Daily demand forecast pipeline in production
Build a demand forecast and run it automatically every day, with data checks, a dashboard and accuracy alerts.
The brief
A forecast is only useful if it arrives every morning and someone notices when it goes wrong. Build the demand forecast, then run it as a scheduled pipeline with data checks, a dashboard of forecasts and actuals, and an alert when accuracy drops. Good looks like: a scheduled, tested pipeline, versioned models, a live dashboard, and alerts that fire on real accuracy drops.
What you will build
- Build and back-test the demand forecast
- Schedule daily data loading and checks
- Register models and run the forecast daily
- Show forecasts against actuals on a dashboard
- Alert when accuracy or data quality drops
- Document how to retrain
Free, open data to build it with
Check each dataset's licence before you publish your work. Your mentor gives your team its own version of the brief (data, scope or region), so every team's project is its own.

