Simulation-Reality Deviation Module (SDMM) defines
the governance
OOF™ Origin Open Foundation™
Independent Methodological Authority
Architecture Ecosystem: Structured Reality Standards™
Architecture Family: SIMULOS® — Simulation Governance Architecture
Parent Standard: Simulation-to-Reality Correlation Standard (SRCS)
Operational Layer: Simulation-to-Reality Correlation Layer
Category: Governance & Enforcement
Subcategory: Simulation Governance
Type: Parent Standard Module
Governed Space: Simulation-Reality Deviation
Version: 1.0
Status: Canonical · Open Module
Effective Date: 19 August 2026
Compatibility: OOF® Methodology OS™ · GOA™ · ORA™ · VALIDOS® · INTEGROS® · ADIT® · AGA™ ·
AIG® · ASGA™ · RIS™
AI-Readable: Yes
Authority: OOF®
Protection: MIP® — Methodological Intellectual Property
Canonical Language: English (UCL™)
Governed Space
Simulation-Reality Deviation
Canonical Definition
Simulation-Reality Deviation Module (SDMM) defines the governance conditionsunder which material differences between simulated behavior, conditions,
states, events, relationships, trajectories, and outputs and their
corresponding Operational Reality observations are identified, measured,
localized, classified, contextualized, traced, and documented.
SDMM establishes the governed deviation layer between:
Simulation-Reality Comparison
and:Reality Alignment.
Its purpose is not to require simulation and reality to be identical.
Its purpose is to determine:
where they differ, by how much, under which conditions, whether the deviation
is systematic or conditional, and which part of the simulation-to-reality
relationship is affected.
Operational Role
SDMM is the third internal governance space of the Simulation-to-RealityCorrelation Standard (SRCS).
The progression is:
SRCM — Simulation-Reality Comparison
Module
→ establishes comparable simulation and reality observations.SCCM — Simulation Correlation Criteria
Module
→ establishes the criteria against which correspondence is assessed.SDMM — Simulation-Reality Deviation Module
→ identifies and structures material differences revealed by that comparison.SRAM — Reality Alignment Module
→ governs how the simulation-reality relationship is reassessed or aligned inresponse to identified differences.
SCUM — Correlation Uncertainty Module
→ governs uncertainty remaining in the correlation relationship.SDMM therefore asks:
Where does the simulation depart from reality, how material is that departure,
and what does the deviation actually affect?
Module Function
SDMM prevents a simulation from being described as simply:“close to reality”
or:
“different from reality”
without identifying the structure of that difference.
A useful simulation-reality assessment must be capable of distinguishing,
where relevant:
what matched,
what deviated,
where the deviation occurred,
under which conditions it occurred,
how large it was,
whether it repeated,
and:
what downstream interpretation may be affected.
Deviation Object
Every material deviation should be capable of being represented as agovernable object.
Conceptually:
Deviation ID
↓Simulation Observation
↓Reality Observation
↓Correlation Criterion
↓Observed Difference
↓Deviation Classification
↓Materiality
↓
Affected Simulation Element
↓Known / Suspected Basis
↓Uncertainty
↓
Governance Status
This prevents significant deviations from disappearing inside aggregatecorrelation metrics.
Deviation Identification
A Simulation-Reality Deviation exists when a materially comparable simulatedand real-world observation differs according to an applicable correlation
criterion.
The difference may concern: magnitude, direction, timing, sequence, frequency,
duration, state, response, interaction, distribution, threshold crossing,
failure mode, recovery behavior, or another materially relevant property.
Deviation ≠ Failure
SIMULOS® establishes:A Simulation-Reality Deviation is not automatically a simulation failure.
A deviation may arise because: the simulation is simplified, reality contains
unmodeled variables, measurements are imperfect, the Operational Reality state
changed, boundary conditions differ, assumptions apply only conditionally,
stochastic behavior exists, the comparison basis is incomplete, or the
simulation genuinely misrepresents reality.
SDMM preserves the deviation before determining its consequence.
Deviation Magnitude
Where measurable, deviation magnitude should establish the size of theobserved difference.
Magnitude may be expressed through: absolute difference, relative difference,
percentage difference, distributional distance, timing offset, positional
error, trajectory error, frequency difference, state mismatch, categorical
difference, or another appropriate measure.
SIMULOS® does not prescribe one universal deviation metric.
