Overview of U.S EPA New Generation Emission Model: MOVES

The U.S Environment Protection Agency (EPA) has released a new generation regulatory mobile emission model, entitled MOVES (Motor Vehicle Emissions Simulator) to replace its current emission models, MOBILE and NONROAD. The transition to MOVES will have important policy implications for regional mobile source emissions, particularly inventories associated with transportation conformity. The methodology of MOVES is based on a discrete binning approach compared to average speed based approach employed in traditional emission models. The scope of MOVES has been expanded to estimate emissions at national, regional and project scales, inclusion of number of pollutants and emission processes, alternative vehicle and fuel types. MOVES has an extensive database reflective of real world driving conditions developed by assessing millions of vehicles for a long period of time. Detailed description of methodology, scope and data of MOVES is presented in this paper. Using a case study of Cook County, Illinois emission estimates of green house gas namely carbon dioxide (CO 2 ) and a criteria pollutant namely nitrogen oxide (NO x ) are estimated by MOVES and compared to its predecessor, MOBILE6.2. The fundamental differences in emission estimation methodology between the two models are the reasons for different estimation results, with MOVES believed to be more accurate owing to its theoretical superiority over MOBILE. sophisticated emission models based on their local specific conditions, emission standards, vehicle and road types, fuel types, inspection/ maintenance programs. In the U.S, emission models can be categorized as macroscopic and microscopic models (1). Macroscopic models use average aggregate network parameters to estimate network-wide energy consumption and emission factors. The primary macroscopic emission models used in the U.S developed for regulatory purposes have been the U.S EPA's MOBILE and California Air Resources Board's EMFAC model. Both these models are conceptually similar as they use network wide average speed as input to produce activity-specific emission factors which when multiplied with vehicle activities such as vehicle miles traveled (VMT) gives the total emission inventories. The main drawback of these models is in the use of a single