Estimation of activity-based model parameters and trip chain demand from vehicle identification data

抄録

This paper proposes a maximum-likelihood method to calibrate the activity-based model and estimate trip-chain demand by using vehicle identification data. Vehicle identification data can be collected by a license plate survey or automatic identification technology, which provides time and the sequence of vehicles along a series of designated survey points. The paper adopts the hierarchy activity-based decision model, in which the upper tier is an activity pattern generation model, and the lower tier is a destination and route choice model which are unobserved. The maximum-likelihood estimation problem is formulated in which the unobserved parameters are calibrated. The paper proposes a method to solve this calibration problem using the EM algorithm. The model and algorithm are then tested. The results illustrate the efficiency of the model and its potential for an application to real-world cases.