Inspiration
In high-traffic airport environments, businesses like restaurants and retail stores often face challenges in predicting peak periods, managing staff, and minimizing waste.
What it does
How we built it
Built using Python, Cloudscraper, Pandas and Streamlit, RestoPort combines predictive analytics and data visualization to deliver actionable insights in a user-friendly interface. Deployment was done using Streamlits very easy web deployment and hosting.
Challenges we ran into
Unconventional API request to fetch data from airport website. The api service did not seem to follow conventional RESTapi methods. Traditional multilayered architecture approach does not work when working with Streamlit, you must lean into the Streamlit development architecture.
Accomplishments that we're proud of
The clean look of the UI in such short time frame. The efficiency of the data collection and live updates.
What we learned
There is always a way to get what you want most times without having to reinvent the wheel. Lean into the benefits of using mature libraries with python.
What's next for RestoPort
Making the sales prediction metric no longer BETA:
- Implementing a data pipeline to start creating a database of flights parameter
- Training a ML model using supervised learning how to predict sales using flights parameters
- Working with stores and restaurants to collect more data and try to predict inventory needs.
Log in or sign up for Devpost to join the conversation.