Altage – Electric Vehicle Charging Analysis

Inspiration

According to the EPA, motor vehicles collectively cause 75 percent of carbon monoxide pollution in the United States. Hawai'i is among the top 10 electric vehicle markets in America. There are currently more than 10,000 electric vehicles. However, there are only 16 operating EV charging stations, 9 of which are located on Oahu. The lack of EV charging stations discourages consumers from converting to an electric vehicle. While the number of cars on the road increases, the air quality decreases with it. Altage wants to contribute to making the local, and eventually global, society more sustainable by creating a product that incentivizes consumers to think sustainable and purchase an electric vehicle instead of a gasoline car. Therefore, we want to contribute to the future by providing organizations such as Hawaiian Electric with a tool to analyze, manage and visually depict processed data to users working with EV charging stations to better understand the collected data after charging cycles and to predict anomalies based on the data given.

What it does

• Predicts anomalies in EV Charging Stations to report to Hawaiian Electric.
• Displays data based on the following reports: Car charging station health, congestion, and energy usage.
• Detects congestion at EV Charging Stations A and B if the data shows that there are more than three consecutive sessions with 7-minute intervals in between each session.

How we built it

We used R to pre-process the data according to Hawaiian Electric's requirements. Furthermore, we used Salesforce to display the data with the use of visually pleasing graphs and charts.

Challenges we ran into

• Data limits between Salesforce and R
• Getting familiar with Salesforce within the short timeframe
• Invalid data points provided by Hawaiian Electric

Method and Tools Used

• SCRUM Agile Methodology
• Pivotal Tracker
• RStudio
• Salesforce

What's next for Altage?

We intend to further develop the application by implementing Machine Learning into the product as a way for the users to predict data anomalies and congestion. Thereby, users will be able to take action early to prevent future issues. Furthermore, the front-end application can be worked with to include more functionalities and details.

Built With

• r
• salesforce