Deviation Direction
Where relevant, SDMM should distinguish whether the simulation:Overestimates Reality
Underestimates Reality
Leads Reality
Lags Reality
Diverges Bidirectionally
or exhibits another defined directional relationship.Direction may be operationally more important than aggregate error magnitude.
Deviation Localization
A deviation should be localized as precisely as the available evidencepermits.
Localization may identify: Simulation Object component, subsystem, agent,
environment region, scenario, parameter range, operating condition, time
interval, state transition, interaction, input region, or output dimension.
Global ≠ Local Correlation
A simulation may correlate strongly with reality overall while failing badlyin a specific operational region.
Therefore:
Strong aggregate correlation must not erase material localized deviation.
Temporal Deviation
Temporal deviation exists where simulated behavior differs from reality in:initiation time, response latency, event ordering, duration, synchronization,
recovery time, transition speed, or temporal pattern.
A simulation may predict the correct event but at the wrong time.
That difference may be material.
Spatial Deviation
Spatial deviation concerns differences in: position, trajectory, distribution,physical location, movement, coverage, geometry, or spatial interaction.
This may be especially important in: robotics, autonomous vehicles, aerospace,
orbital systems, manufacturing, logistics, and physical infrastructure
simulation.
Behavioral Deviation
Behavioral deviation exists where the simulated system responds differentlyfrom the corresponding real system under materially comparable conditions.
Examples include: different action selection, different recovery behavior,
different escalation behavior, different human response, different agent
coordination, or different autonomous decision pathways.
Environmental Deviation
A simulation may reproduce the target system accurately while inadequatelyreproducing its environment.
SDMM therefore distinguishes:
System Deviation
from:Environment Deviation.
Environmental deviation may concern: weather, terrain, friction, radiation,
communication conditions, traffic, human behavior, physical noise,
infrastructure state, or another environmental variable.
Interaction Deviation
Complex systems may behave correctly in isolation but diverge when interactingwith: humans, other agents, infrastructure, external systems, physical
environments, or unexpected events.
SDMM therefore governs deviations in interaction dynamics separately where
material.
Trajectory Deviation
For systems whose behavior evolves over time, a small initial deviation mayproduce substantial downstream divergence.
Conceptually:
Initial Difference
↓State Divergence
↓Behavioral Divergence
↓Trajectory Divergence
↓Material Outcome Difference
SDMM should preserve this progression where observable.Systematic Deviation
A Systematic Deviation repeatedly occurs in the same direction or undercomparable conditions.
Examples:
simulation consistently underestimates braking distance
or:
simulation consistently overestimates communication reliability.
Systematic deviations may indicate a persistent representational or model
relationship requiring further governance.
Conditional Deviation
A simulation may correlate well with reality under normal conditions butdiverge under a specific condition.
For example:
Normal Temperature → Strong Correlation
Extreme Temperature → Material Deviation
The correct conclusion is not necessarily:Simulation Poor
but potentially:Correlation is conditionally bounded.
Intermittent Deviation
Some deviations occur irregularly.This may arise from: stochastic processes, rare interactions, timing, emergent
behavior, nondeterministic systems, or incomplete observation.
Intermittent deviations should not automatically be discarded as noise where
their consequences may be material.
Deviation Frequency
Where relevant, SDMM may govern:how often the deviation occurs.
A small deviation occurring continuously may be more important than a larger
deviation occurring once.
Conversely, a rare deviation may still be critical where its consequence is
catastrophic.
Deviation Materiality
Deviation magnitude alone does not determine materiality.Conceptually:
Deviation Materiality
may depend upon:Magnitude + Context + Frequency + Consequence + Scenario + Reliance Purpose.
A two-percent difference may be immaterial in one simulation and
mission-critical in another.
Materiality Classification
Possible implementation states may include:Negligible
Minor
Material
Major
Critical
or:
Materiality Undetermined.
These classifications are implementation-dependent rather than universal
SIMULOS® thresholds.
Threshold Deviation
Where correlation criteria establish acceptable thresholds, SDMM shoulddistinguish:
Within Correlation Boundary
from:Outside Correlation Boundary.
However, passing a threshold does not prove universal simulation-to-reality
equivalence.
Deviation Clustering
Individual deviations may reveal little when viewed separately.SDMM may therefore identify:
Deviation Clusters
where multiple differences occur around: one subsystem, one scenario family,one operating condition, one environment state, one parameter region, or one
behavioral pattern.
Clusters can expose simulation weaknesses hidden by aggregate metrics.
