Is O-PAS another, better mousetrap?

The Open Process Automation Standard (O-PAS, bit.ly/OPASstandard) provides an open, standards-based and secure architecture for hardware- and software-platform independence. Its value and intent are that it reduces total cost of ownership (TCO), accelerates innovation, and increases flexibility for industrial process control. 

Interoperability (permitting diverse systems or components to work together for a specified purpose) is step one, and requires independent testing to gain market confidence. It appears from its value proposition that O-PAS wishes to go further, and include interchangeability—the ability of one product, process or service to be used in place of another to fulfill the same requirements—which is at least an order of magnitude more difficult.

Likewise, hardware independence isn’t the same as vendor independence. Control system suppliers provide more than hardware and systems; they have many decades of intellectual property experience, and that’s what you really pay for when you partner with them. This is also why we typically see control system suppliers either have systems tailored to specific industries or focus on different industries and markets.

Another example of vertical industry alignment with some crossover to O-PAS is the fieldbus technology and protocols developed in the 1990s. I believe there are many commonalities in the challenges O-PAS faces that were also experienced 30 years ago. 

Fieldbus and similar technologies aim for interoperability, and have their various checkmarks to demonstrate that individual products will work together. Even so, the adoption rate for fieldbus technologies was approximately 10 years after their first products transitioned from concepts defined by the standards and first prototypes to industry acceptance as part of project implementations, which happened in the early 2000s for greenfield installations. 

The 10-year gap for  new process industry adoption typically begins with five years in pilot plants or projects to work out wrinkles and confirm associated, necessary design processes. This is followed by another five years for implementation in a facility, and five more years before full commissioning.

Another factor impacting adoption of any new technology is that, in the hydrocarbon industry, upstream projects have an asset life of 20-25 years. Meanwhile, downstream adoption is 25-plus years if you want to introduce new technology, in most cases greenfield projects, or those depending on some interoperability and expansion. The Achilles’ heel of fieldbus was, even though it reused existing cable, it required changes at both ends of the wire. Changing an I/O card is relatively easy, but it also requires changing every field device. So, despite being a wonderful technology, it didn’t gain as widespread adoption as everyone envisioned. It’s another example of a better mousetrap that didn’t achieve expectations.

Get your subscription to Control's tri-weekly newsletter.

The Open Process Automation Forum (OPAF, bit.ly/OPASforum) started as a home for O-PAS in 2016, so it’s coming up on its first decade. Based on the 10-year technology adoption timeline, we should see several O-PAS plants being designed. However, I’m not aware of any such projects, which makes me think O-PAS is at or nearing a similar crossroads of maybe being another, better mousetrap. There may not be as many process plants being built today for which O-PAS is designed, so that’s a factor. However, expecting an existing facility to replace what is, in their minds, a functioning system that’s worked for decades with new technology is a hard sell. They may replace a field junction box with remote I/O for increased signal density, but that doesn’t require a wholesale technology change because most control system suppliers have a remote I/O product. There are also quite a few control system-agnostic I/O products available, including several that are part of O-PAS.

O-PAS, I suspect, is also victim of engineering think—an obviously good idea, but like the better mousetrap, not always about following the market, particularly the automation suppliers they depend on for ongoing support. Similar to many engineers, we see the obvious benefit that answers the “what’s in it for me?” question for themselves (i.e. operating companies), but not necessarily for other participants in the process.

In conversations with Dick Morley, an angel investor in automation startups, he told me there’s never a shortage of wonderful ideas (better mousetraps)’ However, what differentiates which ones get investment and a chance to go to the next level is their ability to clearly answer benefit all the participants in their target market, and that means marketing.

I believe OPAF and O-PAS have lots of great ideas, and many of them have been incorporated into many control system suppliers’ offerings. Remote I/O and virtualization are, in my mind, two of the main ideas that are now commonly used. To me, the concept with the biggest adoption and impact is virtualization, which enables us to run systems that previously required dedicated hardware with conventional hardware on an abstraction layer, so it can be upgraded without impacting operations. What are your thoughts on how O-PAS has changed our industry?