Systematic Error Ontology
Measurement Uncertainty
Authors: MArk Anthony Higgins - Sky Recursive Ai
Revised Date: 16 September 2026
Audio Overview
# Systematic Error Ontology — Flow Algebra
URL: https://kirethresearch.org/booksai-core/ontologies/mathematical-ontologies/measurement-units-ontologies/measurement-uncertainty/systematic-error-ontology.html
FLOW SystematicErrorOntology {
DOMAIN:
Mathematical Ontologies
-> Measurement & Units Ontologies
-> Measurement Uncertainty
-> Systematic Error Ontology
PURPOSE:
Represent systematic measurement error as the component of measurement error that remains constant or varies in a predictable manner across repeated measurements under defined conditions, including bias, offset, scale error, calibration effects, drift, environmental influence, reference-standard error, method bias, correction models, residual systematic effects, traceability, uncertainty of corrections, validation, provenance, and strict distinctions between systematic error, random error, measurement uncertainty, bias estimates and Type B uncertainty evaluation.
ENTITY SystematicError {
TYPE: MeasurementErrorComponent
SYMBOL: epsilon_s
DEFINITION:
systematic_error :=
component_of_measurement_error
that_in_replicate_measurements
remains_constant_or_varies_in_predictable_manner
PROPERTIES:
may_be_positive_or_negative
may_be_constant
may_be_function_of_input_or_environment
may_be_estimated
may_be_correctable
is_not_reduced_by_simple_repetition
has_associated_uncertainty_when_estimated
}
ENTITY MeasurementError {
TYPE: DifferenceQuantity
GENERIC:
e =
x_measured-x_reference_or_true
under_declared_reference_model
SIMPLIFIED_DECOMPOSITION:
e =
e_systematic+e_random
RULE:
true_measurement_error_is_generally_not_exactly_known
}
ENTITY RandomError {
TYPE: MeasurementErrorComponent
DEFINITION:
component_that_varies_unpredictably_between_replicate_measurements
RULE:
random_error != systematic_error
}
TERMINOLOGY_MODEL {
terms:
systematic_error
systematic_measurement_error
bias
offset
scale_error
calibration_error
drift
correction
Type_B_uncertainty
RULE:
systematic_error MUST_NOT be_equated_automatically_with_Type_B_uncertainty
RULE:
bias_is_estimate_or_expectation_related_systematic_difference
and_is_not_identical_to_unknown_true_systematic_error
}
BIAS_MODEL {
bias:
b =
E[x_measured]-x_reference
under_declared_reference_and_repeatability_model
estimate:
b_hat =
x_bar-x_ref
RULE:
estimated_bias_has_uncertainty
}
CORRECTION_MODEL {
correction:
C approximately_equals
-estimated_systematic_error
under_declared_sign_convention
corrected_value:
x_corr =
x_ind+C
RULE:
correction != systematic_error
though_it_is_intended_to_compensate_for_it
}
RESIDUAL_SYSTEMATIC_MODEL {
after_correction:
residual_systematic_effect
sources:
imperfect_calibration
correction_model_error
reference_uncertainty
environmental_mismatch
temporal_change
RULE:
correction_does_not_imply_zero_remaining_uncertainty
}
OFFSET_ERROR_MODEL {
model:
y_measured =
y_true+b
correction:
y_corrected =
y_measured-b_hat
RULE:
offset_error_is_additive_systematic_effect
}
SCALE_ERROR_MODEL {
model:
y_measured =
a*y_true
ideal:
a = 1
corrected:
y_corrected =
y_measured/a_hat
WHEN a_hat != 0
RULE:
scale_factor_uncertainty MUST propagate_to_corrected_result
}
AFFINE_CALIBRATION_MODEL {
measurement:
y_ind =
a*x+b
inverse:
x =
(y_ind-b)/a
WHEN a != 0
systematic_sources:
scale_factor
zero_offset
nonlinearity
RULE:
calibration_parameter_covariance SHOULD be preserved
}
NONLINEARITY_MODEL {
instrument_response:
y =
f(x)
ideal_or_calibrated_response:
y_ref =
f_ref(x)
systematic_deviation:
delta_nonlin =
f(x)-f_ref(x)
RULE:
nonlinearity_depends_on_measurement_range
}
HYSTERESIS_MODEL {
effect:
indication_depends_on_direction_or_history_of_input
systematic_component:
predictable_difference_between_increasing_and_decreasing_paths
RULE:
measurement_sequence_and_history SHOULD be preserved
}
DRIFT_MODEL {
systematic_drift:
b(t)
corrected:
