Grundfos and Affecto harness the powers of IoT, Big Data, Predictive Analytics and cloud services with innovative pilot project. The goal: Explore and showcase the potential of closing data gaps and further optimize Grundfos’ production of the world’s leading circulator pumps

Even the most successful, industry-leading companies of the world are constantly seeking new ways of leveraging technology to optimize production. At Grundfos – one of the world’s leading manufacturers of circulator pumps – a successful digitalization pilot project has uncovered the benefits of bringing IoT and Big Data Analytics together in industrial production.
“Collecting data from our production lines is not a new concept for us. But we see exponential growth in the volume of generated data while our current IT systems are not capable of handling that amount of data. So we set out to explore other platforms and approaches to determine in which direction to move forward,” explains Kristian Kragh, Manager Production IT at Grundfos.
The company already knew Affecto from an earlier implementation of BI tools for Grundfos’ Business Intelligence team. With Affecto also specializing in digitalization, IoT and Big Data Analytics solutions tailored to manufacturing environments, Grundfos decided to bring Affecto on board once again.


“Our scope was to identify a suitable area in production, define and execute a pilot project around it and, finally, evaluate what worked and what didn’t. It turned out that we already had the perfect guinea pig: A QA project in our silicon wafer production that needed a different approach,” Kristian Kragh says.
Together, Grundfos and Affecto designed an innovative solution that encompasses deployment of IoT sensors and collection of production data from multiple data sources such as MES, PLCs, test files and user data.
Data are enabled and transformed into actionable information by Microsoft Azure cloud services such as Stream Analytics, Power BI and Lake Analytics.
The solution also contains a platform for machine learning and compelling data visuali-zation with dashboards readily available on computers and mobile devices anytime, anywhere.


Bringing IoT and Big Data Analytics to the production line via a robust and scalable cloud platform offers vast business potential. Optimized production with higher quality and minimized waste are among the obvious benefits, while Predictive Analytics and machine learning will add even more value by transforming traditional production into a truly intelligent operation.
A remarkable – and very tangible – example from the pilot is that Predictive Analytics achieved 89 % accuracy in predicting silicon wafer test failure. The ability to, at an early stage, predict failure in a component that is later to be enclosed in a circulator pump a long way down the production line, can lead to tremendous cost savings.


“We are very satisfied with the results. The scope was to explore and showcase the potential and to inspire the entire Grundfos business. Together with Affecto, we have achieved just that. We have caught the attention from all corners of the organization with complete buy-in to the concept from everyone, from process owners to executive management,” says Kristian Kragh.
Grundfos and Affecto will continue to work together, exploring and identifying new IoT/Big Data opportunities and define an implementation strategy and roadmap.


  • Pilot project: Industrial Big Data and Digitalization in Production with Predictive Analytics
  • Collect data from multiple sources: MES, sensor data, test files, PLC data, user data, inventory reference data etc.
  • Introduce affordable and easily deployable “Rapid Deployment IoT Kit”
  • Enable store, analyze and visualize data via Microsoft Azure application stack i.e. Power BI, Lake Analytics etc.


  • Successful proof of concept with complete buy-in to concept from entire organization, from process owners to executive management
  • Inspirational showcase for future IoT/Big Data projects
  • Innovative approach to overcome production challenges faster and more efficiently
  • 89 % accuracy in test failure prediction in wafer production = significant potential cost savings
  • Flexible and scalable platform enables pay-as-you-grow approach
  • Project goals reached in time and on budget
  • Strategy and roadmap for next steps and full scale implementation