Explore the options before committing to one.
Simulation branches the current operational picture, applies alternative courses of action under stated constraints and assumptions, and compares the outcomes — without ever letting a simulated outcome pass for a prediction, or a prediction for a certainty.
- OBSERVE
- UNDERSTAND
- SIMULATE
- RECOMMEND
- AUTHORIZE OR REFUSE
- DISTRIBUTE
- ENFORCE
- EXECUTE
- EVIDENCE
What it does
- Course-of-action comparisonAlternatives evaluated side by side against the same starting picture.
- ConstraintsLegal, operational, resource and time constraints applied explicitly.
- AssumptionsEvery assumption recorded next to the result it produced.
- Second-order effectsConsequences traced through dependencies in the Reality Graph.
- Scenario branchesBranches taken from the graph at a point in time, kept separate from it.
- Reversible explorationNothing in a branch changes the operational picture or triggers action.
- Confidence and uncertaintyOutcome ranges and the method behind them, not single numbers.
- Decision supportComparisons attached to the proposal they informed, for later review.
Three different things
| Term | What it is | What it is not |
|---|---|---|
| Simulation | A worked consequence of stated assumptions applied to a known starting state. | A claim about what will happen. |
| Prediction | A statement about what may happen, with method, inputs and uncertainty. | An observation, or a basis that can be treated as fact. |
| Certainty | Not produced by Simulation. | — |
Simulated and exercise data are flagged at creation and that flag cannot be removed. A simulated record can never be promoted to a real-world observation, distributed as real, or used as evidence for an authorization as if it had been observed.
From comparison to proposal
A simulation informs a recommendation; it does not make one. When an operator proposes an action, the comparisons they relied on are attached to the proposal, so the Authority Kernel decision and its evidence record show which alternatives were considered and under which assumptions.
That link also works in reverse. After an event, reviewers can see whether an assumption that shaped the decision later proved wrong — which is a different finding from a decision taken without considering alternatives at all.