Q8 IM13CFinal

You might also like

Download as doc, pdf, or txt
Download as doc, pdf, or txt
You are on page 1of 21

This is file Q8IM13C.DOC - The third of 5 files for solutions to this chapters problems.

13.

For each of the following control charts, assume that the process has been operating in
statistical control for some time. What conclusions should the operators reach at this
point?

Chapter 13 - Statistical Process Control

Ans
13.

a) The first control chart shows an out-of-control process with a definite


downward trend. The last 4 out of 5 points are below one standard
error away from the mean. The process needs adjustment upward.
b) The second control chart shows an out of control condition, with the
first eight points above the centerline. Then there appears to be a
sudden shift in the process average, putting the next six points below
the centerline. It is possible that the process is being over-adjusted.
It needs to be centered and then watched for out-of-control indications
with no unnecessary operator intervention.
c) The third control chart shows the data hugging the centerline,
indicating that the process is possibly out of control. If the process has
multiple machines or operators, a control chart should be constructed
for each machine to avoid "masking" the variation brought on by
mixing data from several sources.
d) The fourth control chart shows a process that is stable and in
control.

Chapter 13 - Statistical Process Control

e) The fifth control chart shows an out of control condition, with a point
above the upper control limit.
f) The sixth control chart shows that seven out of eight of the most
recent points are below the centerline, indicating that the process is
out of control.
g) The last control chart shows too many points close to the upper and
lower control limits, indicating an out of control condition.
14.

Discuss the interpretation of each of the following control charts:

Chapter 13 - Statistical Process Control

Ans
14.

a) Two points outside upper control limit.


b) Process is in control
c) Mean shift upward in second half of control chart.
d) Points hugging upper and lower control limits.

15.

PCDrives has a manufacturing process that is normally distributed and has the sample
means and ranges for fifteen samples of size 5, found in the worksheet Prob. 13-15. Note
that only sample statistics have been given, instead of the raw data from the samples.
Determine process capability limits. If specifications are determined to be 70 25, what
percentage will be out of specification?

Answer
15.

(See spreadsheet Prob.13-15CP-PCD.xls for details.)


For the Center Lines, CL x : x = 68.567; CLR: R = 21.92
Control limits for the x - chart are:
x

A2 R = 68.567 0.577 (21.92) = 55.92 to 81.22

For the R-chart: UCLR = D4 R = 2.114 (21.92) = 46.34

Chapter 13 - Statistical Process Control

LCLR = D3 R = 0
The limits above apply to sample groups of 5 items each.
Estimated = R / d2 = 21.92 / 2.326 = 9.42
The problem asks that students perform a process capability analysis.
This is only justified if the process is in control. The fact that the
process is thought to be normally distributed does not establish that it
is in control. The x - chart shows that the process is, in fact, out of
control, because of 4 out of 5 samples within samples 2-6 are more
than one standard error away from the center-line. The % outside
calculation can be performed as follows. Note the warning given below
about the assumption that the process is in control.

Percent outside Specification Limits (45 to 95)


% Below LSL: z = LSL

z = 45 - 68.567 = - 2.50 ; P( z < -2.50) = (0.5 - 0.4938) = 0. 0062


that
9.42
% Above USL: z = USL

items will exceed lower limit


x

z = 95 - 68.567 = 2.81 ; P( z > 2.81) = (0.5 - 0.4975) = 0.0025 that


9.42
items will exceed upper limit
Therefore, the percent outside is calculated as: 0.87 %
Although the % outside calculations seem to show that the process has
a relatively small % outside specifications, it should be noted that the
x-bar chart shows that the process is not in control. Hence, the %
outside calculation is going to generate questionable results.

Chapter 13 - Statistical Process Control

Charts showing out-of-control conditions

16.

Suppose that, in a review of the J. Phelps Swim Club data (Prob. 13-03 worksheet), it
was found that the values for the second sample had been recorded incorrectly and should

Chapter 13 - Statistical Process Control

have been 1.26, 1.22 and 1.21. In addition, the last value of the last sample was also
recorded incorrectly. It should have been 1.25. After the two samples were revised, the
process was considered under control, with an upper specification limit of USL = 1.5, and
the lower specification limit at LSL = 1.0. Compute the process capability and the percent
outside of specifications.
Answer
16.

