4 machine control trends

Flattening networks, combining PCs and PLC and self-learning machines are just some of the machine control trends we can expect now and in the near future

By Dave Perkon, technical editor

I always enjoy talking to the many vendors and manufacturers available to control system designers, engineers and programmers. There are so many to choose from. Some represent products from many manufacturers, and others manufacture a wide variety of products. They seem to understand how to help to create a great machine control system, and they have the products to make it happen.

The foundational pillars of a great machine control system are product reliability, performance consistency and ease of integration and support, says Mike Chen, director of the Automation Center at Omron Automation Americas. "These allow manufacturers to guarantee high quality production at an optimum cost,” he explains. “Any hardware, software or services offered need to take these factors into account. The direction that machine control innovation is heading ultimately boils down to these three factors."

The automation and machine control industry is currently flourishing with new innovations and strategies, and it can be difficult to determine what the future will hold, adds Sriram Ramadurai, marketing manager, controllers/components/safety, at Omron Automation Americas. "Machine control in particular is at the heart of numerous new automation strategies because it connects individual machines to the business logistics systems and integrates robotics, vision and other technologies into a comprehensive automation solution," he says. "It also has a significant impact on the security of the overall system.”

Newark is the one-stop source for all the components you need to build a machine control system from sensor information gathering to cloud communications and back, says Danny Weiss, senior product manager at Newark element14. "Because machine control plays such an important role in industrial automation, it’s essential to understand the factors that are driving the evolution of today’s control technology forward," he says. He and his partners at Omron provide some specific trends in industrial automation.

State of Technology Report: Trends in PLCs and PACs

Trend 1—More integration between the control- and enterprise-level networks: "Today, all talk of machine control and automation must include Industrial Internet of Things (IIoT) connectivity," says Weiss. "Traditionally, automation systems have been built according to a pyramid-shaped visualization known as the Purdue Model. This model provides a means for organizing key manufacturing operations into a hierarchy, with the lowest level being the actual physical processes themselves and the highest level being the overarching business logistics systems. In between these two are the intelligent devices, control systems and manufacturing operations systems, with each layer consolidating data from the one below it and feeding the data into the layer above."

This hierarchical model may not be long for this world—that is, the currently blossoming world of the IIoT. "The automation industry is seeing a trend of increased hardware integration between network levels that the Purdue Model designates as separate," says Weiss. "Increased peer-to-peer networking and a flatter approach to manufacturing operations offer the advantage of utilizing fewer hardware pieces but also generate new concerns about security and reliability. It can be hazardous to blindly expose critical process controls to the Internet, so manufacturers seeking to take advantage of more hardware integration are also taking measures to increase security."

Trend 2—Leveraging the benefits of both PCs and PLCs: "Historically, there has been a bit of a back-and-forth regarding the use of PCs versus PLCs," says Weiss. "PLCs are more reliable and more optimized for machine control, but PCs offer greater flexibility and greater processing power. The near future may see a tendency to try to include more complex software functionality, such as algorithms for data analysis, onto PLCs. An example of this idea in action today would be an IPC that has a split-core PLC and PC. This provides all the reliability one expects from an IEC 61131 programmable controller, but it runs in parallel with PC software inside the same piece of hardware and stays immune to Windows OS crashes.

Trend 3—Getting the machines to learn on their own: "Machine learning algorithms are increasingly being leveraged to detect patterns and anomalies in machine function," says Ramadurai. "In some cases, these machine-learning engines run within the machines themselves. It's a trend that in turn is fueling interest in the two trends discussed above. Anomaly detection—the practice of analyzing machine data to recognize outliers relative to normal operation—is an area in which machine learning can be extremely helpful. Whereas traditional periodic maintenance struggles to find the economic optimum between excess service cost versus risk of downtime cost, new predictive maintenance methods can request maintenance tasks in response to trends observed in machine data in real time."

The use of machine learning to improve maintenance and keep an eye on machine functionality poses significant benefits for manufacturing facilities that could suffer excessive amounts of machine downtime when experienced workers leave, continues Ramadurai. "Applying machine learning to maintenance can minimize downtime while new hires are learning the ropes," he says.

Trend 4—Integrated solutions with a single integrated development environment (IDE): "As automation solutions become more capable and more complex, manufacturers are beginning to see great value in implementing integrated solutions from a single supplier," says Weiss. "This strategy dramatically reduces the engineering costs of automation design and integration, and it also makes maintenance much less complicated. In addition, the use of a single IDE for programming all aspects of automation—including sensing, motion, vision, robotics and safety—can cut down on the time required to train operators because they only need to familiarize themselves with a single software interface."

Although there’s a lot happening in the automation industry right now, these four trends seem likely to define machine control for the near future, continues Weiss. "Manufacturers are striving to incorporate more automation into their lines to boost quality and productivity, and this is driving the need for more creative machine control strategies," he says.

In addition, automation is beginning to spread its wings outside the traditional environment of manufacturing and make an impact in various commercial sectors, although at a more tentative pace. "It all comes back to what value you can bring to the customer, and if the juice is worth the squeeze,” says Chen. "It’s an exciting time for this industry, and many new trends, strategies and innovations are likely to pop up in coming years."

ALSO READ: Step up the technology in machine control