Page 361 - Six Sigma Demystified

Part 3 S i x S i g m a To o l S 341

When to Use

Measure Stage

• To quantify the measurement error’s contribution to the variation (com-
mon and special causes) included in process baseline estimates
• To verify adequate gauge discrimination

Control Stage

• To qualify the proficiency of specific measuring equipment

Methodology

When conducting an R&R study, it is imperative to use typical samples from
the operations rather than pristine reference standards or test cases. The ability
of the measurement system to measure these atypical test cases precisely is not
a good indicator of whether the system will measure samples precisely in daily
operations. Remember the purposes of the measurement system: (1) to evalu-
ate the suitability of the process output to meet (internal or external) customer
requirements and (2) to evaluate the stability (i.e., statistical control) and ca-
pability of the process producing the output.
Using actual samples from operations allows us to estimate error associated
with process equipment (such as measurement fixtures used to hold the sample
or gauge during measurement), as well as variation within the sample pieces
themselves. Each of these sources of error contributes to increased repeatability
error on a regular basis. When operations personnel have developed improved
techniques for dealing with the equipment problems or the within-sample

variation, their repeatability error often will be less than that of other apprais-
ers, providing an opportunity that, when recognized, will lead to system
improvements.
These techniques are not limited to mechanical inspections or to material
parts. They also can be applied to optical inspections, chemical analyses, surveys
or rating systems, or any other system designed to measure process perfor-
mance, its output, or its subject matter. In chemical analyses and patient care,
sample preparation may be a critical component of the measurement and
should be incorporated into the measurement systems analysis.
After obtaining a representative sample, process personnel (i.e., appraisers)
will inspect each sample multiple times (each inspection is termed a trial). A
typical study involves having 3 appraisers measure 10 samples, 3 trials each.