top of page

Home / Modernizing a Legacy Steel Production System

Back

Modernizing a Legacy Steel Production System

  Project Summary  

Client
Domain
Role 

​Users

Tools

: Leading Steel Manufacturing Company
: Industrial Process Management / Manufacturing

: Senior Technical Lead – UX Design (Responsible for User Research, Usability and Prototyping)

: Production Plant Engineers and Support staff managing steel processes

: Figma

  User Interiews  
Screenshot 2025-06-04 at 5.04.44 PM.png
  Business Objective  
Modernize a legacy system to improve usability, reduce friction in workflows, and increase employee satisfaction.
  User Roles Involved  

  • Production Plant Engineers  –  Optimising technical processes and Analysing system performance

  • Production Supervisors         –   Monitor real-time progress and manage production schedules.

  • Shift Operators                        –   Input data, Track machinery status, and log production events.

  • Quality Control Inspectors   –   Review and validate steel quality parameters.

  • Maintenance Personnel        –   Log equipment issues and downtime reports.

   Why SUS Survey  
We began with a System Usability Scale (SUS) survey to quickly capture users’ perceived usability of the existing interface. Conducted with 1500 system users using a 5-point Likert scale, the SUS provided a standardized usability score that served as a strong benchmark and helped identify key focus areas for deeper evaluation.
​

1 – Strongly Disagree, 2 – Disagree, 3 – Neutral, 4 – Agree, 5 – Strongly Agree.​​

  SUS Questions  
  1. ​I think that I would like to use this system frequently.
     
  2. I found the system unnecessarily complex.
     
  3. I thought the system was easy to use.
     
  4. I think that I would need the support of a technical person to be able to use this system.
     
  5. I found the various functions in this system were well integrated.
     
  6. I thought there was too much inconsistency in this system.
     
  7. I would imagine that most people would learn to use this system very quickly.
     
  8. I found the system very cumbersome to use.
     
  9. I felt very confident using the system.
     
  10. I needed to learn a lot of things before I could get going with this system.
System Usability Scale Results.png
  Calculated SUS Score (Legacy System) - 49.2 / 100  
This indicates below-average usability — a common threshold is 68. So this score shows clear room for improvement.
  Auditing the Legacy System  
Following the SUS survey insights, we conducted a Heuristic Evaluation to interpret usability concerns from an expert perspective. Using Nielsen’s 10 Usability Heuristics, we systematically reviewed the interface to identify violations related to consistency, feedback, error prevention, Visual Hierarchy, Layout and Error Prevention. This helped us quickly flag critical UI/UX issues that could explain the low usability scores and guided the focus for further user research.
  • Reviewed 40+ operational screens across 5 key modules to map complete user workflows and identify redundant navigation steps.
     
  • Applied Jakob Nielsen’s 10 usability heuristics to uncover 25+ critical issues related to layout inconsistency, missing feedback, and poor error handling.
     
  • Found that key tasks often required 8+ clicks and up to 3 pop-ups—indicating deep navigation paths and excessive user effort.
     
  • Annotated 40+ issues with screenshots and categorised them by core usability themes such as navigation depth, redundant actions, and cluttered layouts.
     
  • Many screens displayed a low visual hierarchy, tightly packed fields, and dense information layouts, reducing clarity and increasing cognitive load for users.
Ladle_Waiting_Converter_1_edited.jpg
 Existing Legacy System  
  User Survey  
Building on the expert findings, we designed a custom user survey to validate and prioritize the issues identified during the heuristic evaluation. Targeting core users such as Plant Engineers, Supervisors, Shift Operators, QC Inspectors and Maintenance persons, the survey explored real-world usage patterns, task challenges, and feature satisfaction. This helped us understand which expert-flagged issues truly impacted users and revealed additional pain points from day-to-day operations.
  User Survey Questions and Feedback for Quantitative Result  
  1. ​​How easy is it to locate key features or functions in the system?
     
