Study On The Identification And Influence Mechanism Of The Heterogeneity Of Car Dependence

The increasing popularity of private car use has brought convenience to residents while causing traffic congestion,environmental pollution and other negative impacts,and reducing car use is the key to solving the problem.However,car dependence hinders the willingness of car users to shift their travel mode,and different dependent groups differ in their preferences for travel mode choices and responses to policy measures,making it more difficult to manage demand and effectively guide car use.Therefore,an in-depth study of car dependence heterogeneity,grasping its group travel characteristics and determinants,and analyzing its influential role in travel mode decisions can help provide a reference basis for transportation planning managers to formulate more refined policy measures for car use.This paper uses the travel survey data of car users in Kunming to conduct research.The main research work and conclusions include:(1)To clarify the concepts of car dependence and car dependence heterogeneity,to consider the subjective preferences and objective needs of car users,and to select and process indicators for car dependence heterogeneity identification from two dimensions: attitude and behavior;using the latent category model(LCM),combining the model fitting effect,interpretability of latent categories and research value to determine a reasonable number of car dependence heterogeneous groups,and classifying the belonging categories of car users based on the posterior probability of latent categories to achieve the purpose of car dependence heterogeneity identification;on this basis,heterogeneity characteristics of potential categories were analyzed from longitudinal and cross-sectional perspectives,respectively.The results of the study showed that there were four heterogeneous groups of car dependence:demand-dependent(9.1%);non-dependent(24.9%);fully dependent(41.3%);and preference-dependent(24.7%).(2)Taking the heterogeneity of car dependence as the research object,based on the MNL model,the influence of the heterogeneity of car dependence in the decision-making of short-distance travel mode is explored,and the characteristics of the heterogeneity of car dependence in the evaluation of transportation mode service quality are analyzed.The results show that considering car-dependent heterogeneity in travel mode choice modeling has an enhancing effect on the model fit;car-dependent heterogeneity has a significant effect on car users’ short-distance travel mode choice,and non-dependent and preference-dependent groups are relatively more likely to shift their travel mode;in addition,the service quality of each transportation mode is evaluated differently by different dependent groups,and comfort becomes a key indicator that prevents car users from switching to green transportation modes.(3)The MNL model and Bayesian network(BN)model were developed to investigate the factors and mechanisms influencing car dependence heterogeneity,respectively.Firstly,the factors with significant influence on car dependence heterogeneity are extracted based on the results of MNL model;secondly,a BN model for car dependence heterogeneity is established,and the mechanism of the influence of car dependence heterogeneity under the interaction of the factors is studied for the correlation relationship and strength between the factors.The results of the study reflect that occupation,driving age,average acceptable daily transportation cost,and convenience of access to public transportation have significant direct effects on car dependence heterogeneity,while age,gender,and annual household income have indirect effects on car dependence heterogeneity.