29-04-2022

What Is Quality?

 “The degree of excellence of a

thing” (Webster’s Dictionary)
 “The totality of features and
characteristics that satisfy needs”
(ASQ)
 Fitness for use
Management of Quality  Quality of design

Icons of Perceptions about Quality

Quality….
The Taj Mahal has not suffered • Fitness to use
from the injury of time, thanks
to the excellence of the • Conformance to Requirements
materials and the perfection of
the workmanship. The two • Totality of features /characteristics of a product/service that
Taj Mahal – The crowning jewel of
hundred and thirty years that bear on its ability to satisfy the given need (ISO)
have elapsed since it was built
Indo-Islamic architecture, the Taj
Mahal is one of the world’s most have not spoiled the luster of • Ratio of performance to Expectations
the marble nor the countless
beautiful and beloved structures. The
monument was built in Agra, India, for wonders of the chiseling of
• What the customer says !
Mumtaz Mahal, the favorite wife of these charming, delicate
Mughal emperor Shah Jahan.
Construction of the tomb began in mosaics.
1632 and employed more than 20,000  F. Devay, 1867, Journal d'un voyage
laborers for 20 years. dans l'Inde anglaise (Journal of a Trip
Through English India)

Meaning of Quality: Dimensions of Quality (Garvin)

Consumer’s Perspective
 Fitness for use 1. Performance
 how well product or service does what  Basic operating characteristics of a product; how well a car is handled
it is supposed to or its gas mileage
 Quality of design
 designing quality characteristics into a
2. Features
product or service  “Extra” items added to basic features; such as a stereo CD or a leather
 A Mercedes and a Ford are equally “fit for interior in a car
use,” but with different design
dimensions 3. Reliability
 Probability product will operate over time; that is, a TV will work
without repair for about seven years

1
 29-04-2022

Dimensions of Quality (Garvin) Dimensions of Quality (Garvin)

4. Conformance 7. Aesthetics
 Meeting pre-established standards/ customer  Look, feel, sound, smell or taste
expectations; 8. Safety
5. Durability  Freedom from injury or harm; assurance that customer will not suffer
injury or harm from a product; an especially important consideration
 Life span before replacement for automobiles
6. Serviceability 9. Other perceptions
 Ease of getting repairs, speed of repairs, courtesy and  Subjective perceptions based on brand name, advertising, etc
competence of repair person

Service Quality 4. Consistency

 Same level of service for all customers;
1. Time & Timeliness is your newspaper delivered on time every morning?
 Customer waiting time, completed on time; how long must a customer wait 5. Accessibility & Convenience
for service, and is it completed on time? is an overnight package delivered  Ease of obtaining service; does service representative
overnight? answer you calls quickly?
2. Completeness 6. Accuracy
 Customer gets all they asked for; is a mail order from a catalogue company  Performed right every time; is your bank or credit card
complete when delivered? statement correct every month?
3. Courtesy 7. Responsiveness
 Treatment by employees; are catalogue phone operators nice and are their  Reactions to unusual situations; how well is a telephone
voices pleasant? operator at able to respond to a customer’s questions?

Meaning of Quality: Meaning of Quality:

Producer’s Perspective A Final Perspective
 Quality of conformance  Consumer’s and producer’s perspectives depend on each
 Making sure product or service is produced according to other
design  Consumer’s perspective: PRICE
 if new tires do not conform to specifications, they  Producer’s perspective: COST
wobble  Consumer’s view must dominate
 if a hotel room is not clean when a guest checks in,
hotel is not functioning according to specifications of its
design

2
 29-04-2022

The Meaning of Quality Cost of Quality

The Meaning of Quality
Cost of achieving good quality
Prevention:
 Planning, Product design,
Producer’s Perspective Consumer’s Perspective Process, Training, Information;
customer assessment, process
Quality of Conformance Quality of Design control, and quality
Production • Conformance to • Quality characteristics Marketing improvement costs to
specifications • Price
• Cost prevent defects from
occurring
Fitness for
Consumer Use

