Design for Manufacturing and Assembly Process Capability Analysis

PROCESS CAPABILITY ANALYSIS
PROCESS CAPABILITY (CP): Process capability is the repeatability and consistency of a manufacturing process relative to the customer requirements in terms of specification limits of a product parameter. This measure is used to objectively measure the degree to which your process is or is not meeting the requirements.

Process capability compares the output of an in-control process to the specification limits by using capability indices. The comparison is made by forming the ratio of the spread between the process specifications (the specification "width") to the spread of the process values, as measured by 6 process standard deviation units (the process "width"). Cp = (USL - LSL) / 6 sigma Cp<1 means the process variation exceeds specification, and a significant number of defects are being made. Cp=1 means that the process is just meeting specifications. A minimum of .3% defects will be made and more if the process is not centered. Cp>1 means that the process variation is less than the specification, however, defects might be made if the process is not centered on the target value.

1

Design for Manufacturing and Assembly Process Capability Analysis While Cp relates the spread of the process relative to the specification width, it does not address how well the process average, X, is centered to the target value. Cp is often referred to as process "potential".

We define process capability analysis as an engineering study to estimate process capability. The estimate of process capability may be in the form of a probability distribution having a specified shape, center (mean), and spread (standard deviation). For example, we may determine that the process output is normally distributed with mean µ = 1.0 cm and standard deviation σ = 0.001cm. in this sense, a process capability analysis may be performed without regard to specifications on the quality characteristic.

2

Design for Manufacturing and Assembly Process Capability Analysis PROCESS CAPABILITY INDICES (CPK): A capable process is one where almost all the measurements fall inside the specification limits. This can be represented pictorially by the plot below:

The Cp, Cpk, and Cpm statistics assume that the population of data values is normally distributed. Assuming a two-sided specification, if and are the mean and standard deviation, respectively, of the normal data and USL, LSL, and T are the upper and lower specification limits and the target value, respectively, then the population capability indices are defined as follows:

The estimator for Cpk can also be expressed as Cpk = Cp(1-k), where k is a scaled distance between the midpoint of the specification range, m, and the process mean, . Denote the midpoint of the specification range by m = (USL+LSL)/2. The distance between the process mean, , and the optimum, which is m, is - m, where

. The scaled distance is

3

Note that the reject figures are based on the assumption that the distribution is centered at Process Capability Ratio (Cp) and Associated . ). is . adjusted by the k factor.Values of the 4 . consider the following plot This can be expressed numerically by the table below: where ppm = parts per million and ppb = parts per billion. To determine the Since . we estimate by . To get an idea of the value of the Cp statistic for varying process widths.Design for Manufacturing and Assembly Process Capability Analysis (the absolute sign takes care of the case when estimated value. . Note that The estimator for the Cp index. it follows that .

Design for Manufacturing and Assembly Process Capability Analysis We have discussed the situation with two spec. This is known as the bilateral or two-sided case. Using one spec limit is called unilateral or one-sided. limits. respectively. There are many cases where only the lower or upper specifications are used. the USL and LSL. Cpu}. This can be represented pictorially by Note that we also can write: Cpk = min {Cpl. The corresponding capability indices are where and are the process mean and standard deviation. respectively Estimators of Cpu and Cpl are obtained by replacing and by and s. 5 . The following relationship holds Cp = (Cpu + Cpl) /2.

0. If possible. is 0. we can do a Process Capability Analysis. is the algebraic equivalent of the min{ .Cpk. But it doesn't. since = 16. s = 2. Last month's e-zine contained an explanation of process capability and introduced one metric (Cp) to measure process capability. We can compute the From this we see that the Note that the formula } definition.6667. From this we obtain This means that the process is capable as long as it is located at the midpoint. The observed process average. m = (USL + LSL)/2 = 14. = 16. 6 . and the standard deviation. so this is not a good process. This month's e-zine introduces a second metric to measure process capability .Design for Manufacturing and Assembly Process Capability Analysis CAPABILITY INDEX EXAMPLE: For a certain process the USL = 20 and the LSL = 8. PROCESS CAPABILITY METRICS: For processes that are in statistical control and that are normally distributed. which is the smallest of the above indices. and reduce the variability or/and center the process. The factor is found by and We would like to have at least 1. .

Design for Manufacturing and Assembly Process Capability Analysis Cpk EXAMPLE FOR CPK : 7 .

