Special Causes
Once we have our data in a capability chart and, if it is over time, split it where
the visual pattern changes, we can then ask whether we have any ‘signals’ – i.e.:
any data points that fall outside of the upper and lower control limits (the predictable
range). In the data below, showing end to end time to decide and pay (or decline)
benefit applications, we can see that some of the data points are above the upper
control limit (the red line). The lower control limit in this chart is 0 days, therefore
no ‘special causes’ are below it, as it isn’t possible to decide and pay (or decline)
in -3 days. Data outside the upper and lower control limits represents ‘special
cause variation’. It happens in 0.03% of the time, once in a blue moon, and we would
exclude it from the calculation to understand what is predictable and typical.
The capability chart shows all ‘special causes’ excluded from the calculation of
control limits. These are represented in CapChart by green stars.
Systems Thinking
in the Public Sector.