---
title: "More Measurement Digits Do Not Supply an Acceptance Decision"
description: "Leviathan Systems — Technical content Published 2026-10-20; updated 2026-10-20."
canonical: "https://www.leviathansystems.co/articles/measurement-resolution-uncertainty-conformity-decision"
last-updated: "2026-10-20"
---

# More Measurement Digits Do Not Supply an Acceptance Decision

Leviathan Systems — Technical content
Published 2026-10-20; updated 2026-10-20.

To turn a near-limit physical measurement into an acceptance decision, establish the result and its uncertainty basis, the applicable requirement and the agreed decision rule. Displayed digits describe a representation; resolution describes a distinguishable change. Neither supplies the uncertainty of the result or chooses the rule. Calibration supports a relation between indication and result under specified conditions; it does not confer blanket acceptance against an application's requirements.

## A finer indication answers only part of the question

Imagine receiving a dimensional report just below a specified maximum. Its export shows several decimal places, so the result looks decisive. Before using it to close a physical requirement, distinguish how the number is printed from what the measurement establishes.

VIM3 entry 4.14 relates resolution to the smallest change in the measured quantity that produces a perceptible indication change. Its note identifies possible dependence on noise, friction and the quantity's value. Entry 4.15 separately describes the smallest difference between displayed indications that can be meaningfully distinguished. These properties do not state the uncertainty of a particular measurement result.

Changing a report's formatting from 49.8 mm to 49.800 mm supplies no new physical evidence. A genuine resolution statement is more useful than extra zeros, but it still answers a limited question about distinguishable change under the relevant conditions. It cannot establish how closely the reported estimate represents the quantity being measured.

References: [V1](https://jcgm.bipm.org/vim/en/4.14.html), [V2](https://jcgm.bipm.org/vim/en/4.15.html)

## Uncertainty belongs to the result and its conditions

The result must concern the quantity the requirement actually names. Which dimension, of which item, in which stated condition does the reported value describe? A well-presented number for a different dimension leaves the intended comparison unanswered.

VIM3 entry 2.9 describes a measurement result as values attributed to the measured quantity together with relevant information; its notes generally express a result through a measured value and uncertainty. Entry 2.26 describes uncertainty through the dispersion of values attributed to that quantity using the available information. Resolution alone does not account for the many components described there, including those associated with systematic effects.

An uncertainty supplied for a calibration relationship is therefore not automatically the uncertainty of every later result obtained with the instrument. The applicable result needs its own supported basis under its actual measurement conditions. That basis may use calibration information, but a copied certificate value does not establish the whole relationship.

Also retain what kind of uncertainty has been reported. Expanded uncertainty comes with a coverage basis; an unexplained plus-or-minus number leaves its interpretation incomplete. The recipient needs that supplied explanation, rather than an uncertainty guessed from the last displayed digit.

References: [U1](https://jcgm.bipm.org/vim/en/2.9.html), [U2](https://jcgm.bipm.org/vim/en/2.26.html), [E1](https://jcgm.bipm.org/vim/en/2.35.html)

## Calibration supports the relationship, not product approval

VIM3 entry 2.39 describes calibration under specified conditions in two linked stages. It relates reference values and their uncertainties to indications and their uncertainties, then uses that information to establish how to obtain a measurement result from an indication. A calibration can include a correction with associated uncertainty.

This is why the indication and reported result need not be interchangeable. The report should make the basis of the supplied result intelligible to its responsible reviewer. The buyer does not resolve that relationship by counting decimal places or seeing the word calibrated.

Even an applicable calibration relation supplies neither the product's requirement nor the decision rule for accepting it. A calibration record can support the measurement evidence while leaving the application decision open. Conversely, an absent supporting record is an evidence question to resolve; it is not itself proof that the measured item is physically defective.

References: [C1](https://jcgm.bipm.org/vim/en/2.39.html)

## Keep the product limit and the decision rule distinct

JCGM 106:2012 distinguishes a tolerance limit on the property's permissible values from an acceptance limit on permissible measured values. Its decision rule documents how uncertainty enters acceptance or rejection. The guide describes both simple acceptance, where the corresponding limits coincide, and guarded acceptance, where an acceptance limit lies inside the tolerance interval.

