Beyond hardware lock-in: How to master software-defined automation for flexible control systems
Michael Kleiner is vice president (VP) of edge AI solutions at OnLogic.
What is the primary focus of software-defined automation (SDA)?
Michael Kleiner, VP of edge AI solutions, OnLogic: The main idea with software-defined automation is to move away from having the control of your automated systems tied directly to specific, one-off pieces of hardware. Instead, you're putting the smarts into the software. This gives you a lot more wiggle room to set things up, tweak them and manage your automation. Ultimately, it's about being more efficient and getting different systems to play nicely together by using software as the main driver and then picking the right hardware for the job based on the particular needs of the project or the environment in which it will live.
What are the primary benefits of software-defined automation?
Michael Kleiner, VP of edge AI solutions, OnLogic: The big wins with software-defined automation are flexibility and being able to quickly change things when your operations evolve. It can also save you money by letting you use more standard computing hardware instead of always needing specialized devices. Plus, scaling up or down your automation becomes much easier. And a huge benefit is better teamwork between different systems, making it simpler to share data and make smarter decisions. Managing and updating everything also becomes a lot less of a headache with software-based control.
How does software-defined automation figure in the convergence of IT and OT?
Michael Kleiner, VP of edge AI solutions, OnLogic: Software-defined automation is a major component of bringing IT and OT closer together. By using software for both control and communication, it helps bridge that traditional gap. It allows your operational tech to integrate much more smoothly with your IT infrastructure, making it easier to get data flowing from the plant floor to the folks doing analysis. This means you can do things like remote monitoring, predict when maintenance is needed and just get a much better overall picture of what's happening, which can help boost efficiency and create opportunities for innovation in your processes and products.
Which standards and protocols will be affected most or increase/decrease in use because of software-defined automation?
Michael Kleiner, VP of edge AI solutions, OnLogic: We're expecting to see more reliance on open communication standards. Things like OPC UA, MQTT and flexible industrial Ethernet solutions are likely to become even more popular because they play well with IT networks. On the flip side, those closed-off communication methods that are tied to specific hardware might become less common as companies look for more open, software-driven ways of doing things. Also, with the need for really reliable, real-time control, standards like time-sensitive networking (TSN) will become more important.
Which components will see the biggest impact from software-defined automation?
Michael Kleiner, VP of edge AI solutions, OnLogic: You'll probably see the biggest changes in traditional hardware controllers, as more of their functions get virtualized in software. Industrial computing platforms and edge devices will become even more crucial because they provide the muscle and adaptability to run these software-defined control systems. Network infrastructure will also need to keep up with more data and real-time communication, pushing the adoption of technologies like TSN and software-defined networking. And, of course, software platforms for managing everything and crunching the data will become central.
In what ways does software-defined automation allow machine builders more flexibility in hardware selection and management?
Michael Kleiner, VP of edge AI solutions, OnLogic: Software-defined automation gives machine builders a lot more freedom when it comes to picking hardware because the control software isn't tied to specific vendors anymore. This means they can choose the hardware that best fits the job and their budget, without getting locked into one company's ecosystem. Managing things also gets easier with software-based setup and updates. Machine builders can often handle and check on their automation systems remotely, making it simpler to scale, fix issues and adapt to new requirements without having to rip and replace a bunch of hardware.
Get your subscription to Control Design’s daily newsletter.
How can machine builders prepare for and leverage software-defined automation?
Michael Kleiner, VP of edge AI solutions, OnLogic: For machine builders looking to get into software-defined automation, a big focus should be on building up their software smarts and system integration skills. This means training teams on the right programming languages, industrial communication methods and virtualization tech. It's also a good idea to embrace open standards and team up with software companies that know industrial automation inside and out. Understanding how edge computing can host these software-defined control systems will be key to designing more adaptable and powerful machines. And getting hands-on experience with pilot projects is always a smart move.
How does software-defined automation build on existing IT and network infrastructure in factories and plants?
Michael Kleiner, VP of edge AI solutions, OnLogic: Software-defined automation is designed to play well with the IT and network infrastructure that's already in place. It uses standard networking protocols and IT best practices, allowing OT systems to connect to regular Ethernet networks using things like TCP/IP. Plus, virtualization techniques that are common in IT can be used to run control software on standard industrial PCs. This means you don't necessarily need completely separate OT networks, and you can apply IT security measures and management tools to the OT side, leading to a more unified and efficient setup.
Tell us about your company’s state-of-the-art product that involves software-defined automation.
Michael Kleiner, VP of edge AI solutions, OnLogic: The backbone of software defined automation is really the computing infrastructure. What we bring to the table is a range of super reliable, high-performance industrial-grade computers—things like our industrial and rugged PCs, panel PCs and edge servers. These are the workhorses that run the software doing all the SDA heavy lifting. They've got the power, the connections, and the toughness needed for automation in a wide range of industrial settings, which aren't often friendly to traditional computer systems, making those adaptable strategies a reality.
