Praveen Manchi

Wind Turbine Blade & Tower Inspection - GE

Served as staff UX/UI designer for GE Renewable Energy Reinspection project using ML (machine learning) and ADR (advanced data retrieval) to automate the inspection of wind turbine blades. Designed a user-friendly application that leverages best user experience design and advanced Layout to optimize time to do inspection by detecting, validating the blades.
Interactive UI Collage
(Withheld due to NDA – Imagine a dynamic mosaic of industrial dashboards, defect detection interfaces, and ML-driven workflow visualizations)

Overview

This platform revolutionizes industrial quality assurance by integrating ML into NDT processes, enabling faster and more accurate reinspections remotely in digital environment using UI which reperent the phiscal world environment in a digital environment and creating a seamless user experience to perform inspection in minimal time.

It supports real-time data (ADR's), automated image reviewing, reporting, and change management tracking, reducing manual errors and accelerating asset lifecycle management.
Team: UX/UI Designer (Me🙋🏻‍♂️), Data Scientists, Engineers, Product Managers, QA, renewable inspectors & engineers.
Timeline: Jul 2022 – Mar 2025
Domain: Industrial AI, Machine Learning, Renewable energy
Project Type: Physical to Digital Workflow Design (Web & Enterprise Applications)
Technologies: Figma, Adobe XD, Mural, sketch Jira,Confluence

The Challenge

Traditional reinspection processes were manual, time-consuming, and prone to human error, leading to delays in site while conduxting in-person inspectiona nd potential safety risks in critical industries.

Complex datasets from inspections were hard to visualize, compliance management was fragmented, and integration with existing enterprise systems was challenging.

Key pain points included: inconsistent defect detection accuracy, lengthy reporting times, and scalability issues for high-volume operations.

Business Goals

🔍Improve Detection Accuracy

Leverage ML to enhance defect identification and reduce false positives/negatives.

⚡Accelerate Workflows

Automate reinspection processes to cut manual reporting and In-person inspection.

📊Enhance Data Visualization

Create intuitive dashboards for complex inspection data analysis to detect, validate the blades.

🛡️Ensure Compliance

Build in robust UI for regulatory adherence to make the inspection process more efficient and accurate.

📈Scale Operations

Designed layout to scale and reduce the time to perform inspection in digital environment.

The Solution

Developed an ML-augmented platform that automates NDT workflows, providing real-time analytics, predictive defect modeling, and seamless integration with enterprise systems.

Core features include interactive data visualizations, automated report generation, and modular interfaces for customizable user experiences.
Solution Interface Highlights
(NDA-restricted – Envision advanced dashboards with ML visualizations, workflow automations, and compliance trackers)

Key Benefits & Impact

GE’s inspection innovations reduce downtime and risk by minimizing physical entry and extended turbine shutdowns; deliver cost savings (up to 25% O&M reduction) through early defect detection; and expand coverage across surface and sub-surface defects.

By blending AI, robotics, and real-time analytics into inspection workflows, GE sets benchmarks for safer, faster, and more reliable lifecycle management—principles reflected in the ML-driven reinspection platform designed in this project.

How Research Informed Our Design Solutions

Performed in-depth workflow analysis and user interviews with NDT engineers to identify pain points and opportunities for ML integration.

Key insights: 70% of time spent on manual data entry, 55% needed better visualization tools, and 80% prioritized mobile accessibility for field operations.
Manual Processes

Time-consuming inspections; automated ML detection streamlined this.

Data Overload

Complex datasets; intuitive visualizations simplified analysis.

Compliance Gaps

Fragmented tracking; built-in modules ensured seamless management.

Scalability Issues

High-volume limitations; cloud integration enabled growth.

User Training

Steep learning curves; intuitive interfaces reduced onboarding time.

Integration Challenges

Legacy systems; API-focused design bridged gaps.

Design Process

Followed an Agile methodology with iterative sprints:
  1. Research: User interviews and workflow mapping.
  2. Ideation: Sketching and low-fidelity prototypes.
  3. Prototyping: High-fidelity designs in Figma/Adobe XD.
  4. Testing: Usability sessions and iterations.
  5. Implementation: Developer handoffs via Jira/Confluence.
Design Process Flow
(NDA-restricted – Visualize wireframes evolving into ML-integrated interfaces)

Responsibilities

  • Led the UX/UI strategy for ML-driven reinspection modules supporting industrial quality assurance.
  • Conducted workflow analysis with NDT engineers/operators to define critical user needs and pain points.
  • Delivered low- and high-fidelity prototypes, interactive dashboards, and intuitive data visualization for complex inspection datasets.
  • Collaborated with ML/data engineering teams to bridge technical constraints with user experience best practices.
  • Established a modular, component-based design system for scalable feature deployment.
  • Conducted iterative usability testing, achieving a 60% reduction in reinspection reporting time for end-users.
  • Worked in Agile sprints, providing detailed handoffs and supporting developer QA for design accuracy.