the ticket types. I’m using a minimum circle size, as otherwise some stations would be practically invisible on the map, as they can see days go by without any passengers – or trains.

Some interesting patterns – blues for many of the airport stations, where off peak tickets generally aren’t available, and most people don’t think to get advance tickets, such as to/from Stansted:

…and almost no-one pays full fare for some of the remotest stations:

Purples on the Welsh valleys lines, showing mainly commuters and peak time users:

Bright greens for stations serving major destinations where advance tickets are readily available, such Newcastle-upon-Tyne:

…popular tourist places, where many people will be visiting outside of the rush hours and at weekends, such as Oxford and Bicester Village retail outlet:

…and areas well covered by discounted travelcards, like Liverpool’s Merseyrail:

Reds where the season ticket holders dominate, such as Chelmsford and Colchester to the north-east of London:

Browns showing an “urban mix” of season ticket commuters and travelcard local journey makers, like in Straford, London:

See this map on TubeCreature.

The second map looks at the change in numbers between 2015/6 and 2016/7 (a major methodological change means I cannot use data from earlier years, for a more complete time series). You can view the absolute numbers for both years, but what is of more interest is looking at the changes. The circle fill colour is the % change (with 100% green for a doubling of numbers and 100% for a halving of numbers). The area of the circle represents the absolute change in numbers. The border colour emphasises whether the change is an increase or decrease. Stations with little change will show up as small circles. The biggest trends are the new lines to Oxford via Bicester, and from Edinburgh to Tweedbank. In both cases, the lines were only open for part of the first year, so an increase would be expected even if the day-by-day numbers were flat:

Big drops show in parts of London – the Goblin line having been closed for much of 2016/7 due to a bungled overhead line installation:

There is also a big drop at Kensington Olympia’s however the source reports says this is due to a methodological change – i.e. it may not have actually been a significant drop at all. This is somewhat puzzling, as there are ticket gates at this station, so in/out numbers should be pretty solid, but it may be relating to due to many fewer people, than previously thought, transferring in-barrier to the sparse District line services at this station. When they do this, they are no longer considered to be National Rail passengers and so have “exited” the station here, from a National Rail perspective.

Most parts of the country see a steady increase (light greens):

The big exception being area served by Southern trains – with them being on strike for much of the second year, the fall in numbers in this region is almost universal:

See this map on TubeCreature. You can also download all the total counts and ticket type counts data from CDRC Data.