When Data Crosses a Control Limit

Suppose that you are monitoring the monthly turnover of your company. The average monthly turnover is 2%, and the standard deviation is 0.25%:

2% ± (3 * 0.25%)

In other words, you don’t want to really worry about turnover unless your turnover is less than

2% - (3 * 0.25%) = 1.25%

or if your turnover is greater than

2% + (3 * 0.25%) = 2.75%

Otherwise, while it may be interesting to know the latest figures, it should not be triggering your attention. Suppose that one month, your turnover is at 3%. Is this reason for alarm? Perhaps. This is when you want to pull up your control chart to see if this is expected. There are a couple of likely scenarios.

Scenario 1: The turnover rate is very abnormal. Figure 4.1 shows what this scenario might look like.

You can see that historically the turnover rate has stayed within the lower control limit (label LCL) and upper control limit (label UCL). Suddenly, in September 2016...