Sign In

Communications of the ACM

ACM TechNews

New Algorithm Finds Best Routes For One-Way Car Sharing

View as: Print Mobile App Share:
A vehicle in Paris' Autolib car-sharing service fleet.

A new algorithm holds the promise of rebalancing one-way vehicle-sharing systems.

Credit: Benoit Tessier/Reuters

Massachusetts Institute of Technology (MIT) researchers have developed a vehicle-routing algorithm designed to help rebalance one-way vehicle-sharing systems.

One-way systems enable drivers to pick up vehicles near their departure point and drop them off near their destination, but an imbalance arises, for example, when stations close to cities fill up with cars while stations near the suburbs are depleted. Some programs, such as Car2Go, attempt to resolve this by employing drivers to shift vehicles to high-demand locations, but this imposes an expense and also tends to become unbalanced.

MIT professor Emilio Frazzoli and his colleagues wrote an algorithm involving a driver who returns a car to a depleted station while taking a customer along like a taxi service. The algorithm identifies the most efficient means of balancing taxi and shuttle trips and reduces wasted trips.

The group found that at least one shuttling driver is necessary for every three fleet vehicles to maintain adequate vehicle availability. Autonomous vehicles would further improve car-sharing, Frazzoli notes. "It would save money and it would decrease the cost of using a vehicle, and allow you to reduce congestion," he says. "So this can have a tangible benefit to people, especially those living in large cities."

From MIT News
View Full Article


Abstracts Copyright © 2013 Information Inc., Bethesda, Maryland, USA


No entries found