So which is the more useful… a Run Chart or a Shewart Chart?
Run Chart SPC
Displays data as a picture that shows variation over time
Simple rules make it easy to spot and interpret patterns
(and not over interpret)
Show signals of change
Is more sensitive and shows special
causes of variation
Enables you to see the individual data points and the average
at the same time
Shows the natural range,
useful in planning
Needs small amount of data
Need 20-30 data points
as a minimum
One way to create Many different types exist
Run Chart decision rules that signal a change1
Rule 1 - Shift
Looks for six or more consecutive data
points above or below the median line.
This is known as a shift. A shift suggests
a non-random pattern of data. Ignore
values that fall on the median and continue
counting.
Note at least 10 data points are required
to use this rule.
Rule 2 - Trend
Ignore the median line and simply look for
five or more consecutive points all going
up or all going down. This is known as a
trend. If two data points have the same
value, count only one of them as part of
your trend. Ignore repeating values.
Note where rules 1 and 2 are met there
is a 95% or more chance that the data
patterns observed are as a result of
changes and not just random occurances.