The problem
Over-ordering spoils stock, while under-ordering loses sales.
What to build
Create a forecaster that accounts for seasonality, weather and promotions and recommends order quantities.
Success looks like
Lower waste than a moving-average baseline in back-tests, with forecast intervals shown.
Suggested technologies
- Time-series
- Gradient boosting
- External signals
- Inventory policy
- Back-testing
An original, independent statement. It is not an official Smart India Hackathon problem statement, and Arventiq Labs is not affiliated with or endorsed by that programme. The scenario is illustrative and does not describe any specific organisation. Free to use with credit.