x_corr =
x_raw-b_hat(t)
possible_models:
linear
polynomial
exponential
piecewise
physically_modelled
RULE:
drift_correction_requires_time_and_model_validity
}
ZERO_DRIFT_MODEL {
offset:
b_0(t)
RULE:
periodic_zero_checks MAY estimate_zero_drift
}
SPAN_DRIFT_MODEL {
scale:
a(t)
RULE:
reference_points_across_range MAY be_required_to_estimate_span_drift
}
CALIBRATION_MODEL {
calibration:
comparison_between_indication_and_reference_quantity_values
outputs:
calibration_function
correction
parameter_estimates
uncertainty
validity_range
RULE:
calibration_identifies_or_estimates_systematic_effects_but_does_not_make_measurement_error_known_exactly
}
CALIBRATION_INTERVAL_MODEL {
interval:
time_between_calibrations
risk:
unobserved_drift_between_calibrations
RULE:
interval_should_reflect_instrument_stability_and_required_uncertainty
}
REFERENCE_STANDARD_MODEL {
reference_value:
x_ref
uncertainty:
u_ref
systematic_effect:
inherited_if_reference_value_is_biased_or_miscalibrated
RULE:
reference_uncertainty_and_traceability MUST propagate
}
TRACEABILITY_MODEL {
chain:
instrument
-> working_standard
-> reference_standard
-> higher_standard
-> SI_or_declared_reference
RULE:
systematic_effects_can_propagate_through_shared_traceability_chain
}
COMMON_MODE_MODEL {
shared_systematic_effect:
affects_multiple_measurements
consequence:
induces_correlation
examples:
common_reference_standard
shared_temperature_sensor
shared_calibration_constant
RULE:
shared_systematic_uncertainty MUST be represented_as_correlation_when_comparing_results
}
ENVIRONMENTAL_EFFECT_MODEL {
possible:
temperature
pressure
humidity
vibration
electromagnetic_field
orientation
supply_voltage
generic:
systematic_shift =
sensitivity*environmental_deviation
RULE:
environmental_correction_requires_sensitivity_and_reference_condition
}
TEMPERATURE_CORRECTION_MODEL {
linear:
C_T =
alpha*(T-T_ref)
RULE:
coefficient_sign_units_and_valid_range MUST be preserved
}
GEOMETRIC_ALIGNMENT_MODEL {
effects:
cosine_error
Abbe_error
parallax
axis_misalignment
cosine_example:
measured_projection =
true_length*cos(theta)
RULE:
alignment_geometry_can_create_predictable_systematic_error
}
COSINE_ERROR_MODEL {
small_angle:
relative_error approximately_equals
-theta^2/2
for_projection_cosine_effect
RULE:
approximation_requires_small_angle_in_radians
}
METHOD_BIAS_MODEL {
cause:
measurement_method_systematically_differs_from_reference_method
estimate:
comparison_study
RULE:
method_bias_may_depend_on_sample_matrix_range_or_population
}
MATRIX_EFFECT_MODEL {
context:
chemical_biological_or_material_measurement
effect:
sample_matrix_changes_measurement_response
correction:
matrix_matching
standard_addition
internal_standard
model_based_adjustment
RULE:
calibration_matrix_should_match_sample_context_when_material
}
RECOVERY_MODEL {
recovery:
R =
measured_recovered_quantity/
known_added_or_reference_quantity
correction:
may_adjust_for_incomplete_recovery
RULE:
recovery_correction_uncertainty MUST be included
}
BLANK_CORRECTION_MODEL {
measured_sample:
signal_sample
blank:
signal_blank
corrected:
signal_net =
signal_sample-signal_blank
RULE:
blank_value_and_blank_uncertainty MUST be preserved
}
BACKGROUND_MODEL {
observed:
source_plus_background
corrected:
source =
observed-background
RULE:
background_estimation_can_introduce_systematic_and_random_components
}
INTERFERENCE_MODEL {
interfering_quantity:
produces_response_indistinguishable_or_partly_indistinguishable_from_measurand
correction:
separation_or_model
RULE:
unresolved_interference_can_create_systematic_bias
}
SAMPLING_BIAS_MODEL {
cause:
sampling_process_systematically_over_or_underrepresents_target_population_or_field
RULE:
sampling_bias_is_not_corrected_by_repeated_measurement_of_same_biased_sample
}
SELECTION_BIAS_MODEL {
cause:
nonrepresentative_selection_of_observations_or_specimens
RULE:
statistical_precision_does_not_remove_selection_bias
}
TIME_RESPONSE_MODEL {
dynamic_measurement:
