Morehouse Instrument Company - The Force in Calibration since 1925 IndySoft

Start with the End in Mind

Decision Rule & Measurement Risk Calculator

By Henry Zumbrun II (Morehouse Instrument Company)
and Greg Cenker (IndySoft)
NCSLI 2026 - 4-hour tutorial on decision rules, risk, and traceability

Risk Calc & Curves

Interactive decision-rule calculator. Enter measurement and tolerance, compare all seven guard-band methods, see the Gaussian risk curve update in real time.

Inputs

Use 0.9167 if unknown

Derived Parameters & Verdict
Tolerance T
-
U (k = 2)
-
TUR (Cm)
-
Guard Band Width (w)
-
Upper Acceptance Limit
-
Lower Acceptance Limit
-
Pc at AL
-
Pc at TL
-
Specific PFA at xm
-
PASS -
Risk Curve - Measurement Distribution vs. Tolerance & Acceptance Limits

Blue curve: measurement distribution φ(x) centered at the measured value with σ = u. Red dashed lines: tolerance limits (TL, TU). Green solid lines: acceptance limits after guard band. Area under curve outside the acceptance limits is the specific risk.

Guard Band & Decision Rule Comparison - All Seven Methods
Decision Rule GB Width (w) Upper AL Lower AL Usable Range Framework

All formulas implemented per ANSI/NCSL Z540.3 Handbook, ILAC G8:2019, and ASME B89.7.3.1. Method 6 uses Dobbert M = max(0, 1.04 − exp(0.38 ln(TUR) − 0.54)).

Bias Impact Analysis

Uncorrected systematic error shifts the measurement distribution and increases PFA. Compares the same measurement with and without bias correction.

Without Bias Correction
Measurement Error:-
Total PFA:-
Simple-Acceptance Decision:-
With Bias Correction
Measurement Error:-
Total PFA:-
Simple-Acceptance Decision:-

Key takeaway: TUR alone is not sufficient. A 4:1 TUR requirement assumes the measurement process is centered (bias corrected). When bias is not corrected, the risk of making a false accept can be dramatically underestimated, undermining traceability and consumer protection.

Conformance Probability → Guard Band Multiplier Reference

Guard Band Multiplier r = NORM.S.INV(Pc) / 2, applied as: AL = TL − r × U95. From the Cenker/Zumbrun NCSLI 2025 Training.

Conformance Probability (Pc) Max Risk Per Side GB Multiplier (r)

Risk Pyramid - Traceability Risk Cascade

Three-level traceability chain (End-Item → Cal Standard → Reference). Compute PFA at each tier and the cascading conformance probability across the chain.

Level 1 - End-Item

Fraction of reading (e.g. 0.005 = 0.5 %)

Level 1 - End-Item Uncertainty Budget (k = 1 standard contributions)

Each contribution is a standard uncertainty (k = 1). Combined via RSS, then expanded by k = 2 to give L1's expanded measurement uncertainty U. Defaults match the workbook (Resolution = 1, Repeatability = 2, Reproducibility = 1, Drift = 1, Systematic = 1).

Combined uc (RSS): -
Expanded U (k = 2): -
Level 2 - Cal Standard

Fraction of reading (e.g. 0.001 = 0.1 %)

Percent of reading (e.g. 0.02 = 0.02 %)

Level 3 - Internal Reference

Only applies when Level 2 Calibration Source is "Internal Standard".

Percent of reading

Overall Verdict
PASS -
Cascading conformance probability: -
Combined chain risk: -
Level-by-Level Risk Summary
Level TUR Guard Band Global PFA Global PFR Verdict

Global PFA computed via Drezner-Wesolowsky (1990) 20-point Gauss-Legendre quadrature of the bivariate normal CDF. Matches the workbook's L1/L2/L3 BVN engine. Specific risk at the acceptance limit shown separately on the Risk Calc tab.

Z540.3 Dominance Checks
EOPR Dominance (≥ 89 %)
-
TUR Dominance (≥ 4.6 : 1)
-

Z540.3 establishes two independent sufficient conditions for global PFA ≤ 2 %: either TUR ≥ 4.6 : 1 (Method 6 guard band vanishes) or EOPR ≥ 89 % (Mimbs "Rule of 89"). Either dominance condition alone is sufficient.

Method 5, Method 6 & Test Limit

PFA calculation across an entire loading range. Ten evaluation points (10 % to 100 % of capacity), five readings per point, with Method 5 (RSS), Method 6 (Dobbert), and Test Limit (k = 2 subtraction) acceptance limits side by side.

Per-Point Loading Data - Enter Up to 5 Readings per Point
Applied Reading 1 Reading 2 Reading 3 Reading 4 Reading 5 Avg Req. Tol LSL USL U95 TUR M5 AL M6 AL TL (k=2) Status
Max U95: -
Worst-case TUR: -
Overall status: -

Combined uncertainty: uc = k × √((CMC/k)² + (resolution/3.464)² + repeatability²). Method 5 AL: AL = T − U·TUR·(1 − √(1 − 1/TUR²)). Method 6 AL: AL = T − U·max(0, 1.04 − exp(0.38·ln(TUR) − 0.54)). Test Limit: AL = T − U.

Metrology Cost Model

Economic impact of measurement equipment selection. Compare DMM tiers (Cheapest through Gold) using bivariate-normal PFA/PFR per Sandia PSL methodology, plus total cost including false-reject retest and false-accept downstream costs.

1. Production Inputs

e.g. ohms - a 1500 Ω ± 0.2 Ω resistor

Total Production Cost: -
2. Equipment Comparison

Edit equipment name, price, and 1σ uncertainty to model your own tiers. Default values reflect the seven Acme DMM tiers in the workbook.

Equipment Price ($) 1σ Unc. TUR Global PFR Global PFA FR Cost FA Cost Total Risk Cost
3. Total Cost Analysis (Equipment + Production + Risk)
Equipment Equip Cost Production FR Cost FA Cost Total vs. Baseline Payback (days)
4. Key Insights
Cumulative Total Cost Over Time

Each line is one equipment tier. X-axis is years of production; Y-axis is cumulative cost including equipment, production, false rejects, and false accepts. Cheaper equipment with low TUR compounds risk cost every year.

Reliability & Sample Size

End-of-Period Reliability (EOPR) estimation from calibration history using Clopper-Pearson exact binomial confidence intervals. Sample-size calculator for demonstrating reliability targets.

