Individual mobility: From conventional to electric cars

Abstract

The aim of this report is twofold. First, to analyse individual (driver) mobility data to obtain fundamental statistical parameters of driving patterns for both conventional and electric vehicles. In doing so, the information contained in large mobility datasets is condensed into compact and concise descriptions through modelling observed (experimental) distributions of mobility variables by expected theoretical distributions. Specifically, the stretched exponential distribution is shown to model rather accurately the distribution of single-trips and their duration, and the scale-invariant power-law with exponential cut-off the daily mobility length, the distance travelled per day. We argue that the theoretical-distribution parameters depend on the road-network topology, terrain topography, traffic, points of interest, and individual activities. Data from conventional vehicles suggest three approximate daily driving patterns corresponding to weekday, Saturday and Sunday driving, the latter two being rather similar. Work trips were found to be longer than average and of longer duration. The second aim is to ascertain, via the limited electric-vehicle data available from the EU-funded Green eMotion project, whether the behaviour of drivers of conventional vehicles differs from the behaviour of drivers of electric vehicles. The data suggest that electric vehicles are driven for shorter distances and shorter duration. Data from the Green eMotion project showed that the median real-life energy consumption of a typical segment A, small-sized, electric car, for example the Mitsubishi i-MiEV and its variants, is 186 Wh/km with a spread of 55 Wh/km. The real-driving energy consumption (per km) was determined to be approximately 38% higher than the type-approved consumption. Moreover, we found considerable dependence of the energy consumed on the ambient temperature. The median winter energy consumption per kilometre was higher than the median summer consumption by approximately 40%. The data presented in this report can be fundamental for subsequent analyses of infrastructure requirements for electric vehicles and assessments of their potential contribution to energy, transport, and climate policy objectives. 1