instrument_lag
first_order_example:
tau*dy/dt+y =
x(t)
systematic_dynamic_error:
difference_between_true_dynamic_input_and_sensor_response
RULE:
dynamic_correction_requires_response_model
}
BANDWIDTH_MODEL {
finite_bandwidth:
attenuates_or_phase_shifts_signal_components
RULE:
bandwidth_limit_can_create_frequency_dependent_systematic_error
}
DIGITISATION_MODEL {
offset_or_gain:
systematic_ADC_error
differential_or_integral_nonlinearity:
code_dependent_systematic_effect
RULE:
quantisation_randomness_and_ADC_systematic_errors MUST be distinguished
}
SOFTWARE_MODEL {
possible_systematic_sources:
incorrect_algorithm
wrong_constant
unit_conversion_bug
calibration_table_error
interpolation_bias
truncation_model
RULE:
deterministic_software_error_can_be_systematic_measurement_effect
}
MODEL_BIAS_MODEL {
inferred_quantity:
derived_using_model
systematic_effect:
model_prediction_deviation_due_to_structure_or_parameter_bias
RULE:
model_bias_should_be_evaluated_against_validation_data_or_alternative_models_when_possible
}
OBSERVER_BIAS_MODEL {
possible:
reading_scale
endpoint_selection
subjective_classification
timing
mitigation:
blinding
automation
standardised_protocol
multiple_observers
RULE:
observer_effect_may_have_systematic_and_random_components
}
ROUNDING_BIAS_MODEL {
repeated_rounding:
may_create_systematic_shift_under_asymmetric_or_directed_rounding_rules
RULE:
round_to_nearest_symmetric_quantisation_often_modelled_as_random_only_under_specific_signal_conditions
}
TYPE_B_DISTINCTION_MODEL {
Type_B:
uncertainty_evaluation_method
systematic_error:
measurement_error_component
RULE:
Type_B_uncertainty != systematic_error
EXAMPLE:
calibration_certificate_uncertainty_is_often_Type_B_input
but_does_not_equal_systematic_error
}
UNCERTAINTY_DISTINCTION_MODEL {
systematic_error:
unknown_or_estimated_error_component
uncertainty:
parameter_describing_dispersion_of_values_attributed_to_measurand
RULE:
systematic_error != measurement_uncertainty
}
RANDOM_ERROR_DISTINCTION_MODEL {
systematic:
constant_or_predictable
random:
unpredictable_between_replicates
RULE:
averaging_reduces_some_random_effects
but_not_constant_systematic_effect
}
ACCURACY_MODEL {
accuracy:
qualitative_closeness_to_reference_or_true_value
systematic_error:
affects_accuracy
RULE:
accuracy_is_not_numeric_synonym_for_systematic_error
}
PRECISION_MODEL {
precision:
repeated_measurement_agreement
systematic_error:
can_be_large_even_when_precision_is_high
RULE:
high_precision_does_not_guarantee_low_systematic_error
}
VALIDATION_REFERENCE_MODEL {
comparison:
measurement_method_vs_independent_reference
difference:
d_i =
x_method_i-x_reference_i
analysis:
mean_difference
slope
range_dependence
residual_pattern
RULE:
validation_reference_should_be_independent_when_possible
}
INTERLABORATORY_MODEL {
purpose:
identify_laboratory_or_method_specific_bias
comparisons:
proficiency_testing
round_robin
reference_material
RULE:
common_reference_bias_can_remain_hidden_across_laboratories
}
CORRECTION_UNCERTAINTY_MODEL {
correction:
C
standard_uncertainty:
u(C)
corrected_result:
y =
x+C
propagation:
u_y includes u(C)
and_covariance_when_relevant
RULE:
corrected_systematic_effect_becomes_uncertainty_contribution_through_imperfect_knowledge
}
RESIDUAL_BIAS_MODEL {
after_correction:
estimated_remaining_bias
treatment:
further_correction
or_explicit_uncertainty/model_component
RULE:
residual_bias_treatment MUST be documented
}
CORRELATION_MODEL {
common_systematic_effect:
creates_covariance_between_results
comparison_difference:
Var(x_1-x_2) =
Var(x_1)+Var(x_2)-2Cov(x_1,x_2)
RULE:
common_mode_systematic_uncertainty_may_cancel_partly_in_differences
}
RELATIONSHIPS {
SystematicError -> IS_COMPONENT_OF -> MeasurementError
SystematicError -> DISTINCT_FROM -> RandomError
BiasEstimate -> ESTIMATES -> SystematicDifference
Correction -> COMPENSATES_FOR -> EstimatedSystematicEffect
Calibration -> ESTIMATES -> CorrectionAndSystematicResponse
