Plain-language summary
- What it solves
- Laboratory analyses arrive every few hours: in between, product quality is unknown.
- Who it can serve
- Anyone working on process quality control, and anyone studying soft sensors.
- What it produces
- A continuous quality estimate with its margin of error, and a warning when the estimate is not reliable. On simulated data, not real.
Show how a soft sensor can estimate quality between laboratory samples while recognising when it should abstain.
3 · What it produces
Estimate/split-conformal interval, domain status, OOD check and frozen-benchmark context.
The cards expose multiple views of the same workflow so result, method, evidence and history remain distinct. The current benchmark uses synthetic data (simulated data).
4 · How it works
Delayed-label estimation + OOD abstention
1process sensors + delayed laboratory with noise/QC
2Linear, PLS, Ridge, Random Forest, Gradient Boosting; Isolation Forest OOD
3RMSE/MAE/bias/coverage, skill versus last lab, split-conformal and truth-vs-lab; risk-coverage diagnostic only
4Persistence, rendering and governed publishing.
5 · Data and signals
| Element | Detail |
|---|---|
| Data and signals | process sensors + delayed laboratory with noise/QC |
| Industrial transferability | soft sensing, inferential measurement and quality monitoring |
| Research | conformal coverage, OOD as domain distance and model error versus laboratory uncertainty |
| Interpretation condition | truth and laboratory are simulated; no real performance claim |
Linear, PLS, Ridge, Random Forest, Gradient Boosting; Isolation Forest OOD
RMSE/MAE/bias/coverage, skill versus last lab, split-conformal and truth-vs-lab; risk-coverage diagnostic only
Python, pandas, scikit-learn, SQLite, Matplotlib
leakage from true values or a soft sensor answering outside its domain
Workflow technical dossier
Methods, checks, stack and transferability conditions.
Data and signals
process sensors + delayed laboratory with noise/QC
Inference / calculation
Linear, PLS, Ridge, Random Forest, Gradient Boosting; Isolation Forest OOD
Validation
RMSE/MAE/bias/coverage, skill versus last lab, split-conformal and truth-vs-lab; risk-coverage diagnostic only
Stack
Python, pandas, scikit-learn, SQLite, Matplotlib
Limitations
leakage from true values or a soft sensor answering outside its domain
See recent outputs
The Publications page collects local previews and, on request, the public feed.
Potential transferability
Have a problem similar to Virtual Analyzer?
A first discussion can start from the objective, available data and success criterion without sending confidential information. A result observed here is not assumed to transfer automatically to another context.