SPC專業(yè)術(shù)語

1、管制上限/管制下限：UCL / LCL Control limit. A line (or lines) on a control chart used as a basis for judging the significance of the variation from chart point to chart point. Variation beyond a control limit is evidence that assignable causes are affecting the process. Control limits are calculated from process data and are not to be confused with engineering or administrative specifications. A control chart usually displays a lower control limit (LCL) and an upper control limit (UCL). 管制界線。 用在管制圖中當(dāng)作點(diǎn)到點(diǎn)之間變異的衡量基準(zhǔn)。 超出管制界線的變異可以證明某些原因會影響制程。 管制界線是從制程的數(shù)據(jù)來計算，不要和規(guī)格界線混淆。 通常管制圖會顯示管制下限 (LCL) 及管制上限 (UCL) 。 注 : 一般而言 , UCL 與 LCL 的位置為平均值的正負(fù)三個標(biāo)準(zhǔn)差。

2、制程準(zhǔn)確度：Ca The deviation of the process mean from the midpoint of the specification limits. 衡量制程之實(shí)際平均值與規(guī)格中心值之一致性自生產(chǎn)中所獲得產(chǎn)品資料的實(shí)際平均值與規(guī)格標(biāo)準(zhǔn)值間偏差的程度。

3、制程精密度： Cp / Capability of process. The greater the number, the better the process is capable of meeting the specification limits. A process is judged to be capable when CP is at least 1.0. 衡量制程之變異寬度與規(guī)格公差范圍相差之情形。 一般而言 , CP 至少大于一才可以開始量產(chǎn) , 否則因變異產(chǎn)生的成本會過高。

4、制程能力指數(shù)： Cpk Process Capability index related to both dispersion and centeredness. 關(guān)于分散與集中的制程能力指針。 同時考慮變異寬度與制程標(biāo)準(zhǔn)值之綜合性制程能力指針。 綜合 Ca 與 Cp 兩值之指數(shù)。

5、CpL Distance between the process mean and the lower specification limit scaled using the capability sigma. 制程標(biāo)準(zhǔn)值與規(guī)格下限之間的差距值。

6、CpM Process capability index that is affected by the difference between the sample mean and the target value. 受樣本中心值與規(guī)格中心值之差所影響的制程能力指針 。

7、CpU Distance between the process mean and the upper specification limit scaled using the capability sigma. 制程標(biāo)準(zhǔn)值與規(guī)格上限之間的差距 , 通常用在單邊規(guī)格。

8、CR The reciprocal of CP. The smaller the number, the better the process is capable of meeting the specification limits. CR*100% is the percentage of the specification band used up by the process. CP 的倒數(shù)。這個數(shù)愈小，制程愈能和規(guī)格限制相交。 CR 百分比則代表制程中所耗損的規(guī)格寬度。

9、Pp Most commonly used process performance index. This is a more reliable measure of process performance for folded normal distribution since folded normal distribution is one-sided. 此為對于交叉常態(tài)分配更可信賴的制程能力量測指針，因為交叉常態(tài)分配呈現(xiàn)單邊分配。

10、PpK Process performance index related to both dispersion and centeredness. This index is not suitable for folded normal distribution, which is one-sided. 同時考慮變異寬度與制程標(biāo)準(zhǔn)值之綜合性制程能力指針。 但，這指針不適合交叉常態(tài)分配，因它是單邊分配。

11、PpL Distance between the process mean and the lower specification limit scaled using the process standard deviation. This index is not suitable for folded normal distribution, which is one-sided. 制程標(biāo)準(zhǔn)值與規(guī)格下限之間的差距值。 但，這指針不適合交叉常態(tài)分配，因它是單邊分配。

12、Ppm Process performance index that is affected by the difference between the sample mean and the target value. 受樣本中心值與規(guī)格中心值之差所影響的制程能力指針。

13、PpU Distance between the process mean and the upper specification limit scaled using the process standard deviation. This index is not suitable for folded normal distribution, which is one-sided. 制程標(biāo)準(zhǔn)值與規(guī)格上限之間的差距。但，這指針不適合交叉常態(tài)分配，因它是單邊分配。

14、PR PR. Reciprocal of PP. PP 的倒數(shù)。

15、峰度 Kurtosis A measure of peakedness. The population kurtosis is given by the fourth moment over the square of the variance. This ratio equals 0 for any normal distribution. A positive ratio indicates an excess of values in the neighborhood of the mean (with a depletion of the "flanks" of the curve representing the distribution). and a negative ratio corresponds to a curve with a flatter top than the normal curve has. 尖峰的量測指針。 最常用的峰度是四倍的動差乘上變異數(shù)的平方。 對于任何常態(tài)分配而言，這比率為 0 。 正機(jī)率意指超出平均數(shù)的鄰近值，而負(fù)機(jī)率則指超出常態(tài)的曲線。

16、偏態(tài) Skewness A measure of departure from normality. A distribution is skewed if it isn't symmetric but has more cases (a "tail") toward one end of the distribution than the other. If the tail is toward larger values, the distribution is positively skewed, or skewed to the right. If the tail is toward smaller values, the distribution is negatively skewed, or skewed to the left. 衡量從常態(tài)分配的偏移量。 若分配不勻稱則會偏斜，朝向某一邊，呈現(xiàn)單尾。 若偏向較大的值，則分配呈現(xiàn)正偏差，反之則呈現(xiàn)副偏差。