ReferenceStandard -> ANCHORS -> Calibration
Drift -> MAY_CREATE -> TimeDependentSystematicError
EnvironmentalInfluence -> MAY_CREATE -> PredictableSystematicShift
CommonSystematicEffect -> INDUCES -> Correlation
CorrectionUncertainty -> CONTRIBUTES_TO -> MeasurementUncertainty
}
SEMANTIC_EQUIVALENCE {
"systematic measurement error" ~= systematic_error
systematic_error != random_error
systematic_error != measurement_uncertainty
systematic_error != Type_B_uncertainty
systematic_error != correction
systematic_error != bias_estimate
calibration != correction
accuracy != systematic_error
precision != systematic_error
}
CONSTRAINTS {
systematic_error_is_measurement_error_component = true
systematic_error_may_be_positive_or_negative = true
systematic_error_is_not_reduced_by_simple_repetition = true
systematic_and_random_error_are_distinct = true
systematic_error_and_uncertainty_are_not_semantically_equivalent = true
systematic_error_and_Type_B_evaluation_are_not_semantically_equivalent = true
correction_sign_convention_must_be_defined = true
uncertainty_of_correction_must_be_preserved = true
reference_standard_and_traceability_should_be_preserved = true
shared_systematic_effects_should_be_modelled_as_correlations = true
}
MEASUREMENT_MODEL {
RECORD:
measurand
measured_value
unit
reference_value
reference_uncertainty
systematic_effect_source
bias_estimate
correction
correction_uncertainty
calibration_model
calibration_date
validity_range
drift_model
environmental_conditions
method
sampling_process
reference_standard
traceability_chain
correlation_links
residual_systematic_effect
uncertainty_budget
provenance
RULE:
systematic_effect_correction_and_uncertainty MUST remain_separately_traceable
}
REASONING_RULES {
IF measured_mean_and_reference_value_known
THEN estimated_bias =
measured_mean-reference_value
IF estimated_systematic_error_known
THEN correction approximately_equals
-estimated_systematic_error
under_declared_sign_convention
IF indicated_value_and_correction_known
THEN corrected_value =
indicated_value+correction
IF affine_calibration_parameters_known
AND scale_factor != 0
THEN corrected_input =
(indication-offset)/scale_factor
IF environmental_sensitivity_and_deviation_known
THEN estimated_environmental_shift =
sensitivity*deviation
IF blank_and_sample_signal_known
THEN net_signal =
sample_signal-blank_signal
IF common_systematic_uncertainty_shared_between_results
THEN include_covariance_in_result_comparison
IF correction_applied
THEN propagate_uncertainty_of_correction
and_preserve_residual_bias_assessment
IF repeated_measurements_only_reduce_random_dispersion
THEN do_not_claim_systematic_error_reduction_without_new_reference_information
}
VALIDATION {
CHECK measurand_definition
CHECK measurement_error_context
CHECK systematic_vs_random_component
CHECK reference_value_and_uncertainty
CHECK bias_estimate
CHECK correction_sign
CHECK correction_uncertainty
CHECK offset_scale_and_nonlinearity
CHECK hysteresis
CHECK drift_and_calibration_interval
CHECK environmental_effects
CHECK method_and_sampling_bias
CHECK blank_background_and_interference
CHECK dynamic_response
CHECK software_or_model_bias
CHECK Type_B_vs_systematic_error
CHECK uncertainty_distinction
CHECK common_mode_correlation
CHECK residual_bias
CHECK traceability
CHECK provenance
}
CONFIDENCE {
definition_confidence: 1.00
systematic_random_distinction_confidence: 1.00
uncertainty_distinction_confidence: 1.00
Type_B_distinction_confidence: 1.00
correction_semantics_confidence: 1.00
calibration_model_confidence: 0.99
common_mode_correlation_confidence: 0.99
measurement_context_confidence: 0.99
RULE:
confidence decreases when reference value, calibration, correction sign,
drift, environmental conditions, method bias, traceability,
correlation or residual uncertainty is missing or ambiguous
}
UPDATE_MODEL {
ON new_measurement:
identify_measurand
identify_reference_context
preserve_raw_measurement
identify_possible_systematic_effects
