Glossary: Precision, Repeatability, and Measurement in Metrology

Measurement

Measurement is the process of experimentally obtaining values that can be assigned to a property—called a measurand—of an object or phenomenon. According to the International Vocabulary of Metrology (VIM), measurement always involves comparison with a reference, standard, or protocol. This process is central to traceability, ensuring that results can be linked through an unbroken chain to national or international standards, a necessity in aviation, manufacturing, and laboratory science.

The measurement procedure defines the specific steps, instrument selection, environmental controls, and data handling necessary to minimize uncertainty. Every measurement result must be accompanied by an evaluation of measurement uncertainty, quantifying the reliability of the result. Uncertainty arises from instrument limitations, environmental variability, and human or procedural factors, and is assessed following the Guide to the Expression of Uncertainty in Measurement (GUM).

For example, in aviation, laser trackers are used to align aircraft components. The displayed position is an estimate subject to the device’s uncertainty and environmental influences like temperature. Measurement applies not only to physical dimensions but also to derived values such as altitude or airspeed, provided a quantitative value is assigned.

Measurement underpins calibration labs, quality assurance, flight testing, and research. Adherence to standardized procedures and clear reporting of results, uncertainty, and traceability are essential for valid, comparable data.

Precision

Precision is the closeness of agreement between independent measurement results obtained under stipulated conditions (VIM 3:2.15; ISO 5725-1). It quantifies how closely repeated measurements of a stable quantity agree with one another, reflecting random error and expressed statistically as standard deviation or variance.

Precision is often confused with accuracy, but they are distinct: a method can be very precise (results are tightly grouped) but inaccurate (consistently biased). Precision must always be reported with the conditions under which it was determined: same operator, instrument, environment, and time frame, unless otherwise specified.

In aviation maintenance, a torque wrench’s ability to deliver consistent force across uses is a measure of its precision. In analytical labs, precision determines consistency during method validation. ISO and ICAO standards subdivide precision into repeatability, intermediate precision, and reproducibility, each defined by their conditions.

Repeatability

Repeatability describes the closeness of agreement among measurements made under identical conditions: same operator, instrument, environment, and within a short period (VIM 3:2.21; ISO 5725-1). It is the most controlled subset of precision, isolating the measurement system from many external sources of variability.

Repeatability is evaluated by repeated measurements of the same item under constant conditions, with the standard deviation of these results indicating the repeatability. In aviation, repeatability is critical for instrument calibration, aircraft weighing, or checking airframe coating thickness.

ISO 5725-2 and ASTM E177 define how to assess repeatability, including the number of replicates and handling outliers. High repeatability ensures results are stable under daily conditions, though it does not guarantee accuracy or comparability between different operators or locations.

Intermediate Precision

Intermediate Precision refers to measurement precision within a single laboratory, but with routine variations: different operators, instruments, or over time (VIM 3:2.23; ISO 5725-3). It reflects real-world changes encountered in regular operations.

Intermediate precision is assessed by measuring the same sample, using the same procedure and laboratory, but varying at least one factor such as operator or instrument. This standard deviation is usually higher than repeatability and is vital for labs with multiple technicians or shifts.

For example, in aviation component testing, different inspectors may measure turbine blade diameters over weeks, using the same CMM. The variability reflects both the stability of the process and everyday lab changes. Intermediate precision is essential in method validation (ISO 17025), control limit definition, and identifying training needs.

Reproducibility

Reproducibility measures the agreement among results obtained by different operators, in different labs, using different equipment and locations—often over extended periods (VIM 3:2.25; ISO 5725-1). It is the broadest evaluation of precision, encompassing all random error sources across an industry.

Reproducibility is typically assessed via inter-laboratory studies with standardized samples and protocols. The spread of results quantifies reproducibility. In aviation, it is crucial for standardizing fuel analysis, material testing, or environmental monitoring.

The standard deviation of reproducibility (s_R) is generally larger than repeatability or intermediate precision, reflecting more variability sources. ISO 5725-2 and ASTM E177 guide the design and analysis of reproducibility studies, which are essential for method standardization, regulatory approval, and proficiency testing.

Accuracy

Accuracy is the closeness of agreement between a measured value and the true or accepted reference value (ISO 5725-1). Unlike precision, which addresses consistency, accuracy depends on both systematic error (bias) and random error. High accuracy requires both tightly grouped results and centering on the true value.

