Enterprise UX | Applied case study
ADAIR :
Designing time-critical flight widgets for pilots.
Overview
Project Summary:
ADAIR was a large-scale aviation research project. My contribution focused on conceptualising, designing and testing a dedicated cockpit module for time-critical decision-making. My work centred on one core problem: how do you present time-critical information in a way that improves pilots' situational awareness and makes it easier for them to comprehend that information, without adding to the cognitive load they are already managing?
Project Type: Applied
My Role: Product Designer
Domain: Aviation
UX Concepts: User-centered design, heuristic evaluation, iterative design
Tools used: Figma, Miro, NASA Task-Load Index Questionnaire, User Experience Questionnaire
Results: Improved situational awareness by 42%, experience by 54%, reduced workload by 28%.
PROBLEM
Pilots struggle with compromised situational awareness and high cognitive load when monitoring time constraints.
In most cockpit systems, time-related information is scattered across multiple displays and shown as raw numbers. That means pilots are doing mental calculations in their heads during some of the most demanding moments of a flight. The risk isn't just inefficiency. It's errors when there's no margin for them…
USERS & CONTEXT
What pilots actually need (Job to be done framework - JTBD)
At a glance awareness: When monitoring a flight, pilots need time-critical information presented clearly so they can maintain situational awareness without unnecessary cognitive load.
Everything in one place: When managing multiple parameters, pilots need information grouped logically so they don't have to scan across screens and lose awareness.
Visual over numerical: When tracking time constraints, pilots need visual cues over raw numbers so they can understand impact immediately without mental calculation.
Alerts that can't be missed: When anomalies occur, pilots need alerts that stand out clearly so they can react quickly under pressure.
Guidance not just warnings: When handling anomalies, pilots need direct actionable guidance so they can make informed decisions in high-stakes moments.
APPROACH
How the design evolved...
This wasn't a one-shot design. It went through low and high fidelity prototyping, heuristic evaluation with Human Factors/Human-Computer Interaction experts, and two rounds of usability testing with real pilots. Every iteration was shaped by what we heard and observed in those sessions.
Click on the image to enlarge.
DECISIONS IN ACTION
Design decisions and why they matter…
Each decision came from a specific user need and was validated through testing. Nothing was added without a reason, every decision was purposeful.
A single floating module for quicker comprehension
Pilots were scanning across multiple screens to build a complete picture, fragmenting their situational awareness. I brought everything into one persistent floating module, placed directly in their line of sight on the navigation display. One place to look, not five.
Graphical representation for better monitoring
Raw numbers require interpretation under pressure. I introduced a graphical representation alongside the existing numeric values, giving pilots a faster way to read status at a glance without removing the precision they rely on for confirmation. I designed and evaluated two layouts, linear and radial, to find the most effective way to present this hybrid approach.
Actionable Suggestions for Rapid Decision-Making
When something is off, pilots need to know what to do next, not just that something is wrong. The suggestion box translates system states into clear next steps, written in plain language.
Visual Time Cues to Reduce Mental Calculations
Instead of showing raw time values and leaving pilots to calculate the difference, I introduced visual cues that communicate impact directly. Less math, faster decisions.
Emphasized Alerts for Faster Anomaly Recognition
Anomalies need to be impossible to miss. Colour coding makes critical states stand out from everything else on screen.
Clear wording tells pilots not just what happened but what to do about it.
DESIGN EVOLUTION
How the design changed between iterations
The first usability study was clear. The initial design was adding cognitive load rather than reducing it. The second iteration focused on fixing that, prioritising clarity and faster at-a-glance comprehension over adding anything new.
Iteration 1
What wasn't working?
Mixed “must-know” and “nice-to-know” information, increasing cognitive load.
Wide horizontal layout competed with primary flight content.
Missing clear exit/cancel affordance reduced sense of control.
Visual density made rapid scanning difficult in time-critical moment.
The green rectangular design caused confusion, creating ambiguity and reducing confidence.
The suggestion was too vague and did not clearly support decision-making.
Iteration 2
What changed and why it matters
Focused on essential information only ("must-know"), lowering cognitive effort.
Compact vertical layout freed space and improved readability.
Clear icon-based affordances reinforced user control.
Stronger hierarchy enabled faster at-a-glance comprehension.
The bracket-based design was clearer than a box, improving interpretation and confidence.
The suggestion was rewritten to be more specific, making the next action easier to understand.
Note: The radial design followed the same information model as the linear view, with changes focused primarily on layout rather than visual or functional behavior.
DESIGN RATIONALE
Why I avoided certain common choices
Icons vs. Standard Terminology
Pilots are trained on terms like GS, RTA, and ETA. Replacing them with icons introduces interpretation risk in moments where there's no time to second-guess a symbol. Standard terminology preserves clarity and trust.
Typography
Monospaced fonts support faster scanning of numeric data. Consistent character width reduces misreads and aligns with established avionics standards.
Colour Usage
In aviation, colours carry fixed meanings. Red means critical. Adding more colours risks breaking existing mental models that pilots rely on under pressure. A neutral palette keeps the emphasis meaningful.
Why Not Use Red for Warnings?
Red is reserved for immediate critical actions. Using it for caution states would dilute its meaning. Amber communicates urgency without overstating it.
RESULTS
What changed after the redesign
Measured using SAGAT for situational awareness, NASA-TLX for cognitive workload, and UEQ for user experience.
Situational Awareness
Design 1 (linear-view): 42% improvement
Design 2 (radial-view): 37% improvement
compared to Design (0) baseline.
Pilots understood what was happening around them faster and with more confidence.
Perceived Cognitive Workload
Design 1 (linear-view): 28% reduction
Design 2 (radial-view): 27% reduction
compared to Design (0) baseline.
Pilots processed information and responded to changes with less mental effort.
User Experience
Design 1 (linear-view): 43% improvement
Design 2 (radial-view): 54% improvement
compared to Design (0) baseline.
The interface was easier to use and understand, even under time pressure.
User Preference
Participant comments provided additional insight into what drove preference beyond the rankings.
"The way the information is placed for Design 1 and 2 just makes it easier to take a look and understand what’s happening. This also gives me more time to look at other information present.”
— Pilot A
"I don’t like doing mental calculations; though I am used to the baseline design and to such mental calculations, I would prefer not doing them any day."
— Pilot B
" I love the idea of the suggestion box, it’s very helpful. It directly gives me values for speed change which is great!"
— Pilot C
KEY INSIGHT
The thing that surprised me…
Both designs improved performance over the baseline. But pilots consistently preferred the radial view even though the linear view scored slightly higher on measurable metrics. Visual clarity and perceived ease of use influenced preference just as much as objective performance. Sometimes how something feels to use matters as much as how well it performs.
“Designs 1 and 2 were both fine performance-wise, pretty similar really. But I definitely preferred the look and feel of Design 2; it was just more appealing.”
— Pilot D
RESEARCH PUBLICATION
This work was accepted at the Human Computer Interaction International (HCII) Conference 2026, and will be published in July 2026.