Operate the self-service bike | Mirage News

Everywhere, from Berlin to Beijing, there are brightly colored bicycles that you can borrow to get around the city without a car. These systems, along with electric scooters, provide people with a quick and convenient way to get around urban areas. And at a time when cities are scrambling to find ways to meet their climate goals, they are a welcome tool for city planners.

People want the bikes there when they want to use them, and they’ll only want to use the system if it’s good service.

Ensuring e-bikes and scooters are within easy reach can be a challenge – but it’s also key to supply success, says Steffen Bakker, a researcher in NTNU’s Department of Industrial Economics and Technology Management. which studies ways to make transport greener and more efficient.

“If a system like this is going to be successful, we need to have user satisfaction,” Bakker said. “People want the bikes to be there when they want to use them, and they’ll only want to use the system if it’s good service.”

Bakker was co-author of a recent paper that outlines an optimization model to help cities and businesses do a better job of keeping their bike-share customers happy.

Like shooting at a moving target

Consider the challenges of providing bikes or scooters where and when people want them.

You don’t know when customers will pick up the bikes and where they will put them.

The researchers describe the problem as dynamic, because it is constantly changing, and stochastic, because it changes randomly and is often difficult to predict, Bakker said.

“Users of the bike-sharing system pick up the bikes in one place and move them elsewhere. And then the state of the system changes because all of a sudden the bikes aren’t where they started, which is the dynamic part,” he said. “But on top of that, you don’t know when customers will pick up the bikes and where they will put them. This is the stochastic part. So if you want to plan at the start of the day, you don’t know what’s going to happen. »

Bakker and his colleagues can use the huge trove of data collected by electric bikes and scooters when used to make predictions. But there is no guarantee that the way the bikes were used last Tuesday, for example, will be the same the following Tuesday, he said.

“You have to adapt to things that happen during the day,” he said. “Maybe all of a sudden something happens or the weather changes, and then people don’t use the service and the demand pattern changes, which impacts planning.”

Assemble the pieces

What Bakker and his colleagues developed is an optimization model that can give recommendations on what service operators should do.

This includes what service vehicles should do at the station they are currently at – whether to drop off or pick up bikes, or swap batteries for e-bikes and scooters – and where to go next. The underlying calculations are based on what has happened so far during the day and what is expected to happen in the near future.

It’s very complex, because it’s a big system.

The group’s research is funded under a NOK 10 million project funded by the Research Council of Norway called Future of Micro Mobility (FOMO), with the company Urban Sharing AS as the lead researcher. grant.

“Through Pilot-T, we plan to use existing city bike systems as test bases, and by developing new decision support tools, the goal is to increase the efficiency of 30% rebalancing and 20% bike life,” said Jasmina Vele, project manager at Urban Sharing.“This can be achieved through better decisions related to rebalancing and preventive maintenance, and it will match to a significant reduction in the costs of existing urban bicycle systems.”

Move bikes in the most efficient way

The process of collecting and moving bikes from one bike parking station to another is called “rebalancing”. The use of the optimization model, still in the development phase, makes it possible to send drivers a new map each time they arrive at a bike station.

“You don’t just make one plan at the start of the day, but what we do is we create a new plan every time a vehicle comes to a bike station,” he said. declared. “And when the car arrives at the station, we’ll tell them, ‘Okay, take as many bikes or drop off as many bikes’.”

But this is where the tricky part comes in. It’s important not to be too myopic by focusing only on the current state of the system, Bakker says, especially if some stations are expected to have more demand in the next hour or so.

“It’s very complex, because it’s a big system,” he said. “Maybe there will be a lot of demand at the station in an hour. So you already want to bring bikes there. But at the same time, there may be stations now almost empty, and they need bikes. So you have to understand this trade-off.

It’s also important to coordinate pickups and drop-offs between the different vehicles that serve the bike-share network, he said.

Digital twins and computation time

Bakker and his colleagues are working with NTNU’s computer science department to create a “digital twin,” or computer simulation, of the systems they model, so they can try out different approaches without having to test them in the real world. .

Early testing has shown that the model generated by the group can reduce the number of problems (i.e. either not enough bikes where the user wants one, or too many bikes for the user not to can’t park the bike) by 41% compared to not doing any. rebalancing at all.

Compared to the current rebalancing practices of Oslo City Bikes, which is also an NFR grant collaborator, the number of issues has been reduced by 24%. Bakker says the new versions of the model show even more potential.

Simpler approaches are also possible

Unsurprisingly, the kinds of calculations needed to make the model work are complex, and researchers have to fine-tune the various parameters affecting model performance.

Bakker and his colleagues also worked on a component of the optimization model called criticality scores, which is a little simpler and can be used independently of the larger optimization model.

A criticality score is basically a score given to different self-service bike parking areas based on the number of bikes they contain or currently need. These scores are relatively simple to calculate and can be provided to drivers as they travel around town to rebalance the number of bikes at each station.

“It’s a score that tells the driver which station is the most critical to visit,” Bakker said. “If you can take that to the person driving the car and say those are the stations with the highest criticality score, we can deliver something that’s not the best, but it’s probably good, and much better than what bike sharing companies are doing now.”

According to Urban Sharing’s Vele, using these types of optimization models can help make bike sharing an important part of urban transportation.

“Urban Sharing’s vision for future mobility is a responsive and adaptive transportation system. Using data and machine learning/optimization algorithms, we can combine the best of traditional and modern transport systems and create a resource-efficient system that responds to demand and adapts to individual user needs.” , she said.

Reference:

Marte D. Gleditsch, Kristine Hagen, Henrik Andersson, Steffen J. Bakker, Kjetil Fagerholt. A column generation heuristic for the dynamic bicycle rebalancing problem. European Journal of Operations Research. 2022, https://doi.org/10.1016/j.ejor.2022.07.004.

/Public release. This material from the original organization/authors may be ad hoc in nature, edited for clarity, style and length. The views and opinions expressed are those of the author or authors. See in full here.

Maria D. Ervin