Research On Use Preference Of Shared Autonomous Vehicles Based On Bayesian Efficiency Experiment

With the development of autonomous driving technology,shared autonomous vehicle integrate autonomous driving technology with the concept of sharing,which is expected to contribute key forces in solving urban traffic problems,promoting sustainable urban development and meeting diverse travel demands of residents.The introduction of shared autonomous vehicles not only needs continuous research and development in technology,but also needs wide acceptance and support from travelers.Therefore,from the perspective of travel behavior,it is of great significance for transportation policy departments,autonomous driving technology research and development enterprises and travel service providers to study the acceptance and use preference of travelers to shared autonomous vehicles.The purpose of this paper is to explore the acceptance and use preferences of travelers for shared autonomous vehicles,the key factors that affect the acceptance and use preferences of travelers,and effects,and to classify potential users of shared autonomous vehicles.Discrete choice experiment was designed,and the choice set which includes four types of shared autonomous vehicle named as standard,rest,work and game was set based on travelers' demands.Travel-related attributes and vehicle-related attributes were set in the choice experiment.Using the Bayesian D-efficient design method combined with the pilot survey,24 different stated choice scenarios were generated.Based on the experimental design results,a data investigation program was developed.A questionnaire survey was carried out in Chengdu,and a total of 700 valid questionnaires were collected.The results of the questionnaire were analyzed by descriptive analysis and variance analysis.The results show that the respondents' average acceptance of shared autonomous vehicles is relatively high(69.6%).Many respondents(57.3 %)express that they are willing to give up buying private cars in the future.The degree of understanding of autonomous vehicles,monthly income,family car ownership and frequency of shared trips significantly affect the acceptance of travelers to shared autonomous vehicles and they have a positive effect on the acceptance.The degree of understanding of autonomous vehicles,age,monthly income,driving experience,frequency of shared travels,and family car ownership significantly affect the willingness of travelers to abandon the purchase of private cars,in which driving experience and family car ownership have negative effect and the remaining variables have positive effect.MNL model,LC model and RPL model were established based on the data of travelers' choice behavior,and the key factors affecting the traveler's preference for shared autonomous vehicles were analyzed from the sample population,classified group and individual level respectively.The results show that the significant factors mainly include cost,time,automobile brand,vehicle capacity and Wi-Fi,in which the variables related to cost and time have negative effect,while the variables related to automobile brand,vehicle capacity and Wi-Fi have positive effect.The potential users of shared autonomous vehicles can be classified into "focus on travel efficiency-time-sensitive type","focus on travel experience-cost-sensitive type",and "focus on service configuration-network-dependent type".The class probabilities are 29%,44%,and 27%,respectively,and there is significant preference heterogeneity among the three types of user groups.There is significant heterogeneity in the influence of credit deposit,vehicle cost,in-vehicle time and out-vehicle time on the use preference of travelers,and there is significant difference in the use preference of travelers for different types of shared autonomous vehicle.In addition,the RPL model predicts that the average willingness to pay of travelers is 84 RMB/h.Finally,based on the results of discussion and analysis,relevant suggestions and measures were put forward from the vehicle function design as well as market operation and management of shared autonomous vehicles.