Are Human-Technology Interactions Solving or Causing Problems? Part 1

Process automation engineers’ decisions impact everything from project cost and schedule, to uptime, throughput, quality, and safety. Instruments and valves connect the physics and chemistry of your process, and the decisions made by your automation systems and personnel. Those decisions form the foundation of business and personal job success. Emerson’s Tom Wallace, whom you may recall from earlier human centered design (HCD) and industry standards posts, explores the driving forces behind the need for an HCD approach in product design.

Tom notes that there are three connections each instrument or valve makes—the physical connection to the process, the information connection to the process automation system, and the human connection to those who interact directly with the instruments and valves.

These connections are profoundly important to you, even if you never touch an instrument or valve. The decision-making and performance of project execution ranging from project cost, schedule, and risk to operating plant functions, such as operational quality, reliability, and safety, are affected by the quality and performance of those connections.

He cites this troubling statistic:

If a user is unfamiliar with a task, their error rate can be up to 17X higher than if the user is familiar with the task.

Human Centered Design process

Source: Luma Institute

To address this issue, an HCD approach involves a process of designing human–technology interactions around how people learn, think, and work. Tom observes that it’s obvious but rarely done. Most human–technology interactions are built around the structure and limitations of the technology, often when it’s most limited and least mature. Changes are usually accommodated by adding pieces to the initial structure. It’s like building a one-room log cabin, and expanding it to a skyscraper. At some point, you need to redesign from first principles. The time is now and the first principles are human centered design.

Why human centered design now? In the past, there were fewer device types and simpler devices. Workers did the same things to the same device types many times, building expertise. Today there are more device types, and devices themselves are more complex. In addition, devices are more reliable so worker – device interactions are less frequent, but more varied. The result is unfamiliar human-device interactions and more human error.

Studies show that up to 80% of abnormal situations are caused by human error. Demographic and technical trends are poised to make this worse. Fewer, less experienced personnel, and more complex technology are poised to create a perfect storm of human error, with consequences ranging from minor to catastrophic. Our goal is to optimize the human – technology connection and prevent the perfect storm.

Why Now? The Tipping Point. Demographic and technical trends are placing unprecedented stress on the field device – human connection. This issue is gaining importance because the stress is pushing us quickly to a tipping point.

A tipping point is a point where a small input can create a large and rapid effect, like the straw that broke the camel’s back. The perception is that the small input caused the change. In reality, the cause is a large number of small inputs that occurred in the past. The future was determined long before the final input was applied. Here’s an example.

  • It’s good engineering practice to design mechanical systems with a safety margin. If a beam needs to support a load “X”, it’s designed to support a higher load, say, “2X”.
  • This beam can fail three ways. The first is by increased load. The load grows until it exceeds 2X, and the beam fails. The second way is to increase the span the beam must support. If the span increases from “Y” to “2Y”, the beam will fail. The third way is to decrease the strength of the beam. If the strength is reduced from 2X to 1X, the beam will fail.

Figure 1: Design Load

Figure 2: Increased Span and Load

Figure 3: Decreased Strength – Thinner Support Beam

Increasing the load and span, and reducing the strength until failure is very poor engineering practice. Paradoxically, this is seen as good business practice when it happens with human capacity in most organizations.

In a typical facility, the number of instruments and valves is increasing as more and better information is needed to optimize plant performance. To the people who must configure, maintain, and repair these devices, this represents an increase in load.

Individual field devices are becoming more complex, and the number of different device types is growing. This means more knowledge is needed to configure, maintain, and repair the devices. This represents an increase in span.

Finally, the number of people tasked with supporting this load, and the experience of those people is decreasing. This is reducing the strength of those who must support the load.

The cumulative result is that most organizations are fast approaching, and some have passed the tipping point. It’s not a question of “will collapse occur”; rather, it’s a question of “where, when, and with what consequences”.

Fortunately, this problem is solvable. It was created with the three fundamental causative factors, increased load, increased span, and reduced strength. It’s solved by addressing the same three factors. The foundation of the solution is human centered design.

Tom closes his thoughts in this part 1 post with another thought-provoking statistic:

Did you know? Under time pressure, the chance for human error can be up to 11X higher than when this time pressure is removed.

MP3 | iTunes

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

One comment so far

Leave a Reply