The Changing Face of Automation Engineering
The Changing Face of Automation Engineering
CONTROL Magazine looks at lights-out plants, online as-builts, remote asset management, and all the other things that will be necessary to make up for the lack of trained automation professionals in the next 20 years.
We’ve been saying for some time now that the light process control engineers are seeing at the end of the tunnel is the headlamp of an oncoming train. It’s still coming, and it’s picking up speed.
Converging trends that will converge in the next decade or so will utterly change the way process automation is done. Whether you’re simply a process automation professional, or a system integrator, or a vendor, you’ll need to understand these trends and adopt new stances, or the train will make you as flat as a penny laid on the track. There are big changes ahead that aren’t just technical, but are economical and political, as well as social and demographic.
Big Changes Will Alter How We Work!
These changes will come at you from every possible vector—social, economic, political and technical. They will not only alter the way we work, but will affect what we’ll have to know to work in process automation, and maybe even where we’ll have to live, and for whom we may be working.
Best, Brightest Stay Away from Manufacturing
Most control and automation professionals are Baby Boomers. ISA’s membership data shows the average age of its members was 45 in 2000. It’s logical they’re now older than 50 and getting older on average. When we were growing up, it was a great future to work in manufacturing. However, since the late 1960s, it hasn’t been as cool to work in a factory or process plant. The best we have are working in IT, law, banking, and e-commerce—anything to avoid getting dirty in a plant. If it means getting grubby, even the engineers and scientists we’re importing from Eastern Europe, India, and China are unwilling to really pile in, and get the job done.
Factory work is no longer considered a safe career path for anybody. And everybody who was a little kid growing up in the 1980s and 90s is well aware of what happened when dad and mom gave their loyalty to the company—they got laid off, their pensions got stolen, and they had to go back to work doing stuff that paid them much less than they expected, such as being a greeter at Wal-Mart. Furthermore, many of original jobs disappeared overseas as companies locked into short term economic thinking, and pursued labor and regulatory savings in India, China, Korea, Thailand, Indonesia, and Vietnam.
The layoffs of the 1980 and-90s also disposed of many of the people who knew how to run process plants manually. A lot of them had been operating process plants when all of the operations were manual—look at this gauge, turn this valve one quarter turn. They were the people whose calibrated eyeballs could do a sweep of the control room’s “control wall” instrumentation, and get a clear picture of the plant’s operating state. They were the people who helped translate manual operations into closed loops, and turn closed loops into distributed process control systems. To a great extent, these professionals are gone from manufacturing facilities in North America and Western Europe. And, here’s something nobody thinks about, these types of experts mostly never existed in the Third World because, by the time we started building plants there, these facilities were modern automated plants.
WHO'S GONNA REPLACE YOU?
CONTROL Editor in Chief Walt Boyes asks his audience at this year's Honeywell Users Group Symposium to imagine manufacuring after all the Boomers retire.
Those who are left are retiring in droves. The ones that escaped the cost-cutting axes of the last 25 years are now retiring faster than they can be replaced. “My institutional knowledge walks out of the plant every day at 4 p.m.,” say many plant managers. Well, every afternoon, some of that collective expertise walks out, and never comes back! So much has been lost that many process plants have completely lost the chain of knowledge handed down from older to younger employees. The remaining staff simply doesn’t know why the plant is operating the way it does, or why certain things are done, and others are not done.
And the institutional knowledge that older engineers are taking is irreplaceable. For example, in many plants, when multiple alarms go off together, many operators no longer know which are the most critical and which to respond to first. And, in alarm cascades, plant operators that don’t know those things are usually going to do something wrong. We’ve seen this many times in many accidents such as the BP Texas City incidents in 2005.
Process industry companies often are unable to hire even strategically. Just as the process industry companies are beginning to realize exactly what they did to themselves in the 1980s and 90s as a result of layoffs, offshoring, and other screw-jobs, they’re running smack into the post-Boomer demographics. While Boomers were the largest generation in history, those that followed are much smaller, and there simply aren’t enough younger workers to fill all the remaining jobs. Many of these jobs were absorbed by process automation in the late 1980s and 90s, but the important jobs, those requiring institutional knowledge, are going begging. Companies used to have mentorship and training programs to bring younger workers up to speed with the institutional knowledge of their more senior co-workers, but those programs are typically gone now. Even bringing in workers from offshore isn’t enough.
Since training and mentoring and apprenticeships cost money, and the process industries still are very cost conscious, the first thing they’ve done is to turn to their automation vendors for help. And the automation vendors, having heard the siren’s call to move from being equipment companies to being service providers, are answering the industry’s plea with staffing services, operating and maintenance services, and design expertise. In fact, some of the people laid off by the process industries found new homes in the vendor community, and took their expertise there, too. Vendors know it’s better to have somebody pay them once a month for doing tasks than it is to have somebody pay them once in a while for a new piece of equipment. And, since many process plants don’t even have engineering talent, vendor companies are delighted to do the design engineering and even the integration for them.