For a buyer, the consequence is that a product maximum alone may leave the acceptance comparison unspecified. Establish the applicable rule before relying on the report's pass label. A different rule can change the disposition without changing the item, measurement or product requirement.

The 2012 guide explains these concepts; it does not select the rule governing a particular installation. The responsible requirement and acceptance owners must identify the actual applicable basis. A reviewer should not choose whichever rule produces the preferred answer after seeing the number.

References: [J1](https://www.bipm.org/documents/20126/2071204/JCGM_106_2012_E.pdf)

## MU-248: one report, two alternative outcomes

MU-248 is an entirely fictional report-reading exercise. Stipulate an otherwise valid dimensional report with a measured value of 49.8 mm and an instrument resolution of 0.1 mm under the stated conditions. Separately, the report supplies expanded uncertainty U = 0.4 mm. It explicitly states coverage factor k = 2 and approximately 95% coverage under its stipulated normal measurement model. This coverage basis comes from the hypothetical report, not from the 0.1 mm resolution.

The arithmetic interval around the estimate is 49.8 mm minus 0.4 mm to 49.8 mm plus 0.4 mm: 49.4–50.2 mm. It is not a guaranteed enclosure of the true dimension. Nor does the interval imply that all values within it are equally likely. The report's coverage statement is not a probability of satisfying the product requirement.

Stipulate a product upper tolerance limit of 50.0 mm. Consider only that upper-limit requirement and compare two alternative fictional contract rules. These are separate possible agreements, not two rules to apply selectively within one acceptance decision.

Rule A accepts a reported estimate at or below 50.0 mm. Because 49.8 mm is below 50.0 mm, the reported item is accepted under A for this requirement.

Rule B accepts only a reported estimate at or below 49.6 mm. The same 49.8 mm estimate is above 49.6 mm, so the reported item is not accepted under B. The product upper tolerance limit remains 50.0 mm in both alternatives.

B's acceptance limit is separated from the product limit by 50.0 mm minus 49.6 mm = 0.4 mm. That separation is stipulated. Its numerical equality with the supplied expanded uncertainty is part of the invented example, not a recommended factor or a rule deduced from the interval crossing 50.0 mm.

Neither rule is endorsed here. Not accepted under B does not prove that the actual true dimension exceeds 50.0 mm; no true dimension has been supplied. Acceptance under A likewise does not establish the true dimension with certainty. The changed outcome comes from the alternative rule, while the supplied measurement evidence remains the same.

References: [E1](https://jcgm.bipm.org/vim/en/2.35.html), [I1](https://jcgm.bipm.org/vim/en/2.36.html), [J1](https://www.bipm.org/documents/20126/2071204/JCGM_106_2012_E.pdf)

## Resolve the missing basis before ordering a physical response

If two reviewers disagree over MU-248's 49.8 mm, first establish whether they are applying the same requirement and rule to the same result. Additional digits cannot settle a disagreement about which acceptance limit governs. A request for another measurement also leaves that disagreement intact unless its purpose and decision basis are resolved.

For an actual near-limit report, connect the submitted result and its uncertainty basis to the identified requirement and rule. If one is missing or disputed, return that specific question to the appointed owner and retain the submitted value. Awaiting a decision basis and demonstrated physical nonconformity describe different states.

Once the rule is known, keep the disposition attached to the requirement it addresses. A not-accepted result does not, by itself, identify a defect mechanism or select repair, replacement or rechecking. The responsible technical and acceptance owners determine the appropriate next action from the applicable evidence and project arrangements.

The related structured-cabling QA article explains the evidence categories in a physical handoff. The point-cloud article addresses captured geometry, model interpretation and reference; the timestamp article addresses event boundaries and clock relationships. Those foundations remain useful without supplying the acceptance rule for a near-limit result.

