Table of Contents
The Problem Most Interface Designers Still Ignore
Step inside a contemporary factory or the control room of a power grid, and you‘ll observe a peculiarity. The equipment is state of the art. The sensors are producing terabytes of information. But the monitor on which a technician will be working for 12 hours? It looks like it was put together in 2003 cluttered, difficult to read, arranged according to what the system can display, not the operator.
It is that space where the hand of the human rises above the reach of the technical that human usability faces the challenge of overcoming technological ability that is exactly where human-centered design builds a bridge. Industrial settings are now beginning to give it their attention.
This isn‘t just a UX trend adopted from consumer applications. It‘s turning into a safety necessity, an efficiency booster and, to be frank, a competitive advantage for companies that are willing to put the effort in.
What “Human-Centered” Actually Means in a Factory Setting
It‘s Not Just About Pretty Dashboards
Aperception that the HCD of Industrial Designers is simply making artifacts more stylish or cleaner is inaccurate: this is just scratching the surface.
Human-centered design begins with understanding how humans think under stress, what cognitive load feels like when a fault occurs and where attention naturally lands in the moments after “alarm” at 3 in the morning. It means designing interfaces around humans, not bending humans to fit the interface.
In practice, this shows up as:
- Alarm rationalisation ‘letting some of the non critical alarms go’. Flood of non-critical alarms causes operator to switch off the whole lot. The alarm trail is closed down to a level that‘s meaningful.
- Provision of Contextual information displaying only information that is relevant to the task being performed (rather than all information recorded by the system).
- Clear visual hierarchy –using color, size, layout to direct the eye to important elements and to what really matters.
- A shared mental model of behavior the interactions should behave consistently throughout the system
I have looked at various interface audits from process industries, and the tendency is almost always the same: operators work around the system because the design did not take existing workflow into consideration. It is a design flaw, not a training flaw.
Examples of Human-Centered Design in Industrial Interfaces Leading to Results
Control Rooms Are the Clearest Win
This was one of the first industries to embrace this. Since the late 1990s, the ASM Consortium (Abnormal Situation Management) has been issuing standards for control room interface design. But the pace has picked up over the last five years.
It‘s not complicated: when there‘s an abnormal event and an operator reads the process display wrong, the “cost” isn‘t a hung up call or a bad checkout that‘s a safety incident, a production loss or more.
The companies that have implemented HCD in their redesign of control room interfaces can be identified by their responses to the question:
- Faster fault detection
- Reduced alarm response times
- Less operator fatigue on long shifts
- Fewer procedural errors in high-stress situations
The final one is more significant than it generally receives credit for. Cognitive fatigue does exist, and uses that require sustained interpretation, instead of rapid identification, exacerbate it.
What I Noticed When Looking at Field Device Interfaces
Field interfaces (handhelds, local HMI‘s on machines, portable maintenance devices, etc) are where HCD is the furthest behind. In my observations, most of these controls still put function count over workflow logic the technician gets a screen with 40 options when he only needed two.
What is happening now is adaptability interfaces that adapt or present different interfaces based on role, current task or equipment‘s state a maintenance technician looking at diagnostics will see a completely different interface then a supervisor looking at that same piece of equipment‘s raw historical data.
This is where the relevance of the digital simulation comes in. Once the companies have built out a digital twin of their physical infrastructure (The Rise and Types of Digital Twins is a good read if you‘re interested in the growth of this field), we then face the question of the design of the interface: how do you build an interface that allows a human to meaningfully interact with a virtual representation of a real system without bogging them down?
The answer is human-centered design principles used at every level.
The Shift That‘s Actually Hard Organizational, Not Technical
Why Good Design Gets Killed Before Deployment
But here‘s a piece of news that you don‘t see in most industrial UX articles: most of the friction is not coming from the technology. It‘s coming from the organization.
The engineering teams are directed to the hardware, and are typically responsible for choosing and installing SCADA systems, HMIs, and control screens. They make tradeoffs to decide if it is functionally complete- does it display all the data? Integrate with the PLC? Comply with the spec?
UX considerations does an operator under stress understand this in two seconds? too often come too late into the picture, if they at all.
Progress is being made by those companies that treat interface design as a multidisciplinary problem from the outset; to include operators present in the room during design reviews; usability testing including actual shift workers, not engineers; and admitting to oneself that the best person to test whether a system is intuitive is the person who designed it.
