Industrial Internet of Things deals in big stuff — big companies, big assets, big analytic tasks. Thus, building a user-friendly platform to serve this market poses challenges, number one of which is scale. How will companies perform analytics on large numbers of physical assets in need of digital representation? Machine learning and virtualized replicas can help, according to one executive at General Electric.

Hima Mukkamala, head of Engineering and senior executive of Predix at GE Digital, spoke to Jeff Frick (@JeffFrick), host of theCUBE*, from the SiliconANGLE Media team, during the GE Minds + Machines event about what Predix’s software layers has in store for industrial companies’ analytics needs.

Mukkmala explained how important digital twins are to IIoT and to the Predix platform. “At the end of the day, that’s the bread and butter of the platform,” he said. “It’s a combination of the virtual model of the asset, the analytics that can operate on the asset, and the data that’s coming from the asset, so that you can make some outcomes, right?” he said.

“Just like if you go to Google and say, ‘Find me the closest restaurant that serves Indian food,’ you should be able to ask that digital twin, ‘When was the last time a particular asset, a locomotive of the same type, failed, so that I can do predictive maintenance on it?'” he explained.

He added that machine learning is a key enabler of the technology. “Machine learning, which is a big part of the digital twin, is about scale,” he said.

*Disclosure: GE and other companies sponsor some GE Minds + Machines segments on SiliconANGLE Media’s theCUBE. Neither GE nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.

Watch the complete video interview below, and be sure to check out more of SiliconANGLE and theCUBE’s coverage of GE Minds + Machines.

Photo by SiliconANGLE