Statistical Indicators
Easson2D SPC automatically calculates a variety of statistical indicators from the sample data obtained through consecutive measurements.
These indicators help you quickly determine:
- Whether the measurement data is close to the target value
- Whether the data fluctuation is too large
- Whether there are any out-of-spec results
- Whether the current production process is stable
Mean
The mean represents the average level of a batch of measurement data.
If the mean deviates from the standard value over a long period, the dimensional center of the production process may have shifted.
This can be simply understood as:
The mean is used to observe where the data as a whole is biased.
Maximum and Minimum
The maximum represents the largest measurement result in the current data.
The minimum represents the smallest measurement result in the current data.
Together, they give you a quick idea of the general range of this batch of data.
Range
The range is the difference between the maximum and minimum values.
A larger range means the variation span of this batch of measurement data is larger.
This can be simply understood as:
The range is used to observe how large the data fluctuation is.
Median
The median is the value located in the middle of the data after it is arranged in order of magnitude.
Like the mean, it can be used to observe where the data is generally concentrated.
The median is also used in the XMed-R control chart.
Standard Deviation
The standard deviation is used to represent the dispersion of the measurement data.
In general:
Smaller standard deviation → more concentrated data
Larger standard deviation → more dispersed data
The standard deviation is also an important statistic in the XBar-S control chart.
Cp
Cp is used to evaluate the relationship between the fluctuation of the current process and the product specification range.
In general:
The larger the Cp, the smaller the process fluctuation relative to the allowable tolerance.
Cp mainly focuses on the degree of data fluctuation.
Cpk
Cpk considers the data fluctuation and also whether the data as a whole is biased toward the specification upper or lower limit.
This can be simply understood as:
Cp mainly looks at fluctuation, while Cpk also considers whether the data is biased to one side.
In practice, Cp and Cpk are usually observed together.
Ca
Ca is used to reflect the degree of offset of the data center relative to the specification center.
In general:
The closer Ca is to 0, the closer the data center is to the specification center.
Number of Nonconforming Items and Nonconforming Percentage
When a measurement result exceeds the set specification upper limit or falls below the specification lower limit, it is judged as nonconforming.
The SPC item list displays:
- The number of nonconforming items
- The nonconforming percentage
This gives you a quick understanding of the out-of-spec situation of the current control item.
UCL, CL, and LCL
In SPC control charts, you will see:
- UCL: upper control limit
- CL: center line
- LCL: lower control limit
They are used to determine whether the data variation of the production process exceeds the normal control range.
Note that:
The specification limits are used to judge whether a product is conforming, while the UCL/LCL control limits are used to judge whether the process is stable.
These two are not the same concept.
Simple Reminders
In daily use, you can understand it this way:
- Mean: look at the data center
- Range / Standard deviation: look at data fluctuation
- Cp / Cpk: look at process capability
- Ca: look at center offset
- Nonconforming percentage: look at the out-of-spec situation
- UCL / LCL: look at whether the process shows abnormalities
More detailed usage of UCL, CL, and LCL as well as the different control charts can be found in the next page, "SPC Control Charts".