How does workstation design shape operator performance?

Performance-Shaping Factors Series

NEWSLETTER ARTICLE

Operator performance is a function of numerous interacting variables. This series of newsletters examines each of these variables, or performance-shaping factors (Figure 1). 

Lesson 7 – Workstation Design

Some of you are old enough to remember the early distributed control workstations: three monitors with one dedicated to alarms. There were variations with different vendors, but the overall design didn't differ much. Not too long ago I was in a control room where one operator had 12 monitors. Of course, there was a spare set of 12. I thought I had walked into a Best Buy.

Process plants are not unique to monitor creep. The two pictures below show the original and current designs for control of remotely piloted vehicles. Original Design Current Design

I have a couple of thought questions on this proliferation of monitors. Remember, in the early days of DCS, those two or three monitors displayed a maximum of eight instruments each.

Question: If an operator could control the process with three monitors, why would they need more, given they have a better control system?

One valid answer would be that they weren’t optimizing the process when they only had three monitors. However, 12 is a long way from three. Display systems should be compact. Chris Wickens of the University of Illinois found that when the exact same data is spread over two monitors rather than one, the information processing rate is lower for the two-monitor condition. The result? Slower response time by the operators.

A fundamental principle in human factors and psychology is the limitations of our consciousness, our working memory. George Miller discovered that limit in 1956, published in a paper called, “The magical number 7, plus or minus 2: some limits on our capacity to process information”. Essentially, we can only process about seven chunks of information at a time.

Question: Assuming a screen can be considered a chunk, what would be the upper limit on the number of screens?

Screen proliferation can negatively impact performance in two key ways. First, primary information should be within the operator’s critical viewing area, as tunnel vision is inherent with high stress situations. This could potentially lead to the operator failing to notice important information. Calculation of this area has long been established^[1] and reaffirmed^[2] (note actual line of sight is 15 degrees below horizontal but is shown as equal to horizontal for ease of understanding). Second, display designers may be tempted to use all the monitors possible. This results in information being too spread out, making mental processing more difficult. When I questioned the use of ten monitors for one operator on a project, the engineer said he would find a way to use all the monitors.

When Beville personnel collect data on console operator performance characteristics, we record which display is on which monitor. It is not uncommon for an operator to have the same display on two monitors, sometimes the one right next to him. Operators will often say that it’s not essential in steady state, but they need the extra monitors during an upset (true only for whomever is going to help). When an upset hits, tunnel vision will cause the operator to focus in on a few monitors within their primary visual field. If information indicates that a different display is needed, then they could (a) scan around to all the displays on all the monitors to see if it is already called up in order to make an assessment or (b) they can call up the display on one of the few monitors they already have in view.

. . . And the Answer Is?

So, Mr. Bigshot Human Factoring Person, how many monitors should they have? My answer is four, as the primary visual field is about the area of a two-foot circle. My answer will be short-lived, as the size of the monitors keep growing. The real question now is how many pixels does a console operator need? One 46” monitor will cover about the same area as the old four-packs. Caution: Do not think that you can switch to a 46” monitor with displays built for a four-monitor workstation.

Information systems should be compact. However, span of responsibility is increasing at most facilities. How do you achieve both? The answer to this question will be part of our final topic in this series, interface design.

[Note – If you have actually looked at Figure 1 on the other newsletters, you will see that workstation design was not included, but job requirements were. I found this to be a better topic to discuss. As I said in the first newsletter, there are all sorts of performance-shaping factors that make up human performance.]

[1] Van Cott, H.P., and Kinkade, R.G., Human Engineering Guide to Equipment Design, US Government Printing Office, 1972 p393.

[2] IOC11064-4, Ergonomic Design of Control Centers; Part 4, Layout and Dimension of Workstation. 2004, p6.

Copyright 2023 Beville Operator Performance Specialists, Inc., All Rights Reserved