The goal of this paper is to predict the fraction of time vehicles spend in different operating conditions from readily observable emission specific characteristics (ESC), which include geometric design, roadway environment, traffic characteristics, and driver behavior. We rely on a calibrated micro-simulation model to generate second-by-second vehicle trajectory data and use structural equation modeling to understand the influence of observed link ESC on vehicle operation. Our results reveal that 67 percent of link speed variance is explained by emission specific characteristics. At the aggregate level, geometric design elements exert a greater influence on link speed than traffic characteristics, the roadside environment, and driving style. Moreover, the speed limit has the strongest influence on vehicle operation, followed by facility type and driving style. This promising approach can be used to predict vehicle operation for models like MOVES, which was recently released by the Environmental Protection Agency.