What Does Metrological Traceability Actually Require?

Traceability belongs to the measurement result

Metrological traceability is often treated as an equipment status: an instrument is labelled “calibrated” and is therefore assumed to be traceable. That is not the metrological meaning of the term.

The International Vocabulary of Metrology (VIM) defines metrological traceability as a property of a measurement result. The result must be relatable to a reference through a documented unbroken chain of calibrations, each contributing to measurement uncertainty. The reference may be an SI unit realised in practice, a recognised national or international standard, or, where no suitable SI realisation exists, an agreed reference appropriate to the measurement.

This matters operationally. A laboratory does not establish traceability merely by owning calibrated equipment. It establishes traceability when it can show how a particular reported result - for example, a pressure indication of 499.8 kPa - is connected to a stated reference, with the relevant uncertainty and supporting records preserved.

The calibration chain must be unbroken and documented

A typical pressure calibration chain might be:

1. The customer pressure gauge is compared with a working pressure controller or reference transducer.
2. The working standard has been calibrated against a higher-level reference standard.
3. That reference standard has been calibrated by a competent provider against a national metrology institute standard or another suitable reference.
4. The national metrology institute realises or disseminates the pascal through its recognised measurement standards.

Each link needs evidence. At the laboratory level, this normally includes the calibration certificate or record for the reference standard, its stated uncertainty, identification, calibration status and evidence that it was suitable at the time of use. The laboratory must also retain the records connecting that standard to the customer item: the method used, as-found data, environmental conditions where relevant, corrections applied, uncertainty evaluation and review or authorisation of the result.

An unbroken chain does not mean every instrument must be sent directly to a national metrology institute. It means there must be no unsupported gap between the reported result and its reference. A reference standard calibrated by a competent external laboratory can provide an appropriate link, provided the calibration is fit for the required measurement range, uncertainty and intended use.

A calibration certificate is evidence, not the whole case

A certificate may be necessary evidence, but it does not by itself establish traceability for every result produced using the instrument.

For example, a digital multimeter may have a valid calibration certificate with results at 1 V, 10 V and 100 V DC. A laboratory using it at 1,000 V, under a different loading condition, cannot simply rely on the certificate label. It needs a technical basis for using the instrument at that point. This may come from the certificate scope, specification, verified performance, an appropriate uncertainty model and a procedure that addresses the actual measurement configuration.

Similarly, a certificate does not demonstrate that a reference standard remained suitable between calibrations. Depending on the risk and discipline, the laboratory may need intermediate checks, control charts, comparison history, functional checks or other monitoring. These controls are not substitutes for calibration. They provide evidence that the calibration result remains applicable during use.

Under ISO/IEC 17025:2017 clause 6.5, laboratories must establish and maintain metrological traceability of their measurement results through a documented unbroken chain of calibrations, each contributing to measurement uncertainty, linking them to an appropriate reference. The clause also requires the laboratory to ensure that measurement results are traceable to the SI where this is technically possible, or otherwise to an appropriate reference.

Uncertainty is part of the traceability claim

A chain without stated uncertainty is incomplete. Every calibration link contributes uncertainty, and those contributions affect the uncertainty associated with the final measurement result.

Consider a dimensional laboratory calibrating a 100 mm micrometer. Its gauge blocks have a certificate uncertainty of 0.4 µm. The laboratory also needs to consider contributions from the comparator or micrometer reading, repeatability, thermal expansion, contact effects and resolution. If the combined standard uncertainty is evaluated as 0.8 µm, a typical expanded uncertainty may be 1.6 µm at coverage factor *k* = 2, subject to the laboratory's stated coverage basis.

The gauge block certificate supports one input to that evaluation. It does not justify reporting an uncertainty lower than the laboratory can technically achieve in the actual calibration. Nor does accreditation of the certificate issuer automatically make every downstream uncertainty claim valid.

Traceability and uncertainty therefore need to be considered together when selecting standards. A standard may be traceable but still unsuitable if its uncertainty is too large for the required calibration capability or the customer’s intended tolerance.

What records should support the result?

The evidence required will vary by discipline and method, but a defensible traceability record commonly enables a competent reviewer to establish:

• the identity and status of the reference standards used;
• the relevant calibration results, uncertainties and validity dates for those standards;
• the measurement procedure and the points or ranges covered;
• the measurement conditions and corrections that materially affect the result;
• raw observations, calculations and the uncertainty evaluation;
• evidence of intermediate checks where these are part of controlling continued fitness for use; and
• the identity of personnel performing and reviewing the work.

Records should identify the standards actually used, rather than only listing equipment that could have been used. Where a reference was found out of tolerance, expired, damaged or subject to an unresolved environmental excursion, the laboratory should assess the validity of affected results. The necessary extent of retrospective review depends on the likely impact, not solely on the date printed on a certificate.

Traceability must be appropriate to the intended measurement

The practical test is not whether a certificate exists. It is whether the laboratory can demonstrate a technically valid connection between its reported result and an appropriate reference, with uncertainty understood and evidence retained.

That demonstration depends on the entire measurement process: suitable reference standards, valid calibrations, controlled use, a sound uncertainty evaluation and records that can be reviewed after the event. When those elements are in place, traceability becomes what it is intended to be - a credible basis for confidence in the measurement result.

References

• JCGM 200:2012, *International vocabulary of metrology - Basic and general concepts and associated terms (VIM)*, 3rd edition.
• ISO/IEC 17025:2017, *General requirements for the competence of testing and calibration laboratories*, clause 6.5.
• ILAC P10:07/2020, *ILAC Policy on Metrological Traceability of Measurement Results*.