Customer success story
Edwards transforms predictive maintenance for semiconductor manufacturers
Customer
Challenges
- Legacy data infrastructure: Edwards had been collecting data since the 1990s using a homegrown SCADA system, but increasing the sampling rate for predictive analytics began to reveal bottlenecks in their data storage system.
- Predictive maintenance goals: The aim was to help customers avoid costly vacuum pump failures by forecasting potential issues before they occurred—reducing downtime, avoiding premature replacements, and improving service efficiency.
- Focus on core expertise: Edwards wanted to work with a partner that could provide a scalable platform, allowing their teams to focus on areas where they hold domain-specific expertise.
- High cost of failure: A single vacuum pump failure can scrap over 100 wafers, resulting in potential losses of up to $2 million per year.
- Premature replacements: Customers often replaced pumps too early to avoid failure, adding unnecessary cost and inefficiency.
- Requalification complexity: Breaking vacuum for unplanned maintenance requires a time-consuming requalification process, increasing the need for accurate predictive maintenance.
Outcomes
-
Scalable predictive maintenance: Cumulocity enabled Edwards to deploy predictive models across thousands of assets, supporting large data volumes and advanced analytics.
-
Increased focus on innovation: By offloading data infrastructure, Edwards’ engineering team could concentrate on their vacuum systems expertise and customer-specific features.
-
Faster, more accurate failure prediction: Integrated Streaming Analytics and DataHub helped improve prediction accuracy by combining real-time and historical data.
-
Stronger customer engagement: Real-time insights demonstrated the value of predictive maintenance, building trust and credibility with customers.
-
Faster proposal turnaround: Edwards could present real-world performance proposals more quickly, accelerating customer engagement and decision-making.
-
Improved resource allocation: With Cumulocity managing infrastructure, Edwards could reallocate engineering efforts toward improving pump data extraction and semiconductor-specific visualizations.
We could see that we needed a scalable system that will be able to handle ever larger quantities of data, and we also knew that a partner with specific expertise in this field would be beneficial so we can work on Predictive Modelling and using the data in a way only a domain expert can.
Paul Johnson
Senior Manager, Digital Transformation
Edwards
Details
Edwards had long been collecting product data, but when they began applying it to predictive maintenance, the increased sampling rate exposed limitations in their legacy data system. To move forward, they needed a scalable solution—and a partner who would allow them to focus on areas where their domain expertise could drive value.
Want to learn more about how AIoT is transforming industry?
Learn more about how organizations like yours are using AIoT to improve operations, speed decision-making, and introduce new business models to their customers.
After evaluating their options, Edwards selected Cumulocity to support their predictive maintenance strategy. The platform provided the infrastructure to ingest and analyze large volumes of data in real time, allowing Edwards to build and deploy machine learning models at scale—without the overhead of managing the data stack themselves.
One key advantage was the ability to use historical data to demonstrate the system’s effectiveness. In one case, Edwards showed how their model would have predicted a vacuum pump failure that resulted in the loss of 50 wafers—highlighting how predictive insights could prevent costly downtime.
Cumulocity has freed us up to focus on areas where we require domain expertise to drive value, and spend less of our time on data management solutions, which we don’t aspire to be experts in. We are happy with the results, and are now starting to spend more time looking at the roadmap Cumulocity has, and using that to drive us in directions we wouldn’t have thought of. ** — Paul Johnson, Senior Manager - Digital Transformation, Edwards**
Want to learn more about how AIoT is transforming industry?
Learn more about how organizations like yours are using AIoT to improve operations, speed decision-making, and introduce new business models to their customers.