If there is only one specification. A process can have a Cp > 1.not the standard deviation of the individual measurements.5. which the process capability is based on the lower specification limit. So. Just knowing that a process is capable (Cp > 1. The value of Cpk is the minimum of two process capability indices.5 pounds.0 and produce no product or service within specifications. Cpk is defined as shown in the figure. The specifications for the operation are a minimum of 49. . The value of Cpk is the difference between the process average and the nearest specification limit divided by three times the standard deviation. the lower specification limit (LSL) is 49. the value of Cpk is either Cpu or Cpl. this means that there is some product being produced out of specification.LSL) to the natural tolerance (6s).0) does not ensure that all the product or service being received is within the specifications. It should be noted that the standard deviation is the standard deviation based on a R or s chart . The value of Cp does not take into account where the process is centered. Cpk values above 1. Cp is the ratio of the engineering tolerance (USL . Both Cpu and Cpl take into account where the process is centered. which is the process capability based on the upper specification limit. whichever is appropriate for the specification. If Cpk is less than 1. A bagging operation is designed to place 50 pounds of sand into each bag.0 are desired.5 pounds and a maximum of 50.Design for Manufacturing and Assembly Process Capability Analysis Cp values are not the best indicators of process capability.5 pounds. The upper specification limit (USL) is 8 50.sided specifications. This means that essentially no product or service is being produced above USL or below LSL. The other is Cpl.0. Cpk takes into account where the process is centered. In addition. A better measure of process capability is Cpk. One process capability is Cpu. The figure above shows how the Cpk values are calculated. Cp values can't be calculated for one. Algebraically. As shown last month.

The control chart is shown in the figure and is in statistical control. Since Cpk is the minimum of Cpu and Cpl. The standard deviation (from the range chart data) is 0.05. The subgroup average and range are calculated and plotted on an Xbar-R chart.212.71.Design for Manufacturing and Assembly Process Capability Analysis The operation is being monitored using a Xbar-R chart with a subgroup size of 4. four consecutive bags are weighed. Cpk = 0. 9 . The calculations for Cpk are shown below. Each hour. The average bag weight is 50.

+ X n . etc and finally Xn occurs fn times. Mean is expressed algebrically... the range is difference between the largest observed value and the smallest observed value. average yield. 2  Mode: The mode of a set of data is the value which occurs most frequently. X 1 . Median = n +1 2 If n is even.... If n is odd...... Then n = f 1 + f 2 + f 3 + . + f n The mean is used to report average size. Variance ( σ 2 ): It is defined as the sum of the squares of the deviations from the arithmetic mean divided by the number of observations ‘n’. Range(R): In the control chart... x= X 1 + X 2 + X 3 + .. average percent defective etc. Median: When all the observations are arranged in ascending or descending order. then the median is the magnitude of the middle case. n 2 n  + 1 th value... Variance ( σ 2 (x )= 1 − x + x2 − x + xn − x n ) ( 2 ) ( 2 ) 2 Sample problems 10 . f n f X + f 2 X 2 + f 3 X 3 + .... n Where the symbol x represents the arithmetic mean... X2 occurs f2 times...Design for Manufacturing and Assembly Process Capability Analysis Definitions Mean ( x ): The arithmetic mean of a set of ‘n’ numbers is the sum of the numbers divided by ‘n’. X 3 . Median is average of   th and  Where n = No of observations.. then....... + f n X n x= 1 1 f1 + f 2 + f 3 + . are the n values of the variate X i. x= ∑x n If X1 occurs f1 times..... X n ..... X 2 ...e.

variance.5 15.50 2601.5 45. Order size range 1-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 Class mark(x) 5. the variance for the data used above is calculated as follows: Variance σ 2 = 1 2 (1 + 22 + 02 + 32 + 22 ) .5 85.5 35.5 Frequency (f) 1 2 4 12 13 8 8 1 1 fx fx2 5.25 7310. Calculate the mode and variance. 1. 0.00 15123.5 55.00 26913.6 5 () Using this formula.56 5 σ 2 = 1. To calculate x .5 65.00 34322.0 102.5 30.0 75. 2 Solution: Mode = 2 (it occurs more than the other values) This can be put more succinctly using the summation notation as. 2.1.Design for Manufacturing and Assembly Process Capability Analysis Example 1: The no of orders received for a particular item on each day for five days are as follows. Variance σ 2 = 1 ∑ x−x n ( ) 2 It is possible to rearrange this formula in a way which makes the calculation of the variance much easier in general.62 5 18 = .5 75.04 Standard deviation = (σ ) 2 Standard deviation = (1.0 524.0 ∑ fx 2 = 117122.25 ∑ f = 50 ∑ fx = 2285.00 5700.25 24642. 3.5 444.50 11 .5 25.0 426.5 85.25 480.0 591.we use n 1 + 2 + +0 + 3 + 2 Mean x = = 1 . standard deviation for the given order size data.04) = 1.02 Example 2: Calculate mean.5 31.2. Variance σ 2 = 2 1 ∑ x 2 − x .

