I visited with Steve Apple and Chris Wilson of TiPS in Georgetown, Texas on Wednesday afternoon. Steve is a former Pavilion employee, and has been in the process automation business for a long time.
"We've gotten a lot of very large projects lately," Apple said, "and we think we're making our competitors crazy."
Why is he getting big projects? Apple thinks it is simple. "We know that alarm management isn't rocket science. It is really simple stuff. The key to being successful at it is to figure out who the beneficiaries are, and ask them what they want."
So, who are the beneficiaries of alarm management? "It is all about the operators," Apple said. "It is about creating coherence out of chaos for the operators."
"There's a term of art in safety systems engineering," Apple went on. "It goes like this: obvious to even the most disinterested casual observer (or words to that effect--WB). What we have to do is to make alarm management fit that term of art. Right now, most people are designing alarm management systems for engineers, not operators."
And that, Apple noted, is why there is the notorious roll-off curve for alarm management, advanced process control, and other engineer-imposed "performance improvement" projects.
"We have a customer who says that he has to schedule new projects to get the engineers interested in improving performance of the process again. Maintaining steady state process improvement is boring," Apple said. "So he schedules a new project every time he sees the performance improvement curve starting to trend down."
I asked Apple what he thought one of the new trends would be. "Distributed computing," he said. "When DCSes were first created, the 'D' stood for distributed. But they weren't distributed, they were dumb sensors and dumb controllers with a central processor. So they changed the meaning of the 'D' to 'digital.' Think about this: most transmitters now have more computing power than the original IBM PC-AT...and they have lots of unused processing cycles. Fieldbus and wireless will give us the ability to network those processors together, and move most of the control down to the networked device level...true distributed control."
