Cover Image

Not for Sale



View/Hide Left Panel
Click for next page ( 66


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 65
Validation of Atlanta, Georgia, Regional Commission Population Synthesizer John L. Bowman, Bowman Research and Consulting Guy Rousseau, Atlanta Regional Commission F This paper presents the results of initial base-year and rom the beginning, the Atlanta (Georgia) Regional back-cast validation of the new Atlanta (Georgia) Commission (ARC) took seriously the need to use a Regional Commission (ARC) population synthesizer population synthesizer (PopSyn) properly and (PopSyn), which acts as the conduit of land use infor- insisted on being allowed to validate the synthetic popula- mation to the travel demand model. It takes informa- tion used for travel demand forecasts. Implicit in this insis- tion from the census and the land use model and creates tence is the prerogative to adjust the synthesizer if the a detailed synthetic population consistent with land use validation results are not as expected. With a flexible, forecasts. A travel demand model can then predict travel adjustable PopSyn, validation can then become more of an for this population. The synthetic population includes a iterative tuning procedure. record for each household in the region and a record for The ARC PopSyn works in the following basic steps, each person in the household, so it is well suited for use common to many similar PopSyns. First, it starts with an by travel demand models employing disaggregate estimate of the number of households in each zone, with microsimulation. Although a PopSyn constitutes a pow- details (in the cells of the matrix) for each of several erful tool, it should be used with caution. By design, it demographic categories. It also has population forecasts provides misleadingly precise details about every person for some aggregate categories. These control totals are in the population. Because of limitations of its inputs more accurate but less detailed than the initial estimates. and its synthesizing procedures, at best only some of the An iterative proportional fitting (IPF) procedure adjusts person and household characteristics accurately repre- the detailed distribution to match the control totals. Then sent the population at the regional level of geographic the adjusted numbers of households of each type are aggregation, and many of those characteristics can be drawn from the Public Use Microdata Sample (PUMS). imprecise and inaccurate for very small geographic areas For the base-year population, ARC defined controls such as census tracts. A fundamental goal in the devel- per transportation analysis zone (TAZ), with all the con- opment of a PopSyn therefore is to synthesize as accu- trol values coming from census tables [Summary File 1 rately and precisely as possible, for as disaggregate (SF1), SF3, and Census Transportation Planning Pack- geography as possible, as many variables as possible age]. The synthesizer's design allows flexibility in the def- that determine travel behavior. And a fundamental inition of the matrix cells and control categories so that a requirement in the use of a PopSyn should be to rely on variety of one-, two- and three-dimensional census tables it only for the characteristics it accurately represents can be used to supply controls, thereby enabling the cap- and to aggregate results to a level at which the synthetic ture of valuable joint distribution information available population is precise and accurate. in the census tables. For families, the controls distinguish "with children" from "without." For nonfamilies, the 54