Person-Based Micro-Simulation Demand Model for National Long-Distance Travel in the U.S.A.

Abstract

As the nation and various states engage in funding transportation infrastructure improvements to meet future long-distance passenger travel demand, it is imperative to develop effective and practical modeling methods for analysis of long-distance passenger travel. Evaluating national-level infrastructure improvements requires a reliable analysis tool to model the demand for long-distance travel. The national travel demand model presented in this paper implements a person-level tour-based micro-simulation approach for modeling individuals’ long-distance or national activities in the U.S.A. This paper reviews the model framework, explains the model calibration, and presents applications of the model for policy evaluation and demand prediction. The model was estimated using the latest long-distance travel survey in the U.S.A., which is the 1995 American Travel Survey. As the estimation data is old, and no new long-distance travel survey with appropriate sample size is available to re-estimate the model, model calibration is the solution used to update the model and make it capable of capturing up-to-date travel patterns. Calibrating such a large-scale model can be challenging, because each calibration iteration is very costly. This paper describes the calibration effort conducted on the national long-distance micro-simulation model to showcase how a large-scale travel demand model can be calibrated efficiently. A fuel price scenario is analyzed to show how the national travel demand will change under a national fuel price increase scenario in the future year 2040. Another scenario analysis corresponding to construction of high-speed rail (HSR) is conducted to observe the effects of adding a HSR system to the northeast corridor on travel demand from a national perspective.