The Fifth Elephant 2024 Annual Conference (12th &13th July)

Outline
AB InBev sells a significant share of its beer volume through retailers who take AB InBev’s assistance to configure the best shelf assortment that would maximize their revenues. Planograming is a very important step that retailers perform to allocate their available shelf space to the right products. For large retailers that have 1000s of stores, this becomes a very time consuming and tedious exercise.

Using advanced Machine Learning models and GenAI, we have developed an Allocation Engine that creates a 3D Digital Twin of an actual shelf, performs auto-allocation of products, does space optimization, and self-adjusts to the dynamic business rules that different retailers use in different countries. We model all the scenarios here before going live.

During the session, we will bring you close to how we designed the Allocation Engine and how we are using GenAI together with a custom-built Context Module.

Who is the audience for your session? Machine Learning Engineers, Software Engineers, Data Scientists, Product Managers – who are architecting commercial data science products.

What problem/pain are you trying to solve (for the audience)? Making planograming faster and adaptive to business inputs from retailer.

How will participants benefit from your session? You’d gain understanding on how GenAI and ML can create Digital Twins that can help significant efficiencies to the overall business.

Impact

1. Using the Allocation Engine, time to create planograms has come down from 3-4 days to under 20 minutes (on an average).
2. Business users can run multiple scenarios with the 3D Digital Twin and compare results before go-live
3. Instead of manually adjusting products on a planogram, business users feed their rules into the Allocation Engine in plain English/Spanish/Portuguese.