At the first Engerati Meet on EVs in the Grid, several participants expressed the view that mobility is moving away from being product dominated and with the growth of electric and autonomous vehicles, is moving towards becoming more on-demand and service oriented.
Such a model brings new challenges to vehicle management for both operators and providers such as utilities, for example in ensuring the availability of vehicles with sufficient charge to undertake the required journeys and thereafter if necessary, to reach a charging station and then charge within the scheduled time.
While autonomous vehicles still have to prove themselves as safe and overcome the likely consumer antipathy, EVs are fast growing and such challenges are already starting to be experienced. In a presentation at the Meet, Alistair Clarke, CEO and Founder of the UK mobility consultancy EV Technology, discussed findings from over four years of managing London’s first all-electric taxi fleet.
“To move EVs from early adoption to the mass market, we need a deeper understanding of the benefits,” says Clarke.
EV Technology has built a platform running machine learning algorithms with the goal to investigate what Clarke calls the “four ‘Rs’ of EV management” and to provide visibility to overcome them for market participants such as fleet operators and utilities.
“These are the route optimised for EVs, the range based on predictive battery discharge, the reservation of charge points and remittances for payment,” he says, pointing to the availability of some 1.4m data points collected from telematics devices over more than 400,000km of real-world driving.
“On a city-wide basis, when we know the whereabouts of vehicles when they need to charge, then it is possible to start optimising the where and when of the use of that electricity.”
Taxi EV case study
Referencing the EV taxi use case, Clarke says a key issue is to address the range anxiety which is resulting in poor vehicle utilisation as well as limiting the uptake of EVs as taxis.
“Typical on-demand providers such as Uber allocate a car to a particular customer based on the proximity, which is fine assuming unlimited range,” he points out. “Bring batteries into the equation and there’s a problem because the parameters around limited range aren’t taken into account.”
Data from the EV taxis indicates that this issue is being manifested in sub-optimal charging practices, with the peak in charging station visits seen at battery levels of 30%-40%. Moreover, optimal charging has a current value approximating to £100 per week for the driver.
“We saw that the top driver achieved an average 7.8km/kWh, a third more the worst at 5.9km/kWh,” says Clarke. “This corresponds to range differences of 230km versus 175 km. Add to that the average battery utilisation – ranging between 91% and 38% - and the effective range for the worst driver was 93km.”
In another example, Clarke points to two shifts of the same driver. In the first, the driver recharged the vehicle four times at a minimum charge of 48% and had spent one-third of the shift travelling to charge points and just 28% carrying fares to make an income of £109. In the second shift, with two charges at a minimum charge of 11% and 27% of the shift travelling to charge points, the driver was able to spend 52% on the fares and grossed £162.
And with eight fares in each of the shifts, the latter were clearly of higher value.
“This shows the effect of managing the vehicle optimally and by maximising the km/kWh and minimising the trips to the charger, EV charge management is a key to profitability.”