Understanding the Sources of Variability in Measurement Systems

One of the most important aspects of any qualityprocedures open the door to another difference in
system is the control of monitoring andoperators, "interpretation" of the requirements or
measurement processes. The systematic and randommethodology.
errors that can influence test results or processLaboratory Environment
monitoring results can be from a variety of sources.Some samples and equipment may be susceptible to
It is important to identify the cause and effecttemperature, humidity, atmospheric pressure, and
relationships that can create the variability observedother environmental factors. Because these cannot
in test results. Generally, these sources can bebe controlled perfectly within or between labs, they
summarized into the following categories:provide some contribution to the variation of test
Equipmentresults.
The equipment, whether it is a sophisticatedSamples
automated electronic analyzer or glassware, has beenAny non-uniformity of the sample can cause
manufactured to certain tolerances. The variabilityadditional differences in test results. When conducting
inherent to the equipment specifications will bestudies to determine testing variability, special effort
reflected in the test results. Component wear, failure,must be made to obtain test samples that are as
or inadequate maintenance will increase theuniform or similar as possible.
inconsistencies in test results. Any inconsistencies inTime
calibration verification and/or recalibration will alsoAll of the previously mentioned sources of variation
affect the consistency of the results obtained fromcan themselves change with time. In measurement
the equipment.studies, efforts are usually made to keep the time
Peoplespan as short as practical.
People are almost always a contributor to some ofWhen taken together these many sources can cause
these differences in results simply because none ofsignificant issues with the repeatability and
us are exactly alike. We differ in dexterity, reactionreproducibility of a measurement process. It is
times, color sensitivity, and other ways. Even theessential that steps be taken to minimize the
same operators can perform differently at differentdifferences in test results obtained within a given
times due to different degrees of mental andlaboratory or Quality Control area. Working to reduce
physical alertness. Some operator differences arethe effects of any or all of these sources can
practically unavoidable. Of course, some tests areimprove the consistency of your results and build
more sensitive to the effects of operatorconfidence in your quality system.
differences. Incomplete or confusing test methods or