25 Jul 2024


Amazon makes creating digital twins ‘faster and easier’

Amazon says that its new service will make it faster and easier to create digital twins

Amazon Web Services (AWS) has announced a service that, it says, will make it faster and easier for anyone to create digital twins of real-world systems such as factories, industrial equipment and production lines. Digital twins are virtual representations of physical systems that use real-world data to mimic the structure, state and behaviour of the objects they represent, and can be updated with new data as conditions change.

AWS’ IoT TwinMaker makes it easy to integrate data from sources such as sensors and business applications, and combines the data to create a knowledge graph that models the real world. Amazon says it will allow many more people to use digital twins to build applications that mirror real-world systems and can improve operating efficiencies and cut downtime. There are no upfront fees, and users pay only to access the data used to build and operate the digital twins.

“Sensors for equipment, buildings, and industrial processes are proliferating and generating massive amounts of data,” says Michael MacKenzie, general manager for IoT at AWS. “Customers are increasingly eager to use that data to optimise their operations and processes and one way to do that is using digital twins, but they find that building a digital twin and custom applications is difficult, time-consuming, and prohibitively expensive to maintain.

“With AWS IoT TwinMaker,” he adds, “customers can now derive previously unavailable insights about their operations that inform real-time improvements to their buildings, factories, industrial equipment and production lines, and make accurate predictions about system behaviour with minimal effort.”

The service allows users to gather data from a variety of sources, including Siemens MindSphere cloud platform. It creates a knowledge graph that combines and understands the relationships of the connected data sources, and can update the digital twin with real-time information from the system being modelled. Users can import existing 3D models such as CAD files, to create 3D visualisations of the physical system and to overlay the data from the knowledge graph onto the visualisations to create the digital twins.

Once the digital twin has been created, a Web-based application can be generated that displays the twin on devices that plant operators and maintenance engineers use to monitor and inspect their plants. Developers can then set up rules to alert the operators if anomalies are detected and display them on a 3D representation of the plant, helping the operators to make quick decisions on predictive maintenance before equipment fails.

Amazon Web ServicesTwitter  Facebook