Simulations used to estimate carbon dioxide (CO2) emissions and fuel consumption of medium- and heavy-duty vehicles over prescribed drive cycles often employ engine fuel maps consisting of engine measurements at numerous steady-state operating conditions. However, simulating the engine in this way has limitations as engine controls become more complex, particularly when attempting to use steady-state measurements to represent transient operation. This paper explores an alternative approach to vehicle simulation that uses a “cycle average” engine map rather than a steady state engine fuel map. The map contains engine CO2 values measured on an engine dynamometer on cycles derived from vehicle drive cycles for a range of generic vehicles. A similar cycle average mapping approach is developed for a powertrain (engine and transmission) in order to show the specific CO2 improvements due to powertrain optimization that would not be recognized in other approaches. While a related paper  focuses primarily on simulation results, this paper summarizes the associated experimental results and development of the concept. Test results for an engine and a powertrain demonstrate the cycle average map is a smooth, continuous surface. The paper also compares accuracy of using the steady-state engine map versus the cycle average engine map. Considerations for using the cycle average map for vehicle simulation in the context of CO2 regulations are also discussed with the conclusion that the cycle average map appears to be a technically viable approach to utilizing engine fuel consumption performance in a vehicle regulation.