## Bring the physical question to a defined verification scope

Leviathan's Physical-Layer Testing & Installation Verification service is the starting point for discussing agreed visual, cable-ID, labeling and physical-connectivity checks, defect records and handoff evidence. Any instrument measurement requires confirmation of the actual equipment, qualified people, method, coverage and criteria before a commitment is made.

Use the service and project-inquiry links below with a nonconfidential outline of the affected items, the physical question, available result basis and applicable requirements. Identify who will resolve uncertainty and rule questions and who can accept the returned scope. This gives the scope discussion a concrete decision to support.

This article explains report interpretation. Actual physical checks require their own confirmed scope and authorization; the fictional dimensional exercise supplies no inspection technique or approval for work.

## Primary sources

- [V1 — JCGM VIM3, entry 4.14: Resolution](https://jcgm.bipm.org/vim/en/4.14.html). Checked 2026-10-20. Defines resolution through a perceptible indication change and notes its possible dependence on noise, friction and measured value. Useful for separating distinguishable change from result uncertainty.
- [V2 — JCGM VIM3, entry 4.15: Resolution of a displaying device](https://jcgm.bipm.org/vim/en/4.15.html). Checked 2026-10-20. Defines a meaningfully distinguishable difference between displayed indications. Supplies no accuracy or uncertainty specification for a particular instrument.
- [U1 — JCGM VIM3, entry 2.9: Measurement result](https://jcgm.bipm.org/vim/en/2.9.html). Checked 2026-10-20. Definition and notes explain the result's relevant information and its usual expression as a measured value with uncertainty.
- [U2 — JCGM VIM3, entry 2.26: Measurement uncertainty](https://jcgm.bipm.org/vim/en/2.26.html). Checked 2026-10-20. Definition, notes and informative annotation explain uncertainty attached to a result, including components associated with systematic effects.
- [E1 — JCGM VIM3, entry 2.35: Expanded measurement uncertainty](https://jcgm.bipm.org/vim/en/2.35.html). Checked 2026-10-20. Definition and notes relate expanded uncertainty to a coverage factor, probability distribution and selected coverage probability. MU-248 supplies its own fictional values and basis.
- [I1 — JCGM VIM3, entry 2.36: Coverage interval](https://jcgm.bipm.org/vim/en/2.36.html). Checked 2026-10-20. Explains a coverage interval's stated probability and available-information basis. Such an interval is not presented as a guaranteed bound on the true value.
- [C1 — JCGM VIM3, entry 2.39: Calibration](https://jcgm.bipm.org/vim/en/2.39.html). Checked 2026-10-20. Definition and notes explain the calibration relationship under specified conditions, including obtaining a result from an indication and possible corrections with uncertainty.
- [J1 — JCGM 106:2012: The role of measurement uncertainty in conformity assessment](https://www.bipm.org/documents/20126/2071204/JCGM_106_2012_E.pdf). Checked 2026-10-20. 2012 guide, definitions 3.3.4–3.3.12 and clauses 8.1–8.3.2, used for tolerance limits, acceptance limits and decision rules. It does not establish the applicable project rule or recommend MU-248's invented separation.

## Related Leviathan resources

- [Physical-Layer Testing & Installation Verification](/services/network-testing)
- [Project inquiry: define the physical verification decision](/contact?service=network-testing&from=%2Farticles%2Fmeasurement-resolution-uncertainty-conformity-decision)
- [Structured cabling QA: inspection, measurement and acceptance](/articles/structured-cabling-qa-gpu-racks)
- [Existing drawings, point clouds and interpreted models](/articles/gpu-cabling-existing-drawings-point-cloud-survey)
- [Hardware timestamps and clock synchronization](/articles/network-hardware-timestamp-versus-clock-synchronization)

## Define the physical decision your evidence must support

Share a nonconfidential outline of the affected items, available records, required physical checks and acceptance owner. Our team will follow up to discuss scope, requirements and availability.

[Discuss the scope](/contact?service=network-testing&from=/articles/measurement-resolution-uncertainty-conformity-decision)