Using Weather Patterns to Reduce Food Waste

Inspiration

We asked ourselves what tools we could use to help eliminate the problem of food waste plaguing North American retailers, and considered weather patterns to be an exciting potential solution.

What it does

Compares historical sales figures of products to historical weather data in the region, and derives formulas to make sales projections for the near future using relationships with factors like temperature and precipitation.

How we built it

We built a simple GUI in Java that allows the user to input sales data as a file. This data is compared to historical weather data from Environment Canada, and linear regression is applied to determine correlation. Using the Weather Network API we determine the forecast for the next week and determine projected sales, then display the information to the user.

Challenges we ran into

Neither of us had experience using APIs, so it was definitely a learning process on how to incorporate the Weather Network.

Accomplishments that we're proud of

It's our first Hackathon and we managed to finish a product!

What we learned

What's next for Sales Forecasting Using Weather Patterns

Developing more complicated algorithms to find relationships with more obscure weather types.

Built With

• java
• weather-network-api