Electric Grid Software Accessibility
A foundational research project to understand the types of accessibility barriers encountered by software grid operators.
Note: In order to preserve the proprietary nature of this project, I cannot share any specifics about the client, the software, or the research participants. All images on this page are stock photos and any representations of individuals are highly anonymized and fictionalized.
The Problem
Grid operators work in a safety-critical environment with unknown accessibility barriers.
- When confronted with substantial volumes of information, how do grid operators manage the environment and avoid cognitive overload?
- What barriers do operators encounter in the grid software that prevent them from working efficiently and prioritizing safety?
- What functional abilities do operators possess that must be accounted for in the grid software design?
The Research
An in-depth design review followed by several remote interviews and on-site observation.
The total number of grid operators is not very substantial, making it relatively easy to identify participants via the client. The challenge was getting operators to talk about their ordinary workflows in non-technical ways so that we could understand their challenges and barriers.
Participant Profiles: Disability
I initially developed a screening survey that used a "functional needs" approach to identifying accessibility barriers. I hoped to recruit operators to the study who had challenges with vision, color perception, reading comprehension, and memory.
It quickly became clear, though, that it was not possible to recruit as broadly as I hoped. The client served as the intermediary between myself and the grid operators, and I deferred to their access. Ultimately, many of the operators I interviewed had already spoken frequently with other client employees before, so my challenge shifted to finding a way to get these participants to talk about technical details in a non-technical way, and to relate details about their work that they might not consider obviously relevant or important.
It also became clear that, by and large, grid operators almost never have a self-identified or diagnosed disability. While initially a challenge in scoping the research, this became an asset later on.
Remote Interviews
During the first phase of research, I reviewed existing internal research from the client, including past interviews with operators. While these existing materials were not very relevant for the accessibility focus of this project, they were a critical resource for me to learn the professional terminology used by operators and other electric company workers.
Even with this substantial preparation, interviewing grid operators was initially very disorienting. These individuals work in a highly specialized and technical field, and they are used to working with colleagues who share their expertise and lingo (as is the case in any specialized industry). My primary objective was to understand barriers and challenges in operators ordinary (and extraordinary) workflows, not to get into the details of specific technical processes (which, anyhow, were already quite well understood by designers working for the client). However, in order to build rapport with participants, I needed to employ the technical lingo of the electricity grid as authentically as possible—something that I did with widely varying success, but at which I improved over the course of the interviews.
On-site Observation: Developing an Accessibility Focus
Easily the most critical part of the research was when I conducted observation for two days in an electric company's control room. My pre-research and remote interviews gave me a substantial grounding in the risks and intensity of a control room—but not the ordinariness of the daily operations.
On site, I was able to observe how operators deploy the substantial knowledge they had told me about in our interviews. Their workstations are substantial—spanning 10 screens and two separate operating systems in the location I observed—and feed operators a steady flow of information, even on "blue sky" days like when I was there. I had been prepared for the noise in the system, the stressful moments of acknowledging alarms and responding to things as the arise.
But as I shadowed a few different grid operators and dispatchers, I observed how normal all of this was for them. Which is not to say that they are not highly focused—safety is explicitly emphasized throughout their work, and operators are constantly aware of the criticality of their role in keeping people safe. Yet the workflow of the control room is steady, with space for banter and smalltalk, as in any workplace. The relationships between operators emerged as a key agent in the control room operations, as did the relationships between operators and their computers.
And in that ordinariness of the control room is where the accessibility concerns started to emerge. You see processes that are needlessly redundant, systems where information is placed many layers deep in a series of modals, information that lives in the heads of specific individuals (or on their handwritten notepads) that could be invaluable to colleagues. In short, opportunities to reduce the cognitive load that the system places on operators and to redesign the software in ways that do more to ensure ease of operability and, therefore, safety.
Insights and Analysis
Solving for cognitive accessibility will help operators with a variety of access needs and ensure robust usability and safety for the new generations of grid software.
"This job has so many opportunities to slip if your not dialed in." (Terry, senior grid operator)
Insight: Importance of inclusive design for everyone
Grid operators overwhelmingly do not identify as disabled. However, focusing on accessibility and inclusive design can improve their overall experiences with the software, making their work more efficient, safety-focused, and easier.
There are many reasons why an accessibility-first approach to design will help this population:
- Operators are mostly men, meaning that color blindness will be prevalent in the group.
- Invisible disabilities such as ADHD and dyslexia are underdiagnosed in adults and likely prevalent among operators at higher rates that we are aware.
- Disabilities, including cognitive disabilities, can be situational.
- People age into disabilities, such as diminished visual and aural acuity.
Focusing on inclusive design and accessibility ensures that the software is robust enough to support a wide variety of access needs among operators.
Insight: Focus on situational cognitive disabilities
Cognitive disability consists of a large and very diverse group of conditions and access needs, including challenges with language, memory, perception, attention, and many others. Some of the most prevalent cognitive disabilities include ADHD and dyslexia, which are generally thought to be underdiagnosed among adults. This means that the population of grid operators likely have invisible cognitive disabilities at a higher-than-reported rate.
In addition, cognitive disabilities, like all disabilities, can be permanent, temporary, or situational. The classic example of a situational physical disability is a parent holding a baby: for the time they are holding the baby, they only have one arm free to perform other tasks.
Similarly, "cognitive overload" is a kind of situational cognitive disability where a person can have so much information presented to them that they are not able to effectively process and respond to it. Alerts and inputs coming from a variety of sources can make it hard to pay attention to a task. These are hallmarks of the grid operator's normal working environment, and they indicate that operators are very likely dealing with situational cognitive disabilities regularly.
Insight: Apply accessibility design principles for ease of use
Changes of context
Grid operators frequently move between different tasks and different applications in the software. This task switching creates memory gaps as operators return to tasks and may not remember what the next step is. The system needs to support recall across different tasks, and minimize changes of context that the operator did not initiate.
Use color sparingly
Use of color is a major challenge for grid software design, in part because individual utilities have legacy color schemes that they want to maintain. However, the overuse of color makes these color schemes meaningless: operators often did not know what many of the colors in their software meant. Effective color use in the software will use color minimally, use red only in the most urgent cases, and reinforce color with text and symbols.
Manage information efficiently
Operators are bombarded with information from the grid, and they additionally generate tacit knowledge of the system through experience and notes. System knowledge needs to be shared between operators. Further, the grid software should be designed to support Next Best Action decision-making, by performing analysis of information in advance and presenting operators only what they need at a particular point in their workflow.
Conclusion: Distributed Cognition
Electric grid software is the medium by which grid operators manage an almost unimaginably vast system of electricity generation, transmission, and distribution. The cognitive load for operators is often extremely high, which increases the risk that someone will make a mistake under pressure. By applying the principles of inclusive design and accessibility, we can make the operator's job more focused and manageable.
As grid software becomes more advanced, there are new opportunities to reduce cognitive load in novel ways. By thinking of the grid as a cognitive system distributed between the software and the humans who manage it, we can find ways to maximize the capabilities of both software and human. Grid operators bring tremendous expertise to their jobs, but they are not computers and should not be expected to behave as such. We can let the software do what it does best—calculate, analyze data, summarize—and empower operators to act on that information, thus reducing cognitive load, mitigating situational cognitive disabilities, and enhancing safety and security.
