1660244520461 Autofair2017

New data combinations offer deeper insights

Nov. 14, 2017
Frank Kulaszewicz of Rockwell Automation cites data orchestration, mixed reality as key digital transformation investments in Automation Perspectives keynote

"We're on a mission to combine our understanding of market needs, our technology and our domain expertise to make a powerful industrial analytics platform. Powerful, but easy to use." Rockwell Automation’s Frank Kulaszewicz discussed extensions and enhancements to the company’s scalable analytics offering.

At the Automation Perspectives media event this week in Houston, Frank Kulaszewicz, senior vice president, Architecture & Software Division, Rockwell Automation, charted the progress the company is making to improve digital design and operations while placing greater analytics power in the hands of end users.

"Connectivity and access to information in new and existing assets is really important if we're going to achieve that vision of smart manufacturing," Kulaszewicz said during the media event, a day before the company’s annual Automation Fair. "What we're trying to do is create a real-time production environment that's optimized and sustainable," one that improves ease of use for customers across industry, and is resilient in the face of changing conditions.

Digital business outcomes

"Today, everything we have is becoming connected," said Kulaszewicz, from phones and watches to cars to homes, each of which can provide information about itself and when it needs service. "The same thing is happening in industry. It's the same technologies—mobility, wireless, cloud—and they're all making their way into production. Devices on the plant floor are becoming smarter and better connected."

The challenge is to draw data out of isolation in these assets and devices, and into a platform that enables plant teams to derive new and actionable operational insights.

"Much of manufacturing today is still legacy systems, and many of those legacy systems aren't connected," Kulaszewicz observed. "Because of those systems, industry experiences about $20 billion of downtime every year, so there is a significant opportunity there to gain better visibility into plant operations."

Adding to the challenge is that industry has not made uniform progress along its digital manufacturing journey. However, as more digital pilots are run, giving industry a gradual, better understanding of what digital means, Kulaszewicz sees value being generated in long-term revenue and profitability; new connected systems and business models, including new service models for OEMs; and sustainable productivity.

"All of those understandings are driving customers to move ahead with programs to achieve their digital business outcomes," Kulaszewicz said. "And for us, it's driving an understanding of where we should make our investments in the digital space."

The promise of mixed reality

The first of these digital investments is to develop real-time, open, analytics platforms that scale across plant network architectures and which encompass the entire production lifecycle from design to operations.

"Digital transformation enables a digital thread," or trail of information to be collected throughout the entire enterprise, one that captures information throughout the lifecycle of a product, asset, system or process, said Kulaszewicz. "It allows seamless collection and analysis of data, and the ability to gain information from that data."

Ultimately, digital threads combine to enable development of "digital twins" for physical products, assets and processes—emulations that allow products to be tested before they are built, allow production changes to be tested before they're happening, and allows design and performance to be optimized during the production process.

These digital twins increasingly span both design and operations, said Kulaszewicz. In the area of digital design, collaborative design environments are really important to help teams bring better products to market faster, he said. "The environments simplify automation and information design efforts to integrate disparate tools and workflows, and allow complete systems to be simulated."

To this end, Kulaszewicz announced that the Studio 5000 environment from Rockwell Automation offers three new capabilities:

  • Interface to other tools like EPLAN and other CAD systems
  • Improved connectivity for emulation environment so users can do operator training with other systems
  • OPC-UA open unified connectivity for all Rockwell Automation software products

In the area of digital operations, Kulaszewicz stated, "This phase of the life cycle typically lasts a decade or more, which represents a huge opportunity to impact profitability" across three areas:

  • Information management—connectivity, data collection and data orchestration
  • Operations—traditional functions like work orders, production goals and quality assurance
  • Asset reliability—keeping production running in the form of maintenance, troubleshooting, and technical support

One result of the ongoing digitalization of manufacturing is the ability to generate "mixed reality" environments, where project teams can interact virtually with their designs during the design process. "You can validate the design concepts long before they go to production," said Kulaszewicz. "Our machine builder libraries help OEMs move away from traditional methods of programming to modern ways of creating automation systems. The designer shifts from writing code to proven libraries of machine functions that help them design their systems—more drag-and-drop than it is programming."

