A run chart displays observed data as they evolve over time. Just a basic graph that displays data values in a time order. Can be useful for identifying trends or shifts in process but also allows you to test for randomness in the process.
A run chart can reveal shifts and trends, but not points out of control (A run chart does not have control limits; therefore, it cannot detect out of control conditions.) You can turn a run chart into a control chart by adding upper and lower control limits.
Use it to:
- track improvements (and determine success)
- display outputs to look for stability or instability
Run Chart Analysis
(These examples are from the wonderful download Developed by Richard Scoville, PhD. (email@example.com))
Testing a change with a run chart
- Plot the baseline
- Extend the median and begin the test
- Continue to plot data following the change
- Apply the rules
- If there was a signal, re-plot with new median
Run Chart Decision Rules
Signals of an effective change:
- Runs – Are there too many or too few for just common cause variation.
- A run is a series of consecutive points that all lie on the same side of the line.
- Ignore the points exactly on the line!
- Clustering – too few runs.
- Mixtures – too many runs.
- Shift – 6 or more consecutive points above or below the median.
- A general rule of thumb is when seven or eight values are in succession above or below the average of the group, a shift has occurred.
- Trend – 5 or more consecutive increasing or decreasing points.
- A basic rule of thumb is when a Run Chart exhibits seven or eight points successively up or down, then a trend is clearly present in the data.
- Astronomical Point – A dramatically different value.
Run Chart Signals
How Many Runs?
Expected Runs Table
Shewhart’s tests for runs