Choosing and Using Control Charts in Minitab
The evolution of the control chart
A Minitab control chart in its most basic form is the line chart. The line chart is used to correlate measurements over time or number of samples. Typical measurements include down time, defects, delays, people, cost, outages, returns etc...
By adding an average line to the to the line chart we create what is called a run chart. The run chart will show us how the measurements in the line chart compare to the average to illustrate variation from the average over time.
By adding upper control limits (UCL) and lower control limits (LCL) to the run chart we have the fully illustrated Minitab control chart shown here.
Using Control Charts
Control charts are used to analyze variation and process stability over time. Process parameters or customer specifications are set by the designated UCL and LCL. The movement of attributes between the average and the UCL/LCL indicate process variation trends as well as out of bounds conditions.
The C Chart above illustrates a one sided variation analysis meaning that LCL=0. In this case when measuring defects a condition of LCL=0 is optimal or equal to 0 defects.
A two sided variation analysis requires that LCL be greater than zero meaning that out of bounds conditions may exist above UCL or below LCL.
There are seven types of Minitab control charts used in six sigma.
- C Chart - Attribute or counted data (sample size constant)
- Np Chart - Attribute or counted data (sample size constant)
- P Chart - Fraction defective (sample size varies)
- U Chart - Number of defects (sample size varies)
- Individuals Chart - Subgroup size = 1
- XbarR Chart - 2 to 10 subgroups
- XbarS Chart - 2 to 50 subgroups
The use of different control charts is indicated by the type and form of data sample to be analyzed. Use the chart below to properly choose the correct control chart based on data type and form.
To recap: In six sigma we use the control charts to measure process stability over time. This is required to capture and respond to common and special causes of variation in the process.
Combine your control chart analysis with Histogram Cp Cpk Pp Ppk analysis to measure process capability. The combined measurements of Stability (Control Charts), and Capability (Histogram) tell the full story and give us a data driven framework for continuous improvement of the process.
Learn more:
© 2007-2008 QualproXL