Scraping Google Maps to Measure Electric Vehicle Infrastructure: Real-time EV Charger Data V1.1
In late January 2021, I published a medium article providing a method (with the actual code) for how someone could measure a cities EV infrastructure in real-time (see next link for my first article).
Covering the EV market I felt we were spending a lot of time projecting vehicle sales and not enough time analysing the current reality. Data sets are sparse and I was reading many reports stating that charging stations are a barrier to the EV transition, but little analysis on
- where EV chargers are actually located,
- where chargers should be located,
- and where EV chargers are being added.
Using the Google Map Places API, Plotly Chart Studios, and a python notebook, I mapped out the EV chargers currently existing in my home country, Hong Kong.
Last week I re-ran my code.
Now I’m able to show you the change in EV infrastructure in HK between January 2021 and June 2022.
The graph demonstrates the increase in chargers between 2021 and 2022.
While it was fun to teach myself how to write up animated graphs in python, I’ve also put together another graph which is easier to read and interpret.
Now, the way my codes works means that it is possible that a small amount of chargers were not captured in January 2021, so I will keep the numbers I present as rough estimations.
According to my dataset, Hong Kong saw a ~30% increase in available public EV chargers between 2021 and 2022.
And about a quarter of the new chargers were super/quick chargers.
The new chargers are quite evenly spread out across HK and tend to be located close to large roads.
If I took this analysis further I’d like to look at:
- How often is each charger used?
- What’s the input voltage of each charger?
- What type/brand of cars can use each charger?
- And for what percentage of the population is each charger well located to serve?
If you, the reader, has any ideas on how to approach any of the above questions please feel free to connect with me at my LinkedIn.
Thank you for taking the time to read.
