Perceptually important points of mobility patterns to characterise bike sharing systems: The Dublin case

Since the first Bike Sharing System (BSS) was introduced in Amsterdam (1965), studies about BSSs have constantly increased. BSSs studies are typically focused on user's socio-economic characteristics, bike sharing patterns and purpose of use in the city. This paper increases the knowledge of bike station classification due to users' mobility patterns based on data mining tools. For this purpose stations will be identified by a code based on joining three ratios: the load factor or number of available bikes ratio, the cumulative trips ratio, and the turnover station ratio. The latter is the new ratio proposed in this paper, which measures the effectiveness degree of each station. The higher the rate, the more effective the station is. Data mining tools to work with these three ratios are used in the proposed algorithm. Specifically, the perceptually important points (PIP) process to represent and index each time series of each station, and a rule set to classify the stations, are used. The results could support planning and operations decisions for re-design and management of BSSs in relation to the spatial implications of the stations and the users' mobility patterns, due to the classification reveals imbalances in the distribution of bikes and lead to a better understanding of the system structure. The proposed method is applied to the Dublin Bikes Scheme with good performance results.

[1]  Christine Fricker,et al.  Classification of the vélib stations using Kmeans, Dynamic Time Wraping and DBA averaging method , 2014, 2014 International Workshop on Computational Intelligence for Multimedia Understanding (IWCIM).

[2]  T. Warren Liao,et al.  Clustering of time series data - a survey , 2005, Pattern Recognit..

[3]  Tak-chung Fu,et al.  Flexible time series pattern matching based on perceptually important points , 2001 .

[4]  Tak-Chung Fu,et al.  Representing financial time series based on data point importance , 2008, Eng. Appl. Artif. Intell..

[5]  Licia Capra,et al.  Measuring the impact of opening the London shared bicycle scheme to casual users , 2012 .

[6]  Céline Robardet,et al.  Shared Bicycles in a City: a Signal Processing and Data Analysis Perspective , 2011, Adv. Complex Syst..

[7]  Brian Caulfield,et al.  Examining user behaviour on a shared bike scheme: the case of Dublin Bikes , 2012 .

[8]  Tak-Chung Fu,et al.  A review on time series data mining , 2011, Eng. Appl. Artif. Intell..

[9]  J. Gutiérrez,et al.  Optimizing the location of stations in bike-sharing programs: A GIS approach , 2012 .

[10]  W. Deng,et al.  Ridership and effectiveness of bikesharing: The effects of urban features and system characteristics on daily use and turnover rate of public bikes in China , 2014 .

[11]  Rafael E. Banchs,et al.  Article in Press Pervasive and Mobile Computing ( ) – Pervasive and Mobile Computing Urban Cycles and Mobility Patterns: Exploring and Predicting Trends in a Bicycle-based Public Transport System , 2022 .

[12]  J. MacQueen.  Some methods for classification and analysis of multivariate observations , 1967 .

[13]  Dirk C. Mattfeld,et al.  Understanding Bike-Sharing Systems using Data Mining: Exploring Activity Patterns , 2011 .

[14]  Nuria Oliver,et al.  Sensing and predicting the pulse of the city through shared bicycling , 2009, IJCAI 2009.

[15]  Michael Batty,et al.  Mining bicycle sharing data for generating insights into sustainable transport systems , 2014 .

[16]  E. Murphy,et al.  The Role of Bicycle-sharing in the City: Analysis of the Irish Experience , 2015 .