Bike sharing is in vogue. It is a deceptively simple service. Walk up to a stand, pick up a bike and return it to the same or different stand. Charges can be based on an annual membership or per trip charges (the latter can appeal to casual users and tourists). Fast Company has even picked bike sharing as one of the most promising low-carbon transportation technologies (The Top Five Transit Technologies For The Low-Carbon Economy, Sep 20):
For those who live too far (e.g. in distant suburbs) from their work, having access to a bike for inter-city trips may be attractive. Thus, the rapid growth of bike-sharing systems around the globe, with more than 400 systems now in operation in Europe alone. Bike-sharing has moved across the Atlantic, with cities such as New York, Vancouver, and Boston launching programs. PBSC Urban Solutions has been cashing in on this growing trend in bike-sharing systems. Their flexible, modular BIXI bike-sharing system–complete with software and solar-powered communication systems–is now in use in 10 cities around the globe, which support nearly 15,000 bikes and 300,000 users.
New York City recently announced that it was going to launch a bike sharing program in conjunction with Alta Bicycle Share of Portland. As with most things in New York, this program will be big — 10,000 bikes and 600 stations.
But what goes into the logistics and operations of a bike sharing network? The system is basically self-service so it is not immediately obvious to users just what it takes to make it all work.
A recent Marketplace report gives some hints at what is happening behind the scenes (Bike-share system to roll out on New York City streets, Sep 14).
In Washington D.C., commuter David Knight says he uses that city’s year-old bike share all the time to get to work, and for other trips.
KNIGHT: I was across town and it wasn’t raining, so I picked it up last minute.
That flexibility has made bike-share unexpectedly popular in D.C. More than 70,000 tourists and other casual users signed up the first year — dwarfing the 15,000 yearly members. That kind of impulse purchase has some unexpected consequences.
Eric Gilliland of Alta Bike Share, which operates D.C.’s system, says its busiest hour ever was after last month’s rare earthquake.
ERIC GILLILAND: It was just impossible to get around any other way.
Cars and buses were at a standstill, metro trains were full.
GILLILAND: In the span of an hour, we pretty much emptied out everything in the downtown area, which all ended up flooding out to the neighborhoods.
That can be a problem even on a regular day. In D.C. in July, bike stations were empty 9,000 times. And there were 5,000 incidences where stations were completely full, so cyclists couldn’t return their bikes. As a result, the bikes have to be loaded on trucks, and moved around.
Almost every city that’s tried bike share has experienced some version of that phenomenon. In London, suburbanites take the train part way and then bike for the rest of their trips. All the bikes end up in the center of town. In Barcelona cyclists go downhill, but not up.
So these systems are subject to asymmetric flows. To put these numbers in perspective, Capital Bikeshare‘s web site, there are 110 bike sharing stations in DC. That means there are 3,410 “station days” in July so the average station is unable to handle some transaction (pick up or drop off) several time per day. And that does not even consider extreme events like earthquakes that really throw the system out of whack.
How does a bike sharing service deal with these imbalances? Not too surprisingly with guys in trucks. Here a description of what happens at BIXI in Toronto (Inside the BIXI bike warehouse in Toronto, BlogTO, Aug 26, 2011)
Darren takes me into the warehouse, a large room crammed bikes in need of repair. “We have people running around the city in a truck with a trailer, and they bring the bikes here,” he says. “We call them ‘bike warriors.'”
He explains that bike warriors go to 80 BIXI docking stations in a day, balancing out stations by moving bikes between them. Inspections are done three times a week on Mondays, Wednesdays, and Fridays, with roughly 45 bikes brought back to the warehouse each week for repairs. Two full-time mechanics repair between six to eight bikes a day. Extra mechanics-in-training and assistants are also on hand to get through the daily flow of bike repairs.
At the stations, Darren says, basic checks are done starting with the brakes, the front dynamo which acts as the generator for the lights, and the bell. If there is more complicated work to be done, such as graffiti removal or badly damaged parts, the bikes are brought back to the LEF.
BIXI has 1,000 bikes in Toronto so this says that a little less than 5% of its fleet are out commission for part of each week.
So what will this look like in New York? Alta Bicycle Share says they will need 200 people in New York to support their operations (see here). They will likely face similar challenges as there are in DC. For example, they should see a healthy tourist market as visitors look for a quick, convenient way to cut across Central Park. But that could lead to an imbalance of flows as, for example, bikes collect at the Museum of Natural History in the morning but users opt for a cab back to their hotel at the end of their visit.
The question is whether there is a creative way to address some of the inherent complexities in managing a bike sharing system. The obvious answer is pricing. They could simply charge more for returning a bike to a station in a different zone. I suspect that firms don’t want to do this simply because it complicates the offer and may deter usage. A more palatable alternative may be offering a discount for a second rental in the same day. That might give some users pause before catching a cab on their return trip.
Many bikeshare systems have public data feeds from which interested parties can check the status of individual stations or an entire system. Over time this can provide a metric for the effectiveness of bike redistribution in matching supply against demand – see http://www.codeline-telemetry.com/maps/depletion.htm for daily stats and maps covering Capital Bikeshare and other mid-size (50-100-500 station) systems.
One problem with price incentives is that they introduce complexity into a system in which simplicity is an attractive feature.
For systems whose daily operating patterns include circadian peaks of use (riding) and non-use (parking), one approach might be for the operator to collect excess bikes from stations that fill up and simply hold them in a local cache until the next phase of the demand cycle, freeing up enough station docks to allow circulation. There’s nothing new about this: bus and subway systems routinely store equipment in layover facilities during overnight and mid-day off-peak periods.
[…] The logistics of bike sharing « The Operations Room Share this:ShareEmailFacebookLinkedInTwitterPrintLike this:LikeBe the first to like this post. This entry was posted in Production Operations, Service Management, Supply Chains, Sustainability. Bookmark the permalink. ← If Amazon took cash! […]
Thanks, great post. I really interested with your topic
[…] to Lincoln Park for the weekend. This basically puts Avis in the position of a bike sharing service having to regularly re-position its inventory. But repositioning cars is much harder. A reasonably sized truck car carry 20+ bikes and bike […]
[…] we have written about before, the need to rebalance capacity is a common problem in bike sharing systems. It is not unlike a […]