On Friday we chat with Benson Hougland during the Base Camp Digital webinar “The Mobil Serv Approach to Prescriptive Maintenance and New Technology for Our Trusty Toolboxes: Accelerate Automation while Boosting Your Scalability, Security and Performance.” Today we preview his presentation and pick his brain on automation adoption, old tech and poll-response vs. publish-subscribe data communications.
Check it out…
Smart Industry: How are current conditions accelerating the adoption of automation?
Benson: As if digital transformation weren’t pushing the automation industry forward already, every week we are seeing new reports on using automation to implement COVID-19 safety guidelines, like digital twinning, smart vision systems, and so on, all so that we can keep people safe at work. It’s just another example of the pressing need for industrial technologies that move and work with data efficiently, and to quickly solve problems as they arise.
Smart Industry: How does dated technology limit full use of IIoT approaches to manufacturing?
Benson: Legacy operations technologies weren’t designed for the level of connectivity that IIoT systems demand. They can inhibit network performance and growth for a variety of reasons, including lack of security and lack of IT-compatibility. These factors also increase the complexity and cost of IIoT systems due to maintenance and licensing of traditional middleware systems. For scalability at the level demanded by the IIoT, we have to achieve a much lower overall cost per I/O point or field device and simplify the typical work-breakdown for integration projects.
Smart Industry: What is one example of the differences between poll-response and publish-subscribe data communications?
Benson: The most concrete metric comes from Cirrus Link Solutions, co-inventors of the MQTT communications protocol and inventors of the Sparkplug B payload specification. They report an 80-90% decrease in data bandwidth consumption by facilities that convert from traditional poll-response communications. That’s due in part to MQTT’s efficient-payload format, but also because, in a publish-subscribe model, data consumers (like OPC servers or HMI/SCADA software) are not required to request updates constantly from data producers (like PLCs or field I/O systems). Instead, MQTT allows these data producers to report tag updates on-change only, significantly reducing the back-and-forth traffic normally required, while maintaining system-wide state awareness. And since MQTT uses a brokered publish-subscribe model, data producers only need to publish their data once in order for all data consumers to be updated.