With data from Problem 3 and USL = 1.5, and LSL = 1.0, from the
Prob.13-16CP.xls spreadsheet we see:
Upper specification 1.50
Lower specification 1.00
Nominal
1.25
specification

Cp
Cpl
Cpu
Cpk

For the Center Lines, CL x :

1.778
1.823
1.733
1.733

= 1.240; CLR: R = 0.079

Note that the spreadsheet uses an estimated standard deviation of:


Estimated = R / d2 = 0.079 / 1.693 = 0.047
From this, we obtain:
Percent outside Specification Limits (1.0 to 1.5)
% Below LSL: z = LSL

z = 1.00 1.240 = - 5.11 ; P( z < -5.11) = (0.5 - 0.4999---*) =


0.0000, for all practical
0.047
purposes, that items will
exceed lower limit
*Note: This value is outside the table limits of the Appendix A table
that only extends to z = 3.09.

Chapter 13 - Statistical Process Control


% Above USL: z = USL

z = 1.5 1.240 = 5.53 ; P( z > 5.53) = (0.5 - 0.4999--) = 0.0000 for


all practical
0.047
purposes, that items will
exceed lower limit
Therefore, the percent outside is calculated as: 0.0 %
These calculations show that the process has a relatively small %
outside specifications. In this problem, one point that showed an
assignable cause was eliminated, so the process should be in control.
The process still needs a small amount of fine tuning in order to
become more capable (ideally the Cp should be 2.0, or better) as
shown by the capability indexes.
The control charts show the following patterns:

Chapter 13 - Statistical Process Control

17.

Suppose that in Prob. 13-08 (part B) for the Hertz Company, after it was revised, the
upper specification limit was set at USL = 475, and the lower specification limit was set
at LSL = 325. Compute the process capability and the percent outside specifications that
may be expected.

Answer
17.

With data from problem 8 and using USL = 475, LSL = 325, from the
Prob.13-08XR.xls (part b) spreadsheet we see process capability
values, as follows:
Upper specification 475
Lower specification
325
Nominal
specification
400

Cp

1.367

Cpl

1.362

Cpu
Cpk

1.373
1.362

Chapter 13 - Statistical Process Control


For the Center Lines, CL x :

10

= 400.29; CLR: R = 30.96

Note that the spreadsheet uses an estimated standard deviation of:


Estimated = R / d2 = 30.96 / 1.693 = 18.29
From this, we obtain:
Percent outside Specification Limits (325 to 475)
% Below LSL: z = LSL

z = 325 400.29 = - 2.43 ; P( z < -2.43) = (0.5 - 0.4925) = 0.0075


that
30.96
% Above USL: z = USL

items will exceed lower limit


x

z = 475 400.29 = 2.41 ; P( z > 2.41) = (0.5 - 0.4920) = 0.0080 that


30.96
items will exceed upper limit
Therefore, the total percent outside is calculated as: 1.55 %
These calculations show that the process has a relatively small %
outside specifications. In problem 8b, points that showed assignable
causes were eliminated, so the process should be in control. The
process still needs some fine tuning in order to become more
capable (ideally the Cp should be 2.0, or better) as shown by the %
outside calculation and the capability indexes.
The control charts were presented in problem Prob.13-08CP.xls (Tab b
in the spreadsheet).
18.

Fujiyama Electronics needs to construct x - and s-charts for circuit boards that are
purchased from an outside supplier. A critical dimension is the distance between two
holes on the board that are supposed to be 5 cm. apart. Use the data, consisting of 30
samples of size 4 found in the worksheet Prob. 13-18.

Chapter 13 - Statistical Process Control

11

a. Compute the mean and standard deviation of each sample, calculate control limits, and
plot them on control charts.
b. Does the process appear to be in statistical control? Why or why not? What
recommendation would you make to management concerning the process?
Answer
18.

See data and control charts below and spreadsheet Prob.13-18XS.xls for
details.
For the Fujiyama, the center line, CL x :

= 5.031; CLs : s = 0.468

Control limits for the x - s charts are:


x

A3 s = 5.031 1.63 (0.468) = 4.27 to 5.79

For the s-chart: UCLs = B4 s = 2.27 (0.468) = 1.06


LCLs = B3 s = 0

Chapter 13 - Statistical Process Control

12

Chapter 13 - Statistical Process Control

13

The process appears to be under control. Management should be advised to continue to


monitor the process to ensure control.
19.

El Toro Grande Restaurante advertises that customers will have their orders taken within
3 minutes after being seated. Management wants to monitor average times, since it is
such an important guarantee for business. Construct x - and s-charts for the data given in
the worksheet Prob. 13-19.
a. Compute the mean and standard deviation of each sample, calculate control limits, and
plot them on control charts.
b. Does the process appear to be in statistical control? Why or why not?
c. Calculate the process capability statistics, using 3 minutes as the upper tolerance limit
and zero as the lower tolerance limit. What recommendation would you make to
management concerning the process, based on these findings?

