Case study

Machine learning-based Sell-In Forecasting for Consumer Electronics

Business Context

The client, a Korean MNC that designs, manufactures and markets Home Appliances such as Refrigerators, Washers, and Microwaves and operates as over 100 subsidiaries globally. With over 2000 products and a channel-focused Supply Chain planning approach, our Client wanted accurate Supply Chain Forecast for optimal product-availability within 8-week lead-times. In CPG, which is highly promotion driven, competitive and seasonal, this could make or break a business.

The client’s Supply Chain Management software, the algorithms embedded within the Planning Systems produced an FA which were unable to pick up crucial patterns in demand and their relationship to causal factors. With a target of a 10% Accuracy Value-Add, it was imperative to explore new-age forecasting methods based on Machine Learning.

How BRIDGEi2i delivered value

BRIDGEi2i’s Forecasting Engine, with over 6 years of R&D, has 32 forecasting algorithms that have been tuned and refined to specifically forecast demand by reading into patterns beyond just trends, seasonality and cycles. It employed specific type of algorithms to forecast demand for specific types of products and the “headroom” for FA improvement for each segment. Our IP algorithm called Trust Index is a mechanism to have the Planner trust the Machine Learning forecasts and build familiarity with it.

Our Forecasting Engine is a Proven Tool to Improve Visibility

Want to know how we improved the accuracy?

Download the case study!

Related Posts