Catch defects thedatasheet limit can't see.
AI-driven lot-relative anomaly detection and physics-informed drift forecasting for ESS burn-in screening — explains every flag to QA engineers with per-feature attribution and Arrhenius trajectory overlays.
Two modules. One explainable risk score.
Every part gets a fused, lot-relative score — not an absolute-limit pass/fail.
Module A — Fused Outlier Score
Isolation Forest catches multivariate pattern drift; ECOD catches per-parameter tail anomalies — parameter-free, O(n·d log n).
Physics Baseline
Arrhenius / log-linear extrapolation grounds every drift forecast in reliability-engineering theory.
Per-Lot Normalization
MAD-based robust z-scoring per lot per timestamp — the mechanism that makes detection dynamic, not static.
Module B — Drift Predictor
XGBoost residual correction on top of the physics baseline forecasts val_168h from val_0h + val_24h, MAE-optimized.
SHAP Explainability
Every flag decomposes into per-feature contribution — never a black-box score.
Recall-First Threshold
Tuned via PR curve — false negatives (missed defects) cost far more than false positives in mission-critical hardware.
Synthetic Injection Validation
Threshold calibration on synthetically injected drift patterns — detector itself stays unsupervised.
QA Dashboard
Flagged parts, physics-curve overlay, lot percentile, accept/reject override log — all in one interactive console.
End-to-end pipeline
Ingest & normalize
Raw ESS logs → per-lot MAD z-score + log-transform
Feature decorrelation
val_0h, early/mid/late slopes — not raw correlated timestamps
Module A ‖ Module B
IF + ECOD fused score, parallel to physics + XGBoost forecast
Risk fusion
Weighted combine → single explainable risk flag
SHAP + physics overlay
QA-facing justification, not a black-box number
O(n·d log n) at fab-lot scale
ECOD and Isolation Forest both scale linearly-ish in features and log-linearly in lot size — unlike LOF's O(n²), this holds up on large ESS datasets.
See it flag a real part
Live data from the screening pipeline. Click a module to see its detection output.
SHAP / ECOD Feature Attribution
Driver: tpd_96h contributed 4.6 of 25.3 total ECOD score — this part sits at the 93.8th percentile within its lot's distribution.
Tech Stack
Recently Flagged
Top flagged parts across 5 lots — click to drill down
| Part ID | Lot | IF Score | ECOD Score | Fused | Status |
|---|---|---|---|---|---|
| #0401-001 | L-0401 | 0.376 | 17.63 | 0.654 | Warning |
| #0401-004 | L-0401 | 0.364 | 16.31 | 0.612 | Warning |
| #0401-006 | L-0401 | 0.211 | 14.84 | 0.503 | Warning |
| #0401-009 | L-0401 | 0.685 | 25.27 | 0.858 | Anomaly |
| #0401-014 | L-0401 | 0.143 | 14.25 | 0.456 | Warning |
| #0401-017 | L-0401 | 0.365 | 12.25 | 0.501 | Warning |
Ready for mission assurance.
Built for ISRO's ESS pipelines — extensible to defense, aviation, and medical device screening.
Launch QA Console