Whenever a process is out-of-control and it is known to us, we use variable control charts to troubleshoot process. It is because; we can focus on each quality characteristic individually using variable control charts.

The lower control limit for u-chart is expressed by,
LCL = \(\bar{u} – 3\sqrt{\frac{\bar{u}}{n}}\)
Putting \(\bar{u}\)=0.074 we get, LCL=-0.0414. As LCL is less than 0, we use LCL=0.

Probability plots are quite useful but, y have a major disadvantage that y are not an objective procedure. There can be many conclusions for same set of data.

As u-chart uses value of average number of nonconformities per unit, center line of u-chart is given by,
\(\bar{u} = \frac{\sum_{i=1}^N \frac{x_i}{n_i}}{N}\)

When destructive testing is required or any or expensive testing procedure is needed, variable control charts are more suitable than attributes control chart.

The center line value for average number of events control chart or h-chart, when re is no standard given, is expressed by,
CL= t/n.

The efficiency of variable control charts is considerable higher than efficiency of p-charts when value of p is small or away from 0.5.

The OC curve of c chart is a curve which plots type II error probability against true mean number of defects in a process output.

In case we don?t have any information about mean, spread and shape of probability distribution of one CTQ characteristic of a process, we can use specifications on CTQ characteristic.

The advantage of probability plotting is that, it is unnecessary to divide range of variable into class intervals, and it often produces reasonable results for moderately small samples (which histogram will not).