copyright © cengage learning. all rights reserved. 16 quality control methods

46
Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

Upload: rhoda-jones

Post on 18-Jan-2018

216 views

Category:

Documents


0 download

DESCRIPTION

3 A defect of the traditional X chart is its inability to detect a relatively small change in a process mean. This is largely a consequence of the fact that whether a process is judged out of control at a particular time depends only on the sample at that time, and not on the past history of the process. Cumulative sum (CUSUM) control charts and procedures have been designed to remedy this defect.

TRANSCRIPT

Page 1: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

Copyright © Cengage Learning. All rights reserved.

16 Quality Control Methods

Page 2: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

Copyright © Cengage Learning. All rights reserved.

16.5 CUSUM Procedures

Page 3: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

3

CUSUM ProceduresA defect of the traditional X chart is its inability to detect a relatively small change in a process mean.

This is largely a consequence of the fact that whether a process is judged out of control at a particular time depends only on the sample at that time, and not on the past history of the process.

Cumulative sum (CUSUM) control charts and procedures have been designed to remedy this defect.

Page 4: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

4

CUSUM ProceduresThere are two equivalent versions of a CUSUM procedure for a process mean, one graphical and the other computational.

The computational version is used almost exclusively in practice, but the logic behind the procedure is most easily grasped by first considering the graphical form.

Page 5: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

5

The V-Mask

Page 6: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

6

The V-MaskLet 0 denote a target value or goal for the process mean, and define cumulative sums by

S1 = x1 – 0

S2 = (x1 – 0) + (x2 – 0) = (xi – 0)

. . . S1 = (x1 – 0) + . . . + (x1 – 0) = (x1 – 0)

(in the absence of a target value, x is used in place of 0).

Page 7: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

7

The V-MaskThese cumulative sums are plotted over time. That is, at time l, we plot a point at height Sl.

At the current time point r, the plotted points are (1, S1), (2, S2), (3, S3),…, (r, Sr).

Page 8: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

8

The V-MaskNow a V-shaped “mask” is superimposed on the plot, as shown in Figure 16.7.

Figure 16.7

(b)

CUSUM plots: (a) successive points (I, Sl) in a CUSUM plot; (b) a V-mask with 0 = (r, Sr);

(a)

Page 9: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

9

The V-Mask

Figure 16.7

(d)

CUSUM plots: (c) an in-control process; (d) an out-of-control process

(c)

Page 10: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

10

The V-MaskThe point 0, which lies a distance d behind the point at which the two arms of the mask intersect, is positioned at the current CUSUM point (r, Sr).

At time r, the process is judged out of control if any of the plotted points lies outside the V-mask—either above the upper arm or below the lower arm.

When the process is in control, the xi’ will vary around the target value 0, so successive Si’s should vary around 0. Suppose, however, that at a certain time, the process mean shifts to a value larger than the target.

Page 11: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

11

The V-MaskFrom that point on, differences xi’ – 0 will tend to be positive, so that successive Sl’s will increase and plotted points will drift upward.

If a shift has occurred prior to the current time point r, there is a good chance that (r, Sr) will be substantially higher than some other points in the plot, in which case these other points will be below the lower arm of the mask.

Similarly, a shift to a value smaller than the target will subsequently result in points above the upper arm of the mask.

Page 12: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

12

Any particular V-mask is determined by specifying the “lead distance” d and “half-angle” , or, equivalently, by specifying d and the length h of the vertical line segment from 0 to the lower (or to the upper) arm of the mask.

One method for deciding which mask to use involves specifying the size of a shift in the process mean that is of particular concern to an investigator.

Then the parameters of the mask are chosen to give desired values of and , the false-alarm probability and the probability of not detecting the specified shift, respectively.

The V-Mask

Page 13: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

13

The V-MaskAn alternative method involves selecting the mask that yields specified values of the ARL (average run length) both for an in-control process and for a process in which the mean has shifted by a designated amount.

After developing the computational form of the CUSUM procedure, we will illustrate the second method of construction.

Page 14: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

14

Example 8A wood products company manufactures charcoal briquettes for barbecues. It packages these briquettes in bags of various sizes, the largest of which is supposed to contain 40 lbs. Table 16.4 displays the weights of bags from 16 different samples, each of size n = 4.

Table 16.4Observations, x’s and Cumulative Sums for Example 8

Page 15: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

15

Example 8The first 10 of these were drawn from a normal distribution with = 0 = 40 and = .5. Starting with the eleventh sample, the mean has shifted upward to = 40.3.