Answer

Chapter 13 - Statistical Process Control

19.

14

See data and control charts below and spreadsheet Prob.13-19XS.xls for
details.
For the El Toro Grande Restaurante, the center line, CL x : x = 1.518; CLs : s
= 0.308
a. Control limits for the x - s charts are:
x

A3 s = 1.518 1.43 (0.308) = 1.08 to 1.96

For the s-chart: UCLs = B4 s = 2.09 (0.308) = 0.64


LCLs = B3 s = 0
b. The process appears to be in statistical control, because values are distributed
randomly about the mean, all values lie within control limits, and there are no unusual
patterns on the x - s charts.

Chapter 13 - Statistical Process Control

15

c. The process appears to be capable.


Note that the spreadsheet uses an actual standard deviation of =
0.330 calculated from all of the sample values. This is not the same
statistic as s .
Process Capability
Upper specification
3.00
Lower specification
0.00

Cp
Cpu
Cpl
Cpk

1.52
1.50
1.53
1.50

Management should be advised that, although the capability is good, they should work to
reduce variability even more, so as to approach a Cp = 2.0.
20.

Construct x - and s-charts for the data for the Hertz Company from Prob. 13-08 (part B)
after the data was revised. Is the process in control?

Answer
20.

See data and control charts below and spreadsheet Prob.1320XSHert.xls for details.

Chapter 13 - Statistical Process Control

16

For the Hertz Co. using data from problem 13-08, the center line, CL x :
x = 400.290;
CLs : s = 16.404
Control limits for the x - s charts are:
x

A3 s = 400.290 1.95 (16.404) = 367.81 to 432.77

For the s-chart: UCLs = B4 s = 2.57 (16.404) = 41.22


LCLs = B3 s = 0
Since the revised data from Prob. 13-08 (b) with 23 samples was used, the process is
under control, with no apparent problems.

Chapter 13 - Statistical Process Control

21.

17

Construct x - and s-charts for the data for Babbage Chips from Prob. 13-12. Is the
process in control? What recommendation would you make to management concerning
the process, based on your analysis?

Answer
21.

See data and control charts below and spreadsheet Prob.1321XSBabb.xls for details. Using the data from Babbage Chips, Prob.1312, we see:
For the center line, CL x :

= 9.170; CLs : s = 1.046

Control limits for the x - s charts are:


x

A3 s = 9.170 1.427 (1.046) = 7.677 to 10.662

For the s-chart: UCLs = B4 s = 2.089 (1.046) = 2.185

Chapter 13 - Statistical Process Control

18

LCLs = B3 s = 0
As in Prob. 13-12a, results from 30 samples of 5 show that both the x
and s charts are apparently in control. However, on the x chart,
samples 12-23 have an unusual pattern. Although it may not be
statistically significant, these points could be seen as having 7 or 8
points hugging the centerline. This pattern bears watching to see if it
recurs.

Chapter 13 - Statistical Process Control

22.

19

An injection molding machine at the Moby Molding Co. used to make plastic bottles has
four molding heads. The outside diameter of the bottle is an important measure of process
performance. The table in the worksheet Prob.13-22 shows the results of 30 samples in
which the data are coded by subtracting the actual value from the nominal dimension.
Construct x - and s-charts and discuss the results.

Answer
22.

See data and control charts below and spreadsheet Prob.1322XSMoby.xls for details on the Moby Molding Co.s plastic molding
process.
a) For the Center Line, CL x :

= 0.0093; CLs : s = 0.0466

Control limits for the x - s charts are:


x

A3 s = 0.0093 1.628 (0466) = -0.0666 to 0.0852

For the s-chart: UCLs = B4 s = 2.266 (0.0466) = 0.1056


LCLs = B3 s = 0

Chapter 13 - Statistical Process Control

20

b) We can see from the x - chart (and possibly the s-chart) that the
process is "hugging the centerline" creating an out of control condition
on the means and their standard deviations. The cause for this condition
may be judged from the structure of the data. It appears that each of the
heads on the molding machine has a separate distribution of data. Thus,
control charts should be prepared for each head, rather than treating the
data as if it came from the same population.

Chapter 13 - Statistical Process Control

21

See file Q8IM13D.DOC - The next file for more solutions to end-ofchapter problems for Chapter 13.

You might also like