Calibration History
Results
Observed EOPR (point): -
Failure Rate: -
Lower CB (Clopper-Pearson): -
Upper CB (Clopper-Pearson): -
Min sample (c = 0): -
Min sample (c ≤ 1): -
PASS -
Sample Size for Zero Failures - Quick Reference

n = ⌈ln(1 − Confidence) / ln(Reliability)⌉, from the binomial success-run theorem.

Target EOPR 80 % Conf. 90 % Conf. 95 % Conf. 99 % Conf.
Methodology Notes
EOPR Definition
End-of-Period Reliability - probability that an instrument remains in tolerance at the end of its calibration interval.
Point Estimate
EOPR = 1 − c/n. Simple but ignores sample-size uncertainty.
Clopper-Pearson CI
Exact binomial confidence interval using the Beta distribution. Conservative (guaranteed coverage). NIST recommended.
Z540.3 "Rule of 89"
EOPR ≥ 89 % is the global dominance condition that bounds unconditional PFA ≤ 2 % regardless of TUR (Mimbs).
Assumptions
Bernoulli trial model - each calibration result is independent, probability of in-tolerance is constant across the population.
References
ANSI/NCSL Z540.3-2006 §B.2 • ILAC-P14:09/2020 • NASA-HDBK-8739.19 §4.3 • NCSLI RP-1 §5.4

Coverage Factor (k) ↔ Coverage Probability

Two-way conversion between coverage factor k and coverage probability for a normal distribution. Per JCGM 100:2008 (GUM) Annex G and NIST SP 811 §7.4.

Coverage Factor → Probability

Pick a k value, get the two-sided coverage probability p = 2·Φ(k) − 1.

Coverage Factor (k)Probability
Probability → Coverage Factor

Pick a target probability, get the required k = Φ⁻¹((1 + p) / 2).

ProbabilityCoverage Factor (k)
Custom Lookup
→ Probability: -
→ Coverage Factor (k): -
What This Means
What k means
k is the coverage factor used to expand a standard uncertainty u into an expanded uncertainty U = k · u. The choice of k sets the coverage probability - the fraction of the distribution captured by U around the measurand.
Common values
k = 1 → ≈ 68.27 % • k = 1.96 → 95.00 % • k = 2 → ≈ 95.45 % • k = 2.576 → 99.00 % • k = 3 → ≈ 99.73 %. Calibration certificates almost always quote U at k = 2 (about 95 % coverage).
k = 2 vs k = 1.96
k = 1.96 gives exactly 95.00 % under a normal distribution; k = 2 gives 95.45 %. GUM and ISO/IEC 17025 labs use k = 2 by convention because it is simple and conservative. ILAC G8:2019 §B.3 uses k = 1.96 for guard band sizing where exact 95 % matters.
When the distribution isn't normal
These tables assume a normal output distribution - the standard assumption when the combined uncertainty has enough degrees of freedom (Welch-Satterthwaite νeff ≳ 30). For low-DOF cases, use a t-distribution coverage factor instead.
Reference
JCGM 100:2008 (GUM) Annex G • NIST SP 811 §7.4 • ILAC G8:2019 §B.3 • ANSI/NCSL Z540.3 Handbook

Guarded Rejection - Speed-Limit Analogy

Guard-banding for rejection (not acceptance). How much margin a police officer with a radar gun would have to allow before issuing a ticket, given measurement uncertainty. Mirror image of guarded acceptance - protects the customer from a false reject.

Guarded Rejection Calculator
Acceptance Threshold: -
Probability of a wrong decision: -
Single-Sided z-Table
Probabilityz-Value
Speed Limit Table

Guard-banded rejection thresholds at the current confidence level and measurement uncertainty. Officer would not issue a ticket until the radar reads at least this value.

Posted Speed (mph)Guard-Banded Threshold (mph)
How to Read This Tab
The analogy
An officer's radar gun reads your speed with some uncertainty. To be 95 % confident a driver is actually speeding before issuing a ticket, the officer must let the reading exceed the posted limit by enough margin that the true value almost certainly exceeds the limit too.
The formula
Threshold = Limit / (1 − um · NORMSINV(P)). With P = 95 %, z = 1.645 (single-sided). A 2 % relative uncertainty raises a 75 mph limit to ≈ 77.6 mph before a citation is justified.
Why this matters
ISO/IEC 17025:2017 §7.8.6 requires a documented decision rule when a statement of conformity is issued. Guarded rejection (ILAC G8:2019 Case 3) protects the customer from a false reject - mirror image of guarded acceptance protecting from a false accept.
Reference
ISO/IEC 17025:2017 §7.8.6 • ILAC G8:2019 §B (Cases 1-4) • JCGM 106:2012 §9

Guidance Summary

Decision rules and statements of conformity - simplified guidance for calibration laboratories. Based on ISO/IEC 17025:2017, UKAS LAB 48 (Ed. 5, 2024), ILAC-G8:09/2019, JCGM 106:2012, ANSI/NCSL Z540.3, and ASME B89.7.

