Dhairya Shah completed Predictive Maintenance: Industrial ML for Fault Detection

Built a working rolling-bearing fault detector end to end: extracted RMS and kurtosis-style time-domain features, used envelope-spectrum energy at the supplied defect frequencies to localize an outer-race fault at BPFO, and trained a compact depth-capped decision tree on training bearings only. They then evaluated on held-out bearings, reported catch and false-alarm rates, and defended a cost-based alarm threshold in a maintenance memo using their own numbers.

• Leakage-Free Feature Engineering
• Honest Held-Out Evaluation
• Physics-Based Fault Localization
• Cost-Based Operational Decision-Making