6144synthetic TCAD DOE rowsUNKNOWNInspect metric contract
6device contextsUNKNOWNInspect metric contract
0live server runsUNKNOWNInspect metric contract
SIMULATION / SA-DEV-TCAD-OPT-004
TCAD DOE and surrogate optimization replay
TCAD DOE
A browser CSV replay for reading TCAD-style design-of-experiments, surrogate uncertainty, extrapolation risk, validation posture, and decision gates before treating an optimizer output as evidence.
EXPLORE
Start with the observable.
Use this page when the simulation is the primary artifact. The linked lesson explains the concept; the linked lab keeps the visual replay surface available.
- Filter device context, design lever, DOE plan, surrogate family, objective priority, and validation posture before reading a pareto-readiness score.
- Compare mesh stability, calibration anchor strength, sample efficiency, surrogate uncertainty, extrapolation risk, constraint violation, objective fit, and pareto readiness together.
- Use the decision gate to separate mesh-convergence review, calibration-anchor review, surrogate-uncertainty review, extrapolation guard, constraint review, owner review, bounded teaching candidates, and reference-bounded cases.
- Keep the public boundary visible: this is not a TCAD deck, calibrated solver result, foundry-node ranking, device qualification, or production optimizer.
BROWSER INTERACTIVE LAB
Filter the saved CSV. Inspect matching replay rows.
No server solver is running here. The browser filters a precomputed table and redraws the plot locally.
Loading replay rowsReading the saved CSV in this browser.
mesh stability scorecalibration anchor scoresample efficiency scoresurrogate uncertainty proxyextrapolation risk proxy
Each metric is normalized independently for shape inspection. Vertical positions do not compare magnitude across metrics.
n/amesh stability score
n/acalibration anchor score
n/asample efficiency score
n/asurrogate uncertainty proxy
n/aextrapolation risk proxy
Change one or more controls to find a saved row. No unrelated CSV value is substituted.
| case id | device context label | design lever label | doe plan label | surrogate family label | objective priority label | validation posture label | mesh stability score | calibration anchor score | sample efficiency score | surrogate uncertainty proxy | extrapolation risk proxy | constraint violation proxy | pareto readiness score | decision gate | next safe check |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Loading saved replay rows... | |||||||||||||||
RUN CARD
Conditions before conclusions.
The browser filters saved CSV rows. It does not run TCAD, SPICE, process simulation, field solving, optimizer loops, uploaded deck parsing, calibration, device qualification, or server-side models.
Primary knobs
- device context
- design lever
- DOE plan
- surrogate family
- objective priority
- validation posture
Saved outputs
- mesh-stability score
- calibration-anchor score
- sample-efficiency score
- surrogate-uncertainty proxy
- extrapolation-risk proxy
- constraint-violation proxy
- objective-fit score
- pareto-readiness score
- decision gate
EXPORT
Download the public artifacts.
These exports are the supported public files for review. PDK files and restricted third-party material are not redistributed.
TCAD DOE surrogate replay CSV
/data/tcad-doe-surrogate-optimization-replay-v90.csv
Open file ->Typed web JSON
/data/tcad-doe-surrogate-optimization-replay-web-v1.json
Open file ->Model card
/data/tcad-doe-surrogate-optimization-replay-model-card-v1.md
Open file ->METHODS
How the number was made.
Methods are written for citation discipline: the extraction rule matters as much as the plotted value.
Generate a Cartesian synthetic sweep across device context, design lever, DOE plan, surrogate family, objective priority, and validation posture.
Keep this method attached when reusing the figure or metric.
Compute normalized educational proxies for mesh stability, calibration anchor strength, sample efficiency, surrogate uncertainty, extrapolation risk, constraint violation, objective fit, and pareto readiness.
Keep this method attached when reusing the figure or metric.
Apply decision gates that route the learner to mesh convergence, calibration, surrogate uncertainty, extrapolation, constraint, owner review, bounded teaching candidate, or reference-bounded states.
Keep this method attached when reusing the figure or metric.
Store CSV, typed JSON, model card, source links, and public boundary together with no copied decks, solver code, examples, PDKs, process recipes, measured silicon, or optimizer services.
Keep this method attached when reusing the figure or metric.
LIMITS
What this page does not claim.
The MVP is useful because the limits are visible. These statements prevent a teaching simulation from being cited as silicon evidence.
No live TCAD, SPICE, process, electrothermal, RF, field-solver, or optimizer execution runs in the browser or on the server.
Carry this boundary into any derivative note, paper draft, or slide.
No TCAD deck, mesh, structure file, material file, PDK, compact model, process recipe, solver transcript, calibration dataset, measured silicon, or vendor benchmark is copied.
Carry this boundary into any derivative note, paper draft, or slide.
No device ranking, foundry-node claim, product recommendation, qualification, signoff, or production optimization claim is made.
Carry this boundary into any derivative note, paper draft, or slide.
Rows are original synthetic teaching proxies and require owned solver decks, mesh convergence, measured I-V/C-V, uncertainty validation, and device-owner review before engineering reuse.
Carry this boundary into any derivative note, paper draft, or slide.
EVIDENCE
Trace the claim back to artifacts.
The evidence route remains the catalog-level browser; this page is the simulation-level reading card.
CSV replay with 6144 rows and explicit device-context, lever, DOE, surrogate, objective, validation, metric, decision-gate, evidence-state, and next-check fields.
SA-DEV-TCAD-OPT-004
Typed JSON summary with source anchors, axes, decision-gate counts, example rows, average proxy summaries, and public boundary.
SA-DEV-TCAD-OPT-004
Model card saved beside the public dataset and experiment record.
SA-DEV-TCAD-OPT-004
Source spine uses official/public solver, DOE, surrogate, and validation pages as outbound anchors only; it does not copy papers, documentation examples, code, decks, figures, measurements, or vendor benchmarks.
SA-DEV-TCAD-OPT-004
Source spine
These links anchor the public teaching model. They do not convert the replay into a qualified process recipe or signoff result.
DEVSIM TCAD Semiconductor Device SimulatorOfficial public project page for an open-source TCAD device simulator. SemiAgora uses it as a link-only solver vocabulary anchor.DEVSIM JOSS paperPeer-reviewed software paper metadata for DEVSIM. No paper body, examples, or code are copied.Synopsys TCAD overviewOfficial public overview anchoring TCAD process/device simulation, calibration, and optimization vocabulary.NIST response surface designsNIST handbook page anchoring response-surface vocabulary for curvature, interactions, and quadratic models.SciPy quasi-Monte Carlo samplingOfficial SciPy documentation for Latin hypercube, Sobol, Halton, and other low-discrepancy sampling tools.scikit-learn GaussianProcessRegressorOfficial scikit-learn API reference anchoring Gaussian-process regression and predictive uncertainty vocabulary.TCAD-generated surrogate optimization paperPublic arXiv metadata for a TCAD-generated surrogate model used with differential-evolution optimization.NIST semiconductor multiscale modeling and validationNIST project page anchoring validation, uncertainty, and machine-learning acceleration vocabulary.
Public execution boundary
No server-side live ngspice execution, uploads, accounts, payments, comments, newsletter signup, or job/event submissions are active on this page.