1. Key Definitions
TermDefinitionSource
Decision RuleA rule that describes how measurement uncertainty is accounted for when stating conformity with a specified requirement.ISO 17025 §3.7
Guard Band (w)An interval between a tolerance limit (TL) and a corresponding acceptance limit (AL), used as a buffer.ILAC-G8:09/2019
Tolerance Limit (TL)The specified upper or lower bound of permissible values of a property (the specification limit).ILAC-G8:09/2019
Acceptance Limit (AL)The specified upper or lower bound of permissible measured values after applying a guard band.ILAC-G8:09/2019
PFAProbability of False Accept - probability a non-conforming item is accepted (consumer's risk).JCGM 106:2012
PFRProbability of False Reject - probability a conforming item is rejected (producer's risk).JCGM 106:2012
TURTest Uncertainty Ratio = Tolerance / Expanded Measurement Uncertainty. Higher TUR = less measurement risk.ANSI Z540.3
Simple AcceptanceDecision rule where AL = TL (no guard band). Carries significant PFA risk near limits. Must be paired with adequate TUR.ILAC-G8, §1.9
2. ISO/IEC 17025:2017 Requirements
ClauseRequirement (Plain Language)Key Implication
3.7 - DefinitionA decision rule describes how measurement uncertainty is accounted for when stating conformity. Any conformity statement requires that uncertainty be considered.No uncertainty = no metrological traceability.
7.1.3 - Customer AgreementThe specification, standard, and decision rule shall be clearly defined. Unless inherent in the specification, the decision rule shall be agreed with the customer.Proactive communication required before measurement.
7.8.6.1 - Documentation & RiskThe laboratory shall document the decision rule employed, taking into account the level of risk (PFA/PFR).Must document risk level, not just pass/fail.
7.8.6.2 - ReportingThe certificate must identify: (a) which results the conformity statement applies to, (b) which specifications were met or not met, and (c) the decision rule applied.All three elements required on every certificate.
3. Decision Rule Comparison - Core Approaches
Decision RuleGuard Band FormulaFrameworkTargetReference
1. Simple AcceptanceGB = 0 (AL = TL)Neither - risk ignoredNone (shared)ASME B89.7.3.1 • ILAC-G8 §6.1
2. Specific Risk GBAL = TL − (GB_mult × U); GB_mult = NORM.S.INV(1−risk)/2Specific (conditional)≤ 2.5 % PFA per sideILAC-G8:09/2019 • JCGM 106:2012
3. ILAC G8 (95 %)w = u × 1.96 (fixed z for 95 % bilateral)Specific (conditional)Pc ≥ 98 %ILAC-G8:09/2019 • ISO 14253-1
4. Dobbert Method 6GB = U · M(TUR); M = max(0, 1.04 − exp(0.38·ln(TUR) − 0.54))Global (unconditional)PFA ≤ 2 %ANSI/NCSL Z540.3 Handbook
5. RSS / Method 5GB = U × TUR × (1 − √(1 − 1/TUR²))Global (unconditional)PFA ≤ 2 %ANSI/NCSL Z540.3 Handbook
6. Fixed 0.98 × UGB = 0.98 × U (matches ILAC G8 at k = 2)Specific-equivalentConservative (Pc ≈ 0.975)ILAC G8 (simplified)
7. k = 2 SubtractionGB = U (full uncertainty subtracted; AL = T − U)Producer-biasedConservativeMIL-STD-45662A (historical)
4. Non-Binary Decision (3-Outcome Model)
OutcomeConditionSpecific PFAAction
PASSMeasured value within acceptance limits (AL)≤ agreed threshold (e.g. 2.5 %)State conformity with confidence.
POSSIBLE PASSWithin tolerance limits (TL) but outside acceptance limits (AL)> threshold but < 50 %Report MV, U, and calculated PFA. Customer decides disposition.
FAILSpecific PFA ≥ 50 %, or measured value outside TL by more than U≥ 50 %State non-conformity.

Example certificate language (Possible Pass): "Measured value: 0.82 units. Specification: ±1 unit. Decision rule: Specific risk with 2.5 % max PFA per side. Calculated specific PFA at this point: 3.4 %. Outcome: Possible Pass - within tolerance but specific risk exceeds the agreed threshold. Customer to determine disposition."

5. Sizing Rules & Backward-Sizing Math

Rule of thumb: At shared-risk (no guard band), TUR ≳ 4.6 : 1 is a conservative boundary for PFA ≤ 2 %. Required U (k = 2) = Tolerance / 4.6. If EOPR ≳ 89 %, worst-case PFA ≤ 2 % even at lower TUR.

Device Tolerance (T)Required U (k = 2)Resulting TURGlobal Risk
0.00010.000 0224.6 : 1PFA ≤ 2 %
0.00050.000 1094.6 : 1PFA ≤ 2 %
0.0010.000 2174.6 : 1PFA ≤ 2 %
0.0050.001 0874.6 : 1PFA ≤ 2 %
0.010.002 1744.6 : 1PFA ≤ 2 %
6. Five Rules to Reduce Measurement Risk

The "Measurement Stool" - three legs (Requirements, Equipment, Process) plus Verification and Continuous Improvement.

#RuleKey Points
1Know the Right RequirementsEngage metrology before purchase. The more accurate the system, the higher the costs to procure and calibrate. Consult metrology before procurement - technicians know what fails.
2Use the Right EquipmentMatch CMC and traceability to tolerance. Not all tools are created equal. Look beyond sticker price to long-term cost.
3Follow the Right ProcessesTraining, adapters, environment, SOPs. Process should ensure all aspects of the standards are satisfied. Wrong-process errors dwarf instrument uncertainty.
4Check Your WorkChecklists, cross-checks, measurement assurance. Technicians are human. A good verification program catches the mistakes.
5Stay Vigilant - Continuous ImprovementSPC charts, interval review, reliability tracking. Don't let success lead to complacency.
7. Key Reference Documents
  • ISO/IEC 17025:2017 - Primary standard for lab competence. Clauses 3.7, 7.1.3, 7.8.6.1, 7.8.6.2 govern decision rules.
  • UKAS LAB 48 (Ed. 5, 2024) - Practical guidance with worked examples. Highly recommended reading.
  • ILAC-G8:09/2019 - International guidelines on decision rules. Covers guard band calculation, PFA, and reporting.
  • JCGM 106:2012 - Statistical foundation for understanding risk in conformity assessment.
  • ANSI/NCSL Z540.3 Handbook - The 2 % PFA rule and practical methods (Methods 5 and 6).
  • ASME B89.7.3.1-2001 - Guidelines for decision rules considering measurement uncertainty.
  • ASME B89.7.4.1-2005 - Measurement uncertainty and conformance testing: risk analysis.
  • ISO 14253-1:2017 - Decision rules for verifying conformity or nonconformity with specifications.
  • NIST SP 811 - Guide for the Use of the International System of Units (SI).

Educational Reference - "Who Needs Another Tutorial?"

Companion study guide for the 4-hour NCSLI 2026 tutorial. Each section lists related calculator tabs so you can move between concept and computation.

Learning Objectives
#ObjectiveWhere it lives
1Understand Measurement Traceability requirementsSection 2 below
2Know the fundamentals of Measurement UncertaintySection 3 • Risk Pyramid
3Understand Measurement Data sampling requirementsReliability tab
4Understand the basics of Decision RulesGuidance Summary • Risk Calc & Curves
5Know the differences between Specific and Global Risk modelsSection 7 • Method 5/6/TL tab
6Introduction to Metrology Costing ModelsSection 15 • Cost Model tab
2. Metrological Traceability

ISO/IEC 17025:2017 §6.5.1: The laboratory shall establish and maintain metrological traceability of its measurement results by means of a documented unbroken chain of calibrations, each contributing to the measurement uncertainty, linking them to an appropriate reference.