g. as for instance the shape of the distribution of personal incomes. 12 . but occurs.7) 2 50 = 45. These generally represent a purely temporary process condition. but unlike the normal curve they are not symmetrical. and serve as a guide to detecting the presence of some unusual factor like defective material. Coefficient of Skewness= Mean − Mode σ Several measures of skewness have been proposed. (e. Figure 1 Skewed curves Their extreme values occur more frequently in one direction from the centre than in the other. Positive skewness is also quite common. etc. These curves are like normal curves in that the frequencies decrease continuously from the centre to extreme values. Another example is the distribution of the time intervals between randomly occurring events. but are rarely used in practice.5 . The normal distribution is the most commonly occurring symmetrical frequency distribution. The simplest way of describing skewness is to quote the mean. or abnormal machining conditions. in the distribution of times to failure of certain types of equipment.Design for Manufacturing and Assembly Process Capability Analysis Mean x = 2285 50 117122.96 Standard deviation = 253.(45. They appear like “disturbed normal” curves and hence are called “skewed curves”. the median.7 Variance σ 2 = = 253.) tool chatter. which does not follow the shape of the normal curve. and. Negative skewness is less common. tool vibration.96 = 15.9 Skewness: The curve. for instance. such as the arrival of customers at the ends of a queue.

This is very noticeable for the distribution of personal incomes. Coefficient of Skewness= When mode is ill-defined Coefficient of Skewness= Mean − Mode σ 3( Mean − Mode) σ (b) Measure of Skewness based on moments: With the help of moments Skewness can be determined. expressed as Mean-Mode=3(Mean-Median). which will in turn be less than the mean. Measure Of Skewness: (A) Absolute Skeweness (a) Absolute Sk = Mean-Mode (when mode is not ill-defined) (b) Absolute Sk = 3(Mean-Median) (B) Relative Skewness (a) Karl Pearson’s coefficient of Skewness.Design for Manufacturing and Assembly Process Capability Analysis where possible the mode. When the distribution is moderately. these three measures will be in the reverse order. Moments (i) Moments about Mean µ1 = µ2 = ∑(X − X ) N ∑( X − X )2 N 13 = σ 2 or σ = µ2 . Karl Pearson suggested β 1 as Measure of Skewness. For positively skewed distributions. For symmetrical distributions. For negatively skewed distributions. there is an approximate relationship between the three measures. µ32 β1 = 3 µ2 For a symmetrical distribution β 1 = 0. the mode will be less than the median. these three measures will approximately coincide. The differences between the measures give some indication of the extent of the skewness.

(ii) Moments about arbitrary origin A µ 11 = µ 21 = ∑ ( X − A) N ∑ ( X − A) 2 N ∑ ( X − A)3 N ∑ ( X − A) 4 N µ 31 = µ 41 = In a frequency distribution the moments about an arbitrary origin will be calculated as follows: µ 11 = µ 21 = µ 31 = µ 41 = ∑ fd × i N or or or or ∑ f ( X − A) × i ∑ ∑ ∑ N 2 f ( X − A) N 3 f ( X − A) N 4 f ( X − A) N ∑ fd 2 N ∑ fd 3 N ∑ fd 4 N × i2 × i3 × i4 ×i2 × i3 ×i4 Where ‘ i ’ is the class interval and l=   X − A . µ1= µ11-µ11 =0 µ2 = µ21-( µ11)2 µ3 = µ31-(3µ11 µ21)+2 ( µ11)3 µ4 = µ41-(4µ11 µ31)+6 ( µ11)2 µ21 -3( µ11)4 14 . etc.Design for Manufacturing and Assembly Process Capability Analysis µ3 ∑(X − X ) = N 3 µ4 = N ∑( X − X )4 In case of frequency distribution µ1 = ∑ f (X − X ) N =0 µ2 ∑ f (X − X ) = N 2 = σ 2 .  i  Order to simplify calculations the moments are first calculated about an origin A. They can then be converted with the help of the following relationships to obtain moments about mean.

334 σ =  −   100   100        15 .Design for Manufacturing and Assembly Process Capability Analysis Example 3: Calculate any measure of skewness from the following data: X f 0 12 1 27 2 29 3 19 4 8 5 4 6 1 7 0 Solution: Since the question is to calculate any measure of skewness. Calculation of Coefficient of Skewness X 0 1 2 3 4 5 6 7 N=100 f 12 27 29 19 8 4 1 0 x-2 d -2 -1 0 +1 +2 +3 +4 +5 fd -24 -27 0 +19 +16 +12 +4 0 ∑ fd = 0 2 fd2 48 27 0 12 32 36 16 0 ∑ fd = 178 x = A+ ∑ fd N = 2 + ( 0 / 100 ) = 2   ∑ fd 2 Standard deviation σ =    N    ∑ fd  −     N  2       178   0  2   = 1. mode and standard deviation. The formula is: Coefficient of Skewness= Mean − Mode σ Hence for calculating skewness we have to determine the values of mean. we should prefer Karl Pearson’s coefficient of skewness because it is considered to be the best measure for calculating skewness.78 = 1.

