This paper builds a two-stage model to predict the Urban Rail Transit (URT) share rate of the cities with no rail presently. The first stage will predict the proportion of the transport structure with no rail using transfer curve. At the second stage, some influencing factors are selected as characteristic variables of the utility function to forecast passengers transferring from other traffic modes to rail based on multinomial logit model (MNL). Maximum likelihood estimation is adopted to calibrate the parameters. Finally, take Suzhou as an example to apply the model and analyze the sensitivity of each factor. The results shows that the URT share rate can be improved by reducing waiting time, access time, rail fares and increasing the speed of rail, among which reducing rail fares works best, reducing access time a little less and reducing waiting time the least.
Forecasting URT Share Rate Based on MNL Model
11th International Conference of Chinese Transportation Professionals (ICCTP) ; 2011 ; Nanjing, China
ICCTP 2011 ; 677-688
2011-07-26
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
A New Approach for Urban Resident Trip-Mode Share Rate Forecasting
British Library Conference Proceedings | 2009
|Application of MNL Model in Forecasting Mode Share in Transportation Corridor
British Library Conference Proceedings | 2009
|