01697548

Component

This study addresses the problem of calibrating activity-based travel demand model systems. The authors formulated this problem as a simulation-based optimization problem with an analytical expression for the objective function. Then the authors proposed a stochastic gradient-based solution algorithm to solve it. The solution algorithm is able to utilize microsimulation to calculate the unbiased estimator of the simulated aggregate statistics in order to evaluate the objective function. Additionally, the authors derived the expressions to compute the gradient of the objective function, making the solution algorithm explicitly utilize the structure of the model-system to accurately and efficiently calculate the gradient. Finally, the authors show that the proposed solution algorithm outperforms the other simulation-based stochastic gradient algorithm in terms of computational efficiency, stability, and convergence. This study has the potential to facilitate wider and easier application of activity-based model systems.

This paper was sponsored by TRB committee ADB40 Standing Committee on Transportation Demand Forecasting.

19-05633

English

8p

2019

Transportation Research Board 98th Annual Meeting

Digital/other

Figures; References; Tables

Activity choices; Algorithms; Calibration; Microsimulation; Optimization; Travel demand

Planning and Forecasting; Transportation (General)

Transportation Research Board Annual Meeting 2019 Paper #19-05633

TRIS, TRB, ATRI

Dec 7 2018 9:31AM