Munchi.AI

Munchi.AI

1. Inspiration

One small pebble can cause a ripple to echo across the pond. The same can be seen in food waste, when a person throws food away, it builds up, leading to destructive effects such as an increase in greenhouse gas emissions, economic loss, an increase in landfills, and biodiversity loss. More often than not, food is thrown away due to oversupply and unpredictable demand in major food waste contributors like supply chains and cruises. Thus, Munchi.AI was designed to stop this wasteful behavior and decrease the behavior's detrimental effects on Earth.

2. What Does It Do

Overall, Munchi.AI aims to create an accurate prediction of how much food should be prepared and ordered in fast food chains/cruise establishments and food supply chains using reinforced learning to reduce food waste. To start, Munchi.AI goes through several weeks of an "observation period", where the program is fed information that best helps Munchi.AI make food predictions. The type of data gathered may vary depending on which industry is selected. For instance, in a restaurant, establishment owners will record the mass of food prepared initially and what they had after rush hour. As time passed, the results were more popular. More in-depth information about the observation period is available in the Google Sites link embedded.

3. How We Built It

We built our project demo on Google Slides as a rough representation of how our AI would look like. Through buttons that automatically guide the user to the correct slide, our demo successfully displays what Munchi.AI would do if it were an actual program that utilized AI-reinforced learning. Furthermore, our demo is embedded in a Google Sites website that expands on the app's uses and how it functions.

4. Challenges We Ran Into

One of our biggest challenges was understanding how to utilize our different skill sets so that our ideas could be presented in their best form. We all come from different backgrounds and differ in what we know. Although piecing together new information to strengthen our idea was tough, working together lessened the burden and made the overall experience more fun.

5. Accomplishments We Are Proud of

We're proud to create a project proposal that successfully resolves a global issue while taking advantage of the limited resources available. Furthermore, we are proud to be able to overcome our differences and work efficiently as a team.

6. What We Learned

Within our journey, we learned more about the different models of AI, such as neural networks, and how many industries use AI today. We also expanded our knowledge of how supply chains operated and connected our program to resolve the food chains' unpredictability. Furthermore, we learned how to communicate with each other, so one person's weakness could be covered by another's strength.

7. What's Next for Munchi.AI

In the future, we would like to complete Munchi.AI with an AI program and turn our demo on Google Slides into reality. Once the reinforced learning model is trained, we would like to use the program in situations similar to the food chain problems we are attempting to solve. Once again, we strive for tomorrow, by starting with today's wastes.