Project overview
CoreTec UX Research and Redesign
CoreTec is Hitachi Energy's transformer monitoring platform, a mission-critical system used by engineers and field technicians to track asset health, manage alarms, and interpret trends for high-value grid equipment. The platform had evolved feature-by-feature over time, resulting in an interface where critical data and alarms were scattered across long, dense screens. Navigation was misaligned with operator workflows, and inconsistent visual patterns forced users to scroll, hunt, and mentally stitch together information. This increased cognitive load elevated the risk of slip errors during safety-critical operations and slowed response times during incidents.
My role:
Solo UX researcher and UX designer embedded in the R&D of transformers department.
Problem:
CoreTec had grown feature by feature over time. Critical transformer data and alarms were scattered across long screens, navigation did not follow operator workflows, and visual patterns were inconsistent. Operators had to scroll, hunt, and mentally stitch together information, which increased the risk of slip errors and slowed response during incidents.
Goal:
Create a task centered, safety aware CoreTec experience that:
Surfaces critical alarms and asset health at a glance.
Aligns navigation and information architecture with how engineers think about transformers and events.
Reduce Cognitive Load & Error Risk
Optimize for Multi-Context Use
Reduces time to locate, change, and interpret key data.
Establish Foundational Usability
Scales to 7-inch embedded displays with a usable night mode.
Responsibilities:
Planning and leading mixed methods UX research.
Mapping workflows, journeys, and existing IA.
Running heuristic evaluations and baseline usability tests.
Translating findings into IA and navigation concepts.
Creating low and high fidelity interactive prototypes.
Designing a reusable UI system and 7 inch night mode layouts.
Partnering with PM and engineering for feasibility and handoff.
Constraints and context:
8 month internship, from discovery to design handoff.
Safety critical industrial domain with strong regulatory and reliability requirements.
I was the only UXR and designer on the product, so I had to scope research to be fast, focused, and directly actionable.
All about the user :
User Research
The research was designed to diagnose the root causes of inefficiency and error in CoreTec, not just catalog complaints. I used a convergent, mixed methods approach to triangulate findings, so results were robust, actionable, and defensible to engineering and R&D managers.
Methodologies
Contextual Inquiry & Exploratory Interviews (N=3):
Method: Contextual inquiries and live system walkthroughs with 3 engineers and field technicians.
Purpose: To observe daily routines, pain points, and expectations within the actual software environment, moving beyond hypotheticals to grounded, empathetic understanding.
Output: Foundational insights into operational context, terminology, and initial workflow mapping.
Heuristic Evaluation:
Method: A formal diagnostic against Nielsen’s 10 Usability Heuristics, a recognized standard for evaluating complex systems.
Purpose: To provide an objective, expert-led assessment of systemic design flaws, complementing user data with principled analysis.
Output: Identification of 86 heuristic violations, categorizing failures into systemic themes like lack of feedback, consistency, and error prevention.
Usability Testing & In-Depth Task Analysis (N=8):
Method: Moderated usability tests with 8 participants, focusing on predefined, critical tasks (monitoring, alarm triage, configuration). Employed Think-Aloud Protocol and direct observation.the
Purpose: To meticulously deconstruct interaction sequences, identifying exact points of hesitation, error type, and breakdown. Data was synthesized into detailed workflow diagrams and journey maps.
Output: Pinpointed interaction failures, navigation breakdowns, and a clear map of pain points across key user journeys, which later served as the baseline benchmark for validating the redesigned prototype.
Survey (Quantitative) (N=25):
Method: A structured survey was deployed to 25 engineers and technicians.
Purpose: To quantify the prevalence and perceived severity of identified issues (error frequency, workload, learnability) and capture broader sentiment. This validated qualitative findings and provided scalable metrics.
Output: Statistical validation of pain points, correlation data (e.g., linking error types to user experience levels), and priority metrics.
Usability Testing and Task Analysis
Following the contextual inquiries and heuristic evaluation, I conducted a baseline usability study with 8 participants using a set of critical tasks derived from the earlier task analysis.
