Major project Retail, E-commerce & Consumer Goods Case Studio

Item-level demand forecasting and stock planning

Inspired by publicly reported work at Walmart

Forecast daily demand for thousands of items across stores and states, and turn it into stock decisions.

The brief

Walmart's technology team has described forecasting demand item by item and using it to decide which products to stock where, and Sam's Club has written about a central forecasting service. Build a forecasting service on Walmart's own released sales data: forecast daily demand at item, store and state level, keep the levels consistent, and turn forecasts into safety stock and reorder suggestions. Good looks like: WRMSSE against baselines, reconciled forecasts across the hierarchy, a reorder simulation that shows stock-outs and overstock, and a forecasting API.

Read the case first

What Walmart has said in public about this kind of work, on its own pages:

Make it your own

Plan for a different kind of shop: a kirana chain, a pharmacy chain or a college canteen, where shelf space and cash are tight. Change the cost of a stock-out against overstock to fit that shop and show how the reorder plan changes.

AITeky has no link with Walmart, which has not reviewed or endorsed this brief. The brief only draws on what the company has published, uses open data rather than its data, and is a learning exercise: don't present your project as Walmart's work or system.

What you will build

Explore sales, prices and the event calendar
Build baselines and a global gradient-boosted model
Forecast at item, store and state levels and reconcile them
Evaluate with WRMSSE on a time-based holdout
Simulate reorder points and safety stock from the forecasts
Serve forecasts through an API with a planner dashboard

Free, open data to build it with

It takes a minute with your email. AiTeky Pass (₹999/month or ₹9,999/year) is free for your first 30 days.