Start Date
11-8-2016
Description
We investigate customer mobility behavior by examining free-floating carsharing demand dynamics. For this purpose, we analyze rental data of a major carsharing provider in the city of Amsterdam in combination with points of interest (POIs). Connecting POI data to carsharing trips and stratifying the data along 6-hour intervals allows us to illustrate the spatio-temporal dimensions of carsharing usage, i.e. how carsharing demand changes over time and how it shifts spatially within the provider’s business area. We cluster the point data using kernel density estimation and apply a generalized linear model with Gamma distributed values on the sampled data. Our results indicate that, depending on the hour of the day, different POI categories have different, yet significant, impact on trip destinations. Our insights advance the understanding of when and for what purpose customers use carsharing, enabling providers to predict demand in existing and new business areas.
Recommended Citation
Klemmer, Konstantin; Willing, Christoph; Wagner, Sebastian; and Brandt, Tobias, "Explaining Spatio-Temporal Dynamics in Carsharing: A Case Study of Amsterdam" (2016). AMCIS 2016 Proceedings. 35.
https://aisel.aisnet.org/amcis2016/ITProj/Presentations/35
Explaining Spatio-Temporal Dynamics in Carsharing: A Case Study of Amsterdam
We investigate customer mobility behavior by examining free-floating carsharing demand dynamics. For this purpose, we analyze rental data of a major carsharing provider in the city of Amsterdam in combination with points of interest (POIs). Connecting POI data to carsharing trips and stratifying the data along 6-hour intervals allows us to illustrate the spatio-temporal dimensions of carsharing usage, i.e. how carsharing demand changes over time and how it shifts spatially within the provider’s business area. We cluster the point data using kernel density estimation and apply a generalized linear model with Gamma distributed values on the sampled data. Our results indicate that, depending on the hour of the day, different POI categories have different, yet significant, impact on trip destinations. Our insights advance the understanding of when and for what purpose customers use carsharing, enabling providers to predict demand in existing and new business areas.