John Rezabek is a process control specialist for ISP Corp., Lima, Ohio. Email him at [email protected]Recently, the nearby hydrogen plant that's essential for production of useful chemicals where I work informed us they needed to swap out a faulty controller on the fly. "This should be a non-event, right?" asked our operations manager. I've been doing this for awhile and have broken a few promises to operations over the years, so I knew I'd be remiss to say, "Sure it will be no problem." Instead I said, "It should be a non-event," emphasizing "should."
The risk was not zero. The technician could pull the wrong controller. The 15-year-old backplane could have some unseen brittleness that might be revealed when one starts yanking out cards. The equally aged power supplies might be feeble enough to fumble or induce another fault when subjected to the inrush of a new controller booting up. Although I'd say the likelihood was still very low, this non-zero risk would have to be evaluated against the risk of running for months with no controller redundancy. Which was the better bet?
How many of us thought that as chemical, systems, instrument engineers or technicians, we'd be paid to be professional odds-makers? Sure, we're playing with the house's money, as in our employer's/owner's/shareholders' money, but our career advancement or continued employment can pivot around the choices we make, namely the risks we choose to take on. On the highest level, the finance men and women who make our worlds turn are themselves betting on markets and process plant reliability. If you work on projects, you may think you're safe in your cubicle foxhole, but many projects are themselves a wager on cost, schedule and demand for the new production. Even if we're not the individual placing the bets, we're being driven by the people aiming to hedge theirs.
For the most part, the managers and investors in our process businesses strongly prefer low risk to high risk and abhor uncertainty. Compared to my home neighbors, who are trying to grow corn and soybeans in a wacky weather summer, our enterprise is predictable. For businessmen who dislike risk, we are compelled to ask questions, such as "How much insurance do you want to buy?"
This is a good time to ask. Executives and managers at all levels are being bombarded with the "Internet of Things" marketing buzz, and may even be dimly aware that many plant-level devices have had microprocessors embedded in them for some time. You might take the opportunity to further enlighten them: these devices have the innate ability to pass along a few helpful pieces of data in addition to their traditional measurement or control function.
So, before you're asked "What's our strategy for employing the Industrial Internet of Things?," it may be beneficial to contemplate "Where can I get the most bang for the buck?" Given a chance to exploit the zeitgeist to attract some scarce capital, it's best to propose a strategy that has a good chance to yield some early "saves." It will pay to focus on the devices and systems that have the most impact. Don't just bet on some horse because it has a clever name.
For one systems engineer overseeing a pharma facility with many batches of extremely costly and tightly regulated drugs being processed, it was control system foibles that were coming up snake eyes. So he's invested in a system that sends messages to his mobile phone, so he can detect and respond to system errors. Save a couple million-dollar batches, and you may get a little more funding and maybe even another person.
Whatever your focus, keep in mind you'll need to provide not only networks and management/reporting systems, but also training, procedures and probably some form of auditing or efficacy checking. Take a lesson from those whose intelligent device management systems aren't gathering dust in a forgotten corner of the maintenance building, and maybe your wager on the "IIoT" will pay off in spades.
