Citation

Hoang LT, Poteat VP. Transp. Res. Rec. 1980; 779: 6-10.

Copyright

(Copyright © 1980, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing)

DOI

unavailable

PMID

unavailable

Abstract

The development of estimates of vehicle miles of travel (VMT) by state and local agencies has been an area of concern for many years. The problem of costs versus accuracy has long been a deciding element in developing VMT estimation procedures. In an effort to lower costs and maintain accuracy, the Federal Highway Administration (FHWA) developed a statistically based procedure for VMT calculation. FHWA's product is presented in the Guide to Urban Traffic Volume Counting (GUTVC). The method proposed in the GUTVC was based on random selection and sampling of highway links that were stratified by volume, facility type, and area type. Based on statistical sampling theory, a desired confidence interval could be assumed and a sample size calculated. The sample was randomly selected from the population of links. The selected links were then matched with a randomly selected day for sampling (i.e., taking counts). In comparison, the current method used by the Florida Department of Transportation (FDOT) employs sample counts and gasoline consumption information to arrive at VMT estimates. The purpose of this FDOT study was to test the GUTVC method for applicability as well as economical and statistical reliability under real-world conditions. FDOT found that VMT calculated by using the procedures presented in the GUTVC were reasonably close to VMT estimates currently developed in-house. In addition, by using traffic counts obtained through the sampling process to create data files, it is possible to entertain thoughts of reduced costs over long periods of time for data collection.