Procedure
Participants performed each task using the live CoreTec interface while thinking aloud. Throughout the sessions I captured:
Screen recordings
Interaction sequences
Timing and completion data
Real-time verbal reasoning through think-aloud
This setup provided a complete behavioral trace of each workflow.
Data Collection
After the sessions, I aligned the think-aloud transcripts with each participant’s recording to extract:
Task completion time
Error types and frequency
Hesitation points
Interaction loops and backtracking
This allowed me to refine the original task analysis and construct a detailed user journey based entirely on observed behavior, not assumptions.
Cross-Participant Synthesis
To identify systemic issues, I created an interaction flow for each participant, then:
Overlapped all 8 flows to map common breakdowns
Highlighted shared pain points across participants
Flagged stage-specific failure patterns
Preserved individual variations to capture edge-case issues
This convergence step isolated where the interface consistently worked against user expectations and produced a validated blueprint for the redesign.
Interaction Mapping
I created a complete interaction inventory of the CoreTec interface to understand how every part of the system behaved during real workflows.
What I did
Mapped every interaction in the system, including buttons, input fields, numeric controls, dropdowns, scroll triggered actions, confirmation dialogs, cancel and delete actions, toggles, filters, navigation links, and system state changes.
Each interaction type was color-coded, allowing me to visually distinguish control categories and trace interaction patterns across tasks.
Why it mattered
This end-to-end map helped me:
identify inconsistent patterns and redundant controls
trace how users transitioned between screens and states
flag high-risk elements that frequently produced slips
connect specific UI elements to observed pain points in the usability study
This interaction inventory became the backbone for the refined task analysis, the user journey, and the redesign decisions that followed.
Sruvey's Correlates
To validate the qualitative findings, I deployed a structured survey to 25 engineers and technicians and ran correlation analyses across experience level, session length, error types (slips, lapses, mistakes), perceived workload, and satisfaction.
Sample findings:
Experience and errors
Participants with 1–3 years of experience reported more lapses (forgetting next steps), while slip rates were nearly constant across all seniority levels, including 15+ years. This suggests slips are primarily driven by interface design, not expertise.Usage pattern and slips
Users who log in daily or monthly reported more accidental input changes, often describing tiny input fields and scroll behavior that unintentionally changed selected numeric values. This pointed directly to problematic control patterns in the UI.Session length and satisfaction
Very long sessions (often over 2 hours) combined with frequent help seeking correlated with lower satisfaction and complaints about specific areas such as Alarms and Event Logs, which we flagged as high priority targets for redesign.
Key Insights Across Methods
Synthesizing the contextual inquiries, baseline usability tests, heuristic review, and survey correlations, four systemic problem areas emerged in CoreTec’s experience:
Information Architecture:
Similar data scattered across multiple screens.
Deep navigation paths and long scrolling pages slowed task completion.
Hard for users to maintain a mental map of where they are.
Workflow steps did not match how engineers actually diagnose transformer events.
Baseline testing showed frequent backtracking due to unclear page relationships.
Data Density and Visual Hierarchy:
Critical signals and alarms lost in dense tables and charts.
Competing focal points created ambiguity in what to look at first.
Low contrast and inconsistent typography reduced scan speed.
Trend and status data displayed inconsistently across screens.
Survey correlation showed users with long sessions struggled more with visually overloaded areas.
Alarm presentation and errors:
Severity and state not always visually obvious.
Users could miss or misinterpret alarms during busy periods and long sessions.
Higher risk of slip errors in acknowledgement, configuration, and numeric inputs.
Scroll behavior and small targets make it easy to change values unintentionally.
Interaction patterns and feedback:
Inconsistent controls and layouts from screen to screen.
Weak or delayed feedback on actions, which eroded user confidence.
Critical actions look and feel similar to low risk actions, with minimal confirmation.
Error messages are sparse or overly technical, giving little guidance on what to do next.
