Research on the big data of traditional taxi and online car-hailing: A systematic review

Abstract

The purpose of this paper is to provide a summary of a quick overview of the latest developments and unprecedented opportunities for scholars who want to set foot in the field of traditional taxi and online car-hailing (TTOC). From the perspectives of peoples (e.g., passenger, driver, and policymaker), vehicle, road, and environment, this paper describes the current research status of TTOC's big data in six hot topics, including the ridership factor, spatio-temporal distribution and travel behavior, cruising strategy and passenger service market partition, route planning, transportation emission and new-energy, and TTOC's data extensional application. These topics were included in five mainstreams as follows: (1) abundant studies often focus only on determinant analysis on given transportation (taxi, transit, online car-hailing); the exploration of ridership patterns for a multi-modal transportation mode is rare; furthermore, multiple aspects of factors were not considered synchronously in a wide time span; (2) travel behavior research mainly concentrates on the commuting trips and distribution patterns of various travel indices (e.g., distance, displacement, time); (3) the taxi driver-searching strategy can be divided into autopsychic cruising and system dispatching; (4) the spatio-temporal distribution character of TTOC's fuel consumption (FC) and greenhouse gas (GHG) emissions has become a hotspot recently, and there has been a recommendation for electric taxi (ET) in urban cities to decrease transportation congestion is proposed; and (5) based on TTOC and point of interest (POI) multi-source data, many machine learning algorithms were used to predict travel condition indices, land use, and travel behavior. Then, the main bottlenecks and research directions that can be explored in the future are discussed. We hope this result can provide an overview of current fundamental aspects of TTOC's utilization in the urban area. • Six research topics and their key findings in the traditional taxi and online car-hailing fields were reviewed. • Five research mainstreams of traditional taxis and online car-hailing were concluded. • Five conclusive potential challenges and suggestions of traditional taxi and online car-hailing research were summarized. • Multi-source data, supercomputing, cooperation, and machine learning indicate the promising research directions.