This tutorial will help you set up and interpret an attributes control chart in Excel using the XLSTAT statistical software.
Dataset to generate attribute control charts
An Excel sheet containing both the data and the results for use in this tutorial can be downloaded by clicking here.
The data are from [Pyzdek Th. (2003); The six sigma Hanbook Revised and expanded, McGraw Hill, New York] and correspond to 30 inspections for bruised peaches of 1 crate of a production process. In this example a crate consists of 1000 peaches.
Attribute control chart
The control charts for attribute are an efficient method to decide, if a process is statistically under control or not. There a variety of different control charts for different aims.
The attribute charts tool offers you the following chart types:
- P chart
- NP chart
- C chart
- U chart
A P chart is useful to follow the fraction of non conforming units of a production process.
An NP chart is useful to follow the absolute number of non conforming units of a production process.
A C chart is useful to follow the number of non conforming units per inspection unit of a production process having a constant size of an inspection unit.
A U chart is useful to follow the number of non conforming units per inspection unit of a production process having a non constant size of an inspection unit.
In addition to the control charts connected functions are available:
- Box-Cox transformation
- Process capability
- Tests for normality
- Rules for special causes and Westgard rules
- Run charts
During this tutorial an NP chart will be used.
Setting up a NP attribute control chart
Once XLSTAT is activated, select the XLSTAT / SPC / Attribute charts command.
The SPC attribute dialog box will appear.
In the Mode tab, we chose the combination NP chart.
Then move on to the tag General, select the data on the Excel sheet. The group size in this example is 1000.
Note: There are several ways of selecting data with XLSTAT - for further information, please check the section on selecting data.
In this example, the data start from the first row, so it is quicker and easier to use columns selection. This explains why the letters corresponding to the columns are displayed in the selection boxes.
In the Outputs tab, we activate all options, and we activate all special cause rules.
In the Charts tab, we activate all options.
The computations begin once you have clicked on OK.
You are asked to confirm the number of rows and columns (this message can be bypassed by un-selecting the Ask for selections confirmation in the XLSTAT options panel).
Interpreting a NP attribute control chart
The first result is the estimated mean.
The following tables with their corresponding chart represent the NP control chart including the different control limits and central lines. In the first table the control limits of the NP control chart are mentioned.
Then the data of the NP control chart starting with the bruised peaches (NP) and the group size are displayed. After this the central line (CL), the lower (LCL) and the upper (UCL) control limit and the lower and upper zone limits for the A and B area are displayed for each group.
The next table contains the details for the special cause rules. Only “No” can be read in the table. The data has a good quality with respect to this issue. Summing up you can discover all this in the NP control chart. The group means values are always between the lower and upper control limits.
The control chart lets us come to the conclusion that the process is “statistically under control”.
Further it is interesting to know, if the data follows a normal distribution and therefore the common rules for control charts and process capabilities can be applied. In the next section the results of 4 different normality tests and a Q-Q plot are displayed. All of the 4 tests conclude that the Hypothesis H0 of normal distribution can be accepted. In the Q-Q plot we see that the data are close to the first bisector line. We suppose that the data is normally distributed.
At last the run chart is displayed. We see that the measurements inside a group have a wide range, lying between 23 and 37. The values are within the control limits.