The conventional traffic demand forecasting methods based on revealed preference （RP） data are not able to predict the modal split. Passengers＇ stated intentions are indispensable for modal split forecasting and ev...The conventional traffic demand forecasting methods based on revealed preference （RP） data are not able to predict the modal split. Passengers＇ stated intentions are indispensable for modal split forecasting and evaluation of new traffic modes. This paper analyzed the biases and errors included in stated preference data, put forward the new stochastic utility functions, and proposed an unbiased disaggregate model and its approximate model based on the combination of RP and stated preference （SP） data, with analysis of the parameter estimation algorithm. The model was also used to forecast rail transit passenger volumes to the Beijing Capital International Airport and the shift ratios from current traffic modes to rail transit. Experimental results show that the model can greatly increase forecasting accuracy of the modal split ratio of current traffic modes and can accurately forecast the shift ratios from current modes to the new mode.展开更多
文摘The conventional traffic demand forecasting methods based on revealed preference （RP） data are not able to predict the modal split. Passengers＇ stated intentions are indispensable for modal split forecasting and evaluation of new traffic modes. This paper analyzed the biases and errors included in stated preference data, put forward the new stochastic utility functions, and proposed an unbiased disaggregate model and its approximate model based on the combination of RP and stated preference （SP） data, with analysis of the parameter estimation algorithm. The model was also used to forecast rail transit passenger volumes to the Beijing Capital International Airport and the shift ratios from current traffic modes to rail transit. Experimental results show that the model can greatly increase forecasting accuracy of the modal split ratio of current traffic modes and can accurately forecast the shift ratios from current modes to the new mode.
