What Is Adivarent Control?
Published: July 8, 2020
The Purdue Model of ISA-95 has been widely accepted across process industries as the gold standard for the layout and architecture of process control & enterprise integration since the mid-1990s. However, the technology and computing power available to the process industry has undergone what can only be described as a seismic shift since then – shouldn’t the model for process control have changed by now too? Enter Adivarent Control.
At Griffin Open Systems, our AI Toolkit is the backbone of what we call Adivarent Control, a concept that allows process industries to leverage artificial intelligence (AI) to streamline the data flow between the plant supervisory (level 2) and production control (level 3) levels. Because of this, Adivarent Control enables both control operators & control systems to operate with increased efficiency – and these efficiencies can increase throughput, decrease unplanned downtime, and ultimately optimize processes.
So what exactly is Adivarent Control? We like to say it serves as a real-time assistant to control operators and control systems. But, it’s perhaps easier to dive into what exactly it is and what it isn’t …
Adivarent Control is built on a reliable, open, and customizable AI platform and features a no-code graphical programming interface to enable control operators and engineers to swiftly test, iterate upon, and deploy new process models – all without having to dig into the code & logic in the control system. Because of this, it allows tacit knowledge from control operators to be institutionalized and scaled in a much more agile manner. This allows process improvements & optimizations to be realized rapidly, leading to quicker results on the bottom-line.
Further, it is a closed-loop system that integrates AI at the juncture of the plant supervisory and production control levels. Unlike other systems that implement AI at the production control level as advisor tools for operators, or embed lower AI capabilities into the control layer, Adivarent Control bridges that gap and allows AI to work with greater complexity, often integrating discrete control elements.
More important than what Adivarent Control is, might be what it isn’t.
Adivarent Control is not meant to be a large CAPEX investment in hardware replacing your existing control system or your control operators. Because of its open architecture and vendor agnosticism, it is also not a complex “black box” system that beholds users to specific hardware, I/Os, or control systems. Lastly, it is not a monolithic, pre-packaged solution – industrial processes vary, and as such systems to control & optimize them should be flexible to account for that.
Want to learn more about how to leverage AI to break-down and improve the flow of data & control between levels in the Purdue Model? Contact us today to learn the roadmap to implementing Adivarent Control.