Deviation Propagation
A deviation in one simulated element may propagate through the simulation.For example:
Sensor Timing Deviation
↓State Estimation Deviation
↓Decision Deviation
↓Trajectory Deviation
↓Outcome Deviation
SDMM should preserve known propagation relationships rather than treating eachdownstream difference as independent.
Deviation Dependency
Some deviations exist only when multiple conditions combine.For example:
Low Visibility Sensor Noise High Speed
↓Material Simulation-Reality Deviation
Each factor independently may show acceptable correlation.The combination may not.
Deviation Attribution
Where sufficient evidence exists, a deviation may be associated with aprobable basis such as: Reality Representation, assumption, model structure,
environment configuration, parameterization, input data, scenario design,
execution condition, measurement limitation, Operational Reality change, or
unknown cause.
Attribution Must Not Be Invented
SIMULOS® establishes:Observed deviation does not by itself establish the cause of deviation.
Where causation is not sufficiently supported, the status should remain:
Deviation Cause Undetermined.
This is preferable to false explanation.
Unresolved Deviation
A deviation may remain unresolved when: its cause is unknown, observations areincomplete, measurements conflict, reproduction fails, the real-world basis is
uncertain, or multiple explanations remain plausible.
Unresolved deviations remain part of the governed correlation record.
No Average-Washing Principle
A critical SIMULOS® rule is:Material Simulation-Reality Deviations must not disappear merely because
aggregate averages remain acceptable.
A simulation could perform extremely well across 99% of conditions while
failing in the 1% that matters most operationally.
No Selective Correlation Principle
Simulation-reality analysis should not preserve only those comparisons thatdemonstrate strong correspondence.
Material negative correlation evidence must remain visible.
Reality Change
A deviation may emerge because reality changed rather than because thesimulation changed.
Therefore SDMM should preserve where possible:
Simulation Version
and:Operational Reality State / Time Basis.
This allows later analysis to distinguish:
Simulation Drift
from:Reality Drift
or:Combined Drift.
Deviation Across Versions
Simulation versions may be compared against the same or equivalent realitybasis.
For example:
Simulation v1 → Reality → Deviation Set A
Simulation v2 → Reality → Deviation Set B
This can reveal whether changes improved or worsened the correlationrelationship.
Deviation Trend
Across repeated comparisons, deviations may:Decrease
Remain Stable
IncreaseShift Location
Change Direction
or:Become Uncertain.
Deviation trend can become an important governance signal for SRAM.
Deviation Record
A governed deviation record may contain:Deviation ID
Simulation ID / Version
Reality Reference
Scenario / Condition
Correlation Criterion
Simulation Observation
Reality Observation
Deviation Measure
Deviation Direction
Deviation Magnitude
Deviation Classification
MaterialityFrequency
Affected Element
Known Dependencies
Probable Basis where supported
UncertaintyStatus
Required Follow-Up Relationship to SRCM
SRCM establishes:what simulation observation and reality observation may legitimately be
compared.
SDMM governs:
the material difference revealed by that comparison.
Without SRCM, a deviation may be based on incomparable observations.
Relationship to SCCM
SCCM establishes the correlation criteria and thresholds.SDMM applies those criteria to structure and classify observed differences.
Therefore:
Correlation Criterion → Comparison → Deviation.
Relationship to SRAM
SDMM identifies where the simulation-reality relationship differs materially.SRAM uses that governed deviation basis to determine where and how Reality
Alignment should be examined.
Therefore:
Deviation Discovery precedes Alignment Action.
Relationship to SCUM
Some deviation measurements remain uncertain.SCUM governs uncertainty in the broader simulation-to-reality correlation
relationship.
SDMM must therefore preserve uncertainty rather than converting uncertain
deviation into false precision.
Relationship to ORA™
ORA™ governs Operational Reality.SDMM does not redefine reality.
It consumes sufficiently governed reality observations or descriptions and
governs the deviation relationship between those observations and the
simulation.
Relationship to VALIDOS®
SDMM does not determine whether the simulation is valid for a particularValidation Purpose.
It produces structured Simulation-Reality Deviation evidence that VALIDOS® may
subsequently use.
Relationship to INTEGROS®
SDMM does not own general integrity governance.INTEGROS® may govern integrity properties of the observations, measurements,
records, or artifacts used in deviation analysis.
Autonomous Systems Application
For autonomous systems, SDMM can identify where simulated autonomous behaviordiverges from operational behavior.
Material deviation regions may include: perception under degraded conditions,
agent interaction, human response, recovery behavior, escalation, decision
timing, unexpected environmental conditions, and long-horizon autonomous
behavior.