334 Example: 4 Calculate Karl Pearson’s co-efficient of skewness from the following data: Size Frequency Solution: Calculation of Karl Pearson’s Coefficient of Skewness Size x 1 2 3 4 5 6 7 Frequency f 10 18 30 25 12 3 2 N=100 x-4 d -3 -2 -1 0 +1 +2 +3 fd -30 -36 -30 0 +12 +6 +6 fd2 90 72 30 0 12 12 18 1 10 2 18 3 30 4 25 5 12 6 3 7 2 ∑ fd = −72 ∑ fd = 234 2 Coefficient of Skewness= Mean − Mode σ Mean: x = A+ ∑ fd N = 4 .Design for Manufacturing and Assembly Process Capability Analysis Mode: Since the highest frequency is 29. 16 . 3. Mo =2.152 Mode: Since the maximum frequency is 30.334 Substituting these values in the formula. 2. σ =1.28   ∑ fd 2 Standard deviation σ =    N    ∑ fd  −     N  2     σ =   232   − 72  2    = −   100   100        2 . x =2 .e. Coefficient of Skewness = Mean − Mode σ = 2−2 = 0.28856 = 1.e.(72/100) =3. by inspection the mode is the value corresponding to the frequency 29 i. by inspection the mode is the value corresponding to the frequency 30 i. 1.

518 Substituting these values in the formula. In it the data is normal distributed about the mean the kurtosis will be equal to 3. This force the value to be zero based.28 . Zero-based kurtosis= 1 n  xi − x  ∑ s  − 3  n i =1    4 Note that the value of 3 is subtracted from the kurtosis value. The first is mesokurtic distribution as shown in the Figure 2. As the kurtosis deviates above or below 3.Design for Manufacturing and Assembly Process Capability Analysis x =3. 1 n µ 4 = ∑ xi − x n i =1 Kurtosis = ( ) 4 µ s 4 4 The following equation is commonly used to calculate the zero based kurtosis in statistical analysis computer programming. The following equation incorporates the fourth moment about the mean and the fourth power of the samples standard deviation to measure kurtosis. as opposed to be centered around the number 3.184 1. The peakedness or flatness begins to take a numerical significance as described below.518 Kurtosis: The fourth momentum will provide a numerical value associated with the peakedness or flatness of the data as it is a distributed about the mean also known as “kurtosis”. Mo =3. Mesokurtic: They are three general distributions types used to define nature of kurtosis. Coefficient of Skewness = Mean − Mode σ = 3. Figure 2 Mesokurtic distribution 17 . σ =1. The common approach to quantity kurtosis is that the normal peak distribution is centered about the value 3.28 − 3 = 0.

shown in figure 4. In the data is dispersed about the mean in a manner that is very peaked in nature. Figure 3 Platykurtic distribution Leptokurtic: The third is leptokurtic distribution. Figure 4 Leptokurtic distribution 18 . shown in figure 3. In it the data is dispersed bout the mean in a manner that is flat in nature: the kurtosis will be less than 3.Design for Manufacturing and Assembly Process Capability Analysis Platykurtic: The second is platykurtic distribution. the kurtosis will be greater than 3.

If β 2 is more than 3 the curve is leptokurtic and if it β2 = less than 3 the curve is platykurtic. Example: 5 Calculate first four moments from the following data: Also calculate the values of β 1 and β 2 and comment on the nature of the distribution X Y 0 5 1 10 2 15 3 20 4 25 5 20 6 15 7 10 8 5 Solution: Calculation of moments X 0 1 2 3 4 5 6 7 8 f 5 10 15 20 25 20 15 10 5 N=125 fX 0 10 30 60 100 100 90 70 40 ∑fX = 500 (X-4) -4 -3 -2 -1 0 +1 +2 +3 +4 f(X-4) -20 -30 -30 -20 0 +20 +30 +30 +20 ∑f(X-4) =0 f(x-4)2 80 90 60 20 0 20 60 90 80 ∑ f(x-4)2 =500 f(X-4)3 -320 -270 -120 -20 0 +20 +120 +270 +320 ∑ f(X-4)3 =0 f(X-4)4 1280 810 240 20 0 2 40 10 280 ∑ f(X-4)4 =4700 x= ∑ fX µ1 = 500 =4 N 125 ∑ f (X − X ) = N 19 .Design for Manufacturing and Assembly Process Capability Analysis Measures Of Kurtosis: Kurtosis are measured by the coefficient µ4 µ22 γ 2 = β2 − 3 or For normal distribution β 2 =3.

N = 125 µ1 = µ2 = 0 =0 125 ∑ f ( X − X )2 N = 500 =4 125 µ3 µ4 ∑ f ( X − X )3 = = N N 0 =0 125 125 ∑ f ( X − X ) 4 = 4700 = 37. Example: 6 Using moments.35 16 µ22 β1 = Since β 1 is zero. of Workers 8 11 18 9 4 20 .6 = µ32 02 = =0 µ2 3 43 µ 4 37.Design for Manufacturing and Assembly Process Capability Analysis ∑ f ( X − X ) = 0. the distribution is platykurtic.6 = = 2. calculate a measure of relative skewness and a measure of relative kurtosis for the following distribution and comment on the result obtained: Weekly Wages (Rs) 70 but below 90 “ 110 “ 130 “ 150 “ 90 110 130 150 170 No. the distribution is symmetrical β2 = Since β 2 is less than 3.

