Researchers analyze the usage patterns of bicycles in four large cities in USA to make bicycle sharing systems more efficient —


Bicycle sharing techniques (BSSs) are a well-liked transport system in most of the world’s huge cities. Not solely do BSSs present a handy and eco-friendly mode of journey, in addition they assist cut back visitors congestion. Furthermore, bicycles might be rented at one port and returned at a distinct port. Regardless of these benefits, nonetheless, BSSs can not rely solely on its customers to take care of the provision of bicycles in any respect ports always. It is because many bicycle journeys solely go in a single course, inflicting extra bicycles at some ports and a scarcity of bicycles in others.

This downside is mostly solved by rebalancing, which entails strategically dispatching particular vehicles to relocate extra bicycles to different ports, the place they’re wanted. Environment friendly rebalancing, nonetheless, is an optimization downside of its personal, and Professor Tohru Ikeguchi and his colleagues from Tokyo College of Science, Japan, have devoted a lot work to discovering optimum rebalancing methods. In a research from 2021, they proposed a technique for optimally rebalancing excursions in a comparatively quick time. Nevertheless, the researchers solely checked the efficiency of their algorithm utilizing randomly generated ports as benchmarks, which can not replicate the situations of BSS ports in the actual world.

Addressing this concern, Prof. Ikeguchi has just lately led one other research, along with PhD pupil Ms. Honami Tsushima, to search out extra reasonable benchmarks. Of their paper printed in Nonlinear Principle and Its Functions, IEICE, the researchers sought to create these benchmarks by statistically analyzing the precise utilization historical past of rented and returned bicycles in actual BSSs. “Bike sharing techniques might turn into the popular public transport system globally sooner or later. It’s, due to this fact, an necessary concern to handle in our societies,” Prof. Ikeguchi explains.

The researchers used publicly accessible information from 4 actual BSSs positioned in 4 main cities in USA: Boston, Washington DC, New York Metropolis, and Chicago. Save for Boston, these cities have over 560 ports every, for a complete variety of bicycles within the 1000’s.

First, a preliminary evaluation revealed that an extra and lack of bicycles occurred throughout all 4 BSSs throughout all months of the yr, verifying the necessity for energetic rebalancing. Subsequent, the group sought to grasp the temporal patterns of rented and returned bicycles, for which they handled the logged hire and return occasions as “level processes.”

The researchers independently analyzed each level processes utilizing three approaches, particularly raster plots, coefficient of variation, and native variation. Raster plots helped them discover periodic utilization patterns, whereas coefficient of variation and native variation allowed them to measure the worldwide and native variabilities, respectively, of the random intervals between consecutive bicycle hire or return occasions.

The analyses of raster plots yielded helpful insights about how the 4 BSSs had been used of their respective cities. Most bicycles had been used throughout daytime and fewer in a single day, producing a periodic sample. Curiously, from the analyses of the native variation, the group discovered that utilization patterns had been comparable between weekdays and weekends, contradicting the outcomes of earlier research. Lastly, the outcomes indicated that the statistical traits of the temporal patterns of rented and returned bikes adopted a Poisson course of — a extensively studied random distribution — solely in New York Metropolis. This was an necessary discover, given the unique goal of the analysis group. “We are able to now create reasonable benchmark situations whose temporal patterns of rented and returned bicycles comply with the Poisson course of. This, in flip, can assist enhance the bicycle rebalancing mannequin we proposed in our earlier work,” explains Prof. Ikeguchi.

General, this research paves the way in which to a deeper understanding of how folks use BSSs. Furthermore, by additional detailed analyses, it ought to be doable to realize perception into extra complicated features of human life, as Prof. Ikeguchi remarks: “We imagine that the evaluation of BSS information will lead not solely to environment friendly bike sharing but in addition to a greater understanding of the social dynamics of human motion.”

In any case, making BSSs a extra environment friendly and engaging possibility will, hopefully, make a bigger proportion of individuals select the bicycle as their most popular technique of transportation.