At Recycle Scout AI, we believe that data is the key to creating a greener and cleaner earth. This project calculates the optimal places to invest in new recycling infrastructure to maximize the recycling increases for a given budget.
Our project uses AI to find where investments in recycling have the most impact, which impacts all three project areas. By improving recycling infrastructure, we reduce the amount of plastic pollution that makes its way into our oceans. Driving investment into recycling infrastructure promotes rescuing the earth's resources that have already been used, reducing waste, allowing humanity to live more sustainably. We address the lack of financial resources available to under-developed countries, in two ways: by guiding investment towards them, they will have the financial means to improve their infrastructure, and improve the pollution and garbage that may be a problem in their communities, removing the barrier of the cost to generate the data required to know where to optimally build infrastructure.
The Recycle Scout app works in one main way. The AI developed on the backend of the project is fed data from a number of different databases concerning level of construction/development, Average income, amount of trash produced, and other data sets. All of these data sets are by country, so each country has its own data and recommendations from the AI model.
The website component of this app is primarily used to display the findings of the AI running on the backend. The map seen in the image above can be used to display any of the used datasets on the world map. The main category generated by AI is the ‘Percent Recycled Increase’, which showcases the increase in recycling per country. Countries that will benefit the most are shown in red.
Project Goals
The objective of our project is to generate a list of where to invest in recycling infrastructure, how much to invest, and the impact of that investment.
To do this, we had to create an AI to project the metrics used to determine the impact of recycling, as the data does not exist in many developing countries.
Tech Stack
Web Development: AWS - This is what we used to host our website where anyone can interact with the results of our AI training Flask - The python web-framework we used to build the web-app Languages: Python - The language we used to write the web-app and AI HTML/CSS - Languages used for front end design Libraries: Leaflet/Folium - Libraries used to display an interactive map of our AI generated Data Scikit-learn/Pandas/Numpy - Python Libraries used to train our AI models and generate the data used to calculate the optimal locations to place recycling centers
Data Used Data about the construction cost per m2 from World Population Review found at this link: https://worldpopulationreview.com/country-rankings/construction-cost-index-by-country which was used in the AI model to help find the cost to build a new recycling center.
Log in or sign up for Devpost to join the conversation.