Published in:
2022 | OriginalPaper | Chapter
Short-Term Demand Forecasting Methods for Public Bicycles Under Big Data Environment
Authors: Hongxiao Lin, Hui Sun
Published in: Innovative Computing
Publisher: Springer Nature Singapore
Abstract
As the main form of urban slow traffic system, public bicycles are responsible for solving the “last mile” problem of urban public transportation. However, due to inadequate understanding of travel demand and deficiencies in planning and site selection, many cities still have the problem of no public bicycles to borrow or return to at peak times. The development and application of big data technology provides the possibility for refined analysis and prediction of public bicycle borrowing and repayment needs. In view of this, this paper establishes an autoregressive mobility model based on the historical data of public bicycle rental sites, and proposes a big data-based public bicycle demand forecasting method, and compares the forecast results with the forecast results of the exponential smoothing model. The results prove that, Compared with the exponential smoothing model, the autoregressive model used in this article is more accurate.
Advertisement
