Author’s Accepted Manuscript
Optimizing the level of service quality of a bikesharing system
Ramon Alvarez-Valdes, Jose M. Belenguer,
Enrique Benavent, Jose D. Bermudez, Facundo
Muñoz, Enriqueta Vercher, Francisco Verdejo
To appear in: Omega
Received date: 15 October 2014
Revised date: 22 September 2015
Accepted date: 22 September 2015
Cite this article as: Ramon Alvarez-Valdes, Jose M. Belenguer, Enrique
Benavent, Jose D. Bermudez, Facundo Muñoz, Enriqueta Vercher and Francisco
Verdejo, Optimizing the level of service quality of a bike-sharing system,
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form.
Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. www.elsevier.com
Optimizing the level of service quality of a bike-sharing system
Ramon Alvarez-Valdesa,∗, Jose M. Belenguera, Enrique Benaventa, Jose D.
Bermudeza, Facundo Mun˜oza, Enriqueta Verchera, Francisco Verdejoa aDepartment of Stastistics and Operations Research, University of Valencia, Doctor
Moliner 50, 46100 Burjassot, Spain. Telephone: +34 96354308
Public bike-sharing programs have been deployed in hundreds of cities worldwide, improving mobility in a socially equitable and environmentally sustainable way. However, the quality of the service is drastically affected by imbalances in the distribution of bicycles among stations. We address this problem in two stages. First, we estimate the unsatisfied demand (lack of free lockers or lack of bicycles) at each station for a given time period in the future and for each possible number of bicycles at the beginning of the period. In a second stage, we use these estimates to guide our redistribution algorithms. Computational results using real data from the bike-sharing system in Palma de Mallorca (Spain) are reported.
Keywords: Bike-sharing systems, Demand forecasting, Routing, Heuristics 2010 MSC: 90B06, 90B90 1. Introduction
A public bike-sharing system consists of a set of stations scattered over the city and a set of bicycles available to the system users. A user can take a bicycle ∗Corresponding author
Email addresses: firstname.lastname@example.org (Ramon Alvarez-Valdes ), email@example.com (Jose M. Belenguer), firstname.lastname@example.org (Enrique Benavent), email@example.com (Jose D. Bermudez), firstname.lastname@example.org (Facundo Mun˜oz), email@example.com (Enriqueta Vercher), firstname.lastname@example.org (Francisco
Preprint submitted to OMEGA September 25, 2015 at a station, use it for a short journey, and leave it at the same or any other station. Since the first system was established in Amsterdam in 1965, there has5 been a rapidly increasing number of cities providing their citizens with this type of service, which has many advantages of various kinds: it is an environmentally sustainable and socially equitable mode of transportation, it can be used as part of an intermodal public transport system, it reduces motorized traffic and therefore emissions of contaminants, and it promotes a healthier way of life.10
According to the consultancy company MetroBike LCC (), in July 2014 721 cities had a public bike-sharing system, with a total of approximately 814000 bicycles, and 228 were planned or under construction. These systems range from less than one hundred bicycles in small towns to many thousands in cities like Paris (20600), Hangzhou (78000), or Wuhan (90000).15
The most important factor for the success of a public bike-sharing system is its ability to satisfy the varying demands of the users. Underlying the random variations of everyday demands, there are patterns of demand that have to be identified and estimated and the system has to be planned and managed to maximize the level of customer satisfaction. Situations in which the user20 arrives at a station to take a bicycle and finds the station empty, and those in which he/she arrives at the station to leave the bicycle and the station is completely full, have to be avoided as far as possible. For the bike-sharing system to become a sensible alternative to other modes of transportation, it has to be reliable. The everyday users have to be confident that they will find25 bicycles to start their trips and available lockers to leave them when the trips are finished wherever and whenever they need them. This can be achieved in the three phases of design and operation. First, at a strategic level, the number of stations and their location and size have to be decided. Second, at a tactical level, the number of bicycles in the system has to be determined. Third, at30 an operational level, a bike-repositioning system has to be adopted for moving bicycles from stations with an excess to stations with a shortage in order to satisfy the demands forecast for the next periods.
Repositioning is done by means of light trucks based at one or several depots, 2 that pick up bicycles from stations at which there are too many and move them35 to stations where there are too few. Sometimes there are bicycles at the depots, for instance those that were damaged and have been repaired, and these can also be used when constructing the repositioning routes.
There are two types of repositioning systems. In the static case, the system is considered closed, so the users do not interact with it, its initial state is40 considered known and fixed, and the aim of the repositioning is to get the system to a desired, predefined state. In the dynamic case, the repositioning system operates while the bike-sharing system is being used. Therefore, users are continuously taking and leaving bicycles at the stations, modifying their states.