Strategic expansion planning for ElectrifyAmerica's EV charging stations in Virginia, USA.

---

Description

This case study utilizes Python, K-means clustering, Geoencoding API and NaturalEarth Shape file to identify and display optimal locations for future Electrify America charging stations in Virginia. I've integrated Google Cloud Platform's GeoCoding API for precise location display on my plots. The approach offers strategic insights to support Electrify America's expansion efforts, promoting EV adoption in Virginia.

https://hari255.github.io/Unsupervised-Learning/

Data collection

Extratcted this dataset from Kaggle. https://www.kaggle.com/datasets/saketpradhan/electric-and-alternative-fuel-charging-stations

Dataset Description

• After performing data cleading from on the dataset, I've used only few columns that are required to build this model, I've also filtered the data to only include stations in Virginia.
• This dataset include information about various EV charging providers in USA, like Telsa, EVgo, Electrify America etc.
• Dataset consists of 1210 records and 21 columns.

Column	Non-Null Count	Dtype
Fuel Type Code	1210 non-null	object
City	1210 non-null	object
State	1210 non-null	object
ZIP	1210 non-null	object
EV Level2 EVSE Num	1051 non-null	float64
EV DC Fast Count	205 non-null	float64
EV Network	1210 non-null	object
Geocode Status	1210 non-null	object
Latitude	1210 non-null	float64
Longitude	1210 non-null	float64
Date Last Confirmed	1208 non-null	datetime64[ns]
ID	1210 non-null	int64
Updated At	1210 non-null	object
Owner Type Code	617 non-null	object
Open Date	1208 non-null	datetime64[ns]
EV Connector Types	1210 non-null	object
Country	1210 non-null	object
Groups With Access Code (French)	1210 non-null	object
Access Code	1210 non-null	object
Facility Type	562 non-null	object
EV Pricing	579 non-null	object

Natural Earth shape file

import geopandas as gpd
import matplotlib.pyplot as plt
import plotly.express as px

######################################### Load Virginia shapefile   ######################################################

#################### Link to shape file:  https://www.naturalearthdata.com/downloads/110m-cultural-vectors/110m-admin-1-states-provinces/

virginia_gpd = gpd.read_file('C:\\Users\\Harinath\\Downloads\\ne_110m_populated_places\\ne_110m_populated_places.shp')
virginia = pd.read_excel("Virginia_EV.xlsx")


###################################### Plot EV stations in Virginia using plotly ##############################################
fig = px.scatter_mapbox(virginia, lat='Latitude', lon='Longitude', color='EV Network',
                        color_discrete_map={'Electrify America': 'red', 'Other Providers': 'blue'},
                        hover_data={'EV Network': True},
                        mapbox_style='carto-positron', zoom=6, center={'lat': 38.0037, 'lon': -79.4588})
fig.update_layout(title='EV Stations in Virginia', margin={"r": 0, "t": 30, "l": 0, "b": 0})
fig.show()

By using above code, I've created the interactive ploty visualizations, I was able to create that map with the help of shape file downloaded from Natural Earth that enables to hover and view each charging station at deatils about it.

Google Geo Encoding API

################################# Initialize geolocator with Google Maps Geocoding API key #################################

geolocator = GoogleV3(api_key='*************************')

def get_location_details(latitude, longitude):
    location = geolocator.reverse((latitude, longitude), exactly_one=True)
    if location:
        
        ############## Extract facility type from the types field in the geocoding response ##################
        
        facility_type = next(iter(location.raw.get('types', [])), None)
        facility_description = type_mapping.get(facility_type, facility_type)
        return {
            'Location Details': location.address,
            'Facility Type': facility_type
        }
    else:
        return None

The geo-coding API helps us with the address for the locations our model suggested, I've added this to enhance the visualization, when we hover mouse on each dot, it displays information related to that location.

---

Google. (03.24.). Google cloud platform. https://console.cloud.google.com/apis Kaggle. (03.24.). https://www.kaggle.com/datasets