International Choice Modelling Conference, International Choice Modelling Conference 2017

Font Size: 
A joint route choice model for electric and conventional car users
Anders Fjendbo Jensen

Last modified: 28 March 2017



Anders Fjendbo Jensen, Thomas Kjær Rasmussen

DTU Management Engineering, Technical University of Denmark, Kgs. Lyngby, Denmark, {afjje, tkra}

Carlo Prato

School of Civil Engineering, The University of Queensland, Brisbane, Australia,


Worldwide, governments have committed to reducing air pollution and carbon emissions. With a higher share of renewable sources in the electricity production, battery electric cars (EVs) could play a significant role in maintaining these commitments. Growing literature shows an increasing interest in EVs and their market, but current EV travel demand studies are usually based on data collected from users of conventional gasoline or diesel engine cars (CVs) (see e.g. (Golob and Gould 1998; Pearre et al. 2011; Greaves et al. 2014). EVs are however different from CVs in a number of ways, in particular when it comes to the driving range and the refuelling/recharging which can lead to behavioural changes (Jensen and Mabit 2015). EV users might avoid longer and less-planned trips and, when deciding on a route, they might select roads where the general speed is lower, the trip length is shorter, or the charging facilities are better. On the other hand, over a longer period of time, many users do not need charging other than overnight charging at home in order to keep up with their current behaviour (Christensen et al. 2010) . Thus, the impact on traffic of a large scale EV adoption is not obvious, as it cannot be assumed that CVs currently on the road are simply replaced by EVs and individual behaviour otherwise stays constant.

Understanding the behaviour of EV users is important in a number of ways. Beside potential environmental effects, there is a need to understand other related effects, such as effects on the electricity network and the transport network. The objective of this study is to use revealed preferences (RP) data to investigate differences in route choice behaviour between CV and EV users. To our knowledge, this is the first time that a state-of-the-art route choice model has been estimated on RP EV data. In addition, the level of detail in the data allows for accounting for congestion, reliability, topology, weather and socioeconomic background.


This study exploits a unique and vast dataset consisting of GPS records from a large demonstration project about EVs conducted in Denmark during the period 2011-2013. Households participating in the trial had an EV available for a period of three months during which all trips were GPS logged. Additionally, some of the households GPS logged trips by their CV in the month before and the month after the EV was received. The GPS traces were matched to the very detailed NAVTEQ street network (NAVTEQ 2010). The high level of detail of the network is crucial, as EV users might use smaller roads with lower speeds in order to save energy due to current technological restrictions on driving distances. Following the procedure in Prato et al. (2014), route choice behaviour is modelled with a two-stage approach consisting of choice set generation and model estimation. The first stage used a doubly stochastic generation process to generate a choice set consisting of a maximum of 100 unique alternatives for each observed route. Subsequently, the observations were filtered to exclude observations for which the choice set contained only one alternative route or did not contain any alternative reasonably similar to the observed route. In the second stage, a mixed path size correction logit model was estimated for modelling route choice behaviour, (Bovy et al. 2008). Comparison of EV and CV preferences is made possible by estimating jointly across data from each technology using a logit scaling approach with at least one generic parameter across data (Bradley and Daly 1997).


After the map matching and filtering processes, GPS records were available for about 90,000 EV trips from 379 households. About 6,500 CV trips were logged for about 100 households in the month before and after the EV was used. The sample of households was based on voluntary participation under the condition that the household already owned at least one car and had a dedicated parking space where the EV could be home charged. In the trial period, the household had access to both their CV and EV, but they were encouraged to use the EV as the primary option. The participating households resided in 27 of the 98 municipalities in Denmark and were distributed across the entire country (see Figure 1). For trial participation purposes, one household member filled an online application form with information about the household and its composition. Each trip has been merged with weather information from local weather stations, inducing that information about precipitation, wind speed, temperature and visibility at the time of departure is available. The NAVTEQ network consists of 636,243 links covering the entire country and all road classes from large highways to minor local roads.

Figure 1: Spatial distribution of observations. Amount of observations (EV or CV) on links


Table 1 presents the parameters of the preliminary joint path size correction logit model. The model is going to be enriched with information about congestion and reliability on the large-scale network, weather and socioeconomic characteristics. As congestion has not been included in the model in the current stage, the presented model focuses on off-peak trips (i.e., between 9am and 3pm and between 6pm and 7am). A total of 45,516 EV observations and 3,146 CV observations compose the off-peak sample. The model shows that the parameters are significant and with the expected sign. Also, for all attributes the EV parameter has a higher negative value than the CV parameter, of which the highest difference is for the trip distance. One conclusion, therefore, is that EV users are more sensitive to trip disturbances and trip length than CV users, even after taking into account scaling differences.

Table 1: Estimated parameters of the joint path size correction logit model for EV and CV route choice.


Bovy, Piet, Shlomo Bekhor, and Carlo Prato. 2008. “The Factor of Revisited Path Size: Alternative Derivation.” Article. Transportation Research Record: Journal of the Transportation Research Board 2076: 132–40. doi:10.3141/2076-15.

Bradley, Mark A, and Andrew J Daly. 1997. Estimation of Logit Choice Models Using Mixed Stated Preference and Revealed Preference Information. Article. Edited by Peter Stopher and Martin Lee-Gosselin. Understanding Travel Behaviour in an Era of Change. Oxford: Pergamon.

Christensen, Linda, Ole Kveiborg, and Stefan Lindhard Mabit. 2010. “The Market for Electric Vehicles - What Do Potential Users Want.” CONF. In 12th World Conference on Transportation Research.

Golob, Thomas F, and Jane Gould. 1998. “Projecting Use of Electric Vehicles from Household Vehicle Trials.” JOUR. Transportation Research Part B: Methodological 32 (7): 441–54. doi:

Greaves, Stephen, Henry Backman, and Adrian B Ellison. 2014. “An Empirical Assessment of the Feasibility of Battery Electric Vehicles for Day-to-Day Driving.” JOUR. Transportation Research Part A: Policy and Practice 66 (0): 226–37.

Jensen, Anders Fjendbo, and Stefan Lindhard Mabit. 2015. “Modelling Real Choices between Conventional and Electric Cars for Home-Based Journeys.” Unpublished. In The 14th International Conference on Travel Behaviour Research (IATBR), July 19-23, Windsor, United Kingdom.

NAVTEQ. 2010. “NAVSTREETS Street Data Reference Manual v2.8.”

Pearre, Nathaniel S., Willett Kempton, Randall L. Guensler, and Vetri V. Elango. 2011. “Electric Vehicles: How Much Range Is Required for a Day’s Driving?” Transportation Research Part C: Emerging Technologies 19 (6): 1171–84. doi:10.1016/j.trc.2010.12.010.

Prato, Carlo Giacomo, Thomas Kjær Rasmussen, and Otto Anker Nielsen. 2014. “Estimating Value of Congestion and of Reliability from Observation of Route Choice Behavior of Car Drivers.” Article. Transportation Research Record 2412 (2412). U.S. National Research Council Transportation Research Board: 20–27. doi:10.3141/2412-03.


Conference registration is required in order to view papers.