Public engineering surrogate
Public engineering-surrogate reliability emulator. It estimates optical-link closure probability under declared assumptions and validation limits. It is not mission-certified and does not reproduce internal NASA/JPL pipelines.
Optical Link Reliability Emulator
Closure-probability emulator over the reliability constraint manifold, reliability-regime chamber, threshold layer, and safety layer. Public scenario anchors only.
Headline reliability numbers under the public scenario anchor. P_close gauge, percentile margins, dominant failure mode, and escalation flag.
Closure probability P_close
Headline numbers
| P_close | — |
| P_outage | — |
| Median margin M_50 | — |
| 5th percentile M_05 | — |
| Dominant failure mode | — |
| High-fidelity escalation | — |
| Escalation reason | — |
Sixteen-field scenario form. Defaults come from the selected public scenario anchor. Per-input range / distribution controls feed the uncertainty model on the next tab.
Per-input distribution picker. Supported distributions: fixed, normal, lognormal, uniform, triangular, empirical. Correlations are recorded but not yet sampled.
Correlations
Independence is assumed across inputs unless a correlation list is supplied. The MVP records correlations but does not yet draw correlated samples.
Histogram of sampled link margin (dB) over the Monte Carlo run, and the cumulative distribution. About 50 bins; gauge can be re-rendered from the Summary tab.
Sample statistics
| Mean margin | — |
| Std dev | — |
| p05 | — |
| p25 | — |
| p50 | — |
| p75 | — |
| p95 | — |
| Valid samples / total | — |
P_close gauge plus reliability-vs-range and reliability-vs-aperture sweeps (20 grid points each). Other inputs held at scenario values.
Failure-mode probability across the Monte Carlo samples. The dominant mode card shows which mechanism in the reliability constraint manifold was active most often.
Dominant failure mode
Tornado chart of linear sensitivity (delta margin per fractional input change). Largest-magnitude bars are the most-load-bearing inputs.
Dominant uncertainty contributor
Escalation logic flags scenarios where the public engineering surrogate alone is insufficient and a high-fidelity model is recommended.
Escalation verdict
Top-N escalation candidates
Recommended high-fidelity model
The reliability emulator inherits the lab-wide validation page. Method sanity tests below; full report under the link.
Reproducible reliability certificate. The deterministic result_hash pins inputs + constants + formulas + outputs. The receipt_hash adds the export timestamp.
Hashes
——Show research mapping (research id - public engineering name)
The reliability emulator runs over a constraint-compression framework. The public UI uses these engineering names:
| Research id | Public engineering name | Role |
|---|---|---|
K_6 | reliability constraint manifold | Limiter manifold over the link budget |
F_plus | reliability-regime chamber | Regime classifier (which limiter is active) |
L_threshold | threshold layer | Margin-vs-requirement test |
L_safety | safety layer | False-safe risk + escalation projector |
x, ⊕, ⊗ | full geometry map | Active-branch geometry of the reliability problem |
The active-branch architecture is used as a constraint-compression framework, not as a claim that high-dimensional physics is required for standard optical-link engineering.
Public engineering surrogate - reliability emulator - not mission-certified.