SPC control charts are based on mathematical and physical statistics, using the statistical term “detection of anomalies” as a tool for controlling the production process.

Therefore, “finding anomalies” becomes the basis of SPC control chart. It uses valid data to establish control limits, generally divided into upper control limit (UCL) and lower control limit (LCL), the upper control limit is usually set to the control centerline CL + 3σ, σ is the standard deviation, the lower control limit is set to CL-3σ. If the process is not affected by special reasons, then the observed data will not exceed the control limit.

The probability density function of the normal distribution then yields that roughly 68% of the measurements will be in the interval μ-σ and μ+σ, roughly 95.5% of the measurements will be in the interval μ-2σ and μ+2σ, roughly 99.73% of the measurements will be in the interval μ-3σ and μ+3σ, and a minimum of 0.135% and a maximum of 0.135% falls outside.

Since judging abnormal fluctuations is one of the most important functions of the SPC control chart, what is the basis for judging abnormalities and what are the criteria for judging differences? The international GB/T4091-2001 Conventional Control Chart stipulates 8 kinds of criteria for judging anomalies.

SPC's 8 kinds of criteria for judging abnormality:

1. 1 point falls outside the A-zone

2. 9 consecutive points fall on the same side of the center line

3. 6 consecutive points in increasing or decreasing order

4. 14 consecutive points in which neighboring points alternate up and down

5. 2 out of 3 consecutive points fall outside Zone B on the same side of the centerline

6. 4 out of 5 consecutive points fall outside Zone C on the same side of the centerline

7. 15 consecutive points fall within Zone C on both sides of the centerline

8. 8 consecutive points fall on both sides of the center line and none of them are in Zone C.

Disposition principle of the discrepancy: identify the cause, take measures to eliminate it, and incorporate it into the standard if it does not occur again.

In order to apply these criteria, the SPC control chart is divided into 6 zones, each 1σ wide, labeled A, B, C, C, C, B, A. It should be noted that these criteria are mainly applicable to X charts and single-value X charts, and it is assumed that the quality characteristic X obeys a normal distribution.

Criterion 1 A point falls outside the A region.

Cause of abnormality: Calculation error, measurement error, raw material failure, equipment failure, etc.

Criterion 2 Nine consecutive points fall on the same side of the centerline.

Cause of abnormality: Same as criterion 1.

Criterion 3 6 consecutive points are increasing or decreasing.

Cause of abnormality: The parameter changes over time due to wear and tear of the tool and deterioration of maintenance.

Criterion 4 Neighboring points alternate up and down for 14 consecutive points.

Cause of abnormality: Insufficient data stratification due to alternating use of two units or rotation by two operators.

Criterion 5 Two of the three consecutive points fall outside the B zone on the same side of the centerline.

Cause of abnormality: Process parameter u has changed.

Criterion 6 4 out of 5 consecutive points fall outside area C on the same side of the centerline.

Reason for anomaly: Same as criterion 5.

Criterion 7 Fifteen consecutive points fall within area C on both sides of the centerline.

Reason for anomaly: False data, calculation error, insufficient data stratification.

Criterion 8 8 consecutive points fall on both sides of the centerline and none of them are in the C-zone.

Reason for anomaly: Insufficient data stratification.