Research On Travelers’ Choice Behavior For Long-Distance Carpooling

The increasing improvement of urban residents’ living standards and the expansion of car ownership have led to the increase of car utilization,which leads to more and more serious road congestion,pollution and energy waste.How to better reduce the urban road traffic load,alleviate traffic congestion and reduce environmental pollution has always been an important research topic in the field of transportation.Carpooling,as a popular way of travel,is expected to have a huge potential impact on reducing traffic congestion.It is a sustainable measure in the traffic demand management strategy.Therefore,it is necessary to better understand people’s attitude towards long-distance carpooling and explore the determinants of the choice of long-distance carpooling travel mode through empirical investigation.Considering that drivers and passengers will be affected by different factors in the choice of travel mode due to their different roles,this paper innovatively studies drivers and passengers separately,and introduces the variable of "number of carpooling passengers" into the orthogonal experimental design to design the questionnaire.Firstly,the general travel law of travelers is obtained by using descriptive statistics;Secondly,the latent class model is used to divide the driver and passenger population,and the chi square test is used to analyze the heterogeneity;Finally,for each potential category of drivers and passengers,a mixed logit model reflecting individual heterogeneity preference is introduced to explore the determinants of travelers’ choice of long-distance carpooling travel mode,and make marginal prediction.The main results are as follows:(1)For carpool drivers,the main purpose of travel is to visit relatives and friends;Under different travel purposes,the longer the travel distance,the higher the proportion of carpool drivers.Travel time and the number of carpool passengers have a significant negative impact on the choice of carpool drivers,while travel income has a significant positive impact;The second and third groups are more sensitive to the number of carpooling passengers.For each additional carpooling,the probability of choosing carpooling drivers decreases by 3.54% and 10.57% respectively;When the travel time is reduced by 5 minutes,the proportion of carpooling drivers increases by 3.9% and 7.7%respectively;And both groups are not sensitive to the intensity of subsidies.(2)For carpool passengers,the main purpose of travel is shopping;Under different travel purposes,travel distance has different effects on the choice of travel mode,but the proportion of carpooling is the highest.The increase of travel time and cost significantly reduces the choice of carpooling for the second and third groups,while the impact of additional carpooling passengers is not significant.The second group of people is more sensitive to the change of travel expenses.When there is a 60% or 80%discount,the proportion of carpooling increases by 6.05% and 0.76% respectively.Carpooling transfers some taxi people with price advantage,which has little substitution effect on public transport / urban rail.The third group is more sensitive to the change of travel time.When the travel time is reduced by 5 minutes,the proportion of carpooling increases by 8.00%,and about the average people who transfer taxis and buses / urban rail.In order to encourage carpooling behavior,the following suggestions are put forward:(1)for carpooling drivers,for the second group of people,we can reduce travel time by accurately matching carpooling people to improve the travel proportion of carpooling drivers;For the third group,the number of carpool passengers can be mainly controlled.(2)For carpool passengers,for the second group of people,some coupons can be issued regularly to attract users to use carpool;For the third group of people,we can reduce the detour time by accurately matching the carpooling people,and use the time advantage to attract more users to choose carpooling.