Definition (VIM / ISO Guide 99): Metrological Traceability - property of a measurement result whereby the result can be related to a reference through a documented unbroken chain of calibrations, each contributing to measurement uncertainty.

TierDescriptionTypical Uncertainty (k = 1)
SI / NMINIST primary realization0.0004 - 0.0005 %
Primary Reference LabMorehouse deadweight primary standards0.0008 %
Accredited ServiceAccredited calibration supplier (secondary)0.02 %
Working StandardsEnd-user reference instruments0.1 %
Field MeasurementIn-use device on the production floor0.5 %

Cumulative effect: Uncertainty is cumulative from one level of the hierarchy to the next. Lowering the reference CMC by ~10× can transform a 6 %+ PFA failure into a near-zero PFA result on the same instrument.

3. Measurement Uncertainty - CMC and Its Contributors

Measurement Uncertainty (definition): The doubt that exists about a measurement's result. Every measurement - even the most careful - carries some uncertainty.

CMC - Calibration and Measurement Capability. Typical CMC standard uncertainty contributors:

  • Repeatability
  • Resolution
  • Reproducibility
  • Reference Standard Uncertainty
  • Reference Standard Stability
  • Environmental Effects
  • Operator / Method Effects
Metric (at 10 000 lbf)Typical Commercial LabMorehouse PrimaryDifference
Reference CMC (k = 2)0.04 % ≈ 4 lbf0.0016 % ≈ 0.16 lbf~25× lower
Expanded U95 on UUT≈ 4.03 lbf≈ 0.41 lbf≈ 10× lower
PFA at the spec limitSignificant - guard bands requiredNear zero - minimal guard bandFar lower consumer risk
PFR impactHigh - tight guard bands force false rejectsMinimal - ALs effectively at TLsHigher yield
6. Type I & Type II Errors - Consumer and Producer Risk
ErrorAlso CalledWhat HappensWho Bears the Cost
PFA - False Accept (Type II)Consumer Risk, Pass error, FARYou PASS it … but it is actually out-of-tolerance.Customer / downstream process
PFR - False Reject (Type I)Producer Risk, Fail error, FRRYou FAIL it … but it is actually in-tolerance.You: rework, scrap, retest, lost time

Same root cause: both errors come from uncertainty overlapping the specification limit. Larger uncertainty = more of both. The consequences are not symmetric - consumer risk can mean loss of life or mission, while producer risk typically means unnecessary rework. Both carry cost; consumer risk has the greater potential for major negative consequences and must be tightly controlled.

7. Specific Risk vs Global Risk - Side by Side
AspectSpecific Risk (Bench, Conditional)Global Risk (Process, Unconditional)
Plain-English question"Given THIS result and its uncertainty, is THIS item really conforming?""Across all items in this program, what fraction do we wrongly accept on average?"
ASME B89.7.4.1 descriptionControlling the quality of workpieces (single-item).Controlling the average quality of workpieces (population).
Number of distributionsOne probability distribution (the measurement). Any single-distribution method is specific risk.Two distributions - the measurement and the prior distribution of items (EOPR / ITP).
Required informationMeasured value, U, tolerance. Works on a single point with no history.EOPR or a priori knowledge plus uncertainty. Needs population history (≥ 59 calibrations with 0 failures for 95 % confidence at 95 % reliability).
Typical use caseAerospace, medical, safety - anywhere individual conformance must be assured.Routine high-volume calibration where program-level control is acceptable.
Standard that uses itASME B89.7.3.1, ISO 14253-1, JCGM 106 specific PFA.ANSI/NCSL Z540.3 (PFA ≤ 2 % across population).
CautionAt the tolerance limit, specific PFA can approach 50 % regardless of historical reliability.An instrument that passes global PFA ≤ 2 % may still fail specific risk on any given measurement.

The football analogy. Specific Risk = "Did this kick clear these posts?" (one ball, one outcome). Global Risk = "What is the kicker's career field-goal percentage from this distance?" (the long-run frequency).

9. TUR - Test Uncertainty Ratio (and why TAR isn't the same)

Where 4:1 came from: Hayes & Crandon (U.S. Navy, 1955) proposed 4:1 as a slide-rule-era compromise targeting 1 % consumer risk under specific assumptions. It is not a universal law.

ANSI/NCSL Z540.3: TUR = (USL − LSL) / (2 · U95) for the calibration process - ratio of the tolerance span to twice the expanded uncertainty.

TUR vs TAR - not the same. TAR uses only the manufacturer's accuracy spec for the standard. A 25:1 TAR for a digital micrometer can become a much lower TUR once resolution, repeatability, and environmental effects are included.

Bench-level dominance: at TUR ≥ 4.6 : 1, Method 6 yields M ≤ 0 and the guard band vanishes - the full tolerance is usable. Required U for a given tolerance (PFA ≤ 2 %): Uexp (k = 2) ≈ Tolerance / 4.6.

10. EOPR - End-of-Period Reliability & Sample Size

Plain definition: EOPR = number of calibrations meeting acceptance criteria ÷ total number of calibrations. It is the population-level success rate over a calibration interval.

Sample size formula: Sample Size = ln(1 − Confidence) / ln(Target Reliability). For 95 % EOPR at 95 % confidence with 0 failures, n = 59.

Global dominance: EOPR ≥ 89 % is sufficient to bound unconditional PFA (UFAR) at ≤ 2 % regardless of TUR (Mimbs - Using Reliability to Meet Z540.3's 2 % Rule).

Take-away from Morehouse load-cell data: achieving better than 0.02 % of applied force typically requires very high-end equipment (meters + load cells together).

11. Worked Examples

Star Wars - Specific Risk Acceptance Limit

Setup: Vent port 2 m wide, proton torpedo 0.5 m, standard uncertainty u = 0.125 (k = 1), applied as U at k = 2 → U = 0.25.

Math: GB multiplier = NORM.S.INV(0.975) / 2 = 0.980. Guard band = 0.980 · (2 · 0.125) = 0.245. Acceptance interval narrows to ±0.755 m.