This matters because a system can demonstrate excellent average simulation
performance while diverging precisely in the operational conditions where
autonomous governance is most important.
Space and Mission Simulation Application
Space systems often depend heavily on simulation because many missionconditions cannot be repeatedly reproduced in Operational Reality.
SDMM can govern differences revealed through: hardware tests, flight data,
telemetry, prior missions, environmental measurements, subsystem tests, and
operational mission observations.
A simulation may correlate strongly for nominal trajectory behavior while
materially deviating in: thermal extremes, radiation response, communication
degradation, sensor behavior, propulsion anomalies, subsystem interactions, or
autonomous recovery.
The deviation record makes those limits explicit rather than allowing nominal
correlation to represent the entire mission envelope.
Minimum Implementation Framework
1. Establish the Comparison BasisUse governed simulation-reality comparisons from SRCM. 2. Apply Correlation
Criteria
Determine which SCCM criteria govern the
comparison. 3. Identify Deviations
Detect materially relevant differences. 4.
Measure the Difference
Determine magnitude, direction, timing, distribution, or another applicabledeviation measure. 5. Localize the Deviation
Identify the affected: condition, scenario, component, state, interaction,
environment, or output.
6. Classify the Deviation
Determine whether it is: systematic, conditional, intermittent, localized,
cumulative, or another relevant type.
7. Determine Materiality
Assess whether the deviation materially affects the intended
simulation-reality relationship. 8. Identify Dependencies and Propagation
Determine whether other deviations or conditions contribute to or result from
it. 9. Assess Attribution
Identify probable causes only where supported.
Otherwise preserve:
Cause Undetermined. 10. Preserve
Uncertainty
Do not express deviation more precisely than the evidence permits. 11.Document the Deviation
Create a governed Simulation-Reality Deviation Record. 12. Transfer to Reality
Alignment
Provide the structured deviation basis to SRAM.
Governance Outputs
SDMM may produce: Simulation-Reality Deviation ID, Simulation-RealityDeviation Record, Deviation Measurement Record, Deviation Magnitude
Assessment, Deviation Direction Assessment, Deviation Localization Record,
Temporal Deviation Record, Spatial Deviation Record, Behavioral Deviation
Record, Environmental Deviation Record, Interaction Deviation Record,
Trajectory Deviation Record, Systematic Deviation Record, Conditional
Deviation Record, Intermittent Deviation Record, Deviation Frequency
Assessment, Deviation Materiality Assessment, Threshold Deviation Record,
Deviation Cluster Record, Deviation Dependency Map, Deviation Propagation Map,
Deviation Attribution Record, Unresolved Deviation Record, Reality Change
Record, Cross-Version Deviation Comparison, Deviation Trend Record, and
Governed Simulation-Reality Deviation Basis.
Use Case 1 — Autonomous Vehicle Simulation
An autonomous driving system performs strongly in simulation and demonstrateshigh aggregate correspondence with road testing.
SDMM reveals, however, that under the combined conditions of:
heavy rain + low illumination + partially obscured road markings
the simulated perception stack consistently identifies lane boundaries earlier
than the real system.
The deviation is classified as:
Conditional + Systematic + Material.
Instead of concluding that the entire simulation is inaccurate, SDMM localizes
the problem to a specific operational region and transfers the governed
deviation to Reality Alignment. Use Case 2 — Spacecraft Autonomous Recovery
A spacecraft simulation predicts successful autonomous recovery from temporary
communication loss across a broad mission scenario set.
Operational telemetry from comparable subsystem and mission conditions shows
that real recovery begins materially later when:
low power state + communication degradation + thermal constraint
occur simultaneously.
SDMM identifies a conditional temporal and behavioral deviation that is
largely invisible in nominal mission correlation statistics.
The deviation becomes a governed input to SRAM without prematurely declaring
either the simulation or mission system invalid.
Architectural Position
SDMM is the third module of the Simulation-to-Reality Correlation Standard(SRCS).
The internal progression is:
SRCM — Simulation-Reality Comparison
↓SCCM — Simulation Correlation Criteria
↓SDMM — Simulation-Reality Deviation
↓SRAM — Reality Alignment
↓SCUM — Correlation Uncertainty
SDMM occupies the critical point where:observed correspondence becomes structured knowledge about non-correspondence.
Foundational Principle
Simulation similarity is not reality equivalence, and aggregate correlationmust never conceal the specific conditions under which simulation and reality
materially diverge.