Aparameterassociatedwiththeresultofameasurement,thatcharacterisesthedispersionofthevaluesthatcouldreasonablybeattributedtothemeasurand.Aparameterassociatedwiththeresultofameasurement,thatcharacterisesthedispersionofthevaluesthatcouldreasonablybeattributedtothemeasurand.Definition:MeasurementuncertaintyMainReferenceReference:VAMProject3.2.1DevelopmentandHarmonisationofMeasurementUncertaintyPrinciplesPart(d):ProtocolforuncertaintyevaluationfromvalidationdataJanuary2000LGC/VAM/1998/088contractwiththeDepartmentofTradeandIndustryaspartoftheNationalMeasurementSystemsValidAnalyticalMeasurement(VAM)ProgrammeOtherReferencesOtherReferencesILACGuide17IntroducingtheConceptofUncertaintyinTestinginAssociationwiththeApplicationoftheStandardISO/IEC17025(2002)(2002)(2000)(2000)(12/2003)(12/2003)‘sneedsEstimationCalculationObjectivesObjectivesReasonableestimationofuncertaintyresources.Ensurancenotgivingawrongimpressionabouttheuncertainty.RequiredbythetestmethodRequiredbytheclientRequiredbecauseoftheexistanceofnarrowlimitsonwhichdeciciousonconformancetoaspecificationarebased ????1.1.2.2.4.4.3.3.Theupperlimitisexceededwith95%propabilityifthedifferencebetweenthemeasuredvalueandtheupperlimitisbiggerthanthecriticaldifference.Theupperlimitisexceededwith95%propabilityifthedifferencebetweenthemeasuredvalueandtheupperlimitisbiggerthanthecriticaldifference.PrinciplesofmeasurementuncertaintybiasPrinciplesofmeasurementuncertaintyReportingtheResultCd=0.120.02mg/kg(expandeduncertainty,coveragefactor=2;95%confidenceinterval)Check:Existanceofnormativerequirementsontheuncertaintyofmeasurement?RegulationsinthespecialworkinginstructionCheck:„Conflict“oftestresultswithlimitsofspecificationsbecauseofuncertainty?Check:Doestheclientrequiretheestimationofuncertainty?Noestimationofuncertainty1.2.3.yesyesnononoyesStepsIdentificationofallcomponentsofuncertaintyCalculationand/orestimationoftheuncertaintycomponentsEvaluationoftheuncertaintycontributionstothecombineduncertaintyCalculationofthecombined/Expandeduncertainty4.5.6.7.StepsSourcesofuncertaintyOperatoreffectsRandomeffectsSamplingStorageconditionsInstrumenteffectsReagentpurityAssumedstoichiometryMeasurementconditionsSampleeffectsComputationaleffectsBlankcorrectionIdentificationofsourcesofuncertaintyFishboneanalysis–causeandreason-exampleU1U2U3U4U5U6U7U8MeasurementUncertaintyMeasurementAllunknownuncertaintiescannotbecorrected.dispersionofpossibleresultsIdentificationofsourcesofuncertaintySignandmagnitudeoftheMeasurementUncertaintyisunknown,attempttoquantifytheextentofouruncertaintySignandmagnitudeoftheMeasurementUncertaintyisunknown,attempttoquantifytheextentofouruncertaintyFortunately,theMeasurementUncertainty“isastatisticalvariablefollowingaprobabilitydistribution.Fortunately,theMeasurementUncertainty“isastatisticalvariablefollowingaprobabilitydistribution.EstimationofMeasurementUncertaintyInchemicalanalysismeasurement‘srepeatabilityfollowsusuallyanormaldistribution.Inchemicalanalysismeasurement‘srepeatabilityfollowsusuallyanormaldistribution.BundesanstaltfürMaterialforschunMeasurementuncertaintyandnormaldistributionToquantifiesthepossiblediversionaroundtheTrueValue.StandardUncertaintyMeasurementUncertaintyisoftenreferredtoMeasurementUncertaintyStatisticaltestsMeasurementrisksOtherdecision-relatedactivitiesStandardUncertaintyUncertaintiesundercontrol,e.g.controlchartsStandardMethodsMainsourcesofuncertaintiesareknownDefineduncertaintiesApprochesforuncertaintyevaluationThestandarddeviationallowsquantificationofalltheUnknownuncertaintiesintheanalyticalprocess.ThisiscalledStandardUncertainty(=uncertaintyofameasurementexpressedasastandarddeviation)StandardUncertaintyMethodsofevaluationofstandarduncertaintyTypeAusingstatisticalanalysisofaseriesofobservationsTypeBusingotherapproachesthanstatisticalanalysisTheclassificationonlyindicates2differentwaysanddoesnotclassifydifferencesinthenatureoftheuncertaintycomponents.MeasurementMeasure