Machine Learning Times
Machine Learning Times
Three Best Practices for Unilever’s Global Analytics Initiatives
    This article from Morgan Vawter, Global Vice...
Getting Machine Learning Projects from Idea to Execution
 Originally published in Harvard Business Review Machine learning might...
Eric Siegel on Bloomberg Businessweek
  Listen to Eric Siegel, former Columbia University Professor,...
Effective Machine Learning Needs Leadership — Not AI Hype
 Originally published in BigThink, Feb 12, 2024.  Excerpted from The...

2 years ago
Wise Practitioner – Predictive Analytics Interview Series: Oscar Porto and Fábio Ferraretto at DHAUZ


In anticipation of their upcoming presentation at Predictive Analytics World Industry 4.0, Las Vegas, June 19-24, 2022, we asked Oscar Porto, Founding Partner at DHAUZ, and Fábio Ferraretto, Business Analytics Senior Executive at DHAUZ, a few questions about their deployment of predictive analytics. Catch a glimpse of their presentation, Stockout Predictions Using Machine Learning in a Building Material Multinational Company and see what’s in store at the Predictive Analytics World Industry 4.0 conference.

Q: In your work with predictive analytics, what behavior or outcome do your models predict?

A: A predictive algorithm for identifying out-of-stock risks for each SKU/CD combination for the next 7 and 14 days for the Brazilian operation of Saint-Gobain.

Saint-Gobain is a company that designs, manufactures and distributes materials and services for the construction and industrial markets and is present in 75 countries.

Q: How does predictive analytics deliver value at your organization – what is one specific way in which it actively drives decisions or operations?

A: We are an Advanced Analytics company that works on all our projects with an integrated vision of Data Science, Business Processes and Technology knowledge.

Q: Can you describe a quantitative result, such as the predictive lift of your model or the ROI of an analytics initiative?

A: +80% accuracy and 0.7 F1 for main SKUs and major CDs. When we compare the model result with stock-based business rules, for example, we have a lift of 1-(0.82 / 0.6) = 36% accuracy. In terms of F1: 1 – (0.67 / 0.45) = 48%.

Q: What surprising discovery or insight have you unearthed in your data?

A: Out-of-stock predictions on SKUs that weren’t monitored so closely.

Q: Sneak preview: Please tell us a take-away that you will provide during your talk at Predictive Analytics World.

A: Predictive Analytics is able to solve the problem of forecasting out-of-stock situations, generating visibility for the next 14 days, allowing the company to act in order to minimize them and, consequently, generating additional revenues.

Don’t miss Oscar and Fábio’s presentation, Stockout Predictions Using Machine Learning in a Building Material Multinational Company at Predictive Analytics World Industry 4.0, Wednesday, June 22, 2022, from 2:15 PM to 3:00 PM. Click here to register for attendance.

Leave a Reply