Research On The Linear And Nonlinear Relationship Between Bus Perceived Service Quality And Passenger Loyalty

Developing public transportation is an effective way to solve urban traffic congestion,reduce carbon emissions,and maintain sustainable urban development.This is also why many countries have implemented various measures to improve the performance of public transportation systems.The service quality and service level of public transportation are important reasons that affect whether passengers are willing to and repeatedly choose public transportation travel.Improving passenger satisfaction and loyalty on the basis of improving the quality of public transportation service is an effective way to increase the passenger flow.Although a large number of studies have investigated how passenger perceived service quality affects their loyalty,and proposed key determinants of improving passenger loyalty to bus,there is still a lack of research on whether the form of the relationship between perceived service quality and passenger loyalty is independent of the types of methods used in different contexts.Therefore,this study proposed and used Total Accumulated Local Effects(TALE)to analyze the results of Structural Equation Model(SEM)and Gradient Boosting Decision Tree model(GBDT),and explored the essential relationship between perceived service quality and overall loyalty.It aimed to enrich the literature in the field of public transportation service quality and met people’s demand for public transportation service quality,and improving the public transportation travel sharing rate.First of all,on the basis of reading,analyzing,and summarizing a lot of domestic and foreign literature,according to the theoretical basis and research status of perceived service quality,passenger satisfaction and loyalty,this thesis selected four latent variables of perceived service quality of transport-related technical attributes,perceived service quality of transport comfort and safety,passenger satisfaction and passenger loyalty to build the passenger loyalty model based on SEM and GBDT models.On this basis,a survey was conducted on the situation of passengers taking public transportation in Xiamen through a questionnaire,and the reliability and validity of the questionnaire data were tested using SPSS 24.0.Secondly,this thesis proposed TALE based on the interpretable method Accumulated Local Effects(ALE),which took into account the impact of the actual sample score distribution on the results while having the advantage that ALE could handle relevant variables.By using TALE to compare the results of SEM and GBDT models,it was found that there was no obvious difference between the normalized TALE of perceived service quality in SEM and GBDT models,reflecting a similar nonlinear relationship.Therefore,the results showed that there was a nonlinear relationship between passenger perceived service quality and overall passenger loyalty,and the nonlinear relationship was independent of the type of method used.Finally,based on the Importance-Performance Analysis(IPA)and TALE method,this thesis identified the key influencing factors to improve the overall loyalty of passengers.The results showed that the waiting time of bus stops and the temperature in the bus were the service attributes that need to be improved.Bus service time,transfer convenience,walking distance between the departure point and the station and the punctuality rate of bus operation were also important service attributes conducive to improving the overall loyalty of passengers.On this basis,the thesis summarized and analyzes the existing problems and reasons of public transport in Xiamen,and put forward suggestions to improve passenger loyalty.