There is a tendency to associate the profession of User Experience with consumer-facing Web sites and applications. And why not? After all, social media and ecommerce experiences are a constant part of users’ lives—even those who are also UX designers. These experiences represent desirable activities such as buying products and interacting with friends and family. There is high demand for such experiences, which, in turn, draws the collective focus of the UX community. Fair enough. But the profession of User Experience also provides value in unexpected places that exist at the periphery of modern consumerism. In this three-part series, I’ve discussed one such unexpected place—the industrial environment. Humans—who help manufacture the goods we enjoy—must be productive and are no less deserving of experiences that make them more efficient, effective, and satisfied in their jobs.
In Part 1 of this series, I explained that industrial automation is more human facing than you might think. Then, I described how the industrial environment itself presents difficult challenges for UX designers to overcome when designing software for human-machine interfaces (HMIs), covering both plant-floor and control-room environments. Finally, I shared some key principles of effective HMI design that apply to both environments.
In Part 2, I explained some methods that industrial-automation UX designers can use to combat the challenges of designing for isolated customer sites and localizing information for global users.
Now, in Part 3, I’ll share some methods that industrial-automation UX designers can use to overcome challenges relating to
designing for designers
getting access to users
handling the complexities of competitor or legacy products
Designing for Designers
Many industrial-automation software products target personas representing people who design their own systems. One such persona is the Human Machine Interface (HMI) designer who creates visualizations for downstream users such as operators, maintenance engineers, and plant managers—often outside the direct influence of UX professionals. No direct UX involvement is a bad thing, right? But it’s the cold reality that industrial-automation UX designers must find ways to achieve good outcomes by working through HMI designers. I’ll share some challenges that occur in working with this unique persona and describe how industrial-automation UX designers can extend their influence by contributing to the HMI designer’s success—and thus, the user’s success.
Matching the User’s Mental Model
Industrial-automation UX designers must help HMI designers create experiences that will fit the user’s mental model, not the system model. While industrial systems are complex, the user’s experience and perception of a system need not be complex. So why make it more complex than necessary? A literal depiction of a system inundates users with unwanted information that negatively impacts their cognitive load.
For example, Piping and Instrument Diagram (P&ID) drawings, similar to that shown in Figure 1, are common in industrial environments. They show the roles and relationships of piping and control vessels in the process flow. They are useful as reference material when testing system continuity and during the initial build. Unfortunately, many HMI designers simply repurpose such schematics when creating running HMI visualizations—putting their own convenience above that of users. While P&IDs do provide a visual representation of the system and process, using these complex drawings can be problematic because they aren’t designed to support the mental model of the user monitoring a system.
In contrast, Figure 2 shows a system visualization that is more intentional in its design and helps users quickly understand the state of the system, as well as what information is most important or needs their attention. Users who are monitoring a process can better understand a visualization that arranges related content into logical groupings—using Gestalt principles—that the human brain can more efficiently parse. Varying the intensity, contrast, size, and shape of the elements on the display helps make its overall information architecture more apparent.
Image source: The High Performance HMI Handbook, by Bill Hollifield, Eddie Habibi, Dana Oliver, and Ian Nimmo
Of course, not all HMI designers have the same background, and their level of experience and expertise varies. They perform their work in an office cubicle like any software designer or developer—not on the plant floor where the application will ultimately run. Therefore, it’s the responsibility of the industrial-automation UX designer to educate those who create visualizations for downstream users who are unlike themselves.
Help HMI designers to focus their visualizations on the user’s environment, goals, and tasks. Users on the plant floor—who are responsible for a single machine—don’t need a detailed HMI visualization that provides information that is not related to that specific machine. In contrast, operators in a control room need a broader visualization because the scope of their focus is the entire operation. However, they don’t benefit from seeing unnecessary information either. Operators in a control-room setting usually monitor several displays at once, so HMI designers must design the information in such a way that it improves operators’ ability to comprehend all of it.
Unfortunately, there is no silver bullet to help HMI designers create a design that matches the way users mentally perceive a system. This is where observing users at work or conducting a contextual inquiry can help. However, it’s not always practical for Industrial-automation UX designers to gain access to users—or even for HMI designers. I’ll discuss that challenge in “Getting Access to Users.”
Giving HMI Designers a Little Help
Many scenarios in the industrial-automation domain involve HMI designers laying out and styling topological system views, buttons, state indicators, and radial gauges. While it’s not possible to look over the shoulder of all HMI designers to ensure they’re designing responsibly, it is helpful to arm them with smart defaults that aid their success—the design of which is in the industrial-automation UX designer’s sphere of influence.
A typical HMI design application gives the user—in this case, the HMI designer—the ability to drag and drop various predesigned widgets—perhaps including depictions of tanks, vessels, pumps, buttons, and gauges—onto the main canvas that will become the runtime display the operator will view when monitoring a process. It is important to ensure that these predesigned objects offer the proper contrast, scale, luminance, and text sizes by default.