The problem, of course, is that the vendor companies are having the same staffing problems that the process industries experienced. The unpopularity of manufacturing jobs, the economic and career uncertainty of the past generation, and the smaller demographics of younger workers in North America and Western Europe have created problems for the vendors’ grand plan to provide all this intellectual property and institutional knowledge as fee-based services.
Flattening World, Inescapable Data
Of course, that metaphorical train is still coming. As knowledge workers become more pervasive worldwide, and as sensor and control technology continue to improve, we can do vastly more with far fewer “gauge watchers” and “valve-handle-turners” than we’ve ever been able to do before.
The simple division of the world’s economies into First, Second and Third Worlds, or Developed versus Undeveloped, is coming unraveled. There are 250 million Indians, who live at the same standard of education, training, and income as in the U.S., and another 800 million people in India who live far below this level.
Everywhere you go in the world, you can see new tech coexisting with old tech and even ancient tech. Many underdeveloped countries are installing the latest technologies in process plants, just as they bypassed wired telephone systems in favor of cellular technology. So, while you can see a goatherd talking on a cellphone while herding his charges today, his son will be a process engineer tomorrow. We’re going to have to live in a increasingly connected and much smaller world.
Jobs are migrating, but people are not. Economic dislocation isn’t limited to North America and Western Europe. Even Korea and China are finding that it’s no longer possible to build some things there, and stay competitive on the global market. In 1905, more than half the U.S. population worked in agriculture. In 2005, less than 10% still does. Even though the population has greatly increased, there’s still a relatively low rate of unemployment. As jobs migrate, new jobs are created to fill the need. The tsunami of globalism is going all the way around the globe.
Changing Work Practices and Culture
For years, TLA consultants have been preaching Real-Time Control for business and processes. And, for years, captains of industry have been paying lip service to it. However, this change is finally upon us. Even SAP has come to the public realization that you can’t optimize your enterprise with just an automated balance sheet. With Microsoft and SAP supporting ISA95 and ISA88, it’s a sure bet that the Great Divide between the business systems and the plant floor will be erased by the end of the decade.
Even system integrators are changing, becoming more flexible, and moving into higher order integration. Companies like the multi-integrator Automation Alliance Group are creating new business models for expanding and providing services on a global basis, while remaining, in essence, local companies.
Some firms also are changing how they treat employees. While companies in some industries still treat their human resources like completely replaceable commodities, such as steel and nuts and bolts, other firms are beginning to understand the value of knowledge workers. They’re realizing how much more profitably they can operate by giving workers more control over their jobs, and by decentralizing decision making to the lowest possible level. In process automation, this means that the newest plant operator or technician will need to know much of what instrument engineers knew in the 1980s. To attract and keep qualified people in a job market that will be a perennial seller’s market for at least the next generation, companies will have to offer benefits and perks like they used to in the 1950s.
Game-Changing, Global Economies
Vendor and process companies alike have already figured out what it means to “think globally, and act locally.”
Phoenix Contact, for example, has essentially split its entire company into three equal locations: Germany, U.S., and China. The company’s G&A, R&D, and manufacturing will all be split, with many tasks done at each location because now they’ll be able to work 24/7. Phoenix’s people even joked about it: “The sun will never set on our TransNational company!”
Invensys, ABB, Honeywell, Siemens, all the major vendors are doing much the same thing. Yokogawa is so serious about proving that it’s not just a Japanese company that it has relocated its entire international division to Singapore, and sent one of its chairman’s key lieutenants to run it.
Brazil and Venezuela have GDPs growing at 17%, and most these increases is in exports to China. We can’t plan on domestic versus ROW anymore. A U.S. company with exports that account for only 10-20% of total revenue is in serious need of a coffin.
Global Unrest Adds Spice
The Chinese are correct when they say that “interesting times” are scary and fraught with danger. Revolutions never happen when the oppressed are the most downtrodden, but usually occur after they’ve been given some relief, albeit “not fast enough.” Most of the world is considerably more stable than it used to be, but there are still places, such as Venezuela, Brazil, China, Indonesia, and the Middle East, where the economies aren’t large enough yet to support a sizeable middle class, and provide jobs for all the available workers. For the first time, these workers are being exposed to all the goodies of Western Civilization. Some are happy and want the goodies; some are unhappy and don’t want them; and some are even jealous enough to want to steal them.
