Analysis And Modelling Of Residents' Car Dependence Under The Impacts Of Residential And Workplace Built Environment

With urbanization process accelerating in China,Chinese cities are experiencing urban sprawl and rapidly changing built environment.Meanwhile,the improvement of people's living standards and long travel distances have led to explosive growth of car ownership and use for daily travel.The aimless expansion of cities and rapid motorization caused a series of city diseases,such as traffic congestion,environmental pollution and energy consumption.In order to solve the aforementioned problems caused be car dependence,it has become an important agendas of city administrators and scholars to reduce car ownership and use through promoting urban planning strategies and transport policies.Owing to lock-in effects of the city spatial structure and land use,optimizing the urban built environment is viewed as the radical solution to the city diseases caused by car dependence.Therefore,it is necessary explore the influence mechanism of the built environment on car dependence from a individual perspective.Based on the influence mechanism,city and traffic planning strategies can be developed to reduce car ownership and use,which are helpful for mitigating traffic congestion,saving energies and reducing emissions.This study analyzes residents' car dependence based on the theoretic basis of the connection between land use and travel behavior.This study is designed following“analyzing the key research problems—modeling the influences—conducting empirical analysis” as the main research line.The relationship between the built environment and three key components of car dependence(i.e.,car ownership,car use and mode shift)is explored from a static and dynamic combination perspective.First,car ownership behavior is analyzed.The relative importance and nonlinear influences of the built environment attributes at residential and workplace locations on car ownership are explored.And the threshold effects of the built environment attributes are revealed.Second,considering the mediating effects of car ownership,we explored the influences of the built environment attributes at residential and workplace locations on car commuting,in which the spatial heterogeneities are put emphasis on.Third,the causal relationship between the built environment and car commuting is focused on.The influences of the built environmental changes on commuting mode shift is explored and the gender differences between household couples in the influences are revealed.The main contents of this study include five aspects:(1)Car ownership behavior is analyzed.Considering the built environment attributes at residential and workplace locations,the relative importance and threshold effects of the built environment attributes are explored.A gradient boosting decision tree model is built to conduct the empirical analysis based on Changchun household travel survey data.The empirical analysis compares the differences in the influences of the built environment residential and workplace locations by presenting the relative importance of the influential factors.And the nonlinear influences of the built environment attributes are revealed.(2)Using commuting trips as the basic analysis unit,the role that the built environment at residential and workplace locations plays is explored considering the mediating effects of car ownership.Based on a Bayesian cross-classified structural equation model,the Changchun household travel survey data is used to explore the influences of the built environment on car commuting.Meanwhile,the results confirm the mediating role that car ownership plays in the relationship between the built environment and car commuting.And the spatial heterogeneities at both residential and workplace locations are captured.(3)Focusing on the casual relationship between built environment and travel behavior,this study explores the influences of changes in residential built environment and life events on commuting mode shift from a dynamic perspective.Meanwhile,the different roles that changes in residential built environment and life events play in influencing commuting mode shift between household couples are explored considering the spouse similarity.Based the binary Probit regression model and seemingly unrelated Probit regression model,data from China Labor-force Dynamics Surveys is used to conduct the empirical analysis.The results confirm the casual relationship between changes in built environment,life events and commuting mode shift.Moreover,the differences in the influences of the influential factors between household couples are identified.(4)In order to reduce car dependence in the whole process,this study systematically discusses the influences of the built environment on car ownership,car commuting and commuting mode shift from an individual perspective.By combining the static and dynamic relationships,this study builds an analytical framework between the built environment and car dependence,which can provide empirical evidence from rapidly urbanizing contexts and enrich the theoretical analysis system between the built environment and travel behavior.It is expected to enrich the research on the influence mechanism between the built environment and travel behavior from a theoretical perspective.Meanwhile,it can provide scientific basis for mitigating the city diseases(e.g.,traffic congestion,energy shortages and environmental pollution)caused by car dependence by city and traffic planning.