While giving HMI designers some control is desirable—and people in HMI design roles are highly competent—simplifying their work is often a better approach. HMI designers would be all too happy to adorn their designs with details such as color, gradients, and animations. So, while primitive default objects should be sufficient, the components that make up an object should provide contextual information that conveys its appropriate meaning in the most efficient way. Often, this means doing more with less.
Doing More with Less
When designing visualizations, boil your designs down to their most essential elements. Simplicity is the ultimate sophistication—and, as I discussed in Part 1, simple doesn’t just mean more whitespace. A well-executed, simple, functional design will generate its fair share of excitement in the boardroom.
The work of industrial-automation UX designers is in perpetual competition with demoware, which usually comprises fancy displays, with gradient-laden depictions of physical plant assets. While these displays may go over well in customer demos, they don’t communicate data as efficiently as they should.
Figure 3 shows four examples of a vessel widget a designer might use in an HMI visualization. While each example occupies the same amount of screen space, what example conveys the most information in an effective, efficient way? Not the depiction of the shiny, burst vessel, complete with jagged edges, at the far left. It’s the version at the far right, which combines the general shape of a tank, with upper and lower limits, a historical trend line, and a numeric time unit—2 Hrs.
Image source: The High Performance HMI Handbook, by Bill Hollifield, Eddie Habibi, Dana Oliver, and Ian Nimmo
When designing workflows that let the user manipulate the properties or look and feel of user-interface elements, offer smart defaults that make proper use of luminance contrast, information hierarchy, and scale. This will ensure that the product does right by the user.
Always ask yourself whether you’re letting aesthetics get in the way of creating the most usable experience. A good design may not be glitzy or generate excitement in customer boardrooms. But it’s probably not your job to design attractive demos. Your job is to design the most efficient, effective, satisfying product for the user.
Getting Access to Users
Recruiting users for usability studies is one of the biggest hurdles industrial-automation UX designers face. The specialized nature of users’ work in an industrial environment greatly reduces the pool of candidates for participation in an effective usability study. Plus, industrial users’ roles are highly nuanced. For example, a maintenance engineer has a very different set of tasks and responsibilities from a plant-floor operator or an HMI designer.
The industrial-automation industry also has requirements around safety, security, and intellectual property that few other product niches share. An industrial-automation UX designer who hopes to conduct a usability study generally needs to travel and should expect to sign nondisclosure agreements to protect customers’ intellectual property. Sometimes such requirements can render a field study infeasible, so industrial-automation UX designers must be creative in getting user feedback. The following methods can help.
Trade Shows and Conferences
Annual industrial-automation trade shows and conferences provide an effective means of tapping into a potentially high number of users at a single event. By securing a back room or designated floor space at an event, it’s possible to get quality feedback on a design solution from a relevant pool of participants.
A user-research panel is an efficient way of gaining access to a large pool of users. Despite their requiring some prework and setup, user-research panels pay dividends down the road if users opt in—agreeing to participate in occasional remote studies or surveys. Such panels can also be a byproduct of attending a trade show or other event. The trick is to effectively recruit and incentivize potential participants to join your panel. Sometimes the idea of helping to shape the future of a product they use frequently is enough. Sometimes it isn’t. But the information you obtain—about potential panelists’ job details, the products they use, how frequently they use them, and their overall experience level—can itself be worth the effort. This information will help you to match the right user with the right research or validation activity.
Remote Usability Studies
Why are trade shows and user-research panels useful? Because they help you establish inroads for future usability studies. A remote usability study can be a good option because it avoids on-site entanglements involving safety, security, and intellectual property. Remote collaboration tools such as Skype are more ubiquitous than ever and, depending on the roles of industrial users, they may have access to such tools, which can help you quickly set up usability studies that you can record.
Another benefit of a research panel is that it lends itself well to online surveys. It’s relatively simple to send a targeted survey to an established group of users, in a specific role, that focuses on a topic that doesn’t necessitate a UX professional’s presence at their work site. A simple request for feedback on, say, the effectiveness of an icon system, can make good fodder for a short survey.
If you have reliable access to users, embrace it and use it to your full advantage. You have a valuable opportunity to continually validate and test the effectiveness of your design solutions. However, if you have limited access to users, consider putting together a user-research panel, which can help facilitate occasional surveys or remote studies. You may be able to do this by seeking out events at which many users congregate at once. Go where the users are. If you can’t facilitate an on-site study during the event, the next best thing is collecting potential participants’ information so you can connect with them later on, when you need to conduct a remote study.
Handling the Complexities of Competitor or Legacy Products
The factory floor is a mashup of disparate hardware and software from various vendors and at various stages in their lifecycle. No single vendor supplies a customer’s entire plant, so original equipment manufacturers (OEMs) design and build machines that are interoperable with specific vendor’s hardware devices and software applications. Therefore, industrial-automation UX designers must anticipate users’ daily exposure to competing hardware and software products.
