Automated laboratories improve uptime with analytics

Predictive service and maintenance keeps Siemens Healthineers lab tests running on time

Laboratory testing of blood and other fluids has evolved from a manual, labor-intensive process to an automated, high-volume process that allows modern labs to test thousands of samples per day. Today’s clinical diagnostic systems optimize workflow and clinical performance by automatically reading labels, moving samples along a conveyer belt, running tests, returning vials to storage and compiling results.

Because laboratories rely on these systems to get results back to doctors, clinicians and researchers as quickly and accurately as possible, any amount of downtime is unacceptable. 

That’s why Siemens Healthineers set out to build predictive service and maintenance capabilities into its Atellica Solution, an advanced system for immunoassay and clinical chemistry analysis. Now, malfunctions are a rare exception for users of the system.

In collaboration with data analytics and data science experts mayato, Siemens Healthineers developed the predictive service and maintenance solution to increase system availability and reduce service costs. The solution is based on SAS advanced analytics.

Clinical laboratory diagnostics with a vision

As a leading medical technology company, Siemens Healthineers provides imaging for diagnosis and therapy, as well as clinical laboratory diagnostics. The Atellica Solution optimizes workflow and clinical performance by transporting samples along a bidirectional magnetic conveyer belt and tracking every sample with a multicamera vision system. It intelligently schedules and manages each sample throughout the testing process.

Using these innovative technologies, the system can perform up to 440 tests per hour, the highest productivity per square meter in the industry.

The predictive service and maintenance component of the system, designed using SAS, helps make sure throughput is uninterrupted and productivity remains as high as possible.

Proactive service with SAS

Early in the development phase, product development engineers at Siemens Healthineers defined what data could be relevant for predictive maintenance. This meant necessary sensors were integrated and the system connected to the Siemens Healthineers database, to enable data processing.

Once the first development phase for the product series was complete, the next step was the dedicated development of the predictive service and maintenance solution. Here, Siemens Healthineers chose SAS software and solutions, including SAS Data Management technologies, SAS Visual Analytics and SAS Enterprise Guide.

Project implementation with mayato

For the conceptual design and implementation, Siemens Healthineers chose SAS partner mayato GmbH, because they had already worked successfully together on earlier projects. The mayato team handled the complete process, including:

• Implementing the data loading points.
• Setting up the predictive service and maintenance framework.
• Preparing the data for reporting.
  

To work efficiently, the team took an iterative approach, where the basic requirements were first determined, and then modified during the project as needed. 

“Because Siemens Healthineers use SAS solutions for data analysis, the cornerstone of the project was already set,” explains mayato Project Lead Paolo Vacilotto. “We were therefore able to focus completely on the sustainable design of the framework, to ensure high performance and comprehensive configuration options.”

Moving from reactive to proactive maintenance

The predictive service and maintenance capabilities inside the Atellica solution are fully automated and integrated into a system that Siemens calls the Guardian Program. As part of this program, country-level service organizations for Siemens Healthineers provide maintenance services and interact with the users.

Torben Scaffidi, Head of Lab Operation Analytics at Siemens Healthineers in Erlangen, Germany, explains: “Laboratory diagnostics systems transport many fluids and comprise many different components. That’s why the service and maintenance concept is highly complex. Our goal was to transform a service that had previously been predominantly reactive into a proactive concept.”

The predictive maintenance data is transmitted to respective country databases via the Siemens Healthineers Smart Remote Service following regional regulatory requirements. The data comes from various sources, including prestructured sensor data or log files.

The data is then transferred into SAS and analyzed using models built by Healthineers data scientists, to predict failure probabilities. The results are processed, visualized and made available via a web-based front end.

Automated warnings by email

To simplify and automate the process further, mayato developed an email alerting service that sends automatic warnings with detailed information as soon as anything requires action.

In addition, the solution uses incoming information to analyze key performance indicators and report results with SAS Visual Analytics. Initial results include:

• 58 system and process components are monitored proactively.
• 36% less system downtime compared to the reactive service.
• Optimized service technician deployment, because they know in advance which parts are needed at a particular location.

A successful project

Siemens Healthineers started with a goal of building predictive service and maintenance directly into its solutions. SAS solutions provided an ideal way to integrate analytics into the product and scale as the system changes. And, finally, mayato’s structured implementation ensured tremendous flexibility.

The solution revolves around a continuous expansion of analytical data models, providing a modular structure and intuitive framework so the Siemens Healthineers team can make modifications on their own.

“We enjoy working with the mayato team,” says Scaffidi. “We value the competence and reliability of the consultants. Especially in such a dynamic environment, it’s essential to remain flexible, without losing sight of the structure – and we succeeded again in this project."