Abstract:The Environmental Protection Agency's (EPA) MOBILE6 driving cycles were developed by combining chase car data collected in three cities: Baltimore, Spokane, and Los Angeles, and then organizing the data by facility and level of service (LOS) to eliminate regional variability in driving. EPA's approach to driving cycle construction presumes that regional driving variability is insignificant when controlling for facility type and LOS. In this study we re-visit the issue of regional driving variability and its potential impact on emissions using driving data recently collected in the Bay Area, Sacramento and Stanislaus, California. We begin by examining regional driving characteristics for four types of driving conditions: un-congested and congested freeway driving, and un-congested and congested arterial driving. The results suggest regional similarities in terms of the average speeds and accelerations, but marked differences in frequency, duration, and intensity of both steady state and acceleration modal events. A one-way ANOVA analysis indicates that regional driving variability exists even after controlling for LOS and facility type. We also show how these regional differences can result in driving cycles with significantly different compositions of modal events (i.e., cruise, idle, acceleration and deceleration) using a new method for constructing driving cycles. An examination of the cycles with respect to steady state driving and acceleration/deceleration modal events confirms that regional driving differences are sufficiently large enough to result in important driving cycle differences, which may also translate into important regional variability in vehicle emissions estimation.