Supporting Users’ Existing Knowledge
Many software user-interface (UI) patterns leverage the knowledge users already possess, so competing products likely use those patterns. For example, the proper usage of UI components such as radio buttons, check boxes, drop-down menus, and list boxes is an obvious way to help ensure users don’t encounter any unnecessary friction when using a product. Another way of capitalizing on users’ existing knowledge is adhering to established domain-specific layouts and standards that have gained a foothold over the years. Many industrial systems and applications are decades old, and they live on despite modern software improvements.
So why are decades-old software applications still in use? Because a new software solution or a revision to update an existing application is undesirable in a domain where downtime is the equivalent of hell freezing over. If the installation of new software is unscheduled, it can cost millions of dollars to take a production line or machine offline to upgrade its software. So it’s necessary to make plans to temporarily offload production to another plant-floor asset to keep production humming along. So guess what risk-averse plant managers will do if something isn’t broken? Nothing. As a result, legacy software is endemic in plants everywhere. Plus, plant workers themselves are unlikely to tolerate painful changes that force them to adapt their daily work practices.
So what is an industrial-automation UX designer to do? How is it possible to introduce software improvements to customers who are wary of change?
Embracing the Natives
It would be foolish to believe that a new user experience could be the sole driver provoking behavioral change in risk-averse customers—business strategy is also a key driver. However, there are practical ways in which industrial-automation UX designers can ease the friction inherent in the adoption of new software. When you’re designing an installable desktop or mobile application, one way is to use the familiar native components and patterns of the operating system. It is important to resist the urge to invent new patterns or refactor these native components unless there is a much better approach than what the native operating system offers. If you use familiar components and patterns, users will likely understand their purpose and use. Furthermore, the operating system supports these native components well, and since they’re automatically updated along with the operating system, they’ll generally be more bug free than custom solutions. From the industrial-automation UX designer’s point of view, the refactoring of custom components and patterns to keep pace with modern design paradigms could become unwieldy and take their focus away from what’s most important—designing better workflows.
In an industrial environment, the risk—and cost—of a designer’s going rogue can be high, especially if the resulting experience requires a behavioral shift from users. But it’s possible that it might be even higher in other domains. Plus, a system’s degree of customization can vary and complicate things further. For example, while changing the border color of a drop-down menu component to match your brand’s color palette is benign, the impact of a change that would affect a menu’s behavior would be severe. So where do you draw the line? According to this excellent article by Jamie Appleseed, asking the following question is a good litmus test: “Will the interaction or presentational benefits of a custom UI component offset its implementation and maintenance costs and outweigh the reduction in recognizability?”
When considering how users will experience your product, think about what they might have been doing before using it. To what other products have they been exposed? Might a workflow that precedes the one you’re designing or a competitor’s product that resides outside your sphere of influence impact your user experience somehow? If so, significantly customized patterns or components might not be the best approach because they would force users to learn something new, ultimately adding to their cognitive load. So, unless you’ve devised a groundbreaking new way of doing things that is clearly superior and would engender immediate user acceptance by demonstrating savings in time, effort, and money, stick with the standard approach. Otherwise, your solution might cause user frustration—or users might reject it outright. In any case, users are great at finding ways to circumvent newly imposed solutions that were supposed to help them, but failed to do so.
The industrial-automation domain is fraught with many complexities, but industrial-automation products are more human facing than you might think, so overcoming these complexities is well worth the effort. Not all of the human workers have disappeared from the plant floor—despite the buzz around artificial intelligence (AI) and the Internet of Things (IOT). These human users are no less deserving of quality experiences than someone using a consumer-facing Web site or application. They’re interacting with these complex control systems and machines to manufacture the goods that all of us use.
My intent in writing this three-part series was to give UX professionals—working in any domain—insights into the practical challenges that industrial-automation UX designers must overcome to help their users be more effective and efficient in their jobs. Perhaps you’ve found that you can relate to many of the challenges I’ve described. After all, the challenges of localizing content, designing for offline scenarios, and dealing with competing products are not unique to industrial automation. But there may also be some obstacles you’ve never encountered—such as accounting for distracting environments, overcoming customers’ reluctance to upgrade legacy software, and conjuring up creative ways of getting access to especially hard-to-reach users.
The next time you scour the Web for UX best practices, you’ll no doubt find the usual articles about designing consumer-facing experiences. But I hope you’ll remember this series of articles, which explains how UX professionals can provide value in unexpected places.
Jon has a degree in Visual Design from the University of Dayton, as well as experience in Web development, interaction design, user interface design, user research, and copywriting. He spent eight years at Progressive Insurance, where his design and development skills helped shape the #1 insurance Web site in the country, progressive.com. Jon’s passion for user experience fueled his desire to make it his full-time profession. In 2013, Jon joined Rockwell Automation, where he designs software products for some of the most challenging environments in the world. He is UX lead for a revolutionary analytics appliance for users on the factory floor. In addition to his Fortune-500 experience, Jon has contributed his skills to a real-estate startup. Jon rounds out his time by writing and reading anything he can get his hands on. Read More