| In recent years,online car-hailing has developed rapidly as a new format of "Internet +sharing economy".Online car-hailing has become an important part of the way people travel in the urban transportation system.However,online car-hailing provides an efficient and comfortable ride experience At the same time,it also produced a series of social problems,especially the safety of riding.However,the existing research on online car-hailing mainly stays in the qualitative description,and the micro-field of behavior selection is rarely used to study the travel behavior of online car-hailing users in order to understand the function of online car-hailing in the modern transportation system..As young people who are the main users of online car-hailing,portraying their travel behavior is also an important issue that needs to be resolved.Therefore,this article intends to study the travel choice behavior of university students’ online car-hailing users in the context of urban comprehensive transportation travel,and add subjective well-being factors that are easily overlooked in domestic transportation research,in-depth discussion of different estimation models and different travel needs to affect college students’ travel The key factor in choosing behavior.Based on the travel survey data of university students who used online car hailing in Dalian in 2019,this paper constructs a full sample MNL and NL model,a rigid and flexible travel NL model,discusses the differences of university Students’ travel behavior under different model estimation methods and different travel needs,and analyzes the key factors affecting the choice of travel mode of university students’ online car-hailing users.Based on the results of the Logit model,a sensitivity analysis is carried out to propose reasonable measures to promote the development of online car-hailing.Finally,the decision tree method is used to further analyze the overall service satisfaction of online car-hailing.This study uses R programming to build the Logit model.The results show that:Although the structure of the MNL model is simple,the existence of IIA characteristics will cause a certain deviation of the model,and the NL model can overcome this defect and have higher model accuracy;In the full sample NL model,gender and safety are important factors that influence the choice of online car-hailing,and university students with higher subjective well-being prefer public transportation,in flexible travel,gender and subjective well-being are displayed as the full sample NL model has similar results,but has no significant effect on rigid travel,combining the full sample and sub sample models,it is found that the overall service satisfaction has a certain effect on the choice of online ride-hailing;Through decision tree analysis,it is found that comfort and convenience are the key factors that affect the overall service satisfaction of online car-hailing;Online car-hailing companies can effectivelyincrease the share of online car-hailing by reducing the travel time and costs of online car-hailing,improving safety,and adding rigid and flexible travel options for different travel needs. |
PDF Full Text Request