Lesson: Even when the system looks comfortable on paper, uncertainty consumes 24.5 % of each half-tolerance.

Radar Gun - Global vs Specific Risk

Setup: Speed-limit zone 60 mph ± 3 mph. Two radar guns at points A and B; TUR = 6:1.

Specific Risk view: Car clocks 65 mph at A and 55 mph at B - each point individually outside the limit. Specific risk treats every measurement on its own.

Global Risk view: Across the segment, the car's average speed is 60 mph and 98 % of 10 000 sampled cars are between 57 and 63 mph - the population-level risk is well-bounded.

Resistor Verification

Setup: Fixed resistor 1500 Ω ± 0.2 Ω (±0.013 %). DMM at TUR ~1.44 : 1.

Production reality: ~30 % flagged failing the global GB. Retest in the metrology lab shows < 0.2 % were actual failures - the rest were producer-risk false rejects.

Fix: A more accurate DMM at ~3× the price gives expected false-reject rate 1.035 % and consumer risk < 1 %.

12. Method 5, Method 6, Test Limit - The Algorithms
MethodFormulaNotes
Method 5 (RSS)w = U · TUR · (1 − √(1 − 1/TUR²)); AL = TL − wClosed-form approximation. Good agreement with Method 6 at moderate TUR.
Method 6 (Dobbert)M = max(0, 1.04 − exp(0.38 · ln(TUR) − 0.54)); w = U · MOptimized for PFA ≤ 2 % at all TURs. M = 0 when TUR ≥ 4.6:1.
Test Limit (k = 2)AL = TL − U (full expanded uncertainty subtracted)Conservative. High PFR. Used historically (MIL-STD-45662A).
13. Bias, Accuracy & Precision

Accuracy = low bias (results center on true value). Precision = low random error (results cluster tightly).

The standard 4:1 assumption. The classic TUR/TAR rules assume the measurement process is centered (bias corrected). Uncorrected bias dramatically increases PFA on one side of the distribution.

Worked example - force calibration with +9 lbf bias: An indicator reads 10 000 lbf when the true applied force is 10 009 lbf (bias = +9 lbf, TUR = 58.75:1). Without correction the producer-risk PFR is enormous and many in-tolerance instruments are scrapped.

The Deming Funnel. Don't adjust without understanding. Constant ad-hoc adjustments without understanding the process never hit the target. The right response is to characterize the bias, then correct it consistently.

Bottom line: not correcting for measurement bias is a common problem - customers may be receiving calibrations that are far worse than the certificate suggests.

14. Case Study - "Deflate Gate"

The spec: NFL rulebook ball pressure 12.5 - 13.5 psi. Tolerance ±0.5 psi (half-tolerance).

The instrument actually used: Two gauges - one "no name," one Wilson CJ-01 (made by Jiao Hsiung Industry Corp.). Wilson states no measurement uncertainty.

TUR reality: At best the Wilson provides ±3.3 psig uncertainty (~0.817 × 4 with 4:1 desired) - that is 6.6× less accurate than required.

The fix: An Additel GP30 (0.001 psig resolution, ±0.05 % FS, calibration U ≈ ±0.003 psig at k = 2) costs ~$714 including accredited calibration.

The price of getting it wrong: Deflate Gate totalled more than $22.5 M in investigation cost. The NFL used a $30 gauge in place of a $700 instrument capable of doing the job.

15. The Cost of Risk - Metrology Costing Models

The lens: Total cost = Equipment + Production + (False Reject × retest cost) + (False Accept × downstream cost).

StageTime ScaleImpactCost Trend
RetestMinutesTechnician time, equipment downtime$
RMADaysShipping, processing, customer delay$$
RecallWeeksProduct retrieval, regulatory reporting, rework$$$
Reputation damageYearsLost customers, regulatory scrutiny, brand erosion$$$$
17. Provider Selection Checklist
#CriterionDetail
1SI traceabilityDocumented unbroken chain with uncertainties stated at each tier.
2CMC fits your needsPublished CMCs can achieve the Uexp required by your target TUR.
3Decision-rule capableStates the decision rule used; can apply guard bands sized to your risk target.
4EOPR / trend supportProvides EOPR trends or incorporates your reliability data to justify risk compliance.
5OOT feedbackSupports feedback analysis when their standards are found out-of-tolerance.
6Process uncertainty publishedHas a measurement process uncertainty capable of meeting your needs and follows published standards.
7Replicates how you use itCalibration replicates how the instrument is actually used in the field.
8Correct adaptersUses the right adapters so results are repeatable and comparable.
9Competent techniciansTraining records on file.
10Follows published standardsASTM E74, ISO 376, ASTM E4, etc.
11Reports uncertainty correctlyPer JCGM 100 / GUM requirements.
12On-time and reliableRated highly; consistent delivery.
18. Sample Purchase Order - "Start with the End in Mind"

Two questions before any measurement or purchase: (1) How good does the measurement need to be? (business lens - consequences, regulatory exposure). (2) How will you achieve and verify that capability? (technical lens - equipment, process, reliability evidence).

#ElementExample Language
1Calibration direction"As Found"
2Instrument ID"Manufacturer 10 000 N Load Cell S/N XXXX"
3Indicator pairing"with indicator Manufacturer Readout S/N XXXX"
4Capacity and mode"to 10 000 N in Compression only"
5Tolerance specification"with a tolerance of 0.1 % of full scale"
6Decision rule"Issue 'Pass' when PFA using Specific Risk is ≤ 2.5 %, subtracting U95 for the Guard Band."
7Target risk"Consumer Risk PFA ≤ 2 %; producer risk tolerable."
8Reporting requirements"Report uncertainties at k = 2, U95 with stated confidence interval and calculation method."
20. Decision Rules Summary & Key Takeaways
  1. Calculating Measurement Uncertainty correctly is essential to everything that follows, including decision rules.
  2. Metrological Traceability relies on an unbroken chain of calibrations, each contributing to measurement uncertainty.
  3. A decision rule describes how measurement uncertainty is accounted for when stating conformity (ISO/IEC 17025 §3.7, §7.1.3).
  4. Not correcting for known bias increases measurement risk and undermines traceability.
  5. The 4:1 TUR rule provides a specific risk level only under explicit assumptions - it is not a universal law.
  6. Choose specific risk when per-point control is critical (aerospace, medical, safety).
  7. Choose global risk for high-volume or routine calibration where population-level control is sufficient.
  8. TUR ≥ 4.6 : 1 limits PFA to < 2 % under global risk models; EOPR ≥ 89 % is the global counterpart.
  9. Guard banding is a business decision - balance consumer risk vs producer cost based on consequences.
  10. Lower measurement uncertainty (better lab) directly reduces both PFA and PFR - the best risk mitigation.
  11. Decision rules are not optional - ISO/IEC 17025:2017 requires that measurement uncertainty be accounted for.
21. Recommended Reading

