| With the further expansion of the city and the increasing number of private cars,the problem of road traffic congestion is becoming more and more serious.In order to alleviate the problem and improve the traffic efficiency,P&R facilities were created against the backdrop of strong efforts to develop and promote public transportation.Users who use P&R facilities drive to the P&R facility and park the car in the P&R parking lot,then transfer to public transportation to the destination.The primary users of P&R facilities are commuters,with the urban rail transit as the main connected means of transportation.Related studies have shown that many people in large cities have experienced long-distance suburban commuting,which takes longer time than ordinary commuting and is also more easily affected by road conditions and surrounding environments.In this context,taking Beijing as an example,this thesis investigates the rail transit transfer behavior of suburb commuting and builds an analytical model.Firstly,the related researches on the P&R transfer behavior and choice of transfer mode are systematically summarized.The definition,classification and implementation of P&R facilities are briefly introduced.The influence factors of suburban park-and-ride commuting behavior are analyzed in the perspective of individual and household attributes,traffic characteristic attributes,and transportation policy attributes,with an emphasis on the effects of commuters'travel characteristics and park-and-ride characteristics.Secondly,based on the influencing factors of suburban park-and-ride commuting behavior,the RP and SP questionnaires are designed.A sampling field survey is conducted on the commuters who used typical P&R facilities.Meanwhile,the online questionnaire survey on potential car commuting users is also conducted.Based on the survey results,descriptive statistical analysis is conducted,resulting in the commuting behavior characteristics of participants.Then,a structural equation model is developed to characterize suburban park-and-ride commuting behavior,starting with individual and household attributes,traffic characteristics,and P&R facility characteristics.The Pearson chi-square test is used in SPSS software to remove the variables with poor significance,so as to select significant variables.The AMOS software is used to fit and estimate the model,and finally four evaluation indicators,?~2,GFI,RMSEA and mean square error(RMR),are selected to test the model for fit superiority.The results show that the model is able to describe the influence of each variable on P&R travel choices,and the structural equation model fits well.The utilities of age,monthly income,and degree of self-preference in personal attributes are 0.45,0.71,and 0.52,respectively.The utilities of total commuting hours,transfer hours,and downtown parking costs in the traffic characteristic attributes are 0.75,0.62,and 0.54,respectively.The utilities of transfer parking fees,transfer facility reserved parking status,and satisfaction of facilities in P&R facility attributes are 0.45,0.66,0.47,respectively.Above-mentioned variables all have a significant influence on commuting.Finally,the traditional Binomial Logit Model(BLM)is used to study the commuter's choice behavior of travel mode between the full-drive travel and the travel with car-rail transit transfer.Two latent variables required by commuters for the characteristics of P&R facilities are added into the BLM to develop an mixed BLM considering latent variables.This new model is used to investigate the influence of different factors on the choice behavior of commuting mode,including age,monthly income,preference of full-drive,total commuting time,transfer time,parking fees in the central area,parking fees of P&R facilities,transfer facility reserved parking status and satisfaction of P&R facility.Comparing the imitative effect of two models,the result shows that the accuracy of the mixed BLM with latent variables is 7.79%higher than the BLM without latent variables. |
PDF Full Text Request