Inspiration
Having experienced encounters where local grocery store employees threw away unsold products in large quantities, we were inspired to change the impacts of these actions. These products are good for consumption, yet are sent to landfills for waste because of an oversupply. Through research, we discovered that these wasted products produce large amounts of methane which contributes to global warming, and that grocery stores lose hundreds of billions of dollars in profit every year to this overconsumption.
What it does
EcoSupply is an AI-oriented model that uses sales data and economic indicators to suggest the optimal amount of a product a store should stock. It works by linking itself with the store’s inventory management system, recording data automatically when a product is sold. The retailer can then see the specific data and EcoSupply’s suggestion on how much product should be restocked after a set-up sales period. By incorporating our system, stores can reduce the amount of leftover goods, which is not only more profitable for them but also beneficial to the environment.
How we built it
We created an AI model that was trained using reinforcement, unsupervised, and semi-supervised methods. We began by inputting the relevant data and by training the AI until it began to put out the correct outcomes to a preset situation, we created a model that can accurately predict how much grocery stores should stock of a given item.
Challenges we ran into
When creating our website, we ran into issues creating the page where users can input their data to be run by our AI model. We managed to connect the data input tables to the CMS to resemble what the process of data input would be like for our clients.
Accomplishments that we're proud of
We are extremely proud of our work in how we were able to discover a potential solution to oversupply, benefitting the future of the world environmentally and economically.
What we learned
During the process of constructing EcoSupply, we directly saw how much retailers are wasting, and learned how oversupplying creates a great environmental threat. We also learned how strong modern AI models are, how they are trained, and how they can be utilized to solve problems that would otherwise be impossible for humans.
What's next for EcoSupply
After creating our business EcoSupply, we believe that we can utilize our software and apply it to many other parts of the world outside of the United States. This can significantly reduce waste in all parts of the world, contributing to making our planet less wasteful by minimizing excess consumption and production internationally.
Built With
- capcut
- google-docs
- javascript
- wix
Log in or sign up for Devpost to join the conversation.