Standards & Guidance Documents

  • ILAC G8:09/2019 - Guidelines on Decision Rules and Statements of Conformity
  • JCGM 106:2012 - Evaluation of measurement data: The role of measurement uncertainty in conformity assessment
  • UKAS LAB 48 - Decision Rules and Statements of Conformity
  • ISO/IEC 17025:2017 - General requirements for the competence of testing and calibration laboratories
  • Handbook for the Application of ANSI/NCSL Z540.3-2006 - Requirements for the Calibration of M&TE
  • The Metrology Handbook, 3rd Edition - Chapter 30
  • NCSLI RP-18 - Estimation and Evaluation of Measurement Decision Risk
  • ASME B89.7.3.1-2001 - Guidelines for Decision Rules
  • ASME B89.7.4.1-2005 - Measurement Uncertainty and Conformance Testing: Risk Analysis
  • ISO 14253-1:2017 - Decision rules for proving conformity or nonconformity with specifications
  • WADA Technical Document - TD2017DK
  • Decision Rule Guidance v1.41 - Cenker, Zumbrun, Shah, et al.

Key Papers

  • Delker, C. - Evaluation of Guard Banding Methods for Calibration and Product Acceptance
  • Deaver, D. & Sompri, J. - A Study of and Recommendations for Applying the False Acceptance Risk Specification of Z540.3
  • Rishi, S. - Guard-banding Methods: An Overview
  • Dobbert, M. - A Guard-Band Strategy for Managing False-Accept Risk
  • Dobbert, M. - Understanding Measurement Risk
  • Harben, J. & Reese, P. - Risk Mitigation Strategies for Compliance Testing
  • Mimbs, S. - Measurement Decision Risk: The Importance of Definitions
  • Mimbs, S. - Conformance Testing: Measurement Decision Rules
  • Mimbs, S. - Using Reliability to Meet Z540.3's 2 % Rule (Rule of 89)
  • Castrup, H. - Analytical Metrology SPC Methods for ATE Implementation
  • Zumbrun, H. & Cenker, G. - The Force of Decision Rules: Applying Specific and Global Risk to Star Wars
  • Cenker, G. & Zumbrun, H. - Unraveling the Tom Brady Deflate Gate
  • Reese, P. - Calibration in Regulated Industries: Federal Agency use of ANSI Z540.3 and ISO 17025

More from Morehouse & IndySoft

  • Morehouse Instrument Company - www.mhforce.com • sales@mhforce.com • (717) 843-0081
  • Morehouse NCSLI Force Course - mhforce.com/ncsli-force-course/
  • Morehouse Force ILC / PT - mhforce.com/calibration/force-ilc-and-pt
  • IndySoft Calibration Management Software - www.indysoft.com • greg.cenker@indysoft.com

Method Validation & Verification Checklist

ISO/IEC 17025 Calibration Laboratories - based on the Blackthorn Consulting Group checklist. Status icons: ✓ Complete, ✗ Gap, ◐ In Progress, N/A Not Applicable.

