Abstract
Based on Bayes analysis, a new model of trip mode choice is presented. Trip mode choice is divided into three phases: calculating prior distribution, obtaining conditional distribution by sampling and calculating share rate of trip modes. Supply characteristics of trip modes are taken as prior information. Unity value takes the place of unity function in MNL, and then prior distribution is achieved. Condition distribution is gained from sampling information. Bayes analysis is introduced into calculating posterior distribution. Share rates of trip modes, is calculated by total probability formula. Compared with other choice models, Model proposed in this paper improves the forecast accuracy of share rate without the need of parameter calibration like Logit model.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
