Forecast Daily Demand for an Apparel Supply-Chain Team
Overview
What this challenge is about.
You forecast 14-day demand for 500 SKUs across 200 stores using SARIMA, LightGBM, and a TFT model. Get a verifiable certificate.
The scenario
The retailer (publicly listed, ~5,000 stores worldwide) loses about EUR 80M per year to mid-season markdowns; a 10 percent MASE improvement on the top-500-SKU panel would materially reduce that.
The Brief
What you'll do, and what you'll demonstrate.
Pick the demand-forecasting approach that gives the best MASE improvement over the SARIMA baseline at acceptable training cost.
Earning criteria — what you'll demonstrate
- Apply SARIMA, tree-based, and neural forecasters to real retail data
- Evaluate forecasts with rolling-origin folds (not random split)
- Quantify bias and error by forecast horizon
- Communicate a forecasting recommendation to a supply-chain audience
Program Fit
Where this fits in your program.
Sharpens the same skills your degree expects you to demonstrate.
Aligned coursework coming soon.
Skills
Skills you'll demonstrate.
Each one shows up on your verified credential.
- Time Series Forecasting
Apply time series forecasting to solve real industry problems and demonstrate production-level capability.
- Sarima
Apply sarima to solve real industry problems and demonstrate production-level capability.
- Gradient Boosting
Apply gradient boosting to solve real industry problems and demonstrate production-level capability.
- Neural Forecasting
Apply neural forecasting to solve real industry problems and demonstrate production-level capability.
- Rolling Evaluation
Apply rolling evaluation to solve real industry problems and demonstrate production-level capability.
- Python
Write clean, efficient Python for data processing, automation, and backend services.
Careers
Career paths this challenge builds toward
Completing this challenge demonstrates skills that transfer directly to these roles:
Data Scientist
Forecasting at SKU-store scale with rigorous rolling evaluation is the textbook senior data-scientist project at any large retailer.
This challenge sharpens
- time-series-forecasting
- rolling-evaluation
- sarima
Machine Learning Engineer
Comparing classical, tree-based, and neural forecasters across cost and accuracy is the same trade-off MLEs make for production deployment.
This challenge sharpens
- gradient-boosting
- neural-forecasting
- rolling-evaluation
Applied AI Scientist
Translating forecast error analysis into a supply-chain memo is the bread and butter of applied-AI work in operations-heavy retail.
This challenge sharpens
- time-series-forecasting
- neural-forecasting
- sarima