Accuracy is evaluated by comparing results to certified references and correcting systematic errors. In aviation, it is critical for calibrating flight instruments and complying with regulatory safety margins.

Accuracy is often illustrated by the “target” analogy: tightly grouped but offset results are precise but not accurate; widely scattered results that average to the correct value are accurate but imprecise. The ideal method is both precise and accurate.

Measurement Procedure

A measurement procedure is a documented, step-by-step process specifying how a measurement is performed, including instrument selection and calibration, sample handling, environmental controls, data acquisition, and result calculation. Standardized procedures ensure consistent, traceable, and comparable results.

In aviation, examples include calibrating pitot-static systems, inspecting turbine blades, or analyzing fuel. Procedures follow standards (ISO, ASTM, or national) and are managed within quality systems (ISO 9001, ISO 17025) to support audits and compliance.

Standard Deviation

Standard deviation quantifies the spread of values around their mean. In metrology, it is the principal metric of imprecision, used to express variability in repeatability, intermediate precision, and reproducibility studies. Standard deviation is calculated as the square root of variance.

In aviation quality control, standard deviation defines control limits for dimensions or system performance and underpins measurement uncertainty calculations. Always report standard deviation with the number of replicates and measurement conditions.

Comparative Table: Precision, Repeatability, Intermediate Precision, Reproducibility, Accuracy

Visualizing the Concepts

Archery Analogy:

• High Precision, Low Accuracy: Tightly grouped arrows far from bullseye (consistent but incorrect).
• Low Precision, High Accuracy: Scattered arrows averaging near bullseye (unbiased but imprecise).
• High Precision, High Accuracy: Tightly grouped at bullseye (repeatable and correct).
• Low Precision, Low Accuracy: Scattered, far from target (random, biased).

Diagram of Relationships:

Precision
│
├─ Repeatability (same conditions, short time)
├─ Intermediate Precision (same lab, varied operators/days)
└─ Reproducibility (different labs/operators/instruments)

Guidance on Terminology

Use only internationally standardized terms: “repeatability,” “intermediate precision,” and “reproducibility” (ISO 5725, VIM, ASTM E177). Avoid outdated or informal terms like “internal precision,” which are not recognized and can cause noncompliance. Always specify the context and conditions for reported precision or standard deviation.

Practical Examples and Use Cases

Example 1: Analytical Chemistry Laboratory

• Repeatability: Same analyst, five measurements, same instrument, one session.
• Intermediate precision: Different analysts, days, instruments, same lab.
• Reproducibility: Protocol and samples sent to other labs.
• Accuracy: Mean result compared to certified reference.

Example 2: Manufacturing Quality Control

• Repeatability: Multiple measurements of a bearing by the same system.
• Intermediate precision: Measurements by different operators/shifts.
• Reproducibility: Comparison across factories.
• Accuracy: Checked against calibrated reference gauge.

Example 3: Proficiency Testing

• Repeatability: Multiple runs by one lab.
• Reproducibility: Comparison across all participating labs.

These practices establish method reliability, regulatory acceptance, and international comparability.

Common Pitfalls and Misconceptions

• Precision is not Accuracy: Consistent results may be incorrect if systematic error exists.
• Repeatability ≠ Reproducibility: Repeatability is under identical conditions; reproducibility spans labs, operators, and equipment.
• Context-Free Standard Deviation: Always indicate if it refers to repeatability, intermediate precision, or reproducibility.
• Resolution vs. Precision: Display digits do not guarantee true precision; the instrument and process determine it.

Measurement Conditions and Their Impact

Repeatability gives a minimal variability estimate, intermediate precision accounts for routine laboratory variation, and reproducibility reflects full industry-wide variability.

Statistical Evaluation and Reporting

For each level of precision, use the appropriate statistical measure:

• Standard Deviation (s): Overall spread of repeated measurements.
• Repeatability Standard Deviation (s_r): Under repeatability conditions.
• Intermediate Precision Standard Deviation (s_IP): With operator/instrument/time variation in one lab.
• Reproducibility Standard Deviation (s_R): Across different labs and operators.

Always report the context, number of replicates, and detailed measurement conditions for transparency and compliance.

By adhering to these definitions and practices, organizations ensure valid, reliable, and internationally comparable measurement results—essential for safety, quality, and regulatory compliance in aviation, manufacturing, and laboratory science.