This preview shows pages 1–3. Sign up to view the full content.
257
T
EACHING
S
UGGESTIONS
Teaching Suggestion 17.1:
Japan’s change in status since WWII.
Remind students that Japan began a few decades ago with perhaps
the world’s worst quality and that “Made in Japan” was synony
mous with shoddy products just 45 years ago.
Teaching Suggestion 17.2:
Four interesting quotes from QC
expert Philip Crosby.
1.
“The cost of quality is the expense of doing things
wrong.”
2.
“There is absolutely no reason for having errors or de
fects in any product or service.”
3.
“If quality isn’t ingrained in the organization, it will
never happen.”
4.
“It is much less expensive to prevent errors than to re
work, scrap, or service them.”
Teaching Suggestion 17.3:
Natural vs. assignable variations.
Random chance
l
“natural”
SpeciFc cause
l
assignable
Teaching Suggestion 17.4:
Mean and range charts.
Mean and range charts tell us what we need to know about the
process. Each plays a necessary role.
A
LTERNATIVE
E
XAMPLES
Alternative Example 17.1:
TwentyFve engine mountings are
sampled each day and found to have an average width of 2 inches,
with a standard deviation of 0.1 inch. To set control limits that in
clude 99.7% of sample means (
Z
5
3),
5
2.06 inches
5
1.94 inches
Alternative Example 17.2:
Several samples of size
n
5
8 have
been taken from today’s production of fencing poles. The average
pole was 3 yards in length and the average sample range was
0.015 yard. We Fnd 99.7% control limits for the process below.
A
2
5
0.373 from Table 17.2
LCL
yards
x
XAR
52
5
2
5
2
3
0 373 0 015
2 994
.(
.).
UCL
yards
x
51
5
1
5
2
3
0 373 0 015
3 006
R
5
0 015
.
yard
X
5
3
yards
LCL
xx
XZ
5 2
s
23
0
1 2
5 20
0
6
(./
)
.
UCL
5 1
s
0
0
6
)
.
Alternative Example 17.3:
The average range of a process is
10 pounds. The sample size is 10. Using Table 17.2,
D
4
5
1.78,
D
3
5
0.22.
Alternative Example 17.4:
Based on samples of 20 IRS audi
tors, each observed handling 100 Fles, we Fnd that the total
number of mistakes made in handling Fles is 220. We set 95.45%
limits on this process below:
100 is the size of each sample
Alternative Example 17.5:
There have been complaints that the
sports page of the Dubuque
Register
has lots of typos. The last six
days have been examined carefully, and the number of typos/page
recorded below. Is the process in control, using
Z
5
2?
LCL
pp
pZ
5
2
5
s
011
2 003
005
.
( )( .
)
.
UCL
5
1
5
s
017
.
( )( .
)
.
s
p
5
2
5
(. )
(
. )
.
011 1 011
100
003
p
55
5
total no. mistakes
total no. files
220
100 20
()
(
)
.
LCL
pounds
R
DR
5
3
022 10
22
( )
.
UCL
pounds
R
5
4
178 10
178
.
All days are in control.
S
OLUTIONS TO
D
ISCUSSION
Q
UESTIONS
AND
P
ROBLEMS
171.
The central limit theorem allows us to use the normal
curve regardless of the distribution of the population we are trying
to control.
LCL
or
c
cc
5
2
2
2 5
2 1 58
0 66
0
.
)
.
(
)
UCL
c
5
5
2
1
5
8
5
6
6
.
)
.
c
15 6
2 5
/.
17
CHAPTER
Statistical Quality Control
Day
Number of Typos
Mon.
2
Tues.
1
Wed.
5
Thurs.
3
Fri.
4
Sat.
0
a
72106 CH17 GGS
3/30/05
2:36 PM
Page 257
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document258
CHAPTER 17
S
TATISTICAL
Q
UALITY
C
ONTROL
172.
The ultimate goal of  and
R
charts is to ascertain, by a
sampling procedure, that the process is kept within speciFed upper
and lower bounds. The combination of
and
R
charts allows one
This is the end of the preview. Sign up
to
access the rest of the document.
 Spring '11
 MichaelHanna

Click to edit the document details