Documentation |
A control chart displays measurements of process samples over time. The measurements are plotted together with user-defined specification limits and process-defined control limits. The process can then be compared with its specifications—to see if it is in control or out of control.
The chart is just a monitoring tool. Control activity might occur if the chart indicates an undesirable, systematic change in the process. The control chart is used to discover the variation, so that the process can be adjusted to reduce it.
Control charts are created with the controlchart function. Any of the following chart types may be specified:
Xbar or mean
Standard deviation
Range
Exponentially weighted moving average
Individual observation
Moving range of individual observations
Moving average of individual observations
Proportion defective
Number of defectives
Defects per unit
Count of defects
Control rules are specified with the controlrules function. The following example illustrates how to use Western Electric rules to mark out of control measurements on an Xbar chart.
First load the sample data.
load parts;
Construct the Xbar control chart using the Western Electric 2 rule (2 of 3 points at least 2 standard errors above the center line) to mark the out of control measurements.
st = controlchart(runout,'rules','we2');
For a better understanding of the Western Electric 2 rule, calculate and plot the 2 standard errors line on the chart.
x = st.mean; cl = st.mu; se = st.sigma./sqrt(st.n); hold on plot(cl+2*se,'m')
Identify the measurements that violate the control rule.
R = controlrules('we2',x,cl,se);
I = find(R)
I = 21 23 24 25 26 27