TOP TAKES is IoT Sources’ filtered content channel, bringing you the most important breaking news and notable events surrounding the Internet of Things. Today’s post originated from:

Oracle Corp. has added support for augmented reality, machine vision and digital twins while improving data science capabilities in version 4.0 of its Internet of Things-focused IoT Cloud for Industry, which was announced today.

The IoT Cloud combines platform-as-a-service and software-as-a-service capabilities for industrial companies that want to collect and analyze information from sensors and other connected devices without building out their own infrastructure. The PaaS component competes with General Electric Co.’s Predix platform, which companies can use to gain internal efficiencies and even sell outcomes-as-a-service. SaaS offerings are currently comprised of asset monitoring, with performance monitoring and fleet monitoring in the pipeline, said Atul Mahamuni, vice president of IoT applications at Oracle.

The new digital twin feature enables users to create software replicas of hard assets that can be used for diagnosis, modeling and problem resolution. They can be used to monitor the health of assets and prevent failures before they occur, as well as to run simulations and test scenarios.

Augmented reality combines data with visual displays to give plant managers the ability to view operational metrics and related equipment information in the context of the physical asset. That speeds up troubleshooting and helps with maintenance.

Machine vision uses cameras to conduct non-intrusive visual inspections that detect defects that are invisible to the naked eye. Such systems can also recommend appropriate actions. Oracle’s waveform analysis can also work with audio-only data through extensions provided by the company’s partner network, “We can analyze audio waveforms to diagnose a problem by listening to a motor,” he said. “We can literally drop into a microphone and do sound processing.”

New data science features use artificial intelligence algorithms to continuously analyze things like asset utilization, production yields, inventory levels, fleet performance and potential safety issues. “We take a ‘data scientist in a box’ approach,” Mahamuni said. “You choose parameters and we create a model and test it continuously, evaluating the performance we have coded as logic.”

Predix is a strong platform for companies that need high levels of precision, but Oracle is faster and more interactive, Mahamuni said For example, Predix’s digital twin capabilities are based upon finite element analysis, which is extremely accurate, but “takes hours and hours to run. If you just want to know if you need to do maintenance next week, you don’t need that level of accuracy,” he said.

Oracle’s service is also distinctive because of its integration with the company’s other back-end applications in areas like supply chain, manufacturing, procurement, and fleet management, Mahamuni said. “We use [IoT] insights to optimize the applications,” he said. The company also touts the thousands of members of its partner network, which are continuously adding value to the cloud platform.


You may also like

Leave a comment