Bicycle Shared System (SBCs) imbalance problem: how to solve it?

Bike share systems (SBCs) are becoming more and more popular around the world. From 2007 to 2017, the number of SBC services jumped from 48 to 1.766, and in 2017 alone, 444 new services appeared, i.e., there was a 33% growth in new services, as shown in the figure below [1].

Figures 1 : Growth of the shared bicycle system in the world [1]

A survey carried out by Citibike, a company that provides the service in New York, showed that although 91% [2] of respondents were satisfied and wanted the service to be expanded, 64% [2] pointed out as a major problem the fact that stations are completely full or empty, as they are unable to return or pick up bicycles immediately.

This dissatisfaction is related to the imbalance of bicycles between stations. Imbalance is one of the biggest logistical and financial challenges of this system, representing between 30 and 50% of operating costs [3].

One of the causes of this imbalance may be the terrain of the location, as users generally use bicycles on routes leaving from a higher point to a lower point in the city, and the reverse route will probably be done by bus, taxi or other Mode of transport that requires less effort. Another cause is the routine of the users, usually during the week, people have the commute home – work in the morning and the commute work – home at the end of the day, so, in the morning, it is more likely that the stations residential areas are empty and commercial areas are full, and the opposite happens at night [4].

The unbalance problem is not easy to solve, as the solution is not only to predict the number of bicycles needed per station, but also to minimize the costs and time taken to balance the system. Due to this complexity, companies have used heuristics, algorithms and even “machine learning” to improve demand forecasting [6], that is, the number of bicycles needed, the stations where the bicycles should be located, the period of day on which the relocation must take place, number of vans needed to carry out the replacement, among other parameters.

After performing the mathematical part, the bicycles can be redistributed directly as in trucks, vans or bicycles. And it is the bicycles themselves that have gained prominence in the direct relocation model of the SBCs [5], firstly because they are much cheaper than trucks and vans and secondly because they are able to move more quickly to the stations that need bicycles, especially in large cities with high traffic rates.

In parallel with the direct replacement system, with the aim of further mitigating the problem, companies have used indirect methods such as bonus systems for people who travel against the flow of commuting or free rental minutes for those who return bicycles in places of highest altitude [4].

Bike-sharing companies have gone to great lengths to solve this complex problem. With the increase in the number of people interested in the subject and the advancement of forecast models, it is likely that user satisfaction will increase and further leverage the use of a more ecological means of transport and consequently improve traffic in large cities.

Sources:

[1] https://www.mobilize.org.br/midias/pesquisas/situacao-global-do-transporte-e-mudanca-climatica.pdf

[2] https://www.wnyc.org/story/survey-64-citibike-users-unhappy-about-full-and-empty-docks/

[3] http://itdpbrasil.org/wp-content/uploads/2019/05/2-BSPG_Portugu%C3%AAs-1.pdf

[4] https://www.teses.usp.br/teses/disponiveis/3/3135/tde-22052018-092405/publico/RodolfoCelestinodosSantosSilvaCorr18.pdf

[5] https://bikeportland.org/2016/09/07/portland-now-using-pedal-powered-trikes-to-help-rebalance-bike-share-stations-191007

[6] https://www.citylab.com/transportation/2014/08/balancing-bike-share-stations-has-become-a-serious-scientific-endeavor/379188/