GRR 分析報告範例

這是 GRR 分析報告的一個範例。以圖表配合解說，讓使用者更輕鬆的的解讀 GRR 測試的結果以做出更正確的判斷。

使用的工具和方法參考GRR分析應用工具和如何解讀 GRR 分析結果。

GRR summary

! %GRR = 22.1 :
The measurement system is potentially acceptable.

✓ Number of distinct categories = 6 :
The measurement system can distinguish between parts.

𝒾 Repeatability contributes 54.8% of GRR variance, and reproducibility contributes 45.2%.

✗ Effect of operators is significant, with p-value of ANOVA = 0.
Operator contributes 45.2% of GRR variance.

✓ There is no interaction between operators and parts.

𝒾 Number of operators = 3, Number of trials = 3, Number of parts = 9

Table 1: GRR table
Variance Std. Dev. Study Var. % Study Var. % Contribution % Tolerance
Total R&R 5.014 2.239 13.436 22.1 4.9 20.4
Repeatability 2.746 1.657 9.943 16.3 2.7 15.1
Reproducibility 2.268 1.506 9.036 14.8 2.2 13.7
Part 97.879 9.893 59.360 97.5 95.1 89.9
Total Variation 102.893 10.144 60.862 100.0 100.0 92.2

X̄-R chart

! X̄ chart of this GRR test has 67% of data points out of control. That means the measurement system can but not well detect the parts variation.

✓ R chart of this GRR test has 100% data points inside control limits. The measurement system has good consistency.

Measurement by operators

✗ Operators measuring the parts similiarly: No.
P-value of ANOVA (comparing mean) = 0.

✓ Operators’ measurement have the same variation: Yes.
P-value of Levene’s test (comparing variance) = 1.

Interaction of operator by part

✓ No interaction between operators and parts: Yes.
P-value of ANOVA = 1.

Measurement by parts

✓ Parts are measured with the same variation: Yes.
P-value of Levene’s test (comparing variance) = 0.161.

ANOVA

Df Sum Sq Mean Sq F value Pr(>F)
Operator 2 127.963 63.981 23.298 0
Part 8 7069.268 883.659 321.766 0
Residuals 70 192.240 2.746