  2. Do screen layouts reflect the actual workflow in your plant tasks?
     
  3. Do you repeat unnecessary steps to complete a task?
     
  4. How often do you enter the same data multiple times
     
  5. Does the system clearly explain errors and how to fix them?
     
  6. Do error messages help you resolve issues on your own?
     
  7. Have you noticed layout inconsistencies across modules?
     
  8. Do you find yourself scrolling excessively to complete tasks or find information?
     
  9. Do pop-up windows disrupt your workflow?
     
  10. Are fonts, buttons, and colours easy to read and use?
     
  11. Do you feel overwhelmed by too much information or clutter on any screen?
     
  12. Overall, how satisfied are you with the system’s usability?
Graph.png
  Click to view all questions with feedback
  User Interiews  
Based on survey insights, we conducted in-depth user interviews with core participants to explore the root causes behind key usability challenges. These sessions provided rich context on user workflows, mental models, and specific frustrations, allowing us to uncover gaps not visible through surveys or expert reviews. 
  User Interviews – Key Findings  
user.png
  User Interiews Questions for Qualitative Result  
  1. Can you walk me through your typical day using the current system?
  2. Can you walk me through a recent task you did on the system? Where did it feel smooth or frustrating?
  3. Does the screen match how you actually work on the floor?
  4. Which steps in your workflow feel repetitive or unnecessary?
  5. Can you describe situations where you had to re-enter the same data across screens?
  6. Tell me about a time you saw an error. Did the message help you fix it?
  7. Do you find yourself scrolling too much to complete a task or find information?
  8. Do pop-ups interrupt your work? Any example where it caused trouble?
  9. Is anything hard to read or use during night shifts or under poor lighting?
  10. Which screens feel too cluttered or overwhelming to use efficiently?
  The user interview helped us double-confirm key usability issues identified in survey and expert review,   while also uncovering deeper workflow challenges and unmet user needs that weren’t visible through quantitative and expert review data alone.  
  Triangulated Research Approach  
We followed a triangulated research method—using SUS surveys, heuristic reviews, user surveys, and interviews—to identify UX pain points and guide design improvements.
triangle approach.png
  Wireframe Process  
Based on the key insights from research methods, I designed low-fidelity wireframes in Figma, focusing on Minimised Pop-ups, Reduced Scrolls, Improved Visual Hierarchy, Easy Navigation, Minimalist Layouts, Better Error Prevention Mechanisms, Consolidating Repeated Fields, and Enhancing Screen Flow.

I created 2–3 layout options per task, which were reviewed with Development, BA and Plant Operations teams to ensure alignment with real-world processes.
  Design System Creation (Atomic Design Approach)  
Design System_blur.png
To bring clarity and consistency to the redesigned interface, I developed a modular design system following the Atomic Design methodology—structuring UI components as Atoms, Molecules, Organisms, Templates, and Pages—while aligning with the company’s existing brand guidelines.

I focused on standardising key UI elements—like Buttons, UI Patterns, Font Styles & Sizes, Forms, Tables, and Colours—so the experience felt familiar. This also made it easier to scale designs across modules and helped the dev team implement changes faster.
 
  Sample High Fidelity Design  
Steel-Blur.jpg
Once the wireframes were finalised, I developed interactive high-fidelity prototypes in Figma for key workflows.

These screens followed the updated design system and used real field labels and user data from the legacy system.

The prototypes were shared with IT and plant teams for walkthroughs, helping them visualise the final interface and validate improvements in clarity, navigation, and task efficiency.
  Accessibility Enhancements Based on WCAG 2.1 (Level A)  
As a Senior Technical Lead, I recommend & integrated WCAG 2.1 Level A accessibility standards into the redesigned system, tailoring features to real-world usage scenarios:
 
  1. Keyboard Navigation: Enabled full keyboard operability for users working in hands-busy industrial environments.
  2. Adjustable Text Sizes: Designed flexible typography to accommodate ageing employees or those with visual strain.
  3. Voice Input Support: Proposed audio input options for frequent data-entry fields to reduce manual workload.
     
This approach ensured the interface remained inclusive, context-aware, and aligned with the daily challenges of the steel plant workforce.
  Final Outcome  
The redesign of the legacy manufacturing system resulted in measurable improvements in usability, efficiency, and user satisfaction
  1. Streamlined user workflows by removing redundant steps in frequent tasks like batch updates and quality checks.
  2. Pilot users completed routine tasks (e.g., logging batch records) 20% faster than in the old system—validated through timed task simulations.
  3. Applied consistent layout patterns, increased whitespace, and introduced visual hierarchy—making screens easier to scan under factory lighting and shift conditions.
  4. 90% of pilot users rated the new interface as significantly more intuitive; pain points like repeated data entry and poor readability dropped sharply.
  5. Implemented WCAG 2.1 Level A: keyboard navigation for hands-free use, Audio Input, and text resizing for ageing workers.
  6. The improved interface is projected to save the company over 1,000+ hours annually, reducing error rates and training time.
bottom of page