Challenge 13: Stockout Forecasting
Level: Medium
Description: Claudia, the CEO of a small supermarket chain in the US, is worried about stockouts: a situation that occurs when customer orders exceed the available inventory of an item. She would like to know which warehouses are likely to suffer the most from stockouts, and which types of items are going to be the most problematic: critical, regular, or slow-moving ones. To help Claudia, you decided to create a model to forecast stockouts for the different warehouses and item types, allowing her to interactively check the forecasts. What places and item types are more vulnerable to stockouts?
Beginner-friendly objectives: 1. Read a historical dataset on stockouts for all warehouses and item types and filter it by location "All" and item type "Regular Items". 2. Partition the filtered data, train a forecasting model (you decide which one!), and evaluate the quality of the forecasts. 3. Visualize the forecast results against the real ones.
Intermediate-friendly objectives: 1. Make the selection of location and item type more flexible with widgets, creating an interactive data app that makes visualizations more interactive. 2. Some combinations of warehouse location and item type lead to very few samples, which can turn training a forecast model infeasible. Add error handling techniques to make sure that the data app always executes without errors, and inform users if the number of samples for a certain combination is too small.
Author: Aline Bessa
Dataset: Stockout Data on KNIME Community Hub
Remember to upload your solution with tag JKISeason4-13 to your public space on KNIME Community Hub. To increase the visibility of your solution, also post it to this challenge thread on KNIME Forum.