About Client:
Bluenet creates an advanced Market Basket Analytics platform for a large retailer to analyze best selling SKU combinations A large furniture retailer with more than 100,000 SKUs in the inventory.
Problem Statement:
They need to be able to identify the best selling SKU combinations in order to minimize inventory cost and maximize promotion performance. Stores also need the data for better merchandising.
Challenges:
- Very large volume of SKU pair combinations to be analyzed
- Traditional database systems were insufficient to handle data volume
- The analysis has to be unique for each store
Solution:
- Created unique market basket analysis for each store to analyze best selling SKU combinations for each category and subcategory
- Created unique market basket analysis across categories and sub-categories
- Used large volume in-memory key-value-pair databases to handle the very large volumes of data combinations
- Better promotion planning and merchandising resulted in increased revenue and decreased inventory costs
Tool Stack: IBM DB2, IBM Infosphere, Redis , IBM Cognos