Use case · Live operations
From a Changed Fact to Coordinated Action
When one operating fact changes, trace its consequences through the model, compare feasible responses, and update the people and views that depend on the decision.
Narrated film · 0:58
How Live Data Moves Through a Model
New data enters under explicit permissions, recalculates only the affected results, and carries its source and history into every view.
▶ 0:58
Guided path · One pattern, three fields
One changed fact. Three recognizable consequences.
A shipment route closes. Staffed medical capacity falls. One transport becomes unavailable. The domains are different; the need for an inspectable path from revision to authorized response is the same.
Explore the worked examples and calculations
Supply chain · A route closes
The shipment is moving. Its plan is no longer valid.
Component Package 27 must reach an assembly plant inside a six-hour delivery window. Transportation, production, and customer operations begin with one feasible plan.
Changed fact
ROUTING.corridor_kilo_status
- Source
- ROUTING SOURCE
- Revision
- ROUTE-219
- Observed
- 06:18 local time
What depends on it
- Valid lanes
- Travel time and reserve
- Delivery-window feasibility
- Production sequence
- Customer commitment
Bounded finding
The authored model rejects Route Delta on reserve, marks Route Foxtrot dominated on exposure, and presents Route Echo as feasible within the represented window and reserve constraints.Evidence record. Retain source revision ROUTE-219, observation time, the affected dependency path, evaluated alternatives and reasons, the selected plan revision, dispatcher identity, authorization time, and view acknowledgements.
Claim boundary. Fictional, deterministic illustration. The routes, constraints, timing, and result are authored—not live data, customer outcomes, optimization benchmarks, or deployment evidence.
Medical coordination · Staffed capacity falls
Capacity falls while demand is already moving.
In fictional Exercise MED-27, a regional care network is coordinating 18 modeled demand units while a transfer window is already closed and Care Central reports a staffing loss.
Changed fact
CARE-CENTRAL.staffed_capacity
- Source
- CARE CAPACITY AGGREGATE
- Revision
- MED-BIND R17
- Observed
- 09:26:30Z
What depends on it
- Facility capacity
- Route and transport eligibility
- Allocation alternatives
- Modeled arrival P90
- Plan feasibility
Bounded finding
Alternatives A and B fail represented capacity or route constraints. Alternative C distributes demand across four facilities and is the one bounded feasible alternative in the authored fixture.Evidence record. Retain plan and binding revisions, represented constraints, modeled arrival P90 against the window, authorization claim MED-AUTH-023, accountable actor, policy source, validity interval, and view receipts.
Claim boundary. Fictional, deterministic exercise using modeled demand classes and authored timing. It is not clinical decision support, patient-level allocation, a capacity forecast, or evidence of real-world outcomes.
Mission logistics · One transport disappears
Six movements still need six lifts.
Operations, logistics, and command show fictional Course Echo as ready. Six transports support six movements in one wave, completing at 08:40Z—five minutes inside the movement window.
Changed fact
LOGISTICS.available_transport
- Source
- LOGISTICS CAPACITY MODEL
- Revision
- L-442
- Observed
- 07:12:18Z
What depends on it
- Available transport
- Waves required, 1 to 2
- Completion, 08:40Z to 09:20Z
- Margin, +5 to -35 minutes
- Course feasibility
Bounded finding
The original course and a delayed branch remain infeasible. Branch B restores one-wave completion only if an outside authority reallocates TRANSPORT-ECHO-06.Evidence record. Retain source revision L-442, binding B-031, command-model revision M-205, recommendation DECISION-ECHO-B-205, reallocation evidence REALLOC-R12, actor, authority, and resolved view receipts.
Claim boundary. Fictional, deterministic scheduling fixture with two authored alternatives. It is not a real mission, prediction, exhaustive optimization, deployment claim, or runtime benchmark.
The shared pattern
From revision to coordinated action.
-
Receive a named revision
Record what changed, where it came from, when it was observed, and which source revision carries it.
One authoritative change -
Trace what depends on it
Follow the fact through the rules, constraints, plans, and views that use it.
One visible dependency path -
Test bounded responses
Evaluate authored alternatives against represented capacity, timing, policy, and eligibility constraints.
Feasible, conditional, and rejected options -
Make the result legible
Expose the reason, assumptions, tradeoffs, and uncertainty behind each finding.
An inspectable explanation -
Stop at authority
Present the recommendation to the person or external process authorized to act. Calculation is not authorization.
An attributable decision -
Resolve and remember
Update role-specific views from the authorized revision and retain explicitly modeled evidence connecting the fact, model, actor, authorization, and reported external response without treating that response as an outcome.
Coordinated views and history
Continue at your depth
Try it. Watch it. Evaluate it. Build it.
- 01 · Try Change one fact in Course Echo Reduce available transport from six to five and follow the threshold effect through feasibility, alternatives, authority, and coordinated views. Enter the interactive scenario
- 02 · Watch Every System Works. The System Does Not. One transport change makes three separate views disagree; a shared model exposes the mismatch and tests two predefined options for an officer. Watch the 1:12 film
- 03 · Read How to Evaluate Executable Decision Infrastructure A buyer's guide for turning a compelling demonstration into a bounded evaluation of sources, logic, explanations, authority, interoperability, change behavior, and evidence. Read the white paper
- 04 · Build Follow the decision into Grid Continue into the linked Grid Developers guide for the exact behavior, prerequisites, and limits used by this explanation. Explain and validate a changing decision
Evaluate your own work
Bring one recurring decision.
Start with one fact that changes, one consequential decision it affects, and one accountable owner. Use the worksheet to make the current model and a bounded evaluation visible.
- Which facts change most often?
- Which rules and constraints govern the result?
- Where must a person authorize action?
- What evidence must remain inspectable?
Written analysisRead the full article behind this guided example.
The plan was valid until the world changed
A shipment is delayed. Demand is revised. A route closes. A hospital loses capacity. A supplier changes an allocation.
Most organizations can record the new fact quickly. The harder work is determining what the fact changes—and coordinating a response before separate plans begin to diverge.
A common operating picture shows that something happened. A common operating model shows what the change means. It does not erase the differences between a commercial shipment, a medical exercise, and a mission plan. It makes the coordination pattern visible without pretending the decisions are interchangeable.
Imagine it in your work
The same pattern can support inventory reallocation, financial forecasting, workforce scheduling, medical planning, supply-chain recovery, infrastructure response, or policy administration.
The domain changes. The essential question does not:
When a fact changes, how quickly can everyone reach the same inspectable understanding of what to do next?
What to evaluate
Begin with one consequential fact and one decision it affects. Identify the current copies, handoffs, reconciliation steps, constraints, and approval point. That creates a bounded way to compare the existing workflow with a shared model.
The goal is not to begin with a platform-wide promise. It is to determine whether one recurring decision can be represented, tested, explained, and governed as a shared operating model.