A Run Chart displays trends within obervation points over a specified time period. It can be used
to monitor a process to see whether or not the long-range average is changing.
Run Charts are easy to construct and use. Points are plotted on the graph in the order
in which they become available. It is common to graph the results of a process such as
delays, errors, rework and so on, as they vary over time.
A danger in using a Run Chart is the tendency to see every variation in data as being
important. The Run Chart, like other charting tecniques, should be used to focus
attention on truly vital changes in the process. For example, when monitoring any
process, it is expected that we should find an equal number of points falling above and
below the average. When six or more points "run" on one side of the average, it
indicates a statistically unusual event. Such changes should always be investigated. If
the shift is favorable, it should be made a permanent part of the system. If it is
unfavorable, it should be eliminated.
An alternate type of pattern that can occur is a trend of six or more points steadily
increasing or decreasing with no reversals. Neither pattern would be expected to
happen based upon random chance. Such a significant change calls for some research
and analysis - and possible action.
Putting it to work:
At its most basic, the Run Chart is nothing more than a single line showing a single
measurement over time. Here are several key points:
The horizontal axis represents time. The vertical axis can measure one of any number of
things: defects, delays, waits, complaints, and so forth. A marked point indicates the
measurement or quantity observed or sampled at one point in time.
Data points should be connected for easy use and interpretation.
Collected data should be kept in the order in which it was gathered. Because this tool
helps to track a characteristic over time, the sequence of data points is critical.
The example below shows a year's worth of data. The team is using this Run Chart to
track delays caused by breakdowns in equipment.