We should refer to this as organizational resistance because it is the chief reason good research does not result in deployed products.
Where Infrastructure Meets Interface The Data Center Angle
Data centers are a relatively undervalued example of this design challenge. They are physically complex spaces with heightened monitoring and operator diversity,and significantly high consequence of failure conditions.
This report looks at how the infrastructure is evolving with the promise of AI automation and edge computing. However, just as those environment become more complex, the operator interfaces controlling them are under the same HCD pressure as any industrial environment.
By the time a cooling outage begins to cascade at 2 a.m., the technician on call has to know what‘s going on within seconds not hours after sifting through a wall of sensor data. That‘s a design issue. And it‘s one where more and more human-centered approaches are being applied to DCIM platforms.
Putting into perspective the magnitude of what operators are dealing with sheds some light on the importance. The importance of data centers explains why this infrastructure is important at a systems level – which is precisely why the human interface layer should be given just as much attention as the hardware layer.
Two Insights Most Articles on This Topic Skip
The Alarm Is Not the Problem Architecture Is the Interface
Most conversations on operator mistakes in industrial environments arrive at “alarm fatigue” as the culprit. And alarm fatigue is true. But I contend its a symptom, instead of the cause.
By far the biggest reason lies in interface architecture that has perceive alarms as a set of outwardly-facing output layers bolted onto a data system, instead of thinking about a whole information world built from the ground up around the human‘s perceptions of urgency and priority.
If you design from a human-centered perspective then you ask: what does this operator need to understand immediately, in what order, and at what level of granularity? Alarms are contextual cues in a logical information layer not an overlay that yells at you.
This is a subtle, but interesting difference. It changes the design question from “how do we reduce alarm count” to “how do we create an interface where the operator always know where they stand”.
Mobile and Wearable Interfaces Are Changing the Design Surface
The second poorly-documented change is the (physical) physical. Interfaces no longer have to be all and screen and walls.
Tablets on the floor. Smartwatches to provide maintenance alerts. AR headsets to deliver guided procedures. Each of these form factors demands a total redefinition of information hierarchy, interaction design, and cognitive load management.
During my research I realized that the majority of the companies that were using a wearable interface in an industrial context were keeping the same desktop-centric design mind-set. And the result is always the same way too much information and structure.
The design surface is evolving more rapidly than our design thinking. That gap needs to be bridged over the course of the next five years.
What‘s Actually Coming Next
AI-Assisted Interface Personalization
The upcoming generation is dynamic interfaces that change in real-time depending on characteristics of the behavior of the operator – not just role-based profiles….etc.
In the event that an operator repeatedly monitors a known group of parameters during startup, the system displays that group more prominently next time. In the event that the move attention pattern appears to be processing a certain kind of alert sluggishly, the display flickers.
It‘s not science fiction–and it‘s being prototyped in aviation and nuclear right now. The industrials will be next.
Simulation-First Design and Testing
Another area to watch: the use of digital simulations to evaluate interface designs prior to their implementation. Instead of the ineffectiveness of an interface being revealed when it is actually put into use, designers guide operators through high-fidelity digital simulations of abnormal events along with prototype interfaces.
This ties straight back into the digital twin phenomenon. Improved simulation capability means that it becomes more and more practical to stress-test interface designs in real-world situations safely removed from an actual production environment.
My Opinion After Having Looked into This Area for Some Time
Human-centered design of industrial interfaces isn‘t a new concept. What‘s new is the convergence of factors more complex systems, greater volumes of data, aging workforces, and interfaces that haven‘t kept up with any of it that makes it truly urgent.
Companies that get this right all have a few things in common. They have learned to regard operators as expert users not just end users. They understand the value of investing in observational studies into real shifts rather than just requirements research. They also realise that interface re-design is not a one off project.
Who should pay attention to this:
- Engineers and Product Teams working on SCADA, HMI, DCM Platforms.
- Constantly repeating errors to the way in which operations managers
- Tech professionals & others curious how UX principles might be relevant to work outside of consumer software
- For anyone who is developing tools for complex, high risk systems
The tools are getting smarter (a lot smarter.) The interfaces need to keep pace, and that way lies through designing for the person across the screen from you.
I’m a technology writer with a passion for AI and digital marketing. I create engaging and useful content that bridges the gap between complex technology concepts and digital technologies. My writing makes the process easy and curious. and encourage participation I continue to research innovation and technology. Let’s connect and talk technology!
