3. 3
Overview Of Calibration
• Any Calibration exercise is an extension of our commitment
towards Quality Service. This makes sure that any group of
people who have got calibrated together would be on the
same platform while monitoring transactions.
• On the day of the calibration, auditors get together as a
group (process knowledge experts), audit the transactions and
pull the report that shows significant gaps in evaluation, if any.
• In case of any variations, the host auditor need to share the
variations with all the auditors and give the required feedback.
• If the overall calibration score is less than 90% i.e., greater
than 10% of variance has been observed among the
participants, then it would be considered as fail. After giving
the required feedback, the activity needs to be repeated again
this time using the same sample of audits used in the first
calibration. Two types of Calibration are :
Internal Calibration.
External Calibration.
Collect samples for
calibration exercise
Schedule calibration session
as per calibration calendar
Conduct calibration session
Compile and publish
calibration results
Does the calibration
assessment meet the
pass threshold?
End
Start
Yes
Initiate appropriate corrective
action
No
4. 4
Understanding Repeatability & Reproducibility
• Repeatability or test-retest reliability is the variation in measurements taken by a single person or
instrument on the same item and under the same conditions. A less-than-perfect test-retest reliability
causes test-retest variability. Such variability can be caused by, for example, intra-individual variability
and intra-observer variability.
A measurement/object may be said to be repeatable when this variation is smaller than some agreed
limit.
There are two important aspects of Gauge R & R.
• Reproducibility is the degree of agreement between measurements or observations conducted on
replicate specimens in different locations by different people. Reproducibility is part of measuring
variation. Reproducibility also refers to the ability of an entire study to be reproduced, or by someone
at different timelines.
A measurement/object may be said to be reproducible when this variation is smaller than some agreed
limit.