Many control engineers are looking at ominous signs and trying to make sense of them. Low factory orders, dismal plant utilization, and a mass rush to offshore outsourcing seem to be spelling the end. Control engineering implies control systems, production, and manufacturing. As the U.S. seems to be ceding its manufacturing base, what can we do? A lot, as it turns out.
Many control engineers are looking at ominous signs and trying to make sense of them. Low factory orders, dismal plant utilization, and a mass rush to offshore outsourcing seem to be spelling the end. Control engineering implies control systems, production, and manufacturing. As the U.S. seems to be ceding its manufacturing base, what can we do?
A lot, as it turns out. Understand that the U.S. can only win through high value-added production. Proving your productivity advantage involves more than costs—you won't win on that score, so avoid commoditizing. Your customers will pay for value. Ask what kind of company you are working for, and ask how to guarantee that value.
Measures of success change depending on company and industry. Success could be sales per month, revenue per client, new clients per day, or new products per year—driven from clients, not necessarily intrinsic physical truths. Control engineering choices change with the success measure that is used. Control engineering in the revenue per client model means making your factory excel at parallel processing. Good control engineering for new products per year means downloadable CAD data and assembly lines that stop and start quickly. Learn the fraction that drives your company, and ask other functional areas how they think it should be achieved.
Beware any machine running very close to or very far from 100%. Cost accountants love high utilization because it means local efficiency and sound purchasing. But 100% means bottlenecks, building expensive inventory somewhere, and overall inefficiency. For some time in the 1990s, the most productive automotive plant in the world was the Fiat facility in Melfi, Italy, which should reveal the weakness of misapplied productivity goals. Instead, use queueing theory and marketing to expose the dynamics in your production.
We install and program control systems but fail to ask what they achieve. Often, we assume that one purpose of our systems is to improve quality. This is not necessarily true, and the value of control engineering may lie in its real options. Similar to financial options, real options represent value derived from flexibility or position. Controls drastically change the range of what a facility can do. Find, calculate, and demonstrate the real options that are hidden in your production facility. Learn to exercise those real options to capture cash. Once you know how to do this, it's easy to show management how valuable you are—and that outsourcing destroys the real option and information value.
Look outside the traditional boundaries of the craft to get more value from your efforts. Many of us design or maintain human-interfacing systems without knowledge of utility functions, cognitive science, behavior, or ergonomics. Remember that our product is only as good as who uses it—it's not just about PID, and there is no transfer function that describes how humans make choices. Success involves how you take advantage of the knowledge in the operations staff.
Understand that managers are rewarded for return on equity, which causes some perverse behaviors. Among these is the incentive to underinvest. A facility can improve in an accounting sense while it is actually becoming less and less productive. Break this by showing management that their choices in outsourcing drive customers away. We dread angry customers, but they're our best friends—they lay out the improvement plan. Customers never buy a product or service, only a solution to their pain. You can engineer that solution.
Ensuring that engineering remains a vital, decision-making component of manufacturing means speaking in financial terms and modeling decisions in new ways. Understand that finding a metric for your efforts will be hard, that indirect solutions are often more effective, and that you might not get a direct reward. This will do more than just guarantee your job—it will guarantee that the U.S. remains a healthy, vital place to manufacture.
Robert A. Dunlap is a former controls engineer with UOP. He is currently a final year MBA student at the University of Texas.