Data Intelligence driving vehicle electrification forward

According to the World Health Organization (WHO), air pollution kills an estimated 7 million people worldwide every year. Transportations are a major cause of air pollution. Pollution is an important problem in the major cities all around the world, electric vehicles (EV) are the next step in mobility evolution. Governments are applying laws trying to reduce pollution with restrictions to combustions vehicles, e.g: Accessing London Low Emission Zone costs £12.50 per vehicle per day. But can we adopt electric vehicles now? What if data science could help understand the suitability of electric vehicles for fleets?

Geotab is already helping in road safety with its OBD device and it’s aggregated datasets. After joining with Fleetcarma, Geotab won experience in electric vehicle and is focused on helping the early EV adoption offering their customers an “Electric Vehicle Suitability Assessment”.

Geotab ranked 1st in ABI’s (www.abiresearch.com) “Commercial Telematics Competitive Assessment” collects data from more than 1.5 million connected devices, collecting more than 4 billion data points per day. That data points include from GPS coordinates and speed to hazardous driving, engine measurements and faults, seat belt usage, fuel, etc. With that data Geotab is leading in Smart cities and Urban mobility insights and applications, some of them are collected in open datasets in data.geotab.com. Mike Branch, VP of Data and Analytics explained some of these datasets in Big Data Spain 2017.

Electric charging points distribution is inconsistent in some areas and is making difficult for particular users to acquire an electric car, but most fleets have warehouses or garages where they can store and maintain their vehicles, having less problems to start adopting electric vehicles.

Geotab devices are connected to any type of ground transportation fleet; commercial and rental car among others. With a total of approximately 1.4 billion cars in the world, around 25% of them are commercial vehicles, not including normal vehicles used for passenger transportation or different purposes on fleets. Also, fleet vehicles normally drive more kilometers per day than private owned vehicles and carry more weight, having a huge impact on the pollution generated by transportation. Helping fleets to adopt EVs would generate savings in fuel, maintenance and pollution taxes for them and will reduce drastically the pollution generated by vehicle transportation.

In this presentation Geotab will show the results of adapting the EV suitability assessment in a big data way; analysing one year of driving data from over 200.000 vehicles. A machine learning model capable of identifying a vehicle vocation and VIN (chassis number) decoding will be used to identify the suitable vehicles for the analysis from all Geotab fleet. From those vehicles the key data to be analysed is the fuel consumption and trips information. With the trips it’s possible to understand the patterns in daily distances for a fleet and if they use to store the cars in the same facility overnight. With fuel consumption it is easy to calculate the savings if a vehicle is substituted for an EV. From all this data, EV mileage and fuel and vehicles costs we can extract the features to elaborate a model capable of identifying the suitable vehicles to be substituted by EVs or even consider plug in hybrid vehicles.

The results will be presented in an aggregated and anonymized respecting Geotab customers privacy. There will be an analysis by country, vehicle type and vocation. The main goal is to study the suitability of EVs as for today and raise awareness of its benefits for economics and health. Secondary goal is to understand the minimum electric infrastructure needed, in case that todays suitable vehicle number is low.