| Nowadays,the problem of population aging has become the focus of national and social attention.According to the international standard,China is about to enter the ’deep aging society’.How to meet the travel needs of the elderly has become one of the important goals of the current urban transportation system reform.Considering that the travel behavior of the elderly is mostly non-commuting travel,while the existing urban planning,transportation facility design,and public transportation service system mainly meet the commuting travel needs,and lack of consideration for the travel needs and satisfaction of vulnerable groups such as the elderly.If the change of travel demand and travel purpose caused by the change of population structure cannot be considered,it will inevitably lead to more acute traffic congestion and traffic safety problems.Therefore,it is particularly urgent to study the choice behavior of public transport mode for the elderly.In order to reveal the travel characteristics and laws of the elderly and promote the reform and development of the urban public transport system for the elderly.This paper takes the elderly as the research subject,designs a questionnaire combining SP and RP,obtains the panel data needed for regression analysis,tests the reliability and validity of the data,and analyzes the basic characteristics of the elderly travel.Through the survey,it can be seen that the elderly travel mode is mainly public transportation,and the introduction of the use rate is second only to the public transportation of the car behavior control.Considering the evaluation of the comfort,convenience and reliability of the three travel modes of bus,subway and car as the psychological latent variables,the advantages of each latent variable represented by four index variables are demonstrated,and the problem items of each index variable are described in detail in the questionnaire.The structural equation model(SEM)is used to calibrate the perceived psychological latent variables of the elderly,and the fitting degree index is used to verify that the SEM constructed in this paper has a high fitting degree.The calibrated psychological latent variables are introduced into the random parameter Logit model.The stepwise regression method is used to determine the coefficients of travel fare and walking distance as random parameters,and the SEM-random parameter Logit model is constructed.Comparing SEMrandom parameter Logit model with mixed Logit model,random parameters without SEM and traditional Logit,the AIC and BIC of SEM-random parameter Logit are the smallest and the fitting effect is the best.Finally,combined with the regression results of the model,measures and suggestions for improving the public transportation environment for the elderly are proposed.The research shows that travel fare and walking distance have a significant negative impact on travel utility,but individual heterogeneity leads to different effects.The coefficient value is affected by the elderly individual,travel characteristic attributes and travel perception psychological latent variables,mainly manifested as convenience,travel time,car ownership,daily travel distance,education and other characteristic variables have a significant impact on the fare and walking distance coefficient.Among them,travel time has the greatest impact on travel value and is negative,while convenience has a significant positive impact.Based on comfort,the average relative coefficient formula of travel fare and walking distance is constructed.Through sensitivity analysis,it is concluded that when the travel mode running time increases,the sensitivity of the elderly to fare and walking distance increases.When the convenience of travel mode increases,the elderly are less sensitive to fares and walking distance;when the level of education,daily travel distance and family car number of the elderly increases,the sensitivity of individuals to fares and walking distance decreases. |
PDF Full Text Request