Cost of Quality Cost of Quality

Cost of achieving good quality Cost of poor quality
Internal failure costs: Costs
Appraisal: Costs of activities incurred to fix problems that are
designed to ensure quality or
detected before the product/service is
uncover defects
delivered to the customer.
 Inspection and testing,  Scrap, Rework, Process failure,
Test equipment, Process downtime, Price-
Operator Training downgrading

Cost of Quality  Consider your university as a production system in

which the final product is a graduate. For the system
Cost of poor quality  Define quality from the producers and customer’s
External failure costs: All costs perspectives
incurred to fix problems that are  Develop a fitness for use description for final product
detected after the product/service is quality
delivered to the customer.
 Customer complaints,  Give examples of cost of poor quality (internal & external
Product return, failure costs) and the cost of quality assurance (prevention
Warranty, Product and appraisal) costs.
liability, Lost sales

3
 29-04-2022

Pareto Analysis
Seven Quality Control
Tools
• Was given by Alfredo Pareto, an economist in the year 1848 – 1923.
 Pareto Analysis 1
Process
• Based on the principle – “Majority of wealth is held by
 Flow Chart disproportionately small segment of the population”.
2
3
1 2 3 4 4
Dirt 5
• Used in quality by Juran.
 Check Sheet
Old
Temp
Fault

• Of all problems, only few occur quite frequently, while others seldom
 Histogram x xx
occur – “Vital few &Trivial Many”.
x x
• Pareto principle is given by 80% of the problem are created by 20% of
 Scatter Diagram
x x x
x
x x x
x
x x x UCL
the causes.
 SPC Chart LCL • Helps management to quickly identify the critical areas.
• Guides management in allocating limited resources.
 Cause-and-Effect Diagram
• Problems are arranged in order of importance decided by financial
impact or no. of occurrences of problem.

Pareto Pareto 70
(64)
60
Chart Chart
Percent from each cause
50

NUMBER OF 40
CAUSE DEFECTS PERCENTAGE
30
Poor design 80 64 %
Wrong part dimensions 16 13 20
Defective parts 12 10 (13)
(10)
Incorrect machine calibration 7 6 10 (6)
Operator errors 4 3 (3) (2) (2)
Defective material 3 2 0
Surface abrasions 3 2
125 100 %

Causes of poor quality

DOT.COM APPAREL HOUSE

 Analysis of customer complaints for a large QUALITY ERRORS & PERCENTAGES
dot.com apparel house revealed the
following:
Percent Cumulative % Frequency
 Billing errors: 537; Shipping errors: 2460;
Electronic charge errors: 650; Long delay :
5372; Delivery error: 752 Long delays 54.98% 54.98% 5372
Shipping errors 25.18% 80.16% 2460
Construct a Pareto diagram for these data.
Delivery errors 7.69% 87.85% 752
What conclusion would you reach? Electronic charge 6.65% 94.50% 650
errors
Billing errors 5.50% 100.00% 537

Total 9771

4
 29-04-2022

Pareto Chart for Dot.Com Apparel House

Percent
Flow Chart
Cumulative %

Start/
Percent Within Defect Category

60.00% 120.00% Operation Operation Decision Operation

Cumulative Percent
Finish
50.00% 100.00%
40.00% 80.00%
30.00% 60.00%
20.00% 40.00%
10.00% 20.00% Operation Operation
0.00% 0.00%

Decision Start/
Finish

Defect Categories

Customer Waiter Salad Chef Dinner Chef

N
Is order
Place order complete?

Y
Give soup or salad order to chef Prepare soup or
salad order
Give dinner order to chef Prepare dinner
order
Drink Get drinks for customer

Eat salad or Deliver salad or soup order to customer Give order

soup to waiter

Eat dinner Deliver dinner to customer Give order

to waiter
Receives check Deliver check to customer

Gives payment
Service
Receive payment for meal
to waiter
Blueprint
Credit
Cash or
Credit? for an
Collect change,
Cash
Bring change to customer
A Process Installment
leave tip
Map of Lending
Run credit card through

Fill in tip
Restaurant Operation
Return credit slip to customer
amount Service
Source: Lynn Shostack, “Service Positioning through Structural Change,” Journal of Marketing 51 (January 1987), p.
Collect tip
36. Reprinted with permission by the American Marketing Association

