Process Control with Variable Measurements:

In attribute sampling, we find out whether something is bad or good, fit or does not fit, etc. In variable sampling, we calculate the actual volume, weight number of inches, or other variable measurement. Depend on the measurements; we build up control charts to find out the acceptability or rejection of the process. There are four major issues to address in creating a control chart viz. the size of the samples, frequency of samples, number of samples, and control limits.

Size of Samples

For industrial application in procedure control involving the measurement of variables, this is preferable to keep the sample size small. There are two major reasons. Primary, the sample ought to be taken within a reasonable length of time; or else, the process might change whereas the samples are taken. Second, larger the sample, the more it costs to take.

Number of Samples

Once the chart has been established, each sample taken may be compared to the chart and a decision may be made about whether the procedure is acceptable or not.

Frequency of Samples

It based on the trade-off among the cost of sampling and the benefit of adjusting the system. Typically, it is best to start off with frequent sampling of a procedure and taper off as confidence in the procedure builds.

Control Limits

Standard practice in statistical procedure control for variables is to set control limits that mean three standard deviations above the mean and three standard deviations below. It implies that 99.7% of the sample means are expected to fall in these control limits. Therefore, if one of the sample mean falls outside this wide band, we have strong evidence that the procedure is out of control.