Items, Samples, and Groups

Before using SPC for statistical analysis, you need to understand three basic concepts:

Items, samples, and groups.

In short:

• Item: a measurement dimension that needs SPC analysis
• Sample: the measurement result of one workpiece obtained from one complete program run
• Group: several consecutive samples divided into a group according to the set sample size

After understanding these three concepts, it becomes easier to read the group list and SPC control charts later.

Item

An "item" in SPC can also be understood as a control item.

In Easson2D, as long as a measurement dimension has been set with a standard value and upper/lower tolerances, that dimension automatically becomes an SPC item.

For example, if the measurement program contains:

• Circle 1 diameter
• X coordinate of Circle 1
• Distance 1
• Angle 1

and among them:

• Circle 1 diameter
• Distance 1
• Angle 1

have tolerances set, while the X coordinate of Circle 1 has no tolerance set, then 3 control items appear in SPC.

That is:

Circle 1 diameterDistance 1Angle 1

Each item stores its own measurement data and is analyzed independently.

Item List

The lower part of the SPC Manager window displays the current item list.

The item list mainly contains:

• Item number
• Control name
• Standard value
• Upper tolerance
• Lower tolerance
• Number of samples
• Number of nonconforming items
• Nonconforming percentage

Control Name

Indicates the measurement dimension currently being analyzed in SPC.

For example:

Circle 1 diameter

or:

Distance 1

Standard Value and Tolerances

Displays the values set in Easson2D for the dimension:

standard value, upper tolerance, and lower tolerance.

SPC uses these parameters to determine whether the measurement result of each sample exceeds the specification range.

Number of Samples

Indicates how many measurement samples have been accumulated for the current item.

For example, if it displays:

Number of samples: 20

it means the item has recorded 20 workpiece measurement results.

Number of Nonconforming Items and Nonconforming Percentage

If some measurement results exceed the set upper/lower tolerances, SPC counts:

• The number of nonconforming items
• The nonconforming percentage

These figures help you quickly understand the overall conformity of the current item.

Selecting the Item to View

If multiple items exist in SPC, you can click the item you want to view in the item list.

For example, select:

Circle 1 diameter

The sample list, group list, and various SPC charts in the upper area display the data for "Circle 1 diameter".

Select:

Distance 1

The analysis area switches to the data of "Distance 1".

Therefore, you can think of SPC Manager as:

Select the item to analyze in the lower area, and view the detailed data and statistical results of that item in the upper area.

Sample

A "sample" in SPC represents the measurement result of one workpiece obtained from one complete measurement program run.

For example, if a measurement program contains:

• Circle 1 diameter
• Distance 1
• Angle 1

after measuring the first workpiece and completing the program run, the following is formed:

Sample 1

which contains:

• Circle 1 diameter of Sample 1
• Distance 1 of Sample 1
• Angle 1 of Sample 1

After replacing it with a second workpiece and running the complete program once more, the following is formed:

Sample 2

Continuing the measurement yields, in sequence:

Sample 3, Sample 4, Sample 5…

Therefore:

A sample can be understood as the complete measurement result of one workpiece.

Multiple Items and Multiple Samples

Suppose you need to analyze 3 items and measure 10 workpieces in sequence.

This can be understood as:

Sample	Circle 1 diameter	Distance 1	Angle 1
Sample 1	1st measurement value	1st measurement value	1st measurement value
Sample 2	2nd measurement value	2nd measurement value	2nd measurement value
Sample 3	3rd measurement value	3rd measurement value	3rd measurement value
…	…	…	…
Sample 10	10th measurement value	10th measurement value	10th measurement value

SPC performs statistical analysis for each control item separately.

For example, after selecting "Circle 1 diameter", the analysis is on the 10 diameter measurement results corresponding to the 10 workpieces.

What Is the Sample Size

In the SPC analysis area, you can set the:

sample size

Here, the sample size indicates:

How many consecutive measurement samples each group contains.

The settable range is:

2–20

The default value is:

5

For example, set:

sample size = 5

After measuring 10 workpieces in sequence, SPC automatically divides them into:

Group 1: Samples 1–5

Group 2: Samples 6–10

If you continue measuring:

Group 3: Samples 11–15

and so on.

Why Grouping Is Needed

Grouping is mainly used to observe data changes across different stages of production and to provide the statistical basis for some SPC control charts.

For example, when the sample size is set to 5, the software can calculate for each group:

• The mean
• The range
• The median
• The standard deviation

These within-group statistics are displayed in the group list and used for some control chart analyses later.

For example:

• XBar-R
• XMed-R
• XBar-S

all use grouped data.

X-Rm, on the other hand, is a control chart for individual values. It analyzes the data of each sample individually and does not use this sample grouping method.

Modifying the Sample Size

The sample size can be modified according to the actual measurement requirements.

For example, originally:

sample size = 5

Then:

Samples 1–5 → Group 1Samples 6–10 → Group 2

If you change it to:

sample size = 3

the existing data is regrouped according to the new sample size:

Samples 1–3 → Group 1Samples 4–6 → Group 2Samples 7–9 → Group 3Sample 10 → Group 4

Therefore, after modifying the sample size, there is no need to re-measure. The existing historical samples are regrouped according to the new setting.

When the Last Group Has Fewer Samples

If the number of remaining samples is less than the set sample size, SPC still displays this group.

For example:

sample size = 5

but there are currently only 12 samples, so the following groups are formed:

Group 1: Samples 1–5Group 2: Samples 6–10Group 3: Samples 11–12

The insufficient part of the last group is left blank, and the existing data can still be displayed and used for the corresponding statistics.

As new workpieces continue to be measured, new samples keep joining this group.

Basic Relationships

You can understand the SPC data structure in the following way:

A dimension with tolerances set = an item

One complete program run = a sample

Several consecutive samples = a group

For example:

Circle 1 diameter → one SPC item

Measure 10 workpieces in sequence → samples 1–10 are obtained

Set the sample size to 5 → 2 groups are automatically formed

After understanding the relationships among items, samples, and groups, you can further use the sample list, group list, and simple analysis chart to view the actual measurement data and its statistical variation.