Case Study: Low-Frequency Ultrasound ML
Real-time classification for grinding wheel state and process quality.
Project Snapshot
- Role: ML Solution Lead
- Domain: Industrial process monitoring
- Stack: Python, signal processing, supervised ML, real-time decision logic
- Timeline: Research-to-production deployment
Related quality programs
>60% Defect-Crisis Reduction
Analytics-led quality programs using this ML-first operating style reduced major defect crises by more than 60% (public-shareable resume metric).
Targeted optimization window
>10% Yield Improvement
Yield gains exceeding 10% in programs where real-time classification informed intervention timing.
Production deployment
Real-Time Classification
Model deployed to edge devices for live grinding wheel state classification with sub-second latency.
Technical Architecture
graph TD
subgraph Data_Collection
A[Ultrasound Sensor] --> B[Signal Acquisition]
B --> C[Feature Extraction]
end
subgraph Preprocessing
C --> D[Noise Filtering]
D --> E[Time-Frequency Transform]
E --> F[Feature Engineering]
end
subgraph Model
F --> G[Trained Classifier]
G --> H[Wheel State Prediction]
end
subgraph Deployment
H --> I[Edge Device]
I --> J[Real-Time Alert]
I --> K[Process Control]
end
subgraph Feedback
J --> L[Operator Action]
K --> M[Quality Outcome]
M --> N[Model Retraining Data]
N --> G
end
Architecture: Ultrasound sensors capture acoustic emissions from grinding wheels. Signal processing extracts time-frequency features. A trained classifier predicts wheel state (sharp, dull, loaded) in real-time. Deployed to edge devices for sub-second inference. Feedback loop enables continuous improvement.
Decision Tradeoffs
| Option Considered | Pros | Cons | Decision |
|---|---|---|---|
| Supervised ML (Random Forest) | Interpretable, handles engineered features well, fast inference | Requires labeled data, feature engineering effort | Selected — best balance of interpretability and performance for production |
| Deep Learning (CNN) | Automatic feature learning, potentially higher accuracy | Black-box, requires GPU, harder to debug in production | Rejected — interpretability was critical for operator trust |
| Rule-Based Thresholding | Simple, no training data needed | Can't capture complex patterns, high false positive rate | Rejected — prior attempts showed insufficient accuracy |
Problem
Grinding wheel wear affects product quality and process efficiency. Operators lacked real-time visibility into wheel state, leading to delayed interventions and inconsistent quality outcomes.
Approach
I designed a supervised ML pipeline that ingests low-frequency ultrasound signals, extracts time-frequency features, and classifies wheel state in real-time. The model was trained on labeled data from controlled experiments and deployed to edge devices for production use.
Outcome
The system enabled proactive wheel change decisions, reducing defect crises by over 60% in related quality programs. Real-time classification informed intervention timing, contributing to >10% yield improvements where deployed.
Leadership Contribution
- Architecture: Designed the signal processing pipeline and selected Random Forest over deep learning for interpretability in production.
- Team: Coordinated with plant engineers for sensor installation, data collection protocols, and operator training.
- Governance: Established model validation process including cross-validation, holdout testing, and production monitoring.
- Outcomes: Tracked classification accuracy, false positive rate, and downstream quality metrics for continuous improvement.