Optimization Needs More Than Technology

Most process industry plants operate at less than maximum efficiency. That is a bold blanket statement, but one that is based on many years of plant site visits to train personnel how to get the most out of their automation systems. Again and again I have seen plants filled with expensive high-technology equipment, computers, control systems and smart devices, yet plenty of operational problems still existed.

This article summarizes some major sources of those problems ," improperly tuned control loops, a lack of financial analysis and improperly trained plant personnel ," and outlines the actions necessary to overcome them.

Among the conditions that reduce process efficiency are improper control valve selection, sizing or calibration; instability or oscillation above and below the setpoint; slow response to changing conditions; or a combination of these factors.

Numerous plant audits have shown roughly one-third of all installed control valves operate at substandard levels, even though properly operating valves are essential to overall plant efficiency and product quality.

A study performed by Entech Control Inc., Toronto, identified common problems found in control loops. (See Table 1). Some 80 percent of the control loops studied contributed to increased variability. Table 2 details some common symptoms of control loop problems.

Some systems have inherent design deficiencies, but the majority of automation system problems identified by this study stem from lack of human intervention. Most of the problems with control valves, control loops and measurements can be fixed if plant personnel are trained to identify trouble areas and are motivated to make the changes needed for improved performance.

Because most companies need to increase profits, it is difficult to understand why more management groups have not empowered their personnel with training and motivation to make the corrections. Some plant officials might mistakenly believe that problems will go away and the process will run at peak efficiency once new high-tech equipment is installed.

Many process engineers look at control valves as low-tech iron that simply controls fluid flow. However, control valves are actually among the most complex field devices. They repeatedly are required to perform rigorous tasks, often under extreme service conditions. Because many control valves are in motion continuously, they can reduce process efficiency drastically if improperly designed, installed and maintained.

When a concerted effort is made to understand the performance of these final control elements, process variability can be reduced. Control valves need to respond quickly and accurately to control signals if a corresponding improvement in loop performance is to occur.

Another very common symptom is that controllers are placed in manual instead of automatic operation. Although many other reasons could exist, control valve problems are the root cause in many cases.

Too few technicians really understand the value of improving loop performance. Although many case studies show better performance of a critical loop produces significant operational benefits, less-obvious problems go unnoticed and unattended.

More action is needed for the large number of loops considered less important because improvements made to a large number of these loops could magnify the financial benefit. These improvements become possible and commonplace once plant personnel have the knowledge and skills to troubleshoot problems and implement solutions.

Most process plants have accounting systems that monitor overall plant costs, wastes and revenues. Major process changes can be tracked to show if an investment resulted in a return to match financial projections. But macro accounting does not provide much help in managing costs on the plant floor or in decision-making at levels affecting overall profitability.

In my experience, a micro-management system rarely exists down at the control loop level, yet poorly functioning control loops continue to drain profits.

On a single-loop basis, financial returns usually are considered so insignificant that plant management will not invest time or resources to track or attach importance to them. Is anyone in the plant interested in changing a feedstock flow-control loop to reduce the flow setpoint by even a small amount? The answer if probably no, unless plant officials understand the value and make plant improvements a priority.

Suppose a finished product flow should be 500 gallons per minute (gpm). If that loop is somewhat unstable or is cycling, plant personnel could lower the setpoint to 499 gpm to avoid exceeding other established limits.

What is that 1 gpm worth? At first glance, the flow will change by only 0.2 percent ," seemingly not a big deal. However, if the plant has approximately 500,000 productive minutes a year, multiply the cost of 1 gpm by 500,000 to project the annual return. Do the math to recognize that a significant return is possible from a small change.

It is not that difficult for knowledgeable individuals to make changes that influence loop performance. However, once an improvement is made, measurement of the results usually comes in the form of informal and nonspecific comments such as: "Looks good;" "The chart looks better now;" or "The control valve is not moving as much now." These comments do not demonstrate to management the contribution that small improvements are making to the bottom line.