Design for Manufacturing and Assembly Process Capability Analysis Solution: Weekly wages (Rs) 70-90 90-110 110-130 130-150 150-170 f 8 11 18 9 4 N=50 m. µ4 = µ41-4(µ11 µ31)+6 ( µ11)2 µ21 -3( µ11)4 = 674800-4(-4) (-5440) +6(-4)2(544)-3(-4)4 = 642816 21 .p 80 100 120 140 160 d -2 -1 0 1 2 fd -16 -11 0 9 8 ∑fd =-10 fd2 32 11 0 9 16 ∑fd2=68 fd3 -64 -11 0 9 32 ∑ fd3=-34 fd4 128 11 0 9 64 ∑fd4=212 µ 11 = µ 21 = ∑ fd × i = − 10 × 20 = −4 ∑ fd N 50 2 68 × 400 = 544 N 50 ∑ fd 3 × i 3 = − 34 × 8000 = −5440 1 µ3 = N 50 4 ∑ fd × i 4 = 212 × 160000 = 678400 µ 41 = N 50 × i2 = Moment about Mean µ2 = µ21-( µ11)2 =544-(-4)2=528 µ3 = µ31-3(µ11 µ21)+2 ( µ11)3 = -5440-3(-4) (544) +2(-4)3 = 960.

the distributions platykurtic. ∑ fd =-14 .306 µ 2 2 278784 Since β 2 is less than 3.4 = 33. i=10 14 × 10 = 35 − 1. Example: 7 Calculate coefficient of skewness by Karl Pearson’s method and the values of β 1 and β 2 from the following data: Profits(Rs. N=100. m.p (mof fd3 (Rs.6 100 ∆1 Mode: Mode= L + ×i ∆1 − ∆ 2 x = 35+ By inspection mode lies in the class 30-40 22 .Design for Manufacturing and Assembly Process Capability Analysis µ32 960 2 Skewness= β 1 = = = 0. of companies 18 20 30 22 10 Solution: Calculation of Karl Pearson’s Coefficient of Skewness β 1 and β 2 Profits No.08 528 3 µ23 β2 = µ 4 642816 = = 2. 45)/10 fd fd2 Lakhs) cos m d f 10-20 18 15 -2 -36 72 -144 20-30 20 25 -1 -20 20 -20 30-40 30 35 0 0 0 0 40-50 22 45 +1 +22 22 +22 50-60 10 55 +2 +20 40 +80 2 N=100 fd = −14 fd = 154 fd 3 = −62 fd4 ∑ ∑ ∑ ∑ fd 288 20 0 22 160 4 = 490 Karl Pearson’s Coefficient of Skewness= Mean − Mode Mean: x = A+ ∑ fd × i N σ A=35. Lakhs) 10-20 20-30 30-40 40-50 50-60 0 No.

4 ∑ fd N 100 2 154 × 100 = 154 N 100 ∑ fd 3 × i 3 = − 62 × 1000 = −620 µ 31 = N 110 4 ∑ fd × i 4 = 490 × 10000 = 49000 µ 41 = N 100 × i2 = µ2 = µ21-( µ11)2 =154-(-1.4)3 = 20. i = 10 Mode= 30+ 10 × 10 = 30 + 5.56 =-0.33     100   100     Karl Pearson’s Coefficient of Skewness= 33.6 − 35.04 µ3 = µ31-3(µ11 µ21)+2 ( µ11)3 = -620-3(-1.56 = 35.4)2=152.32 µ4 = µ41-4(µ11 µ31)+6 ( µ11)2 µ21 -3( µ11)4 = 49000-{4(-1.4)2(154)-{3(-1.Design for Manufacturing and Assembly Process Capability Analysis L=30 ∆ 1 = 30 − 20 = 10.33 µ32 Calculation of β 1 = 3 µ2 We will have to calculate moments µ 11 = µ 21 = ∑ fd × i = − 14 × 10 = −1.4) (-620)} +{6(-1.159 12.4)4} = 47327.4) (154) +2(-1.516 23 .56 10 + 8 Standard deviation σ =      ∑ fd 2  N 2    ∑ fd  −  × i     N   2  154   − 14  σ =  −  × 10  = 12. ∆ 2 = 30 − 22 = 8.

