This article was printed in CONTROL's September 2009 edition.
By Greg McMillan and Stan Weiner
Greg McMillan and Stan Weiner bring their wits and more than 66 years of process control experience to bear on your questions, comments and problems. Write to them at firstname.lastname@example.org.
Greg: This month we're talking more about knowledge transfer and what today's engineers need to know. I learned most of what I needed to know on the job and from Shinskey's and Lipták's books and technical short courses, not at school. The guidance from experienced engineers and technicians who weren't afraid to tell it the way it is was most critical. It was more than just shadowing and mentoring. You were on a path to get the best instruments specified, installed and commissioned. Stan was the master at this.
Stan: My track record enabled me to specify instruments based on performance rather than cost. I wasn't trying to make points with managers. I was making sure the project had measurements and valves with the best sensitivity and rangeability, realizing that the benefits were there, even if they were not well-recognized.
Greg: I was fortunate to spend most of my first six years in the field supervising the construction and start-up of major projects. I then moved into technology, where I actually used Nyquist plots to develop equations for the ultimate period and gain for integrating processes and for runaway processes. These equations were particularly useful for analyzing batch reactors (see "Life is a Batch"). In the course of solving control problems, I got to develop dynamic simulations of compressors, neutralizers and reactors that helped me gain insights into process relationships and dynamics.
Stan: Greg and I wrote a lot of books together to transfer some of the more practical stuff in a humorous way. Our most widely read book was How to Become an Instrument Engineer. It was probably the only technical book consistently read cover to cover. It had steady sales for 20 years until ISA made it available only as an electronic copy. Sales plummeted to zero. There's a lesson here.
Greg: Electronic searches are powerful and essential to research a topic, but I think there is still place for books. If you don't know what you are—or should be—looking for, flipping through pages of a hard copy is still the most effective way for me.
Stan: You can find almost anything on the Internet, but it is fragmented. It is difficult to put the whole story together.
Greg: Don't get us wrong. We think putting technical information from internal and external expertise on a website is the most effective way of sharing information to the largest possible audience. My website (www.modelingandcontrol.com/) gets 15,000 hits a month.
Stan: To get an current perspective on knowledge transfer in the chemical industry, we asked Nick Sands, ISA VP of Professional Development.
Greg: Is process control technology advancing?
Nick: Yes—ever so slowly in some plants and more rapidly in others. Control system upgrades and major projects continue to provide some of these opportunities. That's when we get a chance to really make step changes. But there are always improvement programs that drive smaller changes.
Greg: What are the skills needed?
Nick: A wide range, including process understanding, the principles of measurements and what the combination of those two means. Understanding the control system and the way to implement sound control strategies are also important. Tuning is becoming less critical. Optimizing is the trick. From advanced control to interlock design, alarm management practices and, increasingly, the human machine interface (HMI) design play an important role.
Greg: Do you see more or less process, instrumentation or configuration engineers doing process control?
Nick: There are already fewer boundaries. There are control engineers, not instrument engineers and configuration engineers. The process engineers seem to do less in the control system as it has become more complex.
Greg: Will we rely more on system integrators, suppliers and contract engineering firms for advanced control, instrumentation and configuration design and implementation?
Nick: Yes. On major projects the instrumentation and control system equipment is more likely to be contracted to a main automation contractor (MAC). The control design is more likely to be done in-house where the process technology is in-house, so the control design is coordinated with the technology. As advanced control tools, such as model predictive controllers, become more embedded in the control system, they may be easier to support.
Greg: Will process control improvements decline into a status quo?
Nick: No. There is an increased emphasis on monitoring performance and maintaining improvements. The status quo shifts to the improved level
Greg: Will some plants become inefficient or shut down due to losing their process control expertise?
Nick: Maybe. The impact of control system performance is not directly measured, and so it may never be identified as the issue. The gap between what is and what could have been goes unmeasured, too. That could be the difference between profit and loss.
Greg: What can be done to keep process control improving?
Nick: Share success and foster professional development to challenge people to be innovators, always reaching for that next improvement. We should celebrate the Greg Shinskeys and Vern Trevathans, and encourage newcomers to aspire to that level.
Greg: Are engineers who are retiring asked to document their knowledge?
Nick: Yes. There is a renewed emphasis on documenting best practices at all experience levels, though I wonder how much those reports are read.
Greg: What can we do to make sure that knowledge is not lost as our experts leave the workplace?
Nick: I think wikis might be a more flexible media for knowledge capture than traditional reports. It's more collaborative and allows anyone with the expertise to contribute.
Greg: Next month, we discuss if engineers can be video stars or talk show hosts. Meanwhile, Randy Reiss shares his best farewell messages ever.