Case Study

Medical device manufacturer


Project overview

Medical device manufacturer

  • 10,000 employees
  • Operations in more than 30 countries
  • 30 production lines globally
  • $ 10 billion in annual revenue
Rapidly building custom AI applications for manufacturing
A large medical device manufacturer and distributor delivers thousands of units across 30 countries each year.

As a medical device manufacturer, you cannot think of quality in simple economic terms; meeting customer expectations can be a matter of life or death. The costs associated with quality take them apart from other manufacturing companies, as regulatory compliance costs are the most significant component.

According to McKinsey's research, quality costs generally equate to 2-2.5% of sales. Still, it is estimated to be only a tiny portion of the potential liability if all costs of poor quality are evaluated. The liabilities associated with the direct costs of poor quality highlights what can go wrong despite all of the resources focused on prevention and compliance. Hence, “quality” in the medical device world is a two-part equation: assurance and control.

Struggling to unify disparate data meant to measure the quality of its products and identify potential outliers, the company decided to partner with Alteia. With the Alteia platform, they have access to an end-to-end infrastructure for data capture, storage, and analysis. Using video streams, they can leverage AI technologies to identify potential quality issues and take the necessary actions, or perform root cause analysis more efficiently.



gain in efficiency


reduction in defects

Alteia Platform

Project highlights

  • Capture video of every device assembled, at each stage, down a manual assembly line
  • Store video footage on the Alteia platform and easily search and retrieve footage for viewing
  • Turn video into data with AI models to apply analytics and provide complete measurements of manual assembly activity
  • Merge video data with production insights in order to conduct predictive failure assessment

Next Customer