Abstract
The air quality conformity process requires travel speed estimation for a horizon year, where often few roadway characteristics are precisely known. This paper presents the results of a study to validate three relatively common speed estimation techniques comparing predicted speeds with speeds measured on specific roadway links and illustrates how disparity in speed predictions affect nitrogen oxide (NOx) and volatile organic compound (VOC) estimates in the MOBILE6 emissions model. The results are intriguing. On average, the mean absolute errors for the three techniques were similar, ranging between 8 and 12 mph. This large error resulted because free-flow speed and capacity from the long-range model were not always representative of the links being studied. Further, the study suggests that the error associated with any of the techniques could be reduced to 4 mph through judicious altering of the default capacity or default free-flow speed for a given class of links, a feasible approach even for large networks. Finally, MOBILE6 testing with Virginia metropolitan data suggests that the differences in speed predictions yield a 2.5% difference in estimated VOC emissions, with one technique overestimating VOC emissions by 1.75% and two underestimating VOC emissions by 0.71%. All three techniques underestimated metropolitan area NOx emissions by less than 2%. Because link-based speed validation efforts are relatively rare within the conformity analysis process, the methodology discussed here should be of interest to smaller metropolitan planning organizations seeking to verify that accurate speeds are used in conformity determination.
Get full access to this article
View all access options for this article.