Evaluation of Alternative Technologies to Estimate Travel Time on Rural Interstates

Abstract

New technologies are becoming more and more readily available to collect vehicular traffic data in a cost-effective and non-intrusive fashion. The primary objective of this paper is to present the results of an evaluation of alternative technologies to estimate travel time along a segment of Interstate 91 in Western Massachusetts where traffic volumes and corresponding sample sizes are expected to be relatively low. The first means of data collection uses global positions system technology (GPS) technology employed by INRIX, an enterprise that provides traffic-related information including travel time, directions, and other driver services. The second means uses Bluetooth technology and field data collected by another vendor, BlueTOAD, along the I-91 study site. A third means of data collection using a license plate based method was devised by the authors to provide “ground truth” travel time against which the results of the above two technologies were compared and evaluated. The data analysis showed that sufficient sample sizes were collected and that the accuracy of travel times estimated from data provided by both vendors (i.e., GPS-based INRIX and Bluetooth-based BlueTOAD) is acceptable since their mean absolute percentage errors (MAPE) were consistently less than 6 percent.