SwedenConsumer goods2020 – ongoingElectrolux
Machine learning forecasts for promotional sales
Electrolux tripled the accuracy of its promotional sales forecasts with machine learning
Problem
Electrolux sells in 155 markets and moves products through more than 400 distribution centers and warehouses. It needed better sales forecasts to support pricing, campaigns and supply chain planning.
Approach
The company built machine learning models for sales forecasting and supply chain optimization on the Databricks platform. It tracked, versioned and shared models with MLflow, and fed them from production data pipelines. A forecasting dashboard gave non-technical staff views on pricing, competition and campaigns.
Results
Takeaway
The accuracy gain applied to promotional sales, where forecasts are hardest to get right. The company did not say how it measured the improvement.
Categories
- Size
- Large enterprise
- Industry
- Consumer goods
- Business function
- Logistics and supply chainMarketing and sales
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and ML
- Adoption
- Off-the-shelf tool or platform
- Goal
- Quality and risk
- Evidence
- Vendor case study
Sources
Source summaries
Databricks describes how Electrolux uses machine learning for sales forecasting and supply chain optimization across 155 markets. The page reports a threefold increase in forecast accuracy for promotional sales. It also gives scale figures such as 400-plus distribution centers and warehouses.
Full details are in the original sources.