Figure 16.8 displays an x chart with control limits0 3 x = 40 3 (.5/ ) = 40 .75

Figure 16.8

X control chart for the data of Example 8

cont’d

Page 16: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

16

Example 8No point on the chart lies outside the control limits. This chart suggests a stable process for which the mean has remained on target.

Figure 16.9 shows CUSUM plots with a particular V-mask superimposed.

cont’d

Page 17: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

17

Example 8The plot in Figure 16.9(a) is for current time r = 12. All points in this plot lie inside the arms of the mask. However, the plot for r = 13 displayed in Figure 16.9(b) gives an out-of-control signal.

(a) (b)

Figure 16.9

CUSUM plots and V-masks for data of Example 8: (a) V-mask at time r = 12, process in control; (b) V-mask at time r = 13, out-of-control signal

cont’d

Page 18: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

18

Example 8The point falling below the lower arm of the mask suggests an increase in the value of the process mean.

The mask at r = 16 is even more emphatic in its out-of-control message. This is in marked contrast to the X chart.

cont’d

Page 19: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

19

A Computational Version

Page 20: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

20

A Computational VersionThe following computational form of the CUSUM procedure is equivalent to the previous graphical description.

Let d0 = e0 = 0, and calculate d1, d2, d3, . . . , and e1, e2, e3, . . . , recursively, using the relationships

dl = max[0, dl – 1 + (xl – (0 + k))]

el = max[0, el – 1 – (xl – (0 – k))] (l = 1, 2, 3, . . .)

Here the symbol k denotes the slope of the lower arm of the V-mask, and its value is customarily taken as /2 (where is the size of a shift in on which attention is focused).

Page 21: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

21

A Computational VersionIf at current time r either dr > h or er > h, the process is judged to be out of control.

The first inequality suggests that the process mean has shifted to a value greater than the target, whereas er > h indicates a shift to a smaller value.

Page 22: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

22

Example 9Reconsider the charcoal briquette data displayed in Table 16.4 of Example 8.

Table 16.4Observations, X’s and Cumulative Sums for Example 8

Page 23: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

23

Example 9The target value is 0 = 40, and the size of a shift to be quickly detected is = .3. Thus

k = = .15 0 + k = 40.15 0 – k = 39.85

so

dl = max[0, dl – 1 + (xl – 40.15)]

el = max[0, el – 1 – (xl – 39.85)]

cont’d

Page 24: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

24

Example 9Calculations of the first few dl’s proceeds as follows:

d0 = 0

d1 = max[0, d0 + (x1 – 40.15)]

= max[0, 0 + (40.20 – 40.15)]

= .05

cont’d

Page 25: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

25

Example 9 d2 = max[0, d1 + (x2 – 40.15)]

= max[0, .05 + (39.72 – 40.15)]

= 0

d3 = max[0, d2 + (x3 – 40.15)]

= max[0, 0 + (40.42 – 40.15)]

= .27

cont’d

Page 26: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

26

Example 9The remaining calculations are summarized in Table 16.5.

Table 16.5

CUSUM Calculations for Example 9

cont’d

Page 27: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

27

Example 9The value h = .95 gives a CUSUM procedure with desirable properties—false alarms (incorrect out-of-control signals) rarely occur, yet a shift of = .3 will usually be detected rather quickly.

With this value of h, the first out-of-control signal comes after the 13th sample is available. Since d13 = 1.17 > .95, it appears that the mean has shifted to a value larger than the target.

cont’d

Page 28: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

28

Example 9This is the same message as the one given by the V-mask in Figure 16.9(b).

(b)

Figure 16.9

CUSUM plots and V-masks for data of Example 8: (b) V-mask at time r = 13, out-of-control signal

cont’d

Page 29: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

29

A Computational VersionTo demonstrate equivalence, again let r denote the current time point, so that x1, x2 . . . , xr are available. Figure 16.10 displays a V-mask with the point labeled 0 at (r, Sr).

Figure 16.10

A V-mask with slope of lower arm = k

Page 30: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

30

A Computational VersionThe slope of the lower arm, which we denote by k, is h/d. Thus the points on the lower arm above r, r – 1, r – 2,. . . are at heights Sr – h, Sr – h – k , Sr – h, Sr – h – 2k and so on.

The process is in control if all points are on or between the arms of the mask. We wish to describe this condition algebraically.

