Impact Studies Of Transport Interventions On Mode Choices And Travelers' Perceptions Considering Heterogeneity

Different travel groups have different cognitions,attitudes,frequency of usage and preferences for different travel modes,requirements for travel quality,and responses for different transport policies.This is known as traveler heterogeneity.With the rapid development of economy and city urbanization,prople's requirements for travel quality has increased.However,travel services have been developing in an extensive manner and there is a lack of fine management for travelers' differentiated demands.Travel demand management implement appropriate restrictions and guidance for travelers' behavior through hard and soft transport interventions.Because of traveler heterogeneity,travelers make different responses to transport interventions.Moreover,objective indicators such as mode share are always chosen to evaluate transport interventions.Little attention has been paid to the change of travelers' feeling.Travel perceptions evaluate travelers' both cognitive and affective aspects.Sustaining and positive travel perceptions influence travel choices,also bring positive influences to life,work,and health.This research is aimed to provide theorey basis and policy insights for travelers' market segmentation,differentiated traffic demand management,green modes share increase and travelers' perceptions improvement.In responses,impact studies of transport interventions on mode choices and travelers' perceptions considering heterogeneity are conducted.The main work and conclusions are as follows.(1)Questionnaires about travelers' perceptions,two transport interventions including information intervention and improvement of public transport service are designed.Travelers' perceptions are measured by improved Satisfaction with Travel Scale which includes both cognitive and affective dimensions.Information intervention are related with “non-motorized traveling and health”,“pollution caused by motorized traveling” and “improvement of green mode transport service”.Indicators of public transport service improvement include decrease in walking time,waiting time,in-vehicle time,transfer times,and increase in in-vehicle comfort.Questionnaire are distributed to local residents or permanent migrants working and living in Beijing based on online surveys.(2)Travel perception models for different travel modes are established respectively.The descriptive results show that,for non-motorized travel modes,cycling has higher travel perception ratings than walking.For motorized travel modes,car travelers rate higher for travel perceptions than public transport travelers' ratings.Travel mode consonance is positively related with travel perceptions.Transport environment of residential location has positive impacts on travel perceptions.(3)Heterogeneous traveler groups are identified and influencing factors on class membership are analyzed based on latent class cluster analysis.According to goodness-of-fit and model interpretability,three clusters are identified.Cluster 1(20.4%)travel less frequently and prefer both automobile and green modes.Cluster 2(30.3%)travel moderate frequently and prefer automobile.Cluster 3(49.3%)travel more frequently and prefer travel by green modes such as public transport,bicycle,and walking.(4)Travel choice modeling method based on latent class choice models(LCCMs)are proposed.Travel mode choice models under information intervention and improvement of public transport service are established.LCCMs which consider traveler heterogeneity are superior to traditional multinomial Logit model in both goodness-of-fit and behavioral interpretability.After introducing travel perceptions,both the log likelihood value and R-square of LCCMs and multinominal logit model,under information intervention and public transport service improvement,are significantly increased.(5)The decrease of in-vehicle time and increase of in-vehicle comfort for public transport have significant influences on car use reduction of cluster 2 and cluster 3.Moreover,cluster 3 has a greater reduction in car use.Compared with public transport,under five scenarios of public transport service improvement S1(15 minutes less for in-vehicle time),S2(more comfortable),S3(30 minutes less for in-vehicle time and more comfortable),S5(30 minutes less for in-vehicle time),S6(15 minutes less for in-vehicle time and more comfortable),the probability of cluster 3 traveling by automobile are 17%,21%,11%,63%,and 18% of that under reference scenario S4.Under scenarios S1,S2,S3,S6,the probability of cluster 2 traveling by automobile are 32%,25%,18%,and 26% of that under scenario S4.When in-vehicle comfort of public transport increase,cluster 2 and 3 have more positive travel perceptions.(6)After information intervention,cluster 2 has lower probability of choosing car as travel duration and travling fee increase.Compared with public transport,before and after information intervention,car use probabilities of cluster 2 become 63% and 55% of that after travel duration increase one unit.Car use probabilities of cluster 2 are 97% and 88% of that as travel fee increase one yuan.(7)Travel mode choice and travel perceptions are interacted on each other.Walking contributes to higher travel perception ratings.Public transport trips bring about negative feelings such as worrying about being late and feeling tired.Travelers' perceptions at current stage influence mode choice in next trips.Travelers who have higher ratings for travel perceptions during car trips will reduce public transport use in next trips.Travelers who have higher ratings for travel perceptions during green mode traveling will continue increasing green mode use.(8)Travelers are more receptive to soft transport interventions.The percentages of being willing to reduce car use and choose green mode traveling under soft transport interventions(such as public transport service improvement)and hard transport interventions(such as parking fee increase and congestion charging)are 80% and 60% perspectively.(9)During the formulation of transport policies,appropriate traveler group identification,analysis of travelers' characteristics of different traveler groups,and response differences for different transport policies help to pertinence of policy formulation and implementation.The improvements of public transport service help to significantly increase public transport mode choice and travel perceptions of cluster 2 and 3(accounting for 79.6% of all participants).On the basis of reasonable traveler group division,designing different intervention information for different traveler groups can affect different traveler groups' cognition of and attitudes towards travel mdoes.Furtherly,different traveler groups can be guided to choose green transportation mode more frequently.