Product Quality
and Robust Design
Wnter 2015
1
What is your definition of a
Quality Product?




Is there a difference between Quality and
Manufacturability?
What are some of the “quality” concerns
about your senior project?
How does product quality tie back into
Functional Specs?
How are you going to validate your critical
design specs?
Wnter 2015
2
Why is Product Quality Important?
There is only one excuse not to invest time in
improving your Product Development
process:
You do not have now and will never have any
competition (i.e., you are in a permanent monopoly
situation), and your customers must, out of necessity
for all time, continue purchasing your good(s) or
service(s) in sufficient quantity that your company can
continue to make adequate profit even in the face of
cost inflation.
Wnter 2015
3
How Do You Select Among
Competing Suppliers?
1. Customer buys on value
2. Value = Quality relative to Price
3. Quality includes all non-price attributes
Product
Customer service
4. Quality, Price and Value are relative
Product
Quality
Value
Wnter 2015
Price
Customer service
4
How Successful is an
Average Launch?
For every 11 serious ideas……
10
IDEAS
Screening
and
Evaluation
3 enter development,
1.3 are launched,
and 1 succeeds
Business
Analysis
5
Development
Testing
Commercialization
0
0
50
% SUCCESS
100
1 Successful
Product
PDMA New Product Development Survey: Performance and Best Practices. Chicago; 1991
Wnter 2015
5
Where do the Winners Invest
Their Time?
500
S
t
a
f
f
D
a
y
s
400
300
200
100
0
s
t
n
h
is
ll
is
st
is
g
st
ss
dy
es
en
tio
nc
ys
Se
ys
Te
Te
se
l
lys
l
tu
nin As s
c
u
l
r
m
s
t
a
a
a
S
e
e
c
n
ia
n
A
t
La
du
op
re
An
et
Tr
du
ke
ro
lA
om
ch
et
e A ev el
n
k
r
t
o
Sc
a
P
r
k
e
c
r
i
o
a
s
l
r
a
T
c
P
ti
M
D
an
ial
t ia
Cu
er
Ma
uc
rn
e-M elim
ct
Tr
m
d
e
Ini
Fin
u
t
o
r
Pr
m
In
P
od
Pr
Co
Pr
Wnter 2015
Winners
Losers
6
Quality Lever
1000:1
PRODUCT
DEFINITION
100:1
PRODUCT
DESIGN
10:1
PROCESS
DESIGN
1:1
PRODUCTION
PRODUCT
IMPROVEMENT
LOW VISIBILITY
HIGH RETURNS
HIGH VISIBILITY
LOW RETURNS
When do you add Quality in the Design Process?
Wnter 2015
7
In the last 50 years, there has been a dramatic
improvement in product quality.
Before that time, much of product quality was in the ease
of repair.
Examples?
•How many TV Repairs shops exist now vs. 1960?
•How many auto-mechanics shops now exist?
•How many of you have ever had a wrist watch repaired?
•What has changed?
•How do insure your product has high quality?
Wnter 2015
8
Escaping Defects
You can try and inspect for quality
Total Defects/Unit
No matter how good your inspection and test processes are, the
more defects you create, the more defects escape to the customer
Wnter 2015
9
The Inspection Exercise
Count the number of times the 6th letter of the alphabet
appears in the following text. You have 45 Seconds to
complete your inspection.
The necessity of training farm hands for the first
class farms in the fatherly handling of farm live stock is
foremost in the eyes of the farm owners. Since the
forefathers of the farm owners trained the farm hands
for first class farms in the fatherly handling of farm live
stock, the farm owners felt they should carry on with
the family tradition of training farm hands of the first
class farmers in the fatherly handling of farm live stock
because they believe it is the basis of good
fundamental farm management.
Wnter 2015
10
What is your answer?
36
What does this tell you
about trying to inspect in
quality?
Wnter 2015
11
The Inspection Exercise
The necessity of training farm hands for the first
class farms in the fatherly handling of farm live stock is
foremost in the eyes of the farm owners. Since the
forefathers of the farm owners trained the farm hands
for first class farms in the fatherly handling of farm live
stock, the farm owners felt they should carry on with
the family tradition of training farm hands of the first
class farmers in the fatherly handling of farm live stock
because they believe it is the basis of good
fundamental farm management.
Wnter 2015
12
The effect of designed-in quality
s
2
3
4
5
6
Process
Capability
Wnter 2015
PPM
308,537
66,807
6,210
233
3.4
Defects per
Million Opp.
13
A 3s Process means that you will have 93.6% good
units coming out of the process. That sounds pretty
good until you think about the consequences of the
failures.
What types of products can be successful with 95%
quality?
What products require higher quality?
Would you ride on an airplane that had 95% quality?
Wnter 2015
14
What Causes Defects?

