| Ridesharing is a travel mode in which two or more groups of passengers with similar itineraries and times travel in the same vehicle.Ridesharing can improve the efficiency of vehicle utilization and is considered to be an energy-saving and environmentally friendly travel mode.With the development of Internet technology and the popularization of the sharing economy,online ride-hailing has become an emerging travel mode,and the traditional ridesharing between acquaintances is also evolving to the online ridesharing based on strangers.At present,there are two main ridesharing modes,namely ridesplitting and carpooling.However,there is currently little research on the differences between the two ridesharing modes.The differences in travel characteristics between ridesplitting and carpooling are still unclear,and their fuel-saving benefits also lack quantitative comparison and analysis.To this end,we are motivated to use the massive actual operation data provided by Didi Chuxing to study the operational differences between the two ridesharing modes from the perspectives of travel feature mining and fuel-saving benefit estimation.The main research contents of this paper are as follows:(1)Based on a large amount of actual operation data,in-depth exploration of the temporal and spatial characteristics,operational characteristics and user characteristics of the two shared travel modes of ridesplitting and carpooling.This paper proposes a classification method of carpooling drivers,identifies the service characteristics of different types of carpooling drivers,and establishes a distance-saving ratio estimation model.The results show that ridesplitting mainly provides short-distance services for urban internal travel demands,whereas carpooling focuses on providing long-distance services in the administrative region of a city.Commuter-type carpooling drivers maintain the highest average distance-saving ratio.High-frequency carpooling drivers maintain the highest average distance-saving ratio.(2)The three types of co-opetition relationships(competition,extension,and complementation)between online ridesharing and urban rail transit are studied separately,and rail transit stations are classified based on AFC data.This paper focuses on the study of the travel characteristics of the URT-competing ridesharing trips and analyzes the competition intensity.The basic characteristics of the URT-extending ridesharing trips and the travel demands in suburban areas without rail transit coverage are studied.In addition,based on the NMF algorithm,the commuter travel chain of " URT-extending ridesharing + rail transit" is identified.Finally,based on the competition and cooperation relationship,relevant policy suggestions for the development of ridesplitting and carpooling are put forward.The study found that both online ridesplitting and carpooling can effectively make up for the shortcomings of the rail transit network,but there are certain differences in the interaction between the two and the rail transit network.(3)A method for estimating the fuel-saving benefits of online ridesharing is proposed by combining vehicle operation information and fuel economy indicators of various modes of transportation.In addition,the fuel-saving performance of ridesplitting and carpooling is studied from multiple dimensions such as driver classification and mode transfer.Then,the fuel-saving estimation model of the commuter travel chain is constructed,and the fuel-saving benefit of the "URT-extending ridesharing + rail transit" commuter travel chain is analyzed.Afterward,the economic benefits of ridesplitting and carpooling trips were studied from the perspectives of drivers,passengers,and platforms.Finally,a series of policy recommendations are put forward to improve the fuel-saving level of online ridesharing.The results show that compared with traveling alone with a car,both ridesplitting and carpooling have certain fuel-saving capabilities.The average fuel-saving ratio of ridesplitting is lower than that of carpooling.Commuter-type carpooling drivers maintain the highest average fuel-saving ratio.From the perspective of transportation system fuel economy,ridesharing is not considered to be fuel-saving,and its scale should be reasonably regulated.The "URT-extending ridesharing + rail transit" commuter travel chain has strong fuel-saving benefits,and the development of this mode should be encouraged.This study first based on the online ridesplitting and carpooling big data,established a series of innovative models,formed a method for mining the characteristics and estimating fuel savings of ridesplitting and carpooling,systematically studied the operating differences between ridesplitting and carpooling,which has important theoretical value and practical guiding significance for the sustainable development of online ridesplitting and carpooling.This paper contains 51 figures,9 tables and 64 references. |
PDF Full Text Request