POUL.TRY

See our pitch deck at https://drive.google.com/drive/folders/1WBwquJrO0hPiZcpXMYnaZObM7DsNwCio?usp=drive_link

Inspiration

In Canada, the government controls the production of chicken. Chicken farmers own a right and an obligation to produce a certain amount of chicken; however, producing outside of their allotment results in harsh penalties. Farmers regulate production through diligent control of the chickens' feed; nonetheless, this process is prone to error and time-consuming. We can change this: by leveraging data reporting systems already present in most barns, we can predict chicken weights and feed consumption to save farmers tens of thousands of dollars annually.

What it does

POUL.TRY scrapes data from barn and bin systems used by farmers, compares the data with the chicken marketing regulations and guidelines provided by the Canadian government, and outputs corresponding reports and recommendations for farmers to modify their chicken diet and feeding schedule.

How we built it

Backend: We scraped the data for chicken weights and farm feed, cleaned up data, and used Pandas to create a series of data corresponding to each crop of chicken. Lastly, used Scikit-learn to fit a model against the data. Frontend: We brainstormed fonts and colour palettes, then utilized Figma to design the webpage, buttons, and text boxes. We focused on simplicity, considering the use of contrasting colours (for colour blindness) and graphics that represent the data. Lastly, we used Next.js to develop the web app's framework.

Challenges we ran into

Connecting the backend and front end.

Accomplishments that we're proud of

We learned how to use Next.js

What we learned

We learned how to use components and creating instances in Figma that allowed for a more dynamic prototype. Figma's auto layout feature was also a new thing we learnt that let us create a more responsive prototype. We were able to learn how to use Next.js for front end.

What's next for POUL.TRY

We could report more statistics like feed conversion rates, expand the platform to accommodate data from more barn systems, and partner with barn data reporters to directly access the data through an API rather than scraping.