Excess variation due to:



Manufacturing processes and measurement systems
Supplied material variation
Unreasonably tight specifications (tighter than the
customer requires for a robust design)
Inadequate Design
Margin
Unstable
Parts and
Materials
Wnter 2015
Insufficient
Process
Capability
15
Dissecting Design Process Capability
Inadequate
Measurement
Capability
Inadequate
Functional
Capability
Supplied
Material
Variation
LSL
USL
Design Overly
Sensitive to
“Noise”
Poor Support
of Robust
Product
Functionality
Way off
Target
Wnter 2015
Process Capability
Out of
Spec.
16
A Three Sigma Process
Lowe r Spe c
35
Upper Spec
45
55
Process Spread
3s
Tolerance
3s
Wnter 2015
65
17
A Six Sigma Process
Lower Spec
20
Process Spread
Tolerance
Wnter 2015
Upper Spec
30
40
50
60
70
80
3s
6s
18
Ford (Batavia Plant)



Produced Ford-Designed Transaxle
Built in Batavia, OH and in Japan (Mazda)
Both met engineering specifications but……
 Fleet owners stated that some Escorts shifted
smoothly while others were not so good
(based on small sample size).
 Analysis:


Wnter 2015
All “Smooth” shifting Transaxles were built in
Japan
Both met specs, but…….
19
Batavia
Process Capability Analysis for Batavia
LSL
Process Data
USL
ST
LT
30.0000
Target
*
LSL
20.0000
Mean
25.0521
Sample N
USL
500
StDev (ST)
1.21562
StDev (LT)
1.20660
Potential (ST) Capability
Cp
1.37
CPU
1.36
CPL
1.39
Cpk
1.36
Cpm
*
Overall (LT) Capability
Wnter 2015
22
23
Observed Performance
24
25
26
27
Expected ST Performance
28
Expected LT Performance
Pp
1.38
PPM < LSL
0.00
PPM < LSL
16.19
PPM < LSL
14.13
PPU
1.37
PPM > USL
0.00
PPM > USL
23.48
PPM > USL
20.60
PPL
1.40
PPM Total
0.00
PPM Total
39.67
PPM Total
34.73
Ppk
1.37
20
Mazda - Japan
Process Capability Analysis for Mazda
LSL
Process Data
USL
ST
LT
30.0000
Target
*
LSL
20.0000
Mean
24.9810
Sample N
USL
500
StDev (ST)
0.442690
StDev (LT)
0.444599
Potential (ST) Capability
Cp
3.76
CPU
3.78
CPL
3.75
Cpk
3.75
Cpm
*
Overall (LT) Capability
Wnter 2015
23.524.0 24.525.025.5 26.026.5
Observed Performance
Expected ST Performance
Expected LT Performance
Pp
3.75
PPM < LSL
0.00
PPM < LSL
0.00
PPM < LSL
0.00
PPU
3.76
PPM > USL
0.00
PPM > USL
0.00
PPM > USL
0.00
PPL
3.73
PPM Total
0.00
PPM Total
0.00
PPM Total
0.00
Ppk
3.73
21
Robust Design
How do components effect your design?
V+
1K Ώ
1K Ώ
Vout
Vin
What is the Gain of this circuit?
100 Ώ
Wnter 2015
100 Ώ
22
Robust Design
What about Resistor tolerances?
If you specify 10% resistors what range of gain should you expect to see?
.9K< 1K < 1.1K
and
90 > 100 > 110
1.1 / .09 = 12.2 : 12.2 * 12.2 = 150
.9 / 110 = 8.18 ; 8.18 x 8.18 = 66.9
Therefore your gain could vary from 66.9 – 150;
2.24 to 1
How would you reduce the process variation?
Wnter 2015
23
How do you improve Quality?
•The Design
•The Components
•The Process
•It is difficult to test-in Quality. It is also very
expensive. What do you do with failed units?
•It usually costs at least twice as much to repair
failed units. Why is this true?
•What are some of the things you are going in your
designs to make them more robust?
Wnter 2015
24
Homework assignment



Develop a test plan for at least one of the critical
specifications of your senior project. Describe why you
chose these specs. Describe the test, including test
equipment involved, measurement technique, and
expected results. This plan should also be included in
the final report with actual numbers from the test
process. Due next Tuesday, Mar. 3rd.
Get your website or alternate team management site up
to date, so that I can review it next week.
Each team leader needs to send me a valid URL and
password if required to get on your site. Send me the
URL information by end of day on Friday.
Wnter 2015
25
Discussion topic for next
lecture—email me with answer

1.
2.
3.
Vote as a team on the following:
Which would you choose? A sure $100, or a
coin flip for $200 or nothing.
Which would you now choose? A sure loss
of $100, or a coin flip for a loss of $200 or
$0?
What is the difference in the two scenarios?
Wnter 2015
26