Inspiration
I was inspired to do this project and further explore it due to one of my professor's transportation lectures. I learned that The Government of India launched the National Electric Mobility Mission Plan in 2013, intending to boost demand for electric vehicles (EVs) in India and create a self-sustaining ecosystem. A slew of schemes and programmes aimed at increasing demand and establishing a supply chain in India was implemented. Even though India's electric vehicle programme has been in place for eight years, electric vehicles still account for only about 1% of total vehicle sales.
What it does
This project uses data from Germany and the United Kingdom, where electric cars have already been established, as a case study to show how customers in India could benefit from adopting the same behaviour.
How we built it
Data Cleaning Data Visualization
- Correlation plot using the heatmap
- Scatter Plots to find the relation between various parameters
- Joint plot to analyse the relationship between two different variables
- Bar graph was used to analyse each parameter more carefully.
- Random forest was used as the machine learning model to predict the price
- The accuracy metrics of the predicted value was done using MAE, MSE, RMSE and R-squared.
Challenges we ran into
I was also looking for real-time India data but couldn't find it.
Accomplishments that we're proud of
The accuracy of the predictions was reasonably good, achieved an R-Squared value of 0.815 which shows a high level of correlation.
What we learned
Price in Germany and Price in the UK have a higher correlation. Top Speed has a higher correlation with Price in Germany and Price in the UK. A car with a higher speed has a higher price. KWH and range have a higher correlation.
What's next for Predicting the price of Electric Transportation
I'll be using India's data to create a more efficient supply chain in order to increase profits and electric vehicle penetration in the Indian market!
Built With
- datascience
- machine-learning
- python

Log in or sign up for Devpost to join the conversation.