Moving to Higher-Order Technology
It used to be thought that North American and Western European companies, both vendors and integrators, would simply move into the Third World as soon as the undeveloped countries developed enough to make the market worthwhile. While these companies waited, indigenous integrators and vendors developed in many Third World areas, and now are competing straight up for the same projects. The recent announcement of the Global System Integrator Alliance shows one of the reactions of First World companies: make alliances with these indigenous companies.
In the early days, we were all concerned with sensors, measurement, and final control elements—how to build the watch. Later it became imperative that we know how to tell time—how to close control loops. Our skill sets expanded yet again when it became obvious that it wasn’t enough to be able to build the watch and tell time, but now we had to know what the benefits of being on time were—expanding the benefits of control to the entire plant via distributed control systems.
Even though this is where a lot of us stopped, it just isn’t enough. We know how sensors work, we know how loops work, and we know how to control a process. Unfortunately, the required skill set has expanded again. Now we have to understand scheduling. Our primary value is to see to it that information from the plant is transmitted to the enterprise. We’re now working on fourth-order concepts.
Smarter controls need smarter people. Yet schools and universities are graduating fewer and fewer engineers, scientists, and mathematicians. It doesn’t take a rocket scientist to see what will happen if this trend isn’t reversed. System integrators, vendors, and end users alike are asking their people to do more with less, and that means smarter people, with smarter tools, and a clear grasp of the way the process works. This means that the market for knowledgeable automation workers is actually increasing worldwide, but these will be automation workers who understand the entire picture from the sensor to the enterprise.
We’ve talked for the last few years about disruptive technologies. This is what we mean when we talk about the oncoming train. Taken together these disruptive technologies will create tomorrow’s process plant. Small teams of engineers and operators will use integrated simulation to model the plant, and push that data into design/draw software, which will produce plant and the control system designs. AIDC sensors and wireless asset management transmitters will continuously update the “live as-builts” for maintenance and management.
The operator will be a supervisor, making sure that real-time performance management systems, advanced process control systems, and asset management systems are working properly through new inputs via wireless and the ability to go mobile, so the control system can be accessed from anywhere in the plant or elsewhere.
Control Process Automation Hall of Famers McMillan and Blevins Speak Out
Terry Blevins, of Emerson Process Management, and Greg McMillan, a retired control uber-guy from Solutia and a Control columnist, met recently with editor-in-chief Walt Boyes to discussed the future of the control system. Between them they have more than a 100 years of experience in automation and process control. McMillan is a leading expert on pH and other wet chemistry sensors and on advanced process control. Blevins is the most successful distributed control system designer in the world. Both are members of Control’s Process Automation Hall of Fame.
Boyes: We keep hearing that PID is dead. We’re told that numerous sensors, not just the most important one for each single loop, will be connected directly to a database, and advanced-rule engines will extract patterns from that data, and automatically control all the final elements in the plant. Do you see that?
McMillan: What’s missing from the advanced-rule engine is that a process is inherently dynamic. These rules use process snapshots, instead of understanding the dynamic response. Process engineers think steady-state for process design. For control design, it’s necessary to consider control dynamics and unmeasured disturbances. This problem may also be encountered when batch sequencing is use to perform functions better addressed by continuous control techniques such as PID and MPC.
Blevins: For example, to avoid a temperature overshoot, batch logic might “shut the steam valve when the temperature exceeds a specific value.” There’s no direct connection with the desired result of achieving a target temperature.
Boyes: So what do we do? Redesign the control system?
McMillan: No, instead of redesigning the control system, we need to effectively apply traditional and advanced control techniques that account for process dynamics. Why use rules, when a variety of advanced control techniques are embedded in modern control systems?
Boyes: That’s an Emerson product.
McMillan: It’s still a good idea. One reason that advanced process control projects often fail is that model-based control isn’t embedded in the control system, so tools and skills are need to update the control system for process change. Thus, APC may not be maintained, and it starts being ignored.
Blevins: If you build the APC into the control system, instead of layering it on top of the control system, then two things happen. First, the APC is updated when anything changes. Second, the APC’s operation becomes more transparent to the operator, just like the traditional PID controller.
McMillan: Another benefit is that APC is designed to work with the regulatory control. For example, when an engineer changes a control setting or operator during startup, this may put a loop into manual. This embedded APC will notice things like that, and automatically take proper action to provide bumpless transfer when the loop is placed back in automatic control.
The article was written by Walt Boyes, Editor-in-Chief of CONTROL Magazine and ControlGlobal.com. CONTROL mediates information transfer in the community it serves via a printed monthly magazine, its affiliated ControlGlobal.com website, a variety of e-newsletters and other digital media offerings. The end-use market CONTROL serves includes the classically defined process industries—chemicals, petroleum, food, paper, plastics, metals, and textiles—as well as power generation and water/wastewater utilities. Engineering design firms and systems integrators also are represented.