The Mechanism And Modeling Of Travel Shift From Private Car To Multi-mode Public Transport

With the continuous development of the economy and the rapid popularization of private car,traffic congestion has become an international problem.Encouraging travelers to shift from private car to public transport is an effective way to alleviate traffic congestion,ease the contradiction between supply and demand,and establish a sustainable transportation system.Exploring the internal mechanism and influencing factors of the shifting behavior,identifying sensitive groups,specific trips,hot spots,and effective policies to reduce car dependence and usage,are key issues of the private car trip shift research.This paper took the multi-mode public transport system composed of rail transit,bus,and public bicycles as the research object.Based on fully understanding and analyzing the behavior and attitude characteristics of private car users,constructed theoretical models and predictive models of shift behavior,revealed the internal mechanism and influencing factors of shifting decision,and proposed guidance strategies for different travelers.The main research content of this article can be divided into five parts:(1)Surveys on the travel shift of private cars to multi-mode public transport.From the four aspects of survey method and object determination,survey content design,survey processing and data collection,questionnaire verification and sample analysis,the survey was systematically and comprehensively designed.The survey included characteristics of individual,travel,built environment,traffic environment,subjective attitude,shift intention and social impact.Designed car psychological dependence questionnaire,multi-mode public transport combination,traffic policy scenario attributes,implementation level,and established built environment factors acquisition methods.Proposed control and consistency standards for checking sample data,screened and analyzed effective samples and conducted sample data.(2)Classification of private car users based on car dependence.Proposed the conceptual framework and market segmentation methods of car dependence.Considered two-dimensional dependence of attitudes and behaviors,the car users were subdivided into dependence,potential dependence,helpless dependence,and non-dependence.Proposed a psychological dependence measurement structure consisting of operation and efficiency,convenience and safety,environment and comfort,hedonic value,perceived cost,perceived value,perceived behavior control,personal and social norms,and used structural equation models to analysis the causal relationship between factors and dependence;Explored the K-Means method to segment the private car users market,analyzed the differences in the psychological dependence and behavioral dependence,individual heterogeneity,behavioral heterogeneity and spatial heterogeneity of each submarket.(3)The travel shift mechanism of the private car to multi-mode public transport.Revealed the mode preferences of private car users with different car dependence under the traffic policy to shift to multi-mode public transport.Combined with the Random Parameter Logit model,constructed different car dependence traveler shift model,identified the key influencing factors of the shift behavior.Based on the elasticity theory,explained the influence mechanism of characteristics attributes on the shift decision of different dependent car travelers.The results proved that the influence of characteristic attributes on the shift behavior of different car dependent groups had significant differences.(4)The travel shift prediction models.Build prediction models based on CART,support vector machines,neural networks,LightGBM and random forests.Tested the optimal parameter combination and analyzed the importance of characteristics through the optimal model.According to the importance of influencing factors,a subset of predictors was selected to find the optimal predictive model suitable for limited predictors.Random forest was the best model for predicting travel shift performance,and had advantages with limited input variables.Built environment characteristics were the most important factor in predicting travel shift,followed by travel,traffic policies and socioeconomic characteristics.The travel distance.The number of rail stations at the origin,the intensity of commercial land at the destination,the number of rail stations at the destination,and the car dependence were the most important factors affecting the accuracy of prediction.(5)Private car travel shift to multi-mode public transport guidance strategies.Proposed a framework from structural strategies and psychological strategies,and summarized the research and application results.Analyzed the social impact of various shift guidance strategies,the probability of shift behaviors,and the implementation effect.Proposed shift guidance strategies for different car dependent groups.Dependent groups should be used public transport administrative strategies.Potential dependent groups should be based on economic strategies,with social culture and public opinion strategies.Helpless dependent groups should focus on public transport administration and technical strategies,and car administration strategies as a supplement.Non-dependent groups should comprehensively use structure strategies,and give play to the social influence and leadership role.