Even though the MOBILE5 model has been widely used in air-quality analysis, it cannot be used in the evaluation of transportation improvements that result in a reduction of vehicle acceleration or deceleration. In this study, least-square regression models were developed to take into account the effect of acceleration or deceleration. To fully capture the dynamics of acceleration or deceleration, the value of acceleration or deceleration in the current period and those in previous periods were included. In addition, the durations of acceleration or deceleration were also considered. Grade was incorporated into the models by using it to adjust the values of acceleration or deceleration. In addition to these independent variables, variables that represent tractive power were also introduced into the models. The results from the calibration of the models indicated that most of the coefficients can be reasonably explained. The emissions model developed in this study was evaluated from microscopic and macroscopic perspectives. The results show that the emissions model developed in this study performs better than two other existing emissions models.