Publication: Taxi dispatches using supply forecasting: A time-series based approach
Date
20-01-2017
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In this paper, we consider the problem of supply forecasting in the context of a mobile application based taxi hailing service. We propose a forecasting methodology which incorporates information from factors which influence supply, such as driver incentive schemes, holidays, hour of the day and the day of the week. In order to improve the forecast accuracy, we also propose techniques to deal with data scarcity by exploiting spatial correlation. The proposed methodology along with the techniques to deal with data scarcity improves the forecast accuracy substantially (greater than 15%) as compared to the raw data forecasting by directly using the appropriate time-series models, for the supply data of the city of Bengaluru, India. Finally, we propose a taxi dispatch algorithm which leverages the supply forecast to reduce the supply-demand mismatch.
Description
Keywords
GPS data, Spatio-temporal data mining, Taxi dispatch, Time series analysis, Transportation