User Personas
To turn the research synthesis into an actionable, human-centered narrative for stakeholders, I developed evidence-based user personas. These personas synthesized behavioral patterns, survey responses, and usability observations into relatable archetypes. Their purpose was to crystallize abstract system flaws into clear human stories, ensuring the team understood not just what was broken, but who was impacted and why it mattered.
Example: The Reliable Operator (Persona D)
Profile: An experienced technician with 6-10 years of tenure, conducting monthly, in-depth diagnostic sessions.
Core Need: Absolute clarity and reliability in data and alarms to make high-stakes decisions with confidence.
Key Pain Point: Despite deep expertise, required frequent peer assistance during setup and alarm verification due to ambiguous system feedback and inconsistent layouts. This highlighted a critical system-induced risk, not a user skill gap.
Design Mandate Derived: This persona directly mandated the redesign of the alarm system into a clear three-tier hierarchy and demanded explicit confirmation for all critical actions, transforming the project goal from a UI refresh to a mission-critical safety enhancement.
Starting the Design
From Research to Design Brief
To move from findings to solutions, I distilled the research synthesis into a design brief that defined what the new CoreTec experience had to achieve.
Design requirements
Grounded in user research and heuristics, the brief captured:
Reduce cognitive load and slip errors in critical workflows.
Make alarm states and transformer types unambiguous at a glance.
Align navigation with real diagnostic workflows, not system modules.
Support both desktop and 7 inch embedded displays in safety critical contexts.
Knowledge repository
I maintained a living synthesis document that:
Consolidated all findings from interviews, heuristics, usability studies, and surveys.
Tagged each issue with evidence, severity, and recommended action.
Served as a shared source of truth for PMs, engineers, and designers, so every design decision could be traced back to validated user needs.
Ideation and Scoping
With the design brief in place, I explored multiple ways to reframe CoreTec’s interaction model before committing to any single solution.
Concept development
Conducted a light competitive scan of industrial monitoring tools to understand common mental models and visualization patterns.
Sketched alternative approaches to key workflows (monitoring, alarm triage, configuration), emphasizing cognitive load reduction, progressive disclosure, and clear task sequencing.
Sketching and visualization
Turned the strongest ideas into paper sketches and low fidelity storyboards, experimenting with different dashboard compositions, alarm hierarchies, and step based flows.
Used these artifacts in quick reviews with PMs and R&D engineers to get early reactions without the weight of “polished UI”.
Feasibility and prioritization
Ran joint working sessions with engineering to assess feasibility, data availability, and implementation risk.
Prioritized concepts using an effort versus impact lens, deciding which directions to prototype first and which to keep as future iterations.
Low Fidelity Prototypes
Next, I translated the strongest concepts into interactive low-fidelity prototypes to validate the structure before visual design.
Structural Design
Built three interactive Figma prototypes, each exploring a different layout logic for navigation and data visualization.
Verified how the redesigned information architecture would scale across the 60 plus screens targeted for the new CoreTec experience.
Early Validation
Walked managers and developers through realistic scenarios using the low fidelity flows.
Feedback confirmed that the revised task sequencing was more predictable and significantly reduced navigation errors and backtracking.
Outcome
The most successful prototype became the structural backbone for all subsequent high fidelity screens, ensuring that visual design work sat on top of a tested IA and flow.
Design System and Custom Iconography
Problem:
Screens used inconsistent controls and visual patterns across modules.
Legacy icons were ambiguous, and operators often misread transformer types and actions.
Without a reusable system, it was difficult to scale the redesign across 60+ screens while keeping behavior consistent.
Solution:
Designed a comprehensive UI component set: cards, tables, filters, inputs, controls, alerts, and layout grids.
Defined a visual system that emphasizes contrast, hierarchy, and legibility in dense operational contexts.
Created a custom iconography set with distinct shapes and metaphors aligned with field terminology, optimized for both desktop and 7-inch embedded displays.
Result:
Interfaces now share consistent patterns, reducing visual noise and ramp-up time for new users.
Icons are immediately distinguishable, lowering the risk of misconfiguration and interaction slips.
The design system gives engineers a stable library to build from, keeping future features aligned with validated UX and human-factors principles.
