Self-directed project
Predictive Maintenance: Industrial ML for Fault Detection
Condition-Monitoring Engineer · AI / ML · August 2026
A ProoV case study · educational project, not employment
Built an end-to-end rolling-bearing fault detector using leakage-free training on bearing vibration features, a compact depth-capped decision tree, and held-out evaluation with catch-rate and false-alarm rate. The candidate also produced a maintenance decision memo that justified a 0.60 alarm threshold using explicit cost trade-offs and translated the result into plain language for operations staff.
Graded against
- Feature engineering (signal to honest numbers)25%
- Held-out evaluation (honest measurement)25%
- Cost-based threshold decision + memo25%
- Compact model + embedded-budget justification15%
- Plain-language explanation10%
Passed · pass mark 60/100
What stood out6
- Correctly identified kurtosis as the time-domain feature that jumps most on the faulty bearing
- Trained a depth-capped decision tree on training bearings only and stated depth and node count against the embedded budget
- Reported held-out catch-rate and false-alarm rate rather than evaluating on training data
- Wrote a maintenance memo that used the student's own threshold, catch-rate, false-alarm rate, and cost figures
- Leakage-Free Feature Engineering
- Honest Held-Out Evaluation
The work I submitted19 tasks
11 tasks · 6.4k characters · Python · 8 written answers · 436 words
- Initial Prediction62 characters
- Example Load Plot HealthyPython · 860 characters
- Load Plot Signal CodePython · 1.8k characters
- Load Plot Signal205 characters
- Example Rms WorkedPython · 697 characters
- Time Features143 characters
- Freq FeaturesPython · 715 characters
- Train Classifier363 characters
- Evaluation Heldout288 characters
- Threshold Decision332 characters
- Decision Memo941 characters
- Initial Prediction9 words
- Example Load Plot Healthy81 words
- Load Plot Signal Code163 words
- Load Plot Signal15 words
- Example Rms Worked72 words
- Time Features10 words
- Freq Features55 words
- Train Classifier31 words
Summarised on purpose — the submitted code and writing stay private so this page cannot be reused as an answer key. The full submission sits behind the verified certificate.