XLSTAT: Subgroup Control Chart in Excel

 

Background

Control charts help determine if a process is statistically in control. The subgroup charts tool includes:

• X̄ chart: Monitors the process mean; shifts are easily detected.

• R chart: Tracks variability using range; highlights differences like those from multiple production lines.

• S and S² charts: Analyze variability using standard deviation (S) and variance (S²).

Notes:

1. Use CUSUM charts for detecting smaller mean shifts.

2. Use individual control charts when only one measurement is available per time point.

3. Use attribute charts for qualitative data (e.g., "OK"/"Not OK").

 

Dataset for Example

The data are from [Pyzdek Th. (2003); The Six Sigma Handbook, McGraw Hill, New York] and correspond to 25 inspections each having 5 measurements of a production process.

Using this example, an X-bar chart and an R chart are generated.

 

Setting Up

• Open XLSTAT.

• Click on the Quality/ Statistical Process Control menu and select the subgroup charts feature.

• The dialog box pops up.

• In the General tab, select the chart type, choose the combination X-Bar – R chart, and the data.

  

• In the Options tab, enable Process capabilities to display key process figures. Enter the values for USL, LSL, and target.

  

• Click on the OK button to launch the computations.

• The results are displayed on a new Excel sheet.

 

Interpreting the Results

The first results include the estimated mean and standard deviation, followed by process capability figures, a chart, and X-bar control limits. The Cp indicator shows the process is not adequate—it does not meet specification requirements.

The X-bar control chart includes the various control limits and central lines.

The central line (CL), the lower (LCL) and upper (UCL) control limits, and the lower and upper zone limits for the B and C areas are displayed for each group.

Like the X-bar chart, the R chart shows all data within control limits, indicating no special causes.

Both control charts confirm the process is statistically under control. The special cause rules table shows only "No," indicating good data quality. Group means remain within control limits throughout. Additionally, four normality tests and a Q-Q plot all support the null hypothesis of normal distribution, validating the use of standard control chart rules and capability analysis.

The Q-Q plot confirms normality as the data closely follow the bisector line. The run chart shows a wide range of individual measurements (91 to 110) within groups, suggesting caution when interpreting group means. This variability, relative to the specification limits, explains the low Cp value and indicates poor process performance.