Thank you for contacting us!

How Mars Wrigley used Quantum ML to mitigate risk and reduce commodity acquisition costs.

Problem Statement

As a global organization with over 150 business units and upwards of $5 billion spend on commodities, Mars Wrigley® is one of the largest and most diversified consumer packaged goods companies in the world. It’s size and variety of products makes it uniquely vulnerable to a wide spectrum of global market and business risks. Our over-arching goal with Quantum ML is to mitigate these risks by leveraging Big Data, AI, and Machine Learning to predict commodity prices, improve market timing, and thereby reduce acquisition costs and improve their bottom line.

Solution

The solution design and process can be organized into four categories.

1. Data - input and target variable data.
2. Modeling - algorithm and feature selection, hyper-parameter tuning, and model training.
3. Strategy Engineering - buying strategy design, backtesting, performance evaluation.
4. Strategy Implementation - real-world evaluation, productionization and change management.

(1) Data

‍

(2) Modeling

(3) Strategy Engineering

(4) Strategy Implementation

Outcome

• Reduced commodity acquisition cost by 3-18%
• 20% improvement in operating margin
• Normalizes and improves buyer performance.