Electric Vehicles in Finland (EVFI)

Introduction

EVFI started as a project for DATA11001 course at University of Helsinki. The idea was to study the electric vehicle (more specifically battery electric vehicles) adoption in Finland using data science methods.

Data

Data source: Tax Administration

Tax information of Finnish municipalities and regions
License: CC BY 4.0

Data source: Traficom

Information about the vehicles in Finland: Dataset 1, Dataset 2
License: CC BY 4.0

Data source: National Land Survey of Finland

Vector maps of Finnish administrative areas
License: CC BY 4.0

Data source: Statistics Finland

Names of Finnish municipalities and regions: Finnish, Swedish, English
Sociodemographics of Finnish municipalities and regions
License: CC BY 4.0

Data source: Sähköautoilijat ry

Information about public chargers: Latauskartta

Findings

The blog post that was the course deliverable can be found here.

Future findings will be published in the blog under the 'evfi' category.

Installation

The following instructions assume that you have Python 3 installed on your computer.

Clone the repository and enter the project directory:

git clone https://github.com/aurala/evfi
cd evfi

Create and activate a Python virtual environment:

python3 -m venv venv
source venv/bin/activate

Install the requirements:

pip install -r requirements.txt

Download the datasets and unzip in the project directory:

Now open up your favourite Jupyter notebook application, choose the virtual environment as the kernel and enjoy data crunching!

Contributing

I will continue working on this project and I'm open for collaboration. If you're interested in electric vehicles, data science or both, feel free to contribute. Here are some ideas how to work together:

Fork the 'evfi' repository on GitHub, do your data science magic and send a pull request.
Drop a comment below (or in the GitHub project) if you have a question to be answered. Or drop me a private message.
Download the datasets and analyze them on your own. (In case you share something publicly, don't forget to attribute the data sources!)

Some things that I have in mind for the future:

Discovering/validating the features that make the machine learning model work
Full utilization of the income data
Analyzing other vehicle types (buses, trucks, etc.)
Using new open data sources
- Fingrid: Price of electricity - how does it affect the EV adoption?
- Traficom: EV makes and models - how are the distributed geographically?
- Statistics Finland: Political orientation of the municipalities - does it affect the EV adoption?
Comparisons to other countries
Building a Streamlit (or similar) application for visualizing the data