Attribute control chart in Excel tutorial

This tutorial shows how to compute and interpret an attributes control chart on quantitative data in Excel using XLSTAT software.

Dataset for the analysis of an attribute control chart

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.

Goal of this tutorial

Control charts by attributes allow to analyze "non-conforming products" or "non-conformities". They are used to control the quality of a product before delivery (manufactured products) or the quality at reception (purchased products). Not all products are necessarily inspected. Inspections are performed in inspection units of a defined size. For example, the size can be 1 in the case of televisions received in a warehouse (each television is inspected). Or a size of 24 in the case of peach crates containing 24 peaches each.

The attribute charts tool offers you the following chart types:

• P chart: it is useful to follow the fraction of non-compliant units of a production process.

• NP chart: it is useful to follow the absolute number of non-compliant units of a production process.

• C chart: it is useful to follow the number of non-compliant units per inspection unit of a production process having a non-constant size of an inspection unit.

• U chart: it is useful to follow the number of non-compliant units per inspection unit of a production process having a non-constant size of an inspection unit.

The P and NP cards allow the analysis of the proportion, respectively the absolute number, of non-conforming products in a production process. For example, one could count the number of non-compliant television sets, or the number of crates with at least one bruised peach.

The C and U cards allow the analysis of the proportion, respectively the absolute number, of occurrences of nonconformities in a controlled unit. One can count the number of non-conforming products in a production process. For example, one could count the number of defective transistors in a controlled unit (there may be several defective transistors in a television), or the number of bruised peaches per crate.

For the aim of this tutorial, the NP chart will be used.

Setting up the dialog box for generating an attribute control chart

After launching XLSTAT, click the Statistical Process Control button on the ribbon and select attribute charts.

The dialog box pops up.

In the General tab, select the chart type, choose the NP Chart, and the data. The group size in this example is 1000.

Then click on the OK button, the computations start.

Interpret the results of the attribute control chart

The first result is the estimated mean.

In the first table, the control limits of the NP control chart are displayed followed by the data of the NP control chart starting with the bruised peaches (NP) and the group size. 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 NP chart follows. The values seem to be randomly dispersed around the control limit. The control chart lets us come to the conclusion that the process is “statistically under control”.

The next table provides the details for the special cause rules. Only “No” can be read in the table. The data is of good quality with respect to this issue. This can be reflected in the NP control chart. The group means values are always between the lower and upper control limits.

Furthermore it is interesting to know if the data follow a normal distribution. Several tests are provided by this feature. We introduce here the results for the Jarque-Bera test. The p-value (0.314) is higher than 5%. So we can 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 the end of the report, 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.