Check Sheet
Histogram
 When there are frequent failures at work, Check
Sheets are used to find the frequency and
location of area of failure.
• Displays large amounts of data that are difficult to interpret in
COMPONENTS REPLACED BY LAB tabular form.
TIME PERIOD: 22 Feb to 27 Feb 2002
REPAIR TECHNICIAN: Bob • Shows centering, variation, and shape.
TV SET MODEL 1013 • Illustrates the underlying distribution of the data.
Integrated Circuits ||||
• Provides useful information for predicting future performance.
Capacitors |||| |||| |||| |||| |||| ||
Resistors || • Helps to answer “Is the process capable of meeting requirements?.
Transformers ||||
Commands
CRT |

5
 29-04-2022

Histogram Scatter Diagram

25
• Supplies the data to confirm a hypothesis that two variables
20 are related.
• Provides both a visual and statistical means to test the
15
strength of a relationship.
10 • Provides a good follow-up to cause and effect diagrams.

0
1 2 6 13 10 16 19 17 12 16 20 17 13 5 6 2 1

Scatter Diagram Control Chart

Y • Focuses attention on detecting and monitoring process
variation over time.
• Distinguishes special from common causes of variation.
• Serves as a tool for on-going control.
• Provides a common language for discussion process
performance.

Control Chart Cause and Effect diagram

27

24
 Developed by Kaoru Ishikawa in 1943.
21
UCL = 23.35  Called as Ishikawa diagram or Fish bone diagram.
Number of defects

18 c = 12.67  Used to identify and systematically list the different

causes that can be attributed to a problem.
15
 Also used to identify, which of the causes has a
12 greatest effect, i.e. identifies the root cause.
9  Main applications
 Cause Enumeration
6 ○ Is the most widely used graphical technique in quality control
LCL = 1.99 ○ Develops through a brainstorming, where all possible types of
3 causes are listed that can influence the problem

2 4 6 8 10 12 14 16
Sample number

6
 29-04-2022

Contd.. Cause-and-Effect Diagram

 Steps
 Define the problem or quality characteristic Measurement Human Machines
 Major causes influencing the characteristic are noted – i.e. Faulty
testing equipment Poor supervision Out of adjustment
men, machine, method, environment etc.
Incorrect specifications Lack of concentration Tooling problems
 Identify the sub causes within each of the major causes.
Improper methods Inadequate training Old / worn
 Each sub causes should also be properly defined and identified
clearly. Quality
 Select appropriate methods of measurement for each causes. Inaccurate Problem
temperature
control Defective from vendor Poor process design
 Each cause is singled out and analysed. Ineffective quality
Not to specifications
 Advantage Dust and Dirt Material-
management
Deficiencies
handling problems in product design
 Process of their construction creates a better understanding
of the components of the process and their relationships
Environment Materials Process

The manager of an apartment complex receives many complaints from

residents. She makes a check sheet of the types complaints for last 12
weeks. Develop a pareto chart using this information. Give
recommendations to the manager w.r.t. the causes of complaints received
using Ishikawa chart. For the chart under each heading at least four
causes may be identified.

Statistical Process
Control

Statistical Process Control Variation

 Common Causes
• Take periodic samples from process Variation inherent in a process
Can be eliminated only through
• Plot sample points on control chart improvements in the system
• Determine if process
is within limits UCL
 Special Causes
• Prevent quality Variation due to identifiable factors
problems LCL Can be modified through operator or
management action

7
 29-04-2022

Natural Variation Assignable Causes

 Also called common causes  Also called special causes of variation

 Affect virtually all production processes  Generally this is some change in the process

 Expected amount of variation  Variations that can be traced to a specific reason

 Output measures follow a probability distribution  The objective is to discover when assignable causes are
present
 For any distribution there is a measure of central  Eliminate the bad causes
tendency and dispersion
 Incorporate the good causes
 If the distribution of outputs falls within acceptable
limits, the process is said to be “in control”

Types of Data Types of Data

Variables Attributes
 Attribute data
 Product characteristic  Characteristics that  Defect-related
evaluated with a discrete choice can take any real characteristics
 Good/bad, yes/no
value  Classify products
 May be in whole or as either good or
 Variable data in fractional bad or count
 Product characteristic that numbers defects
can be measured
 Length, size, weight, height,  Continuous random  Categorical or
time, velocity variables discrete random
variables