04 )2 µ2 Process Capability Analysis: Statistical techniques can be helpful throughout the product cycle. and spread (standard deviation). in quantifying process variability. “Instantaneous” variability. a process capability analysis may be performed without regard to specifications on the quality characteristic.27% of the process output will fall outside the natural tolerance limits. The natural or inherent variability at a specified time.516 β 2 = 42 = = 2. 2.0.27% outside the natural tolerances sounds small. the natural tolerance limits include 99. Figure 5 shows a process for which the quality characteristic has a normal distribution with mean µ and standard deviationσ . σ We define process capability analysis as an engineering study to estimate process capability. However. Two points should be remembered: 1. or put another way.001 cm. σ σ UNTL = µ + 3σ LNTL = µ − 3σ For a normal distribution. but this corresponds to 2700 nonconforming parts per million. Product capability refers to the uniformity of the process. we may determine that the process output is normally distributed with mean µ = 1. and in assisting development and manufacturing in eliminating or greatly reducing this variability.00013 2 (152. The estimate of process capability may be in the form of a probability distribution having a specified shape. If the distribution of process output is nonnormal. That is.Design for Manufacturing and Assembly Process Capability Analysis µ 3 2 (21. that is. .27%. the variability in the process is a measure of the uniformity of output. then the percentage of output falling outside µ ±3 may differ considerably from 0.04 ) 3514581. 2. This general activity is called process capability analysis. For example. The variability over time. Alternatively. The upper and lower “natural tolerance limits” (UNTL & LNTL) of the process fall at µ +3 and0 µ −3 . 24 . in analyzing this variability relative to product requirements or specifications. specifications are not necessary to perform a process capability analysis.32 )2 454. in this sense.1 µ2 µ 47327. There are two ways to think of this variability: 1.0 cm and standard deviation σ = 0.047 (152. It is customary to take the 6-sigma spread in the distribution of the product quality characteristic as a measure of process capability. Obviously. We present methods for investigating and assessing both aspects of process capability.73% of the variable. only 0. we may express process capability as a percentage outside of specifications.54 β1 = 2 = = = 0. center (mean). including development activities prior to manufacturing. respectively.

Thus. 2. the study is a true process capability study. Process capability analysis is a vital part of an overall quality-improvement program. production or manufacturing planning. we can say nothing about the dynamic behavior of the process or its state of statistical control. and there is no direct observation of the process or time history of production. process capability analysis is a technique that has application in many segments of the product cycle. perhaps supplied by the vendor or obtained via receiving inspection. interferences can be made about the stability of the process over time. Predicting how well the process will hold the tolerances. 7. Specifying performance requirements for few equipment. Assisting in establishing an interval between sampling for process monitoring. When the analyst can directly observe the process and can control or monitor the data-collection activity. control charts. vendor sourcing. In a characteristic or the process yield (fraction conforming to specifications). Reducing the variability in a manufacturing process. then the study is more properly called product characterization. Planning the sequence of production process when there is an interactive effect of process on tolerances. including product and process design. Selecting between competing vendors. 3. However. 25 .Design for Manufacturing and Assembly Process Capability Analysis Figure 5 Upper and Lower natural tolerance limits in the normal distribution. 5. because by controlling the data collection and knowing the time sequence of the data. 6. Assisting product developers/designers in selecting or modifying a process. not the process itself. 4. Among the major uses of data from a process capability analysis are the following: 1. A process capability study usually measures functional parameters on the product. Three primary techniques are used in process capability analysis: histograms or probability plots. when we have available only sample units of products. and manufacturing. and designed experiments.

After analysis. The data is then analyzed and bottle-necks and problems identified. The standard deviation indicates how tightly all the various examples are clustered around the mean in a set of data. the project and the process to be improved are defined after which the performance of the process is measured. Six Sigma means that there are 3. Internal Look Existing Process Delay Cycle (days) 20 15 30 10 5 16 Days (Average) “Internal Calibration “= 16 – 8 = 8 Therefore improvement is 50% After conventional improvements (days) 17 2 5 12 4 8Days (Average) 26 .4 defects per million events. and Control). illustrates how the concept of Six Sigma affects different people. Analyze. Example: 7 This sample example at GE. It concentrates on those outputs which are important to customers. After DMAIC circle it is time to define a new project. First. Six Sigma is carried out as projects. Six Sigma is a business method for improving quality by removing defects and their causes in business process activities. This development program is controlled by a management group. Measure. The main goal is continuous improvement. The method uses various statistical tools to measure business processes.Design for Manufacturing and Assembly Process Capability Analysis Six-Sigma: Sigma ( σ ) is a character of the Greek alphabet which is used in mathematical statistics to define standard deviation. In technical terms. improvement program is defined and defects removed. Average Vs Variation Customer expectations: 8 day order to Delay Cycle. Most common type is the DMAIC method (Define. Improve.

