The closeness of agreement between results of successive measurements of the same measurand carried out under identical conditions (same operator, same instrument, same location, short time period).
Repeatability is a measure of how consistently an instrument produces the same result when the same quantity is measured multiple times under conditions that are kept as constant as possible. These "repeatability conditions" include the same measurement procedure, the same operator, the same instrument used under the same conditions, the same location, and repetition over a short period of time.
Repeatability is quantified using the standard deviation of a series of repeated measurements, sometimes expressed as a repeatability limit (the value below which the absolute difference between two successive measurements should fall with 95% probability). It represents the best-case precision of a measurement system because all variable factors are held constant.
In calibration, repeatability is a critical performance characteristic. It is evaluated during calibration by taking multiple readings at each test point. Poor repeatability may indicate mechanical wear, electrical noise, environmental interference, or internal friction. The repeatability standard deviation is often the dominant Type A uncertainty component in the uncertainty budget. Instruments with poor repeatability may need repair or replacement regardless of their accuracy, because inconsistent readings undermine confidence in every measurement they produce.
In aerospace calibration labs, repeatability is critical when calibrating torque wrenches used for aircraft assembly. A technician calibrating a 100 ft-lb torque wrench must obtain measurements within ±0.5% when applying the same reference torque repeatedly within a 2-hour period. Poor repeatability might show readings of 99.8, 100.3, 99.5, and 100.7 ft-lb, indicating instrument drift or environmental instability that could compromise flight safety. Medical device manufacturers face similar challenges when calibrating pressure transducers for infusion pumps. A 300 mmHg transducer should show consistent readings (±0.25%) across five successive measurements at the same pressure point. Inadequate repeatability—readings varying by 2-3 mmHg—suggests calibration equipment issues that could affect patient dosing accuracy. Common audit findings include labs failing to monitor short-term measurement variation, not establishing repeatability limits in procedures, or ignoring environmental factors like temperature fluctuations during calibration cycles. Auditors frequently cite labs where technicians don't recognize that excessive scatter in repeat measurements indicates equipment problems rather than acceptable uncertainty. This leads to unreliable calibration certificates and potential product quality issues downstream.
ISO/IEC 17025:2017 addresses repeatability in Section 7.6 (ensuring measurement validity) and Annex A (measurement uncertainty evaluation). The standard requires labs to evaluate repeatability as part of uncertainty analysis and method validation. AS9100D references repeatability through its calibration requirements in Section 7.1.5.1, mandating aerospace suppliers demonstrate measurement process capability. ISO 13485:2016 Section 7.6 requires medical device manufacturers to validate measurement processes, including repeatability studies for calibration procedures. IATF 16949:2016 Section 7.1.5.1.1 specifically requires automotive labs to conduct measurement system analysis including repeatability studies. ANSI/NCSL Z540.3-2006 Section 9.2 mandates repeatability evaluation during calibration procedure development. GUM (ISO/IEC Guide 98-3) addresses repeatability in Type A uncertainty evaluation (Section 4.2), requiring statistical analysis of repeated observations. Auditors examine calibration records for evidence of repeatability monitoring, review procedures for defined acceptance criteria, verify technician competency in recognizing poor repeatability, and assess how labs incorporate repeatability data into uncertainty budgets. Non-compliance typically involves inadequate documentation of repeat measurements or failure to establish appropriate repeatability limits.
CalibrationOS captures repeatability data through its Measurement Collection module, automatically recording multiple readings during calibration procedures and calculating statistical parameters (mean, standard deviation, range). The system flags excessive variation when repeat measurements exceed user-defined limits, alerting technicians to potential equipment issues before completing calibrations. During certificate generation, the software incorporates repeatability statistics into uncertainty calculations per GUM methodology, ensuring compliant documentation. The Audit Trail feature maintains complete records of all repeat measurements with timestamps, supporting regulatory compliance during assessments. CalibrationOS trending analytics identify patterns in repeatability performance over time, helping labs optimize procedures and detect equipment degradation early. The system's customizable acceptance criteria allow labs to set repeatability limits specific to their quality requirements and industry standards. Reports include statistical summaries of repeatability data, control charts showing measurement consistency, and automated flags for out-of-specification conditions, providing auditors with clear evidence of measurement process control.
Repeatability is the variation in measurements taken by the same operator using the same instrument under the same conditions over a short time. It represents the minimum variability inherent in the measurement system.
Repeatability is tested by having one operator measure the same item multiple times with the same instrument under constant conditions, then calculating the standard deviation of those measurements.
This article is licensed CC BY-SA 4.0. Share, adapt, and reuse with attribution to calibrationos.com/glossary/repeatability.
Industry benchmarks, best practices, and calibration tips — delivered to your inbox.
AIAG average and range method — enter your measurement data and get %GRR, NDC, and pass/fail in seconds.
Open the Calculator