CONTACT US: Contact info
|
Citation |
Cove LL, Clark JE. Transp. Res. Rec. 1993; 1405: 43. |
|
Copyright |
(Copyright © 1993, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing) |
|
DOI |
unavailable |
PMID |
unavailable |
|
Abstract |
A study was undertaken to develop a reliable method for obtaining reasonable estimates of pedestrian volumes on college campuses using short-term volume counts. Pedestrian volume data were collected on the campuses of five colleges and universities in the southeastern United States. The short-term pedestrian volume data collected on the five college compuses were expanded using existing models and analyzed to test the validity of these models. The average sums of the errors between the observed and predicted pedestrian volumes calculated from the existing models were compared with the percentage errors calculated for the additional data collected from the college campuses. The percentage errors were not found to be statistically the same, thus the existing models are not valid for use in expanding short-term pedestrian volume counts on college campuses. Using a random sample of the pedestrian volume count observations from the college campuses, additional expansion models were developed. These models were chosen on the basis of the evaluation of the parameters of the coefficient of determination, R-squared, and the standard error about the mean, SE sub y. These new models were validated using the remaining data that had not been incorporated into the development of the new models. On the basis of the analysis of the data from the college campuses, the following conclusions were made: (a) existing expansion models were not valid for predicting pedestrian volumes on college campuses, (b) the count interval should be started 10 min before the beginning of class periods, (c) accuracy increased as the volume count period increased, and (d) accuracy increased as the prediction volume range increased. Language: en |