It is in transition, unsettled by current events and its own successes. Automation draws together multiple threads of knowledge with little regard for traditional domain boundaries: machine design from mechanical engineering , control theory from electrical engineering, software from computer science, and methods for design and integration from systems engineering .

This synthesis that comprises automation evolved in an unprecedented, cooperative effort on the part of government, industry and academia to build control systems for the aerospace projects of the 1960s and 1970s. Then industry rapidly adopted automation in petroleum refineries, chemical plants , paper mills, water treatment facilities and the like. Automation systems soon became an essential—and largely invisible—part of society’s industrial infrastructure.

But with principles expressed in terms of the calculus or Fourier transforms and practices learned empirically in specialized environments, a coherent treatment of automation was never adequately incorporated into high school, technical school and undergraduate university curricula. The broad foundations necessary for continuity were not developed. And now, the people who developed the conceptual synthesis, as well as those who kept the systems operational, are retiring or have already done so.

Ten Years Out

“We’re experiencing a major resource crunch in the process industries at just about every level: operators, mechanics and engineers,” says Larry O’Brien, research director at ARC Advisory Group Inc. ( www.arcweb.com ), in Dedham, Mass. “I work with the major automation vendors every day, and this is a real problem. We’re looking to the colleges and technical schools for qualified people, but they’re hard to come by.”

At the same time, new automation systems are becoming more complex, particularly from the perspective of the shop floor. “In the last decade, we’ve seen technical advancements that greatly increase the amount of available information,” says O’Brien. This information isn’t always easy to use productively: “The way things are now, process operators spend too much time responding to alarms in the plant and don’t have enough time to work on making the process better.”

The trend to more sophisticated automation systems seems inevitable, and desirable. Such systems are the necessary response to obsolescence and will open new doors. The U.S. Food and Drug Administration Process Analytical Technology (PAT) initiative is only one example of the demand for automation systems that can deliver fine-grained control with solid reliability. The challenge is to create systems that are true allies. “The key to getting the most from technology is focusing on its business value, not on technology alone,” says O’Brien. “That’s where you need to consider the people.”

Professor Raffaello D’Andrea has worked in industry and currently does research in adaptive systems at ETH Zurich (Swiss Federal Institute of Technology). “We now have systems that offer high performance, but are becoming complex,” says D’Andrea. “They can perform very well under certain conditions, but when conditions change, their performance can degrade rapidly. We need, and are developing, adaptive systems that learn. The longer they run, the better they get. But if people don’t know how to handle these systems properly, they will not deliver the promised performance.”

D’Andrea adds: “People who know how to build, deploy and operate these systems will be in high demand. Automation professionals need to be ‘T’ individuals: deeply knowledgeable in a particular technical specialty and educated broadly enough to communicate with other specialists. This is another way to say, ‘systems engineering.’  Automation systems are complicated, complex, high‑performing and challenging. We need people who can manage that complexity.”

Marty Weil , martyweil@charter.net, is an Automation World Contributing Writer.

ARC Advisory Group Inc.
www.arcweb.com

Subscribe to Automation World's RSS Feeds for Columns & Departments