Paper Number
1909
Paper Type
Complete Research Paper
Abstract
Free-floating carsharing services provide customers with a fleet of vehicles distributed within an operation area. These services have gained popularity because of their positive impact on societal and personal mobility. Accurate modeling of customer demand is a key challenge for effective service operation but mainly relies on historically observed bookings. They represent a biased version of demand since censoring issues occur when demand surpasses available supply. In this study, we develop a demand-unconstraining method based on real-world booking and vehicle search request data from a free-floating e-carsharing operator. Building on our unconstrained model, we demonstrate the impacts on service performance in a discrete event-driven simulation of a machine-learning-based vehicle relocation algorithm. Our work contributes by presenting insights into real-world censoring behavior of demand. By applying our approach to service operation, the performance can be increased by up to 6% and relocation efforts can be decreased by up to 22%.
Recommended Citation
Prinz, Christoph; Willnat, Mathias; Rampold, Florian; and Kolbe, Lutz M., "Unconstraining Demand for Effective System Operation: The Case of Vehicle Relocation in Free-Floating E-Carsharing" (2024). ECIS 2024 Proceedings. 14.
https://aisel.aisnet.org/ecis2024/track07_busanalytics/track07_busanalytics/14
Unconstraining Demand for Effective System Operation: The Case of Vehicle Relocation in Free-Floating E-Carsharing
Free-floating carsharing services provide customers with a fleet of vehicles distributed within an operation area. These services have gained popularity because of their positive impact on societal and personal mobility. Accurate modeling of customer demand is a key challenge for effective service operation but mainly relies on historically observed bookings. They represent a biased version of demand since censoring issues occur when demand surpasses available supply. In this study, we develop a demand-unconstraining method based on real-world booking and vehicle search request data from a free-floating e-carsharing operator. Building on our unconstrained model, we demonstrate the impacts on service performance in a discrete event-driven simulation of a machine-learning-based vehicle relocation algorithm. Our work contributes by presenting insights into real-world censoring behavior of demand. By applying our approach to service operation, the performance can be increased by up to 6% and relocation efforts can be decreased by up to 22%.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.