But this effect was not reflected on the customer’s side as they were still getting their products delivered at random as seen from the Figure 6 CUSTOMER LOOK Figure 6 From the above Figure Customer feels no change And once the feedback from the customer was heard. they modified the process to reflect Six Sigma delivery for the customer which resulted in the following: 6Sigma Internal Process 7 9 9 8 7 8 Days (Average) Here the internal look is same. CUSTOMER LOOK Figure 7 27 .Design for Manufacturing and Assembly Process Capability Analysis GE employees claimed that they had achieved Six Sigma capability after improving the delivery time for a medical product by 50% (brining it from an average of 16 days to average of 8 days). But the customer feels Six Sigma (Figure 7).

say Cp = USL − LSL 6σ --------------------------------------------.0099. the process standard deviation σ is unknown and must be replaced by an estimate σ . and σ =0. To estimate σ we typically use either the sample standard deviation S or R /d2 (when variables control charts are used in the capability study). This results in an estimate of the Cp. Usually. 68 We assumed that piston-ring diameter is approximately normally distributed and the cumulative normal distribution table in the appendix was used to estimate that the process produces approximately 20PPM (Parts Per Million) defective. we define the Cp as follows.95mm. For one-sided specifications. quantitative way to express process capability. 05 − 73 .05mm and LSL=73. Cp = USL − LSL 6σ ----------------------------------------- (2-1) Where USL and LSL are the upper and lower specification limits. The piston-ring process uses  1  P =   100  1 . respectively. C pu = USL − µ 3σ (upper specification only)--- -------------. 0099 ) = 1 . The Cp in equation (2-1) has a useful practical interpretation. 5 percent of the specification band. namely  1  P =   Cp  100    ---------------------------------------------- (2-3) Is the percentage of specification band used up by the process.Design for Manufacturing and Assembly Process Capability Analysis Process Capability Ratios: Use and interpretation of Cp It is frequently convenient to have a sample.(2-4) 28 . our estimate of the Cp is Cp = USL − LSL 6σ 74 .(2-2) To illustrate the calculation of the Cp. 95 = 6 ( 0 . The specifications on piston-ring diameter are USL=74. 68  = 59 . Equation (2-1) and (2-2) assume that the process has both upper and lower specification limits. Thus.

90 3.25 226.0009 0. and the case of two sides specification. while the Cp of 1.931 71. 29 .467 6. that the Cp of one implies a fallout rate of 2700 PPM of two sides specifications. assuming the process mean is centered between the upper and lower specification limits.60 35. These assumptions are essential to the accuracy of the reported numbers.70 17.34 1.17 0. Table 2. The process capability ratio is a measure of the ability of the process to manufacture product that means specification table 2.00 1.395 0.614 0.865 35.1 Values of the Process Capability Ratio (Cp) and Associated Process Fallout for a normal distribution process (in defective PPM) A process fallout(in defective PPM) Cp One side Specifications Two-sided specifications 0. respectively.729 0.20 159 318 1.Design for Manufacturing and Assembly Process Capability Analysis C pl = µ − LSL 3σ (lower specification only) -- --------------.934 1. To illustrate the use of table.807 133.00 0.1 shows several values of Cp along with the associated values of process fallout.40 14 27 1.80 8.5 implies fallout rate of 4 PPM of one side specification.30 48 96 1.(2-5) Estimate CpU and Cp L would be obtained by replacing µ and σ in equation (2-4) and (2-5) by estimate µ and σ .628 453. notice.198 16.861 0.0018 Then the table is invalid.03 0.350 2.60 1 2 1. and if they are not true. expressed in defective are non-conforming parts per million.70 0.700 1.50 4 7 1.10 484 967 1.06 2.255 050 66.80 0. This process fallout were calculated assuming a normal distribution of the quality characteristics.

in effect. Process Capability Ratio For An Off-Center Process The process capability ratio (Cp) does not take into account where the process mean is located relative to the specifications. or critical parameter.45 1.8 both have Cp =2. In recent years. For example the top two normal distributions in figure 2. requires that the process capability ratio will be least 2.2. existing process Safety.0. including their suppliers and customers. have adopted similar criteria. or critical parameter.2 Recommended Minimum Values of the Process Capability Ratio (Cp) Two-sided specifications Existing processes New processes Safety. For example.60 Table2. Within Motorola. This. This implies that the Cp should be atleast 1. Cp simply measures the spread of the specifications relative to the 6-sigma spread in the process. say by pouring more glass in the mould. but the process in panel (b) of the figure clearly has lower capability than the process in panel (a) because it is not operating at the midpoint of the interval between the specifications. many companies have adopted criteria for evaluating their processes that include process capability objectives that are more stringent that those of table2. 30 .2 are only recommended minimum. strength. strength.33 1. new process 1. We point out that the values in the table 2.0.45 perhaps one way the Cp could be improved would be increasing the mean strength of the bottles. bottles with inadequate pressure strength may fail and injury customers.50 1.45 1. this has become a corporate quality objective. Motorola’s “six-sigma” program essentially requires that when the process mean in control.Design for Manufacturing and Assembly Process Capability Analysis Table2.2 represents some recommended guidelines for minimum values of Cp the bottle strength characteristics a parameter closely related to the safety of product.25 1. it will not be closer that six standard deviations from the nearest specification limit. Many other organizations.50 One-sided specifications 1.67 1.