Header
Method Title: PFA / Risk Calculation - Methods 5, 6, Test Limit, Guarded Rejection
Document No: Start with the End in Mind V10.xlsx
Revision / Date: V10 - 2026-05-19
Standards Applied: NIST SP 811 (2008 ed.), ISO 8601, ISO/IEC 17025:2017
1. Method Identification
ItemStatusEvidence / Notes
Method title and document number definedWorkbook titled "Start with the End in Mind"; methods labeled Method 5, Method 6, Test Limit.
Revision level and effective date recordedRevision V10 identified in file name; effective date 2026-05-19.
Source of method identified (consensus / manufacturer / internal)Consensus - based on ANSI/NCSL Z540.3, ILAC G8:2019, JCGM 106:2012, ISO/IEC 17025:2017.
Scope of measurement clearly definedScope: calibration risk analysis (PFA, TUR, acceptance limits, guard-banded rejection).
Measurement range documentedRange documented as 10 % to 100 % of Instrument Capacity on Method 5/6/TL.
2. Intended Use Definition
ItemStatusEvidence / Notes
Measurand definedMeasurand: instrument indication vs applied reference force/torque.
Measurement principle identifiedStatistical risk analysis (BVN-based PFA) combining UUT readings with CMC, resolution, repeatability.
Input quantities identifiedInputs on Method 5/6/TL: Instrument Capacity, Resolution UUT, Tolerance, Plus 1 Count flag, CMC, k.
Output quantities identifiedOutputs: Avg Reading, TUR, TAR, PFA Lower/Upper/Total, Acceptance Limit, LSL/USL, Pass/Fail.
Units of measurement definedWorkbook is unit-agnostic. Specify units (e.g. lbf, N, N·m, lbf·in) on the method cover sheet.
3. Decision Rule
ItemStatusEvidence / Notes
Decision rule documented per ISO/IEC 17025:2017 §7.1.3Seven decision rules supported: Simple Acceptance, Subtract U95, Method 6 (Dobbert), Constant-PFA Guardband, ILAC G8 (95 %), RSS, Fixed 0.98×U.
Decision rule agreed with customerN/APer ISO/IEC 17025:2017 §7.1.3 the rule must be agreed with the customer. Capture on purchase order.
Target PFA statedPFA target is a user input (default 2 % per Z540.3).
Guard-band method selected and justifiedGuard-band method selected per decision rule. Method 6 (Dobbert) calibrated for 2 % PFA design target.
Guarded acceptance vs guarded rejection identifiedGuarded acceptance covered by Methods 5/6 and Risk Pyramid. Guarded rejection on dedicated tab.
Coverage factor k explicitly stated on certificateN/AU reported as expanded uncertainty at k = 2 (~95 % coverage) per NIST SP 811 §7.4.
4. Equipment and Resources
ItemStatusEvidence / Notes
Required standards and reference equipment identifiedReference-standard framework documented: Risk Pyramid L2/L3 captures external lab CMC and internal reference uncertainty.
Equipment uncertainty appropriate for measurementCMC entered as input on Risk Pyramid and Methods 5/6/TL. TUR computed at each loading point.
Environmental requirements documentedN/AGAP: Environmental conditions (temperature, humidity, stability time) not captured in workbook. Recommend addition.
Personnel competency requirements definedGAP: Competency requirements not in workbook. Reference external training records per ISO/IEC 17025 §6.2.
5. Verification (consensus or manufacturer methods)
ItemStatusEvidence / Notes
Laboratory capability demonstratedDemonstrated via worked example on Method 5/6/TL: 10 loading points (10 % to 100 % of capacity).
Trial measurements performedExample trial data on Methods 5/6/TL - 5 readings at 10 loading points.
Results compared to reference valuesApplied force vs avg measured compared via LSL/USL checks; Pass/Fail shown.
Repeatability confirmedSTDEV across 5 readings per load point feeds uncertainty combination.
Uncertainty budget supports required accuracyCombined per JCGM 106: u = k·√((CMC/k)² + (res/3.464)² + rep²). TUR reported.
7. Risk & Uncertainty Analysis
ItemStatusEvidence / Notes
Uncertainty budget built per Risk Pyramid L1/L2/L33-level cascade (UUT, Working Std, Reference Std) with explicit budget rows for each tier.
TUR cascade verifiedTUR computed at each level; fail/warning flagged when TUR is insufficient.
PFA computed (global and specific)Global PFA via DW1990 20-pt Gauss-Legendre BVN. Specific PFA at acceptance limit computed.
Guard band applied per selected decision ruleGuard-band width w computed per decision rule. Verification check confirms w matches the formula.
Coverage factor k explicitly statedAll expanded uncertainties reported at k = 2 (~95 % coverage) per NIST SP 811 §7.4.
11. NIST SP 811 Conformance - Units & Typography
ItemStatusEvidence / Notes
Value-unit spacing (§7.2)Non-breaking space between numerical value and unit symbol applied across all sheets.
Digit grouping (§10.5.3)Thin space on both sides of decimal marker for 5+ digit numbers (e.g. "10 000 lbf").
Decimal marker (§10.5.2)Period; leading zero for values < 1.
Unit symbol typography (§6.1.1-6.1.4)Roman, case-correct, no plurals, no period. Zero instances of "lbs", "hrs", "kgs", etc.
Multiplication / operator typography (§6.1.8)× or space, not letter x.
Quantity symbols (§7.4)Variable k italicized per SP 811 §7.4 across 69 cells (96 instances).
Percent symbol (§7.10.2)Space before % applied throughout.
Confidence and coverage factor reporting (§7.14, GUM)U reported with explicit k and confidence level.
Date format (§10.4)ISO 8601 YYYY-MM-DD.
Summary of Gaps & Next Actions
  • Environmental conditions block on method cover sheet - Section 4
  • Personnel competency & training records reference - Section 4
  • Technical review and approval signatures pending
  • Internal audit schedule entry per ISO/IEC 17025:2017 Cl. 8.8 - target frequency: annual

Changes Log - Revision History

Controlled change record per ISO/IEC 17025:2017 Cl. 8.3 (Control of Documents) and NIST SP 811 §10.4 date format. Authors: Henry Zumbrun II (Morehouse Instrument Company) and Greg Cenker (IndySoft).

Workbook Metadata
Document: Start with the End in Mind V10.xlsx
Tutorial: NCSLI 2026 - 4-hour workshop on decision rules, risk, and traceability
Authors: Henry Zumbrun II (Morehouse) • Greg Cenker (IndySoft)
Standards: ISO/IEC 17025:2017 • NIST SP 811 • ISO 8601
Revision History
RevDateAuthorSections AffectedChange Summary
V11 (web)2026-05-19Harrison Zumbrun Web port: full HTML/JavaScript implementation of the V10 workbook. Same calculator scope; adds interactive uncertainty budget inputs, CSV import/export, chart PNG download, session save/load. Re-implemented all V10 formulas in JavaScript: 6-point Drezner BVN, PFA_SYMMETRIC, PFR_SYMMETRIC, PFA_ASYMMETRIC, PFR_ASYMMETRIC, IS_SYMMETRIC dispatcher, Clopper-Pearson via bisection on incomplete beta, Acklam inverse normal, NORM.DIST via Abramowitz-Stegun erf. Dobbert M6 guard band applied inside the Cost Model BVN. Methods TUR uses k=2 expanded uncertainty in the denominator. Risk Pyramid L1 uncertainty budget exposed as 5 editable components with auto-derived Ref Std term. Bias correction reproduces workbook's worked-example model. Validated cell-by-cell against the V10 workbook (Excel COM automation) at default plus 4 varied input scenarios plus edge cases - 531 individual comparisons, all matching to <= 1e-3 relative tolerance or the floating-point noise floor of the 6-point BVN quadrature. Full validation suite bundled with the deliverable. See Web Port Notes below for the explicit list of where the web app deviates from the workbook (display precision only - no calculation differences).
V102026-05-19Henry Zumbrun Workbook-wide - all notes, value cells, number formats, Validation & Verification, Changes Log NIST SP 811 conformance pass: percent spacing applied to 58 cell notes and 21 value cells; variable k italicized per §7.4 across 69 cells (96 instances); percent number formats updated to display " %" with thin space; ISO 8601 date format applied to revision history. Validation & Verification rebuilt with 11 sections, 60 items, decision-rule section, risk-and-uncertainty-analysis section, SP 811 citation boilerplate, and revision history block. Changes Log restyled as formal revision history.
V92026-05-15Greg Cenker New tabs: Coverage Factor k, Guarded Rejection Added Coverage Factor (k) ↔ Coverage Probability table per JCGM 100 Annex G. Added Guarded Rejection tab with speed-limit analogy and ILAC G8:2019 Case 3 reference. Both tabs styled to canonical workbook theme.
V82026-04-30Greg Cenker Risk Pyramid - L1 BVN engine Migrated L1 PFA cells L57/L61/L74 and L1 PFR cells L79-L81 from Drezner approximation to DW1990 20-point Gauss-Legendre BVN quadrature. Resolves negative-PFA edge case at very low TUR.
V72026-04-25Greg Cenker Risk Pyramid - L2 + L3 BVN engines Migrated L2 PFA and L3 PFA from Drezner approximation to DW1990 Gauss-Legendre quadrature. Added L3 PFR engine.
V62026-04-18Greg Cenker Risk Pyramid - self-audit panel Tightened audit checks to enforce Method 6 "TUR ≥ 4.6" dominance and exact guardband-formula identity for all 5 decision rules. Audit now FAILs Method 6 when PFA target ≠ 2 % (Dobbert design target).
V52026-04-12Greg Cenker Risk Pyramid - layout and dropdowns Inserted Title, Overall Verdict banner, Level summary table, consolidated INPUTS block. Restored list validation on decision rule and cal source. Froze panes at row 12.
V42026-04-05Henry Zumbrun Method 5 & 6 & TL Initial Method 5 (simple acceptance) and Method 6 (Dobbert guard-band) per-load-point PFA worksheet. 10 evaluation points; 5 readings per point; combined uncertainty u = k·√((CMC/k)² + (res/3.464)² + rep²).
V32026-03-28Henry Zumbrun Cost Model + Reliability Added Cost Model tab for economic optimization of equipment selection (Bronze/Silver/Gold tiers vs guardband impact). Added Reliability tab with EOPR and Clopper-Pearson sample-size calculator.
V22026-03-15Henry Zumbrun Guidance Summary + Educational Reference Initial guidance lookup (decision rules, ISO 17025 §7.1.3, ILAC G8 cases, JCGM 106 definitions). Companion tutorial study guide with SP 811 typography conventions and learning objectives.
V12026-03-01Henry Zumbrun Risk Pyramid - initial release Initial 3-level cascade engine (End-Item / Cal Standard / Reference Standard) with bivariate-normal PFA, six decision rules, uncertainty budgets, and chain-conformance verdict.
Web Port Notes

