Inspiration
We're a team of data scientists and given the chance to help Sauce Bro's analyze their data and make projections was a exciting challenge.
What it does
Klarity helps restaurants predict sales and product mix to optimize staffing and inventory ordering. By doing this, restaurants can make data-drive decisions about how many staff they need, based on the ML's predictions. This helps restaurants reduce waste, and increase profits.
How we built it
Klarity uses a Time series model called SARIMAX. We tried and tested the data with various machine learning models like linear regression, SARIMAX, Prophet and neural networks
Challenges we ran into
The data that was provided generalizes the product mix into yearly and we would have given much better results with the daily prduct mix recommendation for inventory and supplies if we have information about the product mix per day.
Accomplishments that we're proud of
We're proud that we were able to make a model that was between 80-95% accurate within 24 hours, and display it on a intuitive modern-looking dashboard.
What we learned
We learned that net sales was highly seasonal and day-dependent. Furthermore, accuracy was able to be increased once macro-economic data + weather was incorporated into the model. This project was challenging to get done, with many roadblocks. But we learned a lot, specifically in using data to make business - decisions, and working with actual owners like Redwan to craft product scope and requirements.
What's next for Klarity
We hope to continue working on Klarity outside of HackAI, to help restaurants similar to Sauce Bro's project their sales. This is a real problem, and we're excited to take Klarity as far as we can. Thank you!
Log in or sign up for Devpost to join the conversation.