Sunday, September 18, 2011


Today, the Urbanophile tweeted a link to a "caution" about bike share from Payton Chung. While I agree that caution is in order, I'm disturbed by the framing of Chung's piece, particularly the assumption that a significant amount of "rebalancing" is necessary in any bike share system. Rebalancing is where share bikes are transported, usually in a truck, from "full" stations to "underutilized" ones. Let me go through my ideas in response to that:

First, there's something profoundly depressing about a bike system supported by large gasoline-powered vehicles. I first got the same feeling when I read about the large long-distance bike rides - not tours - that are supported by "sag wagons." Essentially, you ride your bike, but you're paying for someone to drive ahead of you with your stuff. This happens in races, too. A single sag wagon can carry stuff for multiple cyclists, but it still undercuts the whole idea of bicycles as transportation.

Paris's Vélib´system, as the first large-scale bike sharing system, showed that there is an imbalance in the use of the bikes - or at least two, in fact. First, there's a daily pattern where people ride their bikes from home to work and back again. Most bike share stations are not equally balanced in terms of home and work destinations, so the ones that have more homes nearby tend to be emptier in the middle of the day and fuller at night, and vice versa for stations close to jobs. There were similar patterns in terms of travel to shopping, recreation and metro stations.

This one seems like a no-brainer. The system should simply be built with enough flexibility to absorb extra bikes depending on time of day. There should be enough bikes at the home stations in the morning, and at the work stations in the afternoon. This may seem like a waste, but it's preferable to driving the bikes back to the home site for another wave of commuters, and back in the afternoon.

Secondly, the Vélib' managers found that people were quite willing to ride their bikes downhill, and less so uphill. Brian McEntee tweeted that Paris and DC have experimented with building incentives into the pricing structure to encourage people to ride uphill. This seems like a promising response.

It seems to me that there are two primary goals for a bike share system. One is to encourage cycling at the expense of any other mode, thus turning everyone into part-time cyclists and building a constituency of cyclists who will then support other parts of the agenda. The other is to create a composite transportation system that is more flexible, convenient and efficient than private cars. If you really want to increase efficiency overall, you need to look at how much fuel consumption, pollution and carnage a "rebalancing" trip adds to each bike ride.

Does the rebalancing (and all the other fossil fuel use that goes into a bike share system) make it more efficient, cleaner or safer to drive a single-occupant vehicle? What about a bus? Bike share works best with trips that are too dispersed to warrant a transit route, but if you're paying someone to drive a diesel vehicle back and forth, you might as well skip the bike share and just run a bus.


Anonymous said...

I strongly agree that rebalancing is not good. That's why I argued that bikeshare works best in medium-density, mixed-use environments where the bikes naturally circulate themselves, and why I suggest that the NYC system could run into problems.

ant6n said...

Well re-balancing kind of sucks, for the reason you mentioned. On the other hand, one truck can carry a lot of bikes.

One has to look at the positive side of this kind of bike infrastructure, which gives a lot of flexibility, and allows people a very gradual introduction into biking - thus helping to seed bike ownership. In Montreal, there's been a huge increase of bicycling during the last two years, after they installed bixi.

And then when you accept the bike share positives, you have to see that a bike system which has full or empty stations sucks; you cannot rely on getting around and the uncertainty whether you'll get a bike or can drop it off at the end just kills the joy of it all.

Mulad said...

NYC will undoubtedly have significant issues with imbalance between stations, but it'll be interesting to see how much of an issue it really turns out to be.

Minneapolis's Nice Ride system is generally fairly well-balanced, though I honestly feel they have spread out much of their bike network a bit too much -- they probably aren't getting as many riders as they should, and the relatively low ridership may reduce the need for rebalancing. Anyway, they currently have somewhere around 100 stations, yet only about 10 of them typically need rebalancing through the day -- and most of those only need to be rebalanced once or twice daily.

They use a Ford Ranger pickup with a small trailer to do the redistribution. I'm not sure of the capacity of the trailer, but it's somewhere around 8 or 10 bikes. I have no idea what sort of vehicle is planned to be used in NYC, but that number of bikes is not terribly heavy, and Nice Ride's trailer could easily be pulled by almost any car in existence.

Telemaque said...

It seems to me in Boston that any time a station is full or empty, there's a nearby station at the requisite 50% fill rate. It's been suggested that people who take a bike from a full station to an empty one be given a credit. I think at least in Boston, such a policy would prompt teenagers to do this, and reduce the need for rebalancing by truck.

arcady said...

I think some amount of rebalancing is going to be necessary no matter what you do, but it should be seen as a short term solution used until the incentives can be set up so that bikes end up in the right place without having to be moved by truck. And sometimes there are just going to be short term demand spikes where these incentive-based solutions don't have time to work. To some extent, the operators are also going to have to learn to accept some imbalance and install more bike stations in areas where, for example, people go to work, since it's pointless to bring those bikes back out to residential areas to store them during the day and bring them back for the evening commute (or even the lunchtime rush... I can see that being a popular time for bike share use). On the whole, it's an interesting optimization problem with plenty of open questions.

Jonathan said...

Yesterday's water-main break provides a useful reminder that spontaneous events will probably have more of an effect on the overall levels of the system than habitual circulation.

I'd suppose that all of a sudden, straphangers who couldn't get uptown past 59th St would leave the subway, hop on a share bike, and proceed uptown to their destination. Those bikes left at kiosks near closed subway stations from 103rd St to 135th St would be picked up and ridden, many of them back down to 59th St.

But there would definitely be a crest in usage on the Upper West Side, probably getting a little larger in area as people heard about the water-main break and hopped on a bicycle before getting to the closed subway station.

I think that these kinds of spontaneous events would make more of a difference in bike availability than habitual commuting patterns. The large size of the NYC bike share makes it a more enticing option for going to concerts in the park or getting around subway delays.

Anonymous said...

Re: your comment about running a bus instead - why not add bike carrying capacity to buses and use those for rebalancing? Well, it would never happen, since the MTA and the bike share are separate entities, but in an ideal world (or even just one where transportation is not splintered and balkanized as it is here in NYC) you could have bike-riding rebalancers who go to fuller stations, load bikes onto passing buses, which are unloaded by other rebalancers from those buses at emptier ones.