United States Department of Transportation
ROSA P serves as an archival repository of USDOT-published products including scientific findings, journal articles, guidelines, recommendations, or other information authored or co-authored by USDOT or funded partners.
As a repository, ROSA P retains documents in their original published format to ensure public access to scientific information.
i
Simulating household travel survey data in metropolitan areas.
-
2003-09-01
[PDF-141.03 KB]
Details:
-
Creators:
-
Corporate Creators:
-
Corporate Contributors:
-
Subject/TRT Terms:
-
Publication/ Report Number:
-
Resource Type:
-
Geographical Coverage:
-
Corporate Publisher:
-
Abstract:Census data provide a rich range of socioeconomic characteristics from which it is shown that trip characteristics can be simulated. This report summarizes research into the simulation of the trips and trip characteristics for a random sample of households drawn from census data. The simulation source is the 1990 PUMS data from the 1990 Decennial Census of the United States.
A set of categories is defined for the simulation that allows the development of significantly different statistical distributions of trip characteristics, using the 1995 NPTS data. Based on the census data, samples of households are obtained and their trip characteristics in terms of number of trips by purpose, mode, time of departure, and trip length are simulated, using a Monte Carlo type of simulation procedure. This is performed for three regions: Baton Rouge, Louisiana, Dallas-Fort Worth, Texas, and Salt Lake City, Utah. While there are found to be a number of statistically significant differences in the various trip characteristics between the simulation data and actual household travel surveys conducted in 1997 in Baton Rouge, 1996 in Dallas-Fort Worth, and 1993 in Salt Lake City, the numeric differences in many of the characteristics are actually quite small. It is found that the simulation, as currently defined, does not capture trip-length variations that may be attributable to city size, nor does it do as well as might be hoped in capturing effects resulting from differences in household size between cities such as Dallas and Salt Lake. Further refinement of the simulation procedure appears to be warranted.
In the case of Baton Rouge, comparisons are made on the trip rates by purpose with the existing trip generation models (which were borrowed in 1991 for the Baton Rouge area), with national default figures, and with new trip-generation models developed from the 1997 data. The simulation was found to perform much better than the borrowed trip-generation models and the national default figures. In comparison with new trip-generation models, the simulation was found to perform quite well, although the poorest results were obtained with home-based shopping trips.
It is concluded that simulation is a feasible procedure for creating synthetic household travel survey data, using the procedure outlined in this report. A number of new avenues for research are identified, which should enhance the results further.
-
Format:
-
Funding:
-
Collection(s):
-
Main Document Checksum:
-
Download URL:
-
File Type:
