Scalable analytics for IIoT

As connectivity and computing power continues to spread throughout industrial enterprises, the availability of data has revolutionised how companies solve issues and adapt to changes. But producers continue to be challenged by the complexity of making data useful at the right place and time. They also often lack in-house expertise for data management from the device through to the cloud.

Industrial producers cannot be expected to rip and replace all their legacy control and information systems before gaining value from analytics. These scalable and open-architecture capabilities are designed to be extended to a full ecosystem of IIoT data sources. The quick connection to the full range of systems that feed data into a Connected Enterprise includes controllers, MES software and edge devices.

Rockwell Automation has a stated goal of enabling advanced analytics for manufacturing. Project Scio is the next step along that continuum. To make decisions when and where they matter most, new capabilities offered through Project Scio open up access to ad-hoc analytics and performs advanced analysis by pulling structured and unstructured data from virtually any existing source. Project Scio can intelligently fuse related data, delivering analytics in intuitive dashboards – called storyboards – that users can share and view.

“Providing analytics at all levels of the enterprise – on the edge, on-premises or in the cloud – helps users have the ability to gain insights not possible before,” says John Genovesi, vice president of Information Software, Rockwell Automation. “When users gain the ability to fuse multiple data sources and add machine learning, their systems could become more predictive and intelligent. Scio puts analytics to work for everyone. By its addition to the FactoryTalk Analytics Platform, Project Scio gives users secure access to all data sources, structured or unstructured.”

The features of Scio

- Device Auto-Discovery: manually mapping software to each plant-floor device can be a time-consuming and error-prone process. Project Scio can automatically identify Rockwell Automation devices and tags, as well as third-party device data. Additionally, this so-called "auto-discovery" process gives users access to more detailed information than is typically available through manual mapping, such as device name, line location and plant location.

- Centralised data: rather than leave data at its source and take database snapshots, Project Scio brings data into a centralised location and can continually refresh that data. Additionally, connections to data sources only need to be established once. This connection allows users to create custom analytics and refresh them at their preferred rate without the support of a data scientist.

- Flexible machine learning (ML): Project Scio is configurable to support many industry-leading algorithms, including SparkML, MLLib and Python.

Closed-loop analytics: using either ML or pre-defined settings, Project Scio includes capabilities that can monitor operations and automatically trigger control adjustments if processes start to fall outside allowable parameters. This can help users optimise control, improve product quality and consistency, and reduce scrap and waste.

Rockwell Automation plans to develop an “applications marketplace” for in-house and third party applications. Users can also take advantage of Rockwell’s pre-engineered FactoryTalk Analytics applications to monitor common KPIs, such as OEE and quality.

Rockwell Automation continues to collaborate with Microsoft on information projects that can help industrial companies predict performance, boost productivity and maximise asset availability.

On the Microsoft stand, Rockwell Automation demonstrated FactoryTalk Analytics scalability and asset performance management capabilities on a Bradman Lake end-load cartoner machine. Visitors learnt how to use scalable analytics and scalable computing to monitor production, quality and maintenance KPIs. They could also put on a Microsoft HoloLens mixed-reality headset to see detailed machine information and diagnostics at the source, using ThinManager software.