Analysis and Prediction of Carsharing Demand Based on Data Mining Methods
Open Access
- 5 June 2021
- journal article
- research article
- Published by MDPI AG in Algorithms
- Vol. 14 (6), 179
- https://doi.org/10.3390/a14060179
Abstract
With the development of the sharing economy, carsharing is a major achievement in the current mode of transportation in sharing economies. Carsharing can effectively alleviate traffic congestion and reduce the travel cost of residents. However, due to the randomness of users’ travel demand, carsharing operators are faced with problems, such as imbalance in vehicle demand at stations. Therefore, scientific prediction of users’ travel demand is important to ensure the efficient operation of carsharing. The main purpose of this study is to use gradient boosting decision tree to predict the travel demand of station-based carsharing users. The case study is conducted in Lanzhou City, Gansu Province, China. To improve the accuracy, gradient boosting decision tree is designed to predict the demands of users at different stations at various times based on the actual operating data of carsharing. The prediction results are compared with results of the autoregressive integrated moving average. The conclusion shows that gradient boosting decision tree has higher prediction accuracy. This study can provide a reference value for user demand prediction in practical application.Keywords
Funding Information
- National Natural Science Foundation of China ([No.71621001])
- National Natural Science Foundation of China under Grant ([No. 71961137008])
This publication has 21 references indexed in Scilit:
- On-demand high-capacity ride-sharing via dynamic trip-vehicle assignmentProceedings of the National Academy of Sciences of the United States of America, 2017
- Modeling the propensity to join carsharing using hybrid choice models and mixed survey dataTransport Policy, 2016
- Finding Factors that Influence Carsharing Usage: Case Study in SeoulSustainability, 2016
- Empirical analysis of free-floating carsharing usage: The Munich and Berlin caseTransportation Research Part C: Emerging Technologies, 2015
- Time Series Analysis of Booking Data of a Free-Floating Carsharing System in BerlinTransportation Research Procedia, 2015
- Modelling users’ behaviour in inter-urban carsharing program: A stated preference approachTransportation Research Part A: Policy and Practice, 2015
- Predicting new forms of activity/mobility patterns enabled by shared-mobility services through a needs-based stated-response method: Case study of grocery shoppingTransport Policy, 2014
- Modeling Station-Based and Free-Floating Carsharing DemandTransportation Research Record: Journal of the Transportation Research Board, 2014
- How Carsharing Affects the Travel Behavior of Households: A Case Study of Montréal, CanadaInternational Journal of Sustainable Transportation, 2012
- Growth in Worldwide CarsharingTransportation Research Record: Journal of the Transportation Research Board, 2007