Page 410 - Six Sigma Demystified

390 Six SigMa DemystifieD

Control Chart Selection
There are many control charts available for our use. One differentiator between
the control charts is the type of data that will be analyzed on the chart:
• Attributes data (also known as count data). Typically, we will count the
number of times we observe some condition (usually something we don’t
like, such as a defect or an error) in a given sample from the process.
• Variables data (also known as measurement data). Variables data are con-
tinuous in nature and generally capable of being measured to enough
resolution to provide at least 10 unique values for the process being ana-
lyzed.

Attributes data have less resolution than variables data because we only
count if something occurs rather than taking a measurement to see how close
we are to a condition. For example, attributes data for a manufacturing process
might include the number of items in which the diameter exceeds the specifi-
cation, whereas variables data for the same process might be the measurement
of that part diameter. In a service process, we may measure the time it takes to
process an order (variables data) or count the number of times the order pro-
cessing time exceeds a predefined limit (attributes data).
Attributes data generally provide us with less information than variables data
would for the same process. Attributes data generally would not allow us to
predict whether the process is trending toward an undesirable state because it
is already in this condition. As a result, variables data are considered more use-
ful for defect prevention.
There are several attribute control charts, each designed for a slightly differ-
ent use:

• np chart: For monitoring the number of times a condition occurs

• p chart: For monitoring the percent of samples having the condition
• c chart: For monitoring the number of times a condition occurs, when each
sample can have more than one instance of the condition.

• u chart: For monitoring the percent of samples having the condition, when
each sample can have more than one instance of the condition.

There are also several variables charts available for use. The first selection
generally is the subgroup size. The subgroup size is the number of observations,
taken in close proximity of time, used to estimate the short-term variation.