While some folks were recovering from Mardi Gras, soaking up Jamaican sun, or cruising Daytona Bike Week, my boss sent me to Caesar's Palace in Las Vegas for IndusWorld 2003.
Oh, sweet, you're thinking. I'm saying, not so fast. True, the weather was lovely compared to the typical March dreck in Chicago, but the IT folks I rubbed elbows with were as serious and motivated a set of combination bean counters and computer geeks as you'd ever care to meet. Or not meet.
But you should, because these guys have the unenviable task of finding the funds to back up your paycheck, and they're pretty much in charge of the systems you can use to prove you've earned it. The vast majority at the meeting are deeply concerned with keeping power plants, especially nuclear plants, running smoothly, safely, and all the time. And as cheaply as possible. Paper mills, primary metals, other process industries, and commercial enterprises made up the balance.
I didn't see a single one get drunk or so much as pull a handle on a slot machine.
Indus is North America's leading supplier of enterprise asset management and computerized maintenance management software and services. Users, partners, and presenters get together at IndusWorld to share war stories, grouse about system limitations, and in general, learn how to do better IT.
For someone like me, who has only looked at asset management from the process control perspective, it was an eye-opening glimpse at the problems in the rest of the plant. I won't bore you with the end users' tales of woe. ("What do you do? Treat every open oil drum as a warehouse?" "The guy had a $50,000 motor under his bench!" "Fasteners? I don't even want to think about them. They have legs.")
Instead, I want to focus on the direction of this industry, as demonstrated by the technologies on display by Indus partners. From hand-held data-gathering devices and software, to a service that inventories and digitizes the documentation for legacy equipment, to software and integration services powering completely incredible data leveraging and decision-support systems, these companies seem to offer any tool you'd need to get a grip on the huge percentage of the plant that's outside the control system. And by the way, the control system, too.
I'd been nearsightedly believing the uptime and maintenance cost benefits of predictive technologies would obviously flow from self-diagnostics and artificial intelligence/optimization techniques applied to process instrumentation, with a little help from condition-monitoring sensors and systems. But these guys are going after the rest of the story: operations data and maintenance histories.
Add in a good dose of inventory and supply chain efficiency and suddenly you're talking about being able to model and predict the future. For example, Data Systems & Solutions says it can forecast expected plant capacity and availability; the frequency, duration, and cost impact of unplanned downtime; and the effectiveness of design modifications or improvements in operations or maintenance. It can optimize spare parts, predict maintenance resource and budget requirements, and provide quantitative models for allocating resources for the greatest ROI.
Apparently this is all in a day's work for a well-run nuke plant. Wouldn't you like to know exactly when you should check or change that pH probe, valve stem seal, or heat exchanger? How about the bottom-line effect of a control system upgrade?
In this month's cover story on Top Technology Trends, Peter Martin at Invensys points out the need for the engineering and accounting disciplines to get together and define critical performance metrics in common terms. From my brief sortie among Indus users, I'd have to agree.
