A type of variables control chart that shows how correlated, or dependent, variables jointly influence a process or outcome. For example, you can use a multivariate control chart to investigate if temperature and pressure are jointly in control in the production of injection-molded plastic parts.
To minimize type 1 error, you should use a multivariate chart when variables are correlated instead of several control charts, each plotting a single variable, because each control chart has a type 1 error risk that accumulates with the number of charts. Plus, multivariate charts can be easier to interpret than several single control charts because a single control limit determines whether the process is in control.
Minitab offers four multivariate control charts:
Tsquared chart |
Generalized variance chart |
A chart of that uses a special statistic (Hotelling's T2) combining dispersion and mean information from several variables. |
A chart that simultaneously monitors the process variability of two or more related process characteristics. |
Multivariate EWMA chart |
TSquared generalized variance chart |
A multivariate chart of exponentially weighted moving averages. Use to simultaneously monitor two or more related process characteristics. |
A TSquared chart (top) and generalized Variance chart (bottom) displayed in one window. Use to simultaneously assess whether the process mean and variation are in control. |
One drawback to multivariate charts, however, is that the scale on multivariate charts is unrelated to the scale of any of the variables, and out-of-control signals do not reveal which variable (or combination of variables) caused the signal.