-
Welcome page in which you can choose the house you wish to see
-
First half of green report of the chosen house
-
Second half of green report of the chosen house
-
First part of predictions page in which you can see the annual data for electricity and predicted electricity range
-
Second part of predictions page in which you can see the annual data for water and predicted water usage range
-
Last part of predictions page in which you can see the annual data for green scores and the predicted green score range
-
Leaderboard in which you can see the top 25 villas for that month and their green scores, as well as your rank for that month
-
Update profile before updating
-
Update profile after updating
I am generally passionate about sustainability, environment and wildlife. As I thought about this hackathon, I wanted to build something around sustainability around my community (where I live). As I started thinking about ideas, a recent event at our home triggered this project. We started getting unreasonably high bills for water that triggered us to get a smart water meter installed. Now, since we get monthly water and electricity bills via our community app (ApnaComplex), the idea was to create something which can remind community residents of their past and present energy usage and motivate them to save total energy consumption.
I created this Green Citizen Report as a tool that currently works for one community of 100+ houses. The project is fairly extensible to multiple communities or even cities. In its current form, it is comprised of usage report for the current month for a given house, its past usage (by month), a predicted usage (for the next month). It will also show you a comparison of your utilities usage compared to your neighbors/ communities average usage in graphs. It also shows the houses green score, which I have calculated keeping in mind the various factors.
The green score is my idea to combine several factors together. For example, I wanted one score to represent their collective usage of electricity, water and if they use electric vehicles etc. I also wanted to normalize this score based on per resident rather than per household. The current score range is 0 to 100, a higher score is better. I also created buckets based on the score ranging from bad, satisfactory to great or excellent.
In the predictions tab, you can see a graph showing the actual, annual electricity usage, actual annual water consumption and actual annual green score by month. Underneath the corresponding graph is the villas predicted range for their next month’s electricity usage, water consumption and green score. This feature is to help the family try to get better than the predicted range so that they see their trend and can improve on it.
In the update tab, the user gets the option to update their details, letting them update the number of people in their house and the number of vehicles. This feature can come in handy if you buy a new car, or a new family moves in to the house.
There is also leaderboard in which you can see the top 25 green scores of the community for the month. You will see the house number, the month and the green score. This feature is to trigger healthy competition between people to get a better green score so that they see their house on that board.
I built this using Python and in Jupyterlab. I also used multiple libraries to assist me such as Pandas and Seaborn.
I am proud of the fact that I was able to complete it in 48 hours by myself. I am also proud of my data visualizations and predictions because until now I only did very basic visualizations and projects so this hackathon allowed me to learn much more and do much more in this field. I am also proud of coming up with a green score that not only works well for this data but is extensible to any magnitude of data.
I learnt many things such as how to predict future data with data that you have. I learnt many different techniques and tricks in UI. I learnt how to create formulas to calculate the green score. I also learnt different ways to stylize graphs when displaying them such as how to make the bars different colors according to their values.
There were multiple challenges I faced during this project. First, while I have done some elementary graphs through Python, this is a first data visualization project I undertook. So, I had to think about how best to represent data visually. Second, I had to work on the mathematical aspects to normalize data and predict data. It seems straightforward, but I had to invent ways. For example, the very idea of one single metric such as “green score” needs a lot of thought as to how to combine electricity and water usage which are on different scales. Third, I wanted to do something which is possible to code in 48 hours but at the same is a real world problem that can scale to multiple communities later. Finally, coding this UI was a real challenge, may be it is because of the tools I chose to code the UI such as tkinter and Seaborn. I chose them as I was already familiar with them.
I am going to take this project further by expanding it to more housing communities and more cities. I will back it up with a database so more data can be stored and I will also convert it in to an app so that it is more accessible to users.
Log in or sign up for Devpost to join the conversation.