Refining Design
High Fidelity Prototypes
I then translated months of research and structural work into a pixel precise, interactive Figma prototypes.
Key Design Decisions
Human Centered Data Design:
Replaced generic progress bars and scattered numerics with multi layer circular gauges that can display up to ten parameters in one compact visualization.
Gauges aggregated key metrics such as temperature, oil level, and load, enabling technicians to compare states at a glance and spot anomalies faster.
Safety Critical Alarm System:
Introduced a three tier alarm hierarchy (green nominal, yellow caution, red critical) with distinct color and shape coding.
Spatially separated alarms from routine readings to reduce alert fatigue and increase recognition accuracy under stress.
Information Architecture and Workflow Logic:
Reorganized more than 60 screens into a left to right step flow that guides users from monitoring to diagnostics to configuration.
Applied consistent panel structures, global navigation anchors, and persistent headers to help users maintain a stable mental model of where they are in a task.
Terminology and Cognitive Accessibility:
Simplified technical language into plain, action-oriented labels.
Paired text and icons for recognition rather than recall, reducing dependence on manuals and training.
Dark Mode for Low Light Operations:
Designed a contrast-optimized dark theme based on visibility research for night work and confined service areas.
Improved readability and reduced eye strain in field tests during extended, low light monitoring sessions.
7-Inch Embedded Display:
Created a minimal, touch friendly layout for the transformer mounted 7-inch screen, showing only essential indicators with large, high contrast typography.
Sized and spaced touch targets for technicians wearing gloves, keeping the interface reliable in outdoor and harsh field conditions.5
Mockups
High fidelity screens demonstrating the final CoreTec UX experience.
I translated the validated architecture and design system into a high fidelity Figma prototype with 60+ screens for desktop and 7 inch embedded displays. The mockups show how the new components, multi layer gauges, three tier alarm model, and revised workflows come together in real monitoring, diagnostics, and configuration scenarios. Each screen is built from the same reusable system, so interactions, visual hierarchy, and terminology stay consistent across the entire platform.
Outcome
Outcome
The redesign translated months of mixed-methods research into a cohesive, safety-critical interface validated through iterative testing.
Delivered
Research, design, and systems shipped for CoreTec’s UX transformation.
Research & Insight:
Created a single research repository that stores contextual inquiries, heuristic findings, usability videos, correlation diagrams, and synthesized journey maps so future teams can access evidence in one place.
Workflow & IA Blueprint:
Produced complete workflow diagrams, interaction maps, and a reorganized information architecture that describe how technicians actually move from monitoring to diagnostics and configuration.
Design System & Iconography:
Shipped a reusable design system with layout grids, UI components, and a custom icon set tuned for industrial use, encoding human-factors decisions directly into components.
High-Fidelity Prototypes
Built over 60 high-fidelity Figma screens plus dedicated 7-inch dark-mode layouts, ready for implementation and used as the benchmark prototype in post-usability validation.
What I did
Research, design, and systems shipped for CoreTec’s UX transformation.
Led Mixed-Methods Research:
Planned and ran contextual inquiries, heuristic evaluation, usability tests, and a survey, then triangulated the data into prioritized issues and design requirements.
Defined Structure & Flows
Redesigned CoreTec’s information architecture and task flows, including the left-to-right step model, to reduce navigation depth and make workflows predictable.
Designed Safety-Critical UI:
Created multi-layer gauges, a three-tier alarm hierarchy, scroll-safe numeric inputs, and dark-mode / 7-inch layouts that directly target slip, lapse, and comprehension issues.
Drove Cross-Functional Alignment
Facilitated working sessions with PM and R&D, reviewed feasibility, scoped iterations, and handed off specs so engineering could implement the redesigned system with minimal ambiguity.
Impact (Post-Validation):
To validate the impact of the redesign, I conducted a post-usability benchmark with 5 selected experienced technicians, measuring task time, slip/lapse errors, and learnability using the redesigned prototype. This provided a direct, comparable baseline against the original system (Perecentages are rounded.)