C pl = = 2 . the process is centered at the midpoint of the specifications. Note in panel (c) of figure 2.(2-6) Notice that Cpk is just the one-sided Cp for the specification limit nearest to the process average.8 that Cpk =1.8.0.1 to get a 31 . and when Cpk < Cp the process is off-center. we would have C pk = min( C = min( C = min( C pu .Design for Manufacturing and Assembly Process Capability Analysis Figure 8 Relationship of Cp and Cpk This situation may be more accurately reflected by defining a new process capability ratio that takes process centering into account. Several commonly encountered cases are illustrated in figure 2. This quantity is C pk = min( C pu . C pl ) ---------------------------------------------. The magnitude of Cpk relative to Cp is a direct measure of how off-center the process is operating. if Cp= Cpk.C = pl ) pu pu USL − µ µ − LSL . C pl = ) 3σ 3σ 62 − 53 53 − 38 = 1 .0 while Cp=2. For the process shown in figure 8a. One can use table 2. 5 .5 3( 2 ) 3( 2 ) ) Generally.

and they must be designed in conjunction with the application of the specific manufacturing process. there is great risk in having a mismatch between the required tolerance and the capability of a given process. However. Manufacturing Process Capability Metrics: Tolerances are always related to manufacturing processes or to materials used in the manufacture of a product. if we can center the process. The manufacturing engineer must also provide tolerances on the manufacturing process parameters and raw materials to help the team stay well within the tolerances assigned to the component being 32 . As panel (e) illustrates. Panel (d) of figure 8 illustrates the case in which the process mean is exactly equal to one of the specification limits. The design engineer and the manufacturing engineer have to define a common metric that quantifies the relationship that exists between the nominal design specifications. Often during technology development it is necessary to invent and co develop the manufacturing technology required to make the product. their tolerances.0018 PPM. any statistic that combines information about both location (the mean and process centering) and variability. Some authors define Cpk to be nonnegative. Thus.0 and two sided specifications) suggests that the potential fallout is 0. we usually say that PCR measures potential capability in the process.when the engineer finally gets around to selecting one. the entire process lies outside the specification limits.Design for Manufacturing and Assembly Process Capability Analysis quick estimate of potential improvement that would be possible by centering the process. then Cp=2. point estimates of process capability ratios are virtually useless if they are computed from small samples.0 in table 2. when Cpk < 0 the implication is that the process mean lies outside the specifications. It is unwise to wait until the tolerance design phase of a product-commercialization process to select or optimize a manufacturing process. If we take Cp =1. an improvement of several orders of magnitude in process performance. we can estimate the estimate the actual fallout as 1350 PPM. Certainly. Clearly.0 can be achieved. as we will see. Clearly. Only in this way will there be enough time develops necessary relationship between tolerances and manufacturing processes. It is also essential to perform manufacturing-process parameter optimization just as one would for design-component parameters. these ratios need to be used and interpreted very carefully. It is a fundamental precept in concurrent engineering to develop technology concepts or product-design concepts in simultaneously with the necessary manufacturing processes to support the timely and economic commercialization of the desired product. so that values less than zero are defined as zero.1 and read the fallout from the one-sided specifications column. and the variability associated with the measurable output from the manufacturing process. and which requires the assumption of normality for its meaning full interpolation is likely to be misused (or abused). if Cpk < -1. Many quality engineering authorities have advised against the routine use of process capability ratios such as Cp and Cpk (or the others discussed later in this section) on the grounds that they are an oversimplification of a complex phenomenon. leading to Cpk = 0. Furthermore.1 (using Cp=2. If a tolerance band is determined without considering a manufacturing process. while Cpk measures actual capability. and table 2. Capable manufacturing processes must be aligned with the product concept as early as possible.

“Statistical Sources and Techniques” by F. Addison Wesley Publication Company.J. Cp is defined arithmetically as follows: Cp = (USL − LSL) 6σ Where. or Cp(lower limit). “Advanced Practical Statistics” by S.M.Wolf. The focus on manufacturing process set-point tolerances should be directed at keeping the component specifications as close to on-target as possible. 6 σ stands for six times the short-term sample standard deviation of the production measure of part quality in engineering units. The true measure of variability most often used in the alteration of cp is 6s(where σ is the sample standard deviation).Leavenworth. References: 1. Cp(upper limit). 4.Design for Manufacturing and Assembly Process Capability Analysis manufactured. USL=Upper specification limit. Cpk.M. the use of σ is really a misapplication of the population parameter for a standard deviation. 2. “Statistical Quality Control”by Eugene L. . 7th edition Mcgraw hill book co. typically expressed as Cp. Richard S.Grant.P. is the ratio of design tolerance boundaries to the measured variability of the manufacturing process output response.Guptha.. 3. “Tolerance Design” by Creveling C.Rendall and D. LSL= Lower specification limit. 33 . The manufacturing process capability index.

Sign up to vote on this title
UsefulNot useful