Lecture Outline
Meaning of Quality
Quality Management: Total Quality Management
Cost of Quality
ISO 9000
Statistical Process Control (SPC)
Control Charts
1
7/25/2019
Conformance Aesthetics
degree to which a product meets pre–established standards how a product looks, feels, sounds, smells, or tastes?
Durability Safety
how long product lasts before replacement? assurance that customer will not suffer injury or harm from a
product; an especially important consideration for automobiles
Serviceability
Perceptions
ease of getting repairs, speed of repairs, courtesy and competence
of repair person subjective perceptions based on brand name, advertising etc.
Dimensions of Quality:
Dimensions of Quality: Service
Service (cont.)
Consistency
Completeness: is same level of service provided to each customer each time?
is everything customer asked for provided? is your newspaper delivered on time every morning?
is a mail order from a catalogue company complete when
delivered?
2
7/25/2019
Quality assurance (QA) is any systematic A management approach to long-term success through customer
satisfaction.
process of determining whether a product or In a TQM effort, all members of an organization participate in improving
service meets specified requirements. processes, products, services, and the culture in which they work.
Terms quality assurance and quality control Total emphasizes that departments in addition to production are obligated
to improve their operations; management emphasizes that executives are
(QC) are not the same concept. obligated to actively manage quality through funding, training, staffing,
and goal setting.
QA provides the overall guidelines used
anywhere, and QC is a production-focused
TQM enjoyed widespread attention during the late 1980s and early 1990s
process – for things such as inspections. before being overshadowed by ISO 9000, Lean manufacturing, and Six
Sigma.
3
7/25/2019
Information on QFD….
Quality Function Deployment (QFD)
4
7/25/2019
Kano’s Model
5
7/25/2019
Who
Evaluation
This Product
Now
Who vs.
Whats
Whats
(Finding needs of customer)
Now vs
Whats vs
What
Hows
House of Quality
Benefits of QFD
5
Importance
Trade-off matrix
Promotes better understanding of
3 customer demands
Design
characteristics Promotes better understanding of design
interactions
1 4 2
Involves manufacturing in design process
Customer Relationship Competitive
requirements matrix assessment Breaks down barriers between functions
and departments
Provides documentation of design process
6 Target values
Meaning of Quality
Evolution of Quality Management
Meaning of Quality
Fitness for
Consumer Use
6
7/25/2019
Walter Shewart
In 1920s, developed control charts Joseph M. Juran
Introduced term “quality assurance” Followed Deming to Japan in 1954
Focused on strategic quality planning
W. Edwards Deming
Developed courses during World War II to teach statistical quality- Armand V. Feigenbaum
control techniques to engineers and executives of companies that
were military suppliers In 1951, introduced concepts of total quality control and continuous
After war, began teaching statistical quality control to Japanese quality improvement
companies
Deming’s 14 Points
Deming’s 14 Points (cont.)
7
7/25/2019
Quality Circle
Deming Wheel: PDCA Cycle
Organization
8-10 members
Same area
Supervisor/moderator
Training
Presentation Group processes
Implementation Data collection
4. Act 1. Plan Monitoring Problem analysis
Institutionalize Identify
improvement; problem and
continue cycle. develop plan
for
improvement.
3.4 defects per million opportunities (1000000) Requires the use of certain tools and techniques
8
7/25/2019
9
7/25/2019
Random Non-Random
Statistical Process Control
(SPC) common causes special causes
monitoring production process Depends on equipment, Equipment that is out of
UCL
to detect and prevent poor engineering, operator adjustment, defective
quality and system used for materials, change in parts or
materials, broken machinery,
Sample measurement operator fatigue, poor work
subset of items produced to Natural occurrence method, error due to lack of
LCL training.
use for inspection inherent in a process
Control Charts due to identifiable factors
can be eliminated only
can be modified through
process is within statistical through improvements operator or management
control limits in the system action
10
7/25/2019
1 2 3 4 5 6 7 8 9 10
Sample number
The formula for conducting upper and lower limit in control charts
are based on a number of standard deviation ‘z’ from the process
average ( center line) according to the normal distribution.