In the process of working with customers on digital pilot programs, Kulaszewicz said Rockwell Automation is learning which outcomes have the greatest impacts, and is helping to define the "killer apps" that will have the best return on digital investment. These outcomes can be defined in three value streams:

  • Asset management and reliability (i.e., predictive maintenance and prescriptive maintenance)
  • Operational productivity (i.e., OEE and production throughout)
  • Risk management (i.e., safety systems, security, track & trace)

Each of these three value streams represents learnings with customers across industry, via digital programs that deliver the right information at the right time to enable actions that improve enterprise performance.

The power of data orchestration

The second area of digital investment Kulaszewicz covered was a deeper ability to orchestrate disparate data sets to drive stronger operational insights in real time. These data sets range from asset-specific condition monitoring and asset health data to external supply chain and weather/environmental data.

"We're on a mission to combine our understanding of market needs, our technology and our domain expertise, to make a powerful industrial analytics platform,” said Kulaszewicz. “Powerful, but easy to use."

The investment is in a real-time, open, scalable analytics platform that allows plant teams to scale across the entire information architecture—the device, the machine, the production line, the plant and the cloud. This allows decisions to be made closest to the source of the data, where they can have the biggest impact on results.

Leading this investment are a new set of capabilities for the Rockwell Automation FactoryTalk Analytics Platform that focus on new possibilities in data orchestration.

Existing between the twin poles of data ingress (i.e., collection) and data staging (i.e., dashboards and reporting), data orchestration can be defined as the space where end users compare, contrast, re-contextualize and otherwise mash-up disparate data sets, based on plant-specific and project-specific needs.

Rockwell Automation has been exploring these technologies under the collective heading of Project Scio, an initiative dedicated to helping plant teams merge new and existing data to achieve useful, actionable information. At Automation Perspectives, Kulaszewicz was joined by Andrew Ellis, manager of commercial engineering, Rockwell Automation, to review progress made by Project Scio and to run a project demo.

Ellis emphasized that Project Scio is designed to help organizations move from more descriptive- and diagnostics-based analytics into the realm of predictive and prescriptive analytics. Ellis cited data orchestration as the most common barrier for organizations that are trying to move to more advanced analytics tools.

"Advanced analytics is the world of machine learning and artificial intelligence, and is providing game-changing production benefits for manufacturers," said Ellis, adding that the challenge for manufacturers in this area is not a lack of data. "The challenge to achieving advanced analytics is being able to process that data, cleanse it, enhance it, then move it into a tool that can be used to provide insights to customers."

Ultimately, the goal is for platforms like Project Scio to "leverage model-based algorithms to predict future events, and have those events fed back into the control systems in real time," said Ellis, to prevent downtime and sidestep quality issues.

By delivering these capabilities, added Kulaszewicz, Project Scio also is helping to reduce the process of cleaning data, which in many cases consumes 60% of a data scientist's time.

The demo showcased the ability of Project Scio to ingest disparate data sets into an elastic search database, then develop data staging areas based on different mash-ups at the data orchestration layer. One example was the combination of operations data from a fictitious billet manufacturer with lab data on billet chemistry, to more precisely control billet quality throughout the day, across changing environmental conditions like temperature and humidity.

Kulaszewicz and Ellis also demonstrated how Project Scio develops an initial dashboard of KPIs that the platform thinks will be useful to the operator, based on prior use. "Most folks in industry are not data scientists," said Kulaszewicz, "and that audience will be crucial for adoption of these products" where the visualization environments become intelligent.

Plant teams also are free to engage in "active dashboarding" with the Project Scio platform, exploring different custom data visualizations. To enable this, Project Scio features intelligent search, an elastic capability where users can use semantic terms (in English) to explore new analytics and insights. 

Sponsored Recommendations

Measurement instrumentation for improving hydrogen storage and transport

Hydrogen provides a decarbonization opportunity. Learn more about maximizing the potential of hydrogen.

Get Hands-On Training in Emerson's Interactive Plant Environment

Enhance the training experience and increase retention by training hands-on in Emerson's Interactive Plant Environment. Build skills here so you have them where and when it matters...

Learn About: Micro Motion™ 4700 Config I/O Coriolis Transmitter

An Advanced Transmitter that Expands Connectivity

Learn about: Micro Motion G-Series Coriolis Flow and Density Meters

The Micro Motion G-Series is designed to help you access the benefits of Coriolis technology even when available space is limited.