To do so, let

Tl = [ xi – (0 + k)] l = 1, 2, 3, . . . , r

Page 31: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

31

A Computational VersionThe conditions under which all points are on or above the lower arm are

Sr – h Sr (trivially satisfied) i.e., Sr Sr + h

Sr – h – k Sr – 1 i.e., Sr Sr – 1 + h + k

Sr – h – 2k Sr – 2 i.e., Sr Sr – 2 + h + 2k

. . . . . .

Page 32: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

32

A Computational VersionNow subtract rk from both sides of each inequality to obtain

Sr – rk Sr – rk + h i.e., Tr Tr + h

Sr – rk Sr – 1 – (r – 1)k + h i.e., Tr Tr – 1 + h

Sr – rk Sr – 2 – (r – 2)k + h i.e., Tr Tr – 2 + h . . . . . .

Page 33: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

33

A Computational VersionThus all plotted points lie on or above the lower arm if and only if (iff) Tr – Tr h, Tr – Tr – 1 h, Tr – Tr – 2 h, and so on. This is equivalent to

Tr – min(T1, T2, . . . , Tr) h

In a similar manner, if we let

Vr = [xi – (0 – k)] = Sr + rk

Page 34: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

34

A Computational Versionit can be shown that all points lie on or below the upper arm iff

max(V1, . . , Vr) – Vr h

If we now let

dr = Tr – min(T1, . . . , Tr)

er = max(V1, . . . , Vr) – Vr

Page 35: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

35

A Computational Versionit is easily seen that d1, d2, . . . , and e1, e2, . . .can be calculated recursively as illustrated previously. For example, the expression for dr follows from consideration of two cases:

1. min(T1 , . . . , Tr) = Tr , whence dr = 0

2. min(T1 , . . . , Tr) = min(T1 , . . . , Tr – 1), so that

dr = Tr – min(T1, . . . , Tr – 1)

Page 36: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

36

A Computational Version = xr – (0 + k) + Tr – 1 – min(T1, . . ., Tr – 1)

= xr – (0 + k) + dr – 1

Since dr cannot be negative, it is the larger of these two quantities.

Page 37: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

37

Designing a CUSUM Procedure

Page 38: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

38

Designing a CUSUM ProcedureLet denote the size of a shift in that is to be quickly detected using a CUSUM procedure. It is common practice to let k = /2.

Now suppose a quality control practitioner specifies desired values of two average run lengths:1. ARL when the process is in control ( = 0)

2. ARL when the process is out of control because the mean has shifted by ( = 0 + or = 0 – )

Page 39: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

39

Designing a CUSUM ProcedureA chart developed by Kenneth Kemp (“The Use of Cumulative Sums for Sampling Inspection Schemes,” Applied Statistics, 1962: 23), called a nomogram, can then be used to determine values of h and n that achieve the specified ARLs.

This chart is shown as Figure 16.11.

Figure 16.11

The Kemp nomogram

Page 40: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

40

Designing a CUSUM ProcedureThe method for using the chart is described in the accompanying box.

Either the value of must be known or an estimate is used in its place.

Page 41: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

41

Designing a CUSUM ProcedureUsing the Kemp Nomogram

1. Locate the desired ARLs on the in-control and out-of-control scales. Connect these two points with a line.

2. Note where the line crosses the k scale, and solve for n using the equation

Then round n up to the nearest integer.

Page 42: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

42

Designing a CUSUM Procedure3. Connect the point on the k scale with the point on the in-control ARL scale using a second line, and note where

this line crosses the h scale. Then h = ( / ) h.

The value h = .95 was used in Example 9. In that situation, it follows that the in-control ARL is 500 and the out-of-control ARL (for = .3) is 7.

Page 43: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

43

Example 10The target value for the diameter of the interior core of a hydraulic pump is 2.250 in.

If the standard deviation of the core diameter is = .004, what CUSUM procedure will yield an in-control ARL of 500 and an ARL of 5 when the mean core diameter shifts by the amount of .003 in.?

Page 44: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

44

Example 10

Figure 16.11

The Kemp nomogram*

cont’d

Connecting the point 500 on the in-control ARL scale to the point 5 on the out of-control ARL scale and extending the line to the k scale on the far left in Figure 16.11 gives k = .74.

Thus

Page 45: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

45

Example 10So

The CUSUM procedure should therefore be based on the sample size n = 4.

cont’d

Page 46: Copyright © Cengage Learning. All rights reserved. 16 Quality Control Methods

46

Example 10Now connecting .74 on the k scale to 500 on the in-control ARL scale gives h = 3.2, from which

h = (/ ) (3.2)

= (.004 / )(3.2)

= .0064

An out-of-control signal results as soon as either dr > .0064 or er > 0064.

cont’d