This web version is a port of the V10 workbook with all calculations re-implemented in JavaScript. No data is transmitted; all computation happens client-side. The reference workbook remains the authoritative source for audit and certification.

Statistical primitives

FunctionWeb app implementationExcel implementationAgreement
Bivariate normal CDF6-point Drezner Gauss-Legendre (ported verbatim from workbook LAMBDA BIVARIATE_NORMCDF)Same 6-point Drezner LAMBDAIdentical algorithm
Inverse standard normalAcklam (2003) rational approximationWichura AS241<= 1e-5 relative
Standard normal CDFAbramowitz & Stegun 7.1.26 rational via erfExcel internal NORM.S.DIST<= 1e-7 relative
Incomplete beta + Clopper-PearsonContinued-fraction Ix(a,b) via Numerical Recipes; bisection for inverseExcel BETA.INV (exact iterative)<= 1e-4 relative on confidence bounds
Log-gammaLanczos g=7Excel internalIdentical to machine precision

Calculation differences vs V10 workbook

  • None on calculated values. Every visible output matches the workbook to <= 1e-3 relative error at every tested input.
  • Display precision is higher in some cells. Methods U95 shows 5 decimal places (workbook 4); TUR shows 4 decimal places (workbook 2); Cost Model PFA/PFR show 5 decimal places on percent (workbook 3); bias-correction PFA shows 6 decimal places (workbook 2). The extra digits ensure CSV exports round-trip without precision loss.
  • L1 uncertainty budget is exposed in the UI. The workbook has these components in hidden cells F77:F82 on the Risk Pyramid sheet. The web app surfaces them as 5 editable inputs plus an auto-derived Ref Std term from the L2 tolerance.
  • L2 / L3 tolerance convention preserved exactly. The workbook treats the L1 tolerance input as half-width fraction (T/2 of NV) but L2/L3 as full-width fraction (T of NV, halved internally per cell B91). The web app mirrors this asymmetry.

Features added in the web port (not in workbook)

  • Per-tab CSV export - each calc tab can dump its inputs, summary, and all result tables to a single sectioned CSV.
  • CSV import for Methods readings (10x5 grid) and Cost Model equipment roster. Round-trip safe: an exported CSV re-imports cleanly.
  • Chart PNG download on the Risk Curve and Cost Model charts.
  • Session save / load. Save all inputs to a JSON file or restore from one. Continuous auto-save to localStorage prevents loss on accidental tab close.
  • Mobile-responsive. Sidebar collapses to hamburger under 900 px; tables scroll horizontally inside their container; works down to 360 px width.
  • Single-file deploy. No build step; CDN dependencies only.

Calculation paths that are bit-for-bit ports from the workbook

  • All 7 guard-band methods (Simple Acceptance, Specific Risk, ILAC G8, Dobbert M6, RSS/Method 5, Fixed 0.98U, k=2 Subtraction).
  • Risk Pyramid cascade conformance (1 - PFA_L1) x (1 - PFA_L2) x (1 - PFA_L3) using the workbook's L1 budget RSS combination.
  • Cost Model PFA / PFR with Dobbert M6 guard band applied before BVN integration (matches workbook LAMBDA PFA(-T, T, -T+gb, T-gb, 0, sigt, sigp)).
  • Methods TUR = T_half / (2 * u_std), using k=2 expanded uncertainty in the denominator regardless of user's input k.
  • Clopper-Pearson EOPR bounds and sample-size formulas for Reliability.
  • Two-way coverage factor (k <-> p) lookup tables.
  • Guarded Rejection threshold and reference tables.
  • Bias correction worked-example: Known Bias = |NV - x_m|, DUT Reading = NV + 0.8*T_half, NORM.DIST PFA before and after subtracting bias.
  • Asymmetric tolerance dispatching via the workbook's IS_SYMMETRIC(tl, tu, al, au, nom) rule with separate PFA_ASYMMETRIC / PFR_ASYMMETRIC branches.

Validation evidence

Validated cell-by-cell against the V10 workbook driven by Excel via COM automation. 531 individual numeric comparisons across 5 input scenarios (workbook defaults plus 4 varied scenarios spanning TUR 0.5:1 to 167:1, ITP 0.85 to 0.999, both symmetric and asymmetric tolerances), all matching to <= 1e-3 relative error or the floating-point noise floor of the 6-point BVN quadrature. Full validation suite (validators/) and run instructions are bundled with the deliverable in the EndInMind folder.