Comparative study of VSP binning methods for estimating fuel consumptions of LDVs on urban roads

Abstract

The micro-scale fuel consumption model is a tool commonly used to evaluate the effect of traffic flow conditions on vehicle fuel consumptions, which is an essential step in developing traffic management strategies for saving fuels. In developing any micro-scale fuel consumption model, the Vehicle Specific Power (VSP) binning method has been widely adopted in recent years as a fundamental approach. Existing studies have shown that the high distribution frequency when VSP=0 contributes to high fuel consumptions, so a question quickly emerged about whether VSP=0 should be designed as an independent VSP bin in the VSP binning method. This paper strives to compare different VSP binning methods for Light-Duty Vehicles (LDVs) on urban roads in Beijing in terms of their effects on the estimation of fuel consumptions. After collecting and processing field vehicle activity data and fuel consumption data, three VSP binning methods are proposed for the study. Then, total fuel consumptions and fuel consumption factors are calculated by using the proposed VSP binning methods as well as the second-by-second and average travel speed data. Finally, the estimation accuracy of total fuel consumptions and fuel consumption factors based on the three different VSP binning methods are compared for LDVs on urban roads. The study proves that an independent VSP bin at VSP=0 is indeed necessary, which can improve the estimation accuracy on both total fuel consumptions and fuel consumption factors.