Occasionally it is 2 but Most frequently it is 3
Z=2 corresponds to an overall probability of 95%
Z=3 corresponds to a normal probability of 99.74%
95%
99.74%
1. … no sample points outside limits Are used for continuous variables that
can be measured, such as
2. … most points near process average
Weight
3. … about equal number of points above
and below centerline Volume
4. … points appear randomly distributed
11
7/25/2019
The x bar chart is based on the normal If the std dev is not known, then a
distribution. It can be constructed in two ways. table of values based on sample
If the standard deviation of the distribution is ranges is available to develop the
known from the past experience or historical upper and lower limits.
data, then formulas using the std dev can be
used to compute the upper and lower limit.
12
7/25/2019
Second approach:
x-bar Chart; when SD is Unknown
x1 + x2 + ... xk
x= = k
= =
UCL = x + A2R LCL = x - A2R
where
=
x = process avg= average of sample means
13
7/25/2019
5.10 –
5.08 –
UCL = 5.08
5.06 –
5.04 –
R-Chart Example
OBSERVATIONS (SLIP-RING DIAMETER, CM)
SAMPLE k 1 2 3 4 5 x R
The Goliath tool company want to develop an R 1 5.02 5.01 4.94 4.99 4.96 4.98 0.08
chart to control process variablity 2 5.01 5.03 5.07 4.95 4.96 5.00 0.12
3 4.99 5.00 4.93 4.92 4.99 4.97 0.08
4 5.03 4.91 5.01 4.98 4.89 4.96 0.14
5 4.95 4.92 5.03 5.05 5.01 4.99 0.13
6 4.97 5.06 5.06 4.96 5.03 5.01 0.10
7 5.05 5.01 5.10 4.96 4.99 5.02 0.14
8 5.09 5.10 5.00 4.99 5.08 5.05 0.11
9 5.14 5.10 4.99 5.08 5.09 5.08 0.15
10 5.01 4.98 5.08 5.07 4.99 5.03 0.10
50.09 1.15
Example 15.3
14
7/25/2019
0.28 –
0.24 – The R chart in this example suggest that the process is
UCL = 0.243 in control, since none of the ranges for the samples
0.20 –
are close to the control limits. However, the x bar
0.16 – chart in previous exp suggest that the process in not in
Range
R = 0.115
0.12 – control.
0.08 – In fact the ranges for sample 8 and 10 were relatively
0.04 – narrow whereas the means for these samples were
LCL = 0
| | | | | | | | | | relatively high.
0–
1 2 3 4 5 6 7 8 9 10
The use of both chart together provided a more
Sample number
complete picture of the overall process variability.
15
7/25/2019
NUMBER OF PROPORTION
UCL = p (bar) + zp SAMPLE DEFECTIVES DEFECTIVE
NUMBER OF PROPORTION
SAMPLE DEFECTIVES DEFECTIVE total defectives
p= = 200 / 20(100) = 0.10
1 6 .06 total sample observations
2 0 .00
p(1 - p) 0.10(1 - 0.10)
3 4 .04 UCL = p + z = 0.10 + 3
n 100
: : :
: : : UCL = 0.190
20 18 .18
p(1 - p) 0.10(1 - 0.10)
200 LCL = p - z = 0.10 - 3
n 100
20 samples of 100 pairs of jeans
LCL = 0.010
Since p is not known, it can be estimated from the total sample
0.14
Exp: when automobiles are inspected the number of blemishes
0.12 (defects) in the paint job can be counted for each car, but a
p = 0.10 proportion cannot be computed, since the total no of possible
0.10
blemishes is not known. In this case a single car is the sample.
