The fuel consumption of vehicular traffic (and associated CO2 emissions) on a given road section depends strongly on the velocity profiles of the vehicles. The basis for a detailed estimation is therefore the consumption rate as a function of instantaneous velocity and acceleration. This paper will present a model for the instantaneous fuel consumption that includes vehicle properties, engine properties, and gear-selection schemes and implement it for different passenger car types representing the vehicle fleet under consideration. The paper will apply the model to trajectories from microscopic traffic simulation. The proposed model can directly be used in a microscopic traffic simulation software to calculate fuel consumption and derived emission such as carbon dioxide. Next to travel times, the fuel consumption is an important measure for the performance of future Intelligent Transportation Systems. Furthermore, the model is applied to real traffic situations by taking the velocity and acceleration as input from several sets of the NGSIM trajectory data. Dedicated data processing and smoothing algorithms have been applied to the NGSIM data to suppress the data noise that is multiplied by the necessary differentiations for obtaining more realistic velocity and acceleration time series. On the road sections covered by the NGSIM data, we found that traffic congestion typically lead to an increase of fuel consumption of the order of 80% while the traveling time has increased by a factor of up to 4. We conclude that the influence of congestion on fuel consumption is distinctly lower than that on travel time.
Abstract