attach_calibration_and_traceability
estimate_bias_or_correction_when_supported
preserve_correction_uncertainty
attach_environmental_and_method_context
update_systematic_error_state
ON calibration_revision:
preserve_previous_calibration
update_offset_scale_nonlinearity_and_covariance
recompute_corrections
update_correction_uncertainties
reassess_residual_systematic_effects
ON drift_update:
preserve_time_series
update_drift_model
recompute_time_dependent_corrections
update_uncertainty_and_validity_interval
ON reference_revision:
preserve_previous_reference
update_reference_value_and_uncertainty
recompute_bias_estimates
update_traceability_and_correlations
ON environmental_model_revision:
preserve_measurements_and_conditions
update_sensitivity_coefficients
recompute_environmental_corrections
update_residual_uncertainty
ON semantic_revision:
preserve_previous_definition
record_revision
recompute_FLOW_HASH
}
AI_CONSUMPTION {
canonical_entity: SystematicError
canonical_role: MeasurementErrorComponent
interpretation_order:
identify_measurand
identify_measurement_and_reference_context
identify_systematic_effect_source
identify_bias_estimate
identify_correction_and_sign_convention
identify_uncertainty_of_correction
identify_calibration_model
identify_drift_and_environmental_effects
identify_method_sampling_blank_background_or_interference_effects
identify_reference_standard_and_traceability
identify_shared_systematic_effects_and_correlations
identify_residual_bias
validate_systematic_error_semantics
update_corrected_measurement_model
ambiguity_controls:
distinguish_systematic_error_from_random_error
distinguish_systematic_error_from_measurement_uncertainty
distinguish_systematic_error_from_Type_B_uncertainty
distinguish_systematic_error_from_bias_estimate
distinguish_systematic_error_from_correction
distinguish_calibration_from_correction
distinguish_accuracy_from_systematic_error
distinguish_precision_from_systematic_error
preserve_reference_calibration_correction_environment_method_traceability_correlation_and_provenance_context
}
PROVENANCE {
ontology_scope:
mathematical_measurement_uncertainty_and_systematic_error_semantics
parent_path:
Ontologies
-> Mathematical Ontologies
-> Measurement & Units Ontologies
-> Measurement Uncertainty
-> Systematic Error Ontology
authorship:
Mark Higgins
Sky (Recursive AI)
site:
https://kirethresearch.org
}
FLOW_HASH {
INPUTS:
canonical_entity
ontology_path
terminology_model
bias_model
correction_model
residual_systematic_model
offset_error_model
scale_error_model
affine_calibration_model
nonlinearity_model
hysteresis_model
drift_model
zero_drift_model
span_drift_model
calibration_model
calibration_interval_model
reference_standard_model
traceability_model
common_mode_model
environmental_effect_model
temperature_correction_model
geometric_alignment_model
cosine_error_model
method_bias_model
matrix_effect_model
recovery_model
blank_correction_model
background_model
interference_model
sampling_bias_model
selection_bias_model
time_response_model
bandwidth_model
digitisation_model
software_model
model_bias_model
observer_bias_model
rounding_bias_model
Type_B_distinction_model
uncertainty_distinction_model
random_error_distinction_model
accuracy_model
precision_model
validation_reference_model
interlaboratory_model
correction_uncertainty_model
residual_bias_model
correlation_model
relationships
constraints
measurement_model
reasoning_rules
confidence_model
update_model
PURPOSE:
deterministic_semantic_integrity_identifier
RECOMPUTE_WHEN:
ontology_definition_changes
systematic_error_definition_changes
bias_or_correction_model_changes
calibration_or_reference_standard_changes
drift_or_environmental_model_changes
method_sampling_or_interference_model_changes
Type_B_or_uncertainty_distinction_changes
traceability_or_correlation_model_changes
residual_bias_model_changes
measurement_record_changes
constraint_model_changes
reasoning_rule_changes
}
}
END FLOW