Machine Learning Times
Machine Learning Times
EXCLUSIVE HIGHLIGHTS
SHARE THIS:

1 month ago
Wise Practitioner – Predictive Analytics Interview Series: Paul Turner at Stanley Black and Decker

 

By: Steven Ramirez, Conference Chair, Predictive Analytics World for Industry 4.0

In anticipation of his upcoming presentation at Predictive Analytics World for Industry 4.0 Livestream, May 24-28, 2021, we asked Paul Turner, Vice President I4.0 Applications & Analytics at Stanley Black & Decker, a few questions about their deployment of predictive analytics. Catch a glimpse of his presentation, Beyond Platforms and Enablement – Delivering measurable and scalable impact with advanced analytics in manufacturing, and see what’s in store at the PAW Industry 4.0 conference.

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

A: Our use cases tend to be value focused and fall into categories including Safety, Labor Efficiency, Quality, Asset Performance, Materials & Inventory and Material Flow through the factory. Each of our models provide both insights and potential actions based around one of these value levers. We endeavor to not just predict events and provide insights but to also prescribe a set of actions that would mitigate or resolve the issue that the predictive analytics system has identified.

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 predominantly using predictive analytics to identify what we call leakages and opportunities around the value levers mentioned earlier. Leakages represent a cost or efficiency impact due to under-performance and opportunities represent “golden batch” type performance and learning what makes best-of-the-best performance happen. Predictive analytics alone though won’t drive value. We have had instances where actual scrap sensors are not responded to immediately so predicting an hour ahead is great – but it doesn’t fix the responsiveness of operations. For this we are building an effective way of prioritizing and ranking opportunities and leakages and wrapping this in an over-arching change management program supported by automated workflows, collaboration capability, incentives and escalations. We call this our Digital Continuous Improvement Platform.

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

A: We have not published quantitative benefits of predictive analytics but it is significant. We have one example where we are leveraging AI to predict why a machine has gone down (auto-tagging) so that the operators don’t have to type the downtime cause in. For a single shift over a week an operator was saved 80 screen interactions per machine (and they have 3 machines that they cover) so you can see a substantial labor saving in a factory with over 200 machining assets.

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

A: On one application we had a predictive analytics model for product quality. Low yields were a significant problem and the asset was running at near maximum capacity so was a bottleneck. The predictive analytics model showed that by running the asset at a slightly slower speed the yield increased significantly so ironically we managed to increase production by reducing the speed of the asset.

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

A: Analytics in Manufacturing is unlike any other area of data science. I will demonstrate that model accuracy is not the most important factor in successful projects. I will divulge use cases where a model that is only 5% accurate delivered significantly more business value than one that was 95% accurate – for the same use case; explain why models that are 99% accurate can be completely useless and why some models that are 99.5% accurate are actually detrimental to profit margins when a less accurate one is not.

—————————–

Don’t miss Paul’s presentation, Beyond Platforms and Enablement – Delivering measurable and scalable impact with advanced analytics in manufacturing at PAW Industry 4.0 on Monday, May 24, 2021 from 10:20 AM to 11:05 AM. Click here to register for attendance.

By: Steven Ramirez, Conference Chair, Predictive Analytics World for Industry 4.0

One thought on “Wise Practitioner – Predictive Analytics Interview Series: Paul Turner at Stanley Black and Decker

  1. Pingback: Wise Practitioner – Predictive Analytics Interview Series: Paul Turner at Stanley Black and Decker « Machine Learning Times – The Predictive Analytics Times – IoT – Internet of Things

Leave a Reply