How We Work-Human Centered Design for the Operating Plant

Emerson’s Tom Wallace continues his series on the impact of human centered design on the technologies used by automation professionals.

In previous blog posts, I shared some of the science behind Human Centered Design (HCD). In this and the next couple of blog posts, I’ll share ways that HCD is applied to devices in an operating plant.

Reliability, quality, and safety all depend on properly functioning devices. In an operating plant, the greatest gain from human centered design is reduced human error and the elimination of abnormal events those errors can cause. Second is the elimination of routine checks while still verifying proper device operation. Third is the ability to quickly troubleshoot suspected problems.

Status at a glance. It’s been estimated that up to 80% of abnormal situations, including those that ultimately involve a loss, have human error as a contributing cause. HCD can reduce human error by a factor of 40X or more. This reduction in human error can both eliminate, and reduce the severity of abnormal situations leading to a safer overall operating plant.

Routine checks are best eliminated by having an online device asset management system such as Emerson’s AMS Device Manager monitor all devices for alerts, and not conduct device checks unless an alert is detected. This approach is usually adopted over time as experience confirms that such device asset management systems can effectively monitor devices without the need for human checks.

Many facilities don’t have extensive online systems, or have not yet built the confidence to eliminate manual checks. In these cases, the next best thing is to be able to access the device status remotely, and confirm device status at a glance. This is best accomplished by having a device landing screen (the default first screen displayed when the device is accessed) that gives complete device status at a glance.

The most frequent human – field device interaction in an operating plant is a status check. A human centered design approach would make the status display the default-landing page for each device.

There are three key factors to status at a glance. The first is that everything you need to determine device status is on this screen. The second is that color and icon shapes are consistently used to communicate status. Even if the user doesn’t read the text on the screen, color, icon shape and dials communicate that everything is normal, or if something is abnormal, where to go to address the issue.

And by using two modalities (color and shape) to show status, challenges such as poor lighting conditions or colorblind users are addressed. The third is that no detailed knowledge of the specific device or device – human interface is needed to conduct a successful status check in a few seconds. The net result is a dramatic increase in productivity and a dramatic reduction in human error.

Checking device status is the most frequent device interaction, and is usually a necessary first step in conducting other types of device interactions. Therefore, having the necessary information all on the landing page, and using color and other visual clues to make the information easily and subconsciously accessible, is the best first step in improving human – technology interaction.

Orange color and unique icon shape indicates maintenance needed but the device remains operational.

When conditions that can impact operations, quality, reliability, emissions, or safety occur, the most critical factor in minimizing negative consequences is time. Abnormal conditions tend to become worse the longer they persist. Typically, there’s a window of time with minor or no negative consequences. Past this window of time are generally a series of ever-shorter windows of time with ever increasing negative consequences. The key to successful troubleshooting is to resolve the problem before negative consequences occur.

Severity of consequences increase over time. Timely corrective action can minimize or eliminate adverse consequences. Delayed corrective action can lead to severe consequences.

Consequences can propagate across more functions as severity increases. In many cases, consequences of even total failure are limited and propagation stops prior to severe consequences.

If status at a glance indicates an abnormal situation, the next step is to successfully troubleshoot, and if necessary, repair the device. There are four key factors to successful troubleshooting. The first is to be informed there is a problem, and to know the approximate severity of the problem. The second is to localize the problem. The third is to know what’s needed to correct it. The forth is to verify the corrective action was successful in resolving the problem.

Actually solving the problem requires actionable information. It’s key that the level of information provided match the level of corrective action used to solve the problem. Since repair is typically done in the field, and time is important, corrective action should reflect what can be done quickly under adverse conditions. These conditions can include above or below grade, in bad weather, and wearing protective gear. Troubleshooting information should show pictorially what has failed, and how to repair it. It should also provide single-click access to more detailed, printable information.

In this example, a local display is not operational. Guided help shows possible failure points and what parts if any may need replacement. Since a repair technician knows all potential failure points and what field replaceable modules may need replacement, repair time is significantly reduced and the probability of successful repair is significantly increased.

In our next post, we’ll look more closely at guided configuration and setup as well as leveraging productivity across different device types and protocols.

One comment so far

Leave a Reply