ServSmart helps small restaurants cut food waste with ML-powered daily customer forecasts, tailored from day one using business info and refined with their own food usage data.
Inspiration: We saw a problem in our local towns where many small businesses were suffering with revenue because of the excess food waste that was caused by the lack of technology to realize how many people will actually come to the shops.
What it does: We use a machine learning model that predicts the amount of customers that will show up to the shop based on a wide array of factors ranging from weather, day of the week, etc.
How we built it: We used a React based application for the frontend with a Flask and JSON-Server for the backend. For the model, we used Python along with many libraries such as Tensorflow, Numpy, Pandas, etc.
Challenges we ran into: Finding suitable APIs for weather and coordinate data proved to be a bit difficult. We also ran into some time constraints especially nearing the end of the hackathon. We ran into a bit of an issue regarding the Flask server which was solved with ample time remaining.
Accomplishments that we're proud of: We successfully created a working website using frontend and backend. We created a working Machine Learning model that accurately predicted what we wanted.
What we learned: We learned how to delegate tasks among our group in order to have the most efficient and effective outcome.
What's next for ServSmart: We would like to expand this to more than just small businesses. We would like to add automated inventory in the future so that it would be more helpful for the businesses to realize what items they actually need and in what quantities.
Log in or sign up for Devpost to join the conversation.