TRANSFERABILITY AND UPDATING OF DISAGGREGATE TRAVEL DEMAND MODELS

Abstract

In recent years much work has gone into the development of disaggregate travel demand models. However, little has been done to evaluate the ability of these models to predict travel behavior in locations other than the area for which the model was estimated. Unlike aggregate models, the parameters of disaggregate models are not dependent on a particular zonal system and therefore have the potential for transferability. The motivation behind transferring is clear - if a model estimated in one area can be transferred to another, the cost of conducting transportation studies could be greatly reduced. Several possible approaches for transferring are developed and discussed from a theoretical perspective. For an empirical evaluation, a work-trip modal-split model estimated on Washington, D.C., data is transferred to New Bedford, Massachusetts, using each of the proposed approaches. The results of estimating the original model on Los Angeles data are also represented. The most significant result is the exceptional performance of the original Washington work mode choice model on both NEw Bedford and Los Angeles data. This is noteworthy in view of the extreme differences of the means for several variables between these cities. Of the several approaches for transferring that were developed, Bayesian updating based on combining the existing model coefficients with the estimation results from a new sample gave the best overall performance. The results of this study indicate that the potential transferability of disaggregate travel demand models can be realized. /Author/

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 135 citations exist upstream.