0.08
0.06
0.04
2 4 6 8 10 12 14 16 18 20
Sample number
16
7/25/2019
NUMBER
OF
SAMPLE
UCL = c + zc DEFECTS 190
c= 15 = 12.67
c = c 1 12
LCL = c - zc 2 8
3 16 UCL = c + zc
= 12.67 + 3 12.67
: : = 23.35
where : :
15 15 LCL = c + zc
c = number of defects per sample 190 = 12.67 - 3 12.67
= 1.99
c-Chart (cont.)
Control Chart Patterns
24
UCL = 23.35
21
c = 12.67
15 limits.
12
If the sample value display a consistent pattern,
9 even within the control limits, it suggests that this
pattern has a nonrandom cause that might
6
warrant investigation.
3 LCL = 1.99
2 4 6 8 10 12 14 16
Sample number
17
7/25/2019
UCL UCL
LCL LCL
Zone B
= 1
1 sigma = x + 3 (A2R) Rule 1,4,5 violated in exp.
Process
Zone C
=
Company wants to see, If there is a
x
average
Zone C
pattern of nonrandomness exhibited
=
1 sigma = x - 1 (A2R)
3 within the control limits?
Zone B
=
2 sigma = x - 2 (A2R)
3
Zone A
=
LCL 3 sigma = x - A2R
| | | | | | | | | | | | |
1 2 3 4 5 6 7 8 9 10 11 12 13
Sample number
18
7/25/2019
5.10 –
Performing a Pattern Test
5.08 –
UCL = 5.08 (previous exp of x bar chart; x double bar= 5.01; result= no pattern
5.06 – rule appear to be violated)
5.04 –
SAMPLE x ABOVE/BELOW UP/DOWN ZONE
5.02 –
x= = 5.01
Mean
1 4.98 B — B
5.00 –
2 5.00 B U C
4.98 – 3 4.95 B D A
x- bar 4 4.96 B D A
4.96 –
Chart
5 4.99 B U C
Example 4.94 – LCL = 4.94
(cont.) 6 5.01 — U C
4.92 – 7 5.02 A U C
| | | | | | | | | | 8 5.05 A U B
1 2 3 4 5 6 7 8 9 10
Sample number 9 5.08 A U A
10 5.03 A D B
Variable charts require smaller samples Control limits are different than tolerances.
2 to 10 parts in a sample Tolerances are design specification reflecting
product requirement. They specify a range of
Attribute charts require larger sample sizes values above or below a designed target
50 to 100 parts in a sample value, within which product units must fall to
be acceptable.
While control limits provide a means for
determining natural variation in a production
process.
19
7/25/2019
Process capability
range of natural variability in a process what we measure with Pareto Analysis Scatter Diagram
control charts
Flow Chart SPC Chart
Tolerances
Check Sheet Cause-and-Effect
Are design specifications reflecting product requirements
i.e.10+_5 Histogram Diagram
Pareto Chart
Pareto Analysis
70
(64)
60
Percent from each cause
NUMBER OF
CAUSE DEFECTS PERCENTAGE 50
Poor design 80 64 % 40
Wrong part dimensions 16 13
30
Defective parts 12 10
Incorrect machine calibration 7 6 20
Operator errors 4 3 (13)
(10)
Defective material 3 2 10 (6)
(3) (2) (2)
Surface abrasions 3 2 0
125 100 %
20
7/25/2019
15
10
0
1 2 6 13 10 16 19 17 12 16 2017 13 5 6 2 1
X
UCL = 23.35
21 Incorrect specifications Lack of concentration Tooling problems
Number of defects
15 Quality
Inaccurate Problem
12 temperature
control Defective from vendor Poor process design
9 Ineffective quality
Not to specifications management
Dust and Dirt
6 Material- Deficiencies
handling problems in product design
LCL = 1.99
3
2 4 6 8 10 12 14 16 Environment Materials Process
Sample number
21
7/25/2019
ISO 14000
Management systems
Systems development and integration of environmental
responsibilities into business planning
Operations
Consumption of natural resources and energy
Environmental systems
Measuring, assessing and managing emissions, effluents, and
other waste
22