Page 267 - Six Sigma Demystified

Part 3 s i x s i g m a to o l s 247

cycles. If we continue to see no significance of the time factor after eight cycles,
we can either increase the factor level difference (levels such as 25 and 35
minutes) or substitute a different factor.

eWmA Charts

An exponentially weighted moving-average (EWMA) chart is a control chart for
variables data. It plots weighted moving-average values. A weighting factor is
chosen by the user to determine the relative impact of older data to more re-
cent data on the calculated moving-average value. Because the EWMA chart
uses information from all samples, it detects much smaller process shifts than
a normal control chart.

When to Use

Measure Stage
• To baseline a process, particularly when nonnormality of the process is
suspected and rational subgroup size is one

Control Stage

• To control a process, particularly when either nonnormality of the process
is suspected and rational subgroup size is one or when small shifts in the
process must be detected

EWMA charts generally are used for detecting small shifts in the process

mean. They will detect shifts of ½σ to 2σ much faster than Shewhart charts
with the same sample size. They are, however, slower in detecting large shifts
in the process mean. In addition, typical run tests cannot be used because of the
inherent dependence of data points.
EWMA charts also may be preferred when the subgroups are of size n = 1.
In this case, the alternative chart is the individual-X chart, whose standard
control limits are based on the normal distribution. When processes are severely
nonnormal, the individual-X chart’s normal control limits may not fit the pro-
cess data well. When nonnormal control limits are calculated for the individual-
X chart, the data are assumed to be from a controlled process as a requirement
of the curve fitting. The advantage of EWMA charts is that each plotted point