SPC Applied to Services Service Quality Examples

 Hospitals
Timeliness, responsiveness,
 Nature of defect is different in accuracy of lab tests
services  Grocery Stores
Check-out time, stocking, cleanliness
 Service defect is a failure to meet  Airlines
customer requirements Luggage handling, waiting times,
 Monitor times, customer courtesy
satisfaction  Fast food restaurants
Waiting times, food quality,
cleanliness, employee courtesy

8
 29-04-2022

Process Control Control Charts

(a) In statistical
control and capable
Frequency
of producing within
control limits
 Graph establishing process control
Upper control limit
limits
Lower control limit
(b) In statistical  Charts for variables
control but not
capable of producing Mean (x-bar), Range (R)
within control limits
 Charts for attributes
(c) Out of control
p and c

Size
(weight, length, speed, etc.)

Where to Use Control Charts Central Limit Theorem

 Process has a tendency to go out of control Regardless of the distribution of the population,
 Process is particularly harmful and costly if it goes out of control the distribution of sample means drawn from the
 Examples population will tend to follow a normal curve
 at the beginning of a process because it is a waste of time and money to
begin production process with bad supplies
 before a costly or irreversible point, after which product is difficult to 1. The mean of the sampling
rework or correct distribution (x) will be the same as x=m
 before and after assembly or painting operations that might cover defects the population mean m
 before the outgoing final product or service is delivered 2. The standard deviation of the
sampling distribution (sx) will equal
s
the population standard deviation sx =
(s) divided by the square root of the n
sample size, n

Population and Sampling

Distributions Process Control Chart
Three population Distribution of Upper Out of control
distributions sample means control
limit
Mean of sample means = x
Beta
Standard Process
deviation of s
the sample = sx = n
average
Normal
means

Uniform Lower
control
limit
| | | | | | |

-3sx -2sx -1sx x +1sx +2sx +3sx

95.45% fall within ± 2sx
99.73% of all x 1 2 3 4 5 6 7 8 9 10
fall within ± 3sx Sample number

9
 29-04-2022

A Process is In Control Development of Control Chart

if
1. No sample points outside limits
 Based on in-control data
2. Most points near process average
3. About equal number of points  If non-random causes present
above & below centerline discard data
4. Points appear randomly  Correct control chart limits
distributed

Control Charts for p-Chart

Attributes
UCL = p + zsp
 p Charts LCL = p - zsp
Calculate percent defectives in sample where
z = the number of standard deviations from
the process average
 c Charts p = the sample proportion defective; an
estimate of the process average
Count number of sp = the standard deviation of the sample
defects in item proportion
p(1 - p)
sp = n

p-Chart Example
The Normal Distribution
20 samples of 100 pairs of jeans
NUMBER OF PROPORTION
SAMPLE DEFECTIVES DEFECTIVE
1 6 .06
2 0 .00
3 4 .04
: : :
95%
: : :
99.74%
20 18 .18
-3s -2s -1s m=0 1s 2s 3s
200

10
 29-04-2022

p-Chart Example p-Chart Example

20 samples of 100 pairs of jeans 20 samples of 100 pairs of jeans
NUMBER OF PROPORTION NUMBER OF PROPORTION
SAMPLE DEFECTIVES DEFECTIVE SAMPLE DEFECTIVES DEFECTIVE p = 0.10
1 6 .06 1 6 .06
2 0 .00 2 0 p(1.00
- p) 0.10(1 - 0.10)
total defectives UCL = p + z = 0.10 + 3
100
3 4 p = .04 3 4 .04
n
total sample observations
: : : : :
UCL = 0.190 :
: : = 200: / 20(100) : : :
p(1 - p) 0.10(1 - 0.10)
20 18 = 0.10
.18 20 LCL
18= p - z = 0.10 - 3
n.18 100
200 200= 0.010
LCL

p-Chart 0.20 Problem

0.18 UCL = 0.190

0.16
 Twenty-five samples of size 50 are chosen from a plastic-
0.14
injection molding machine producing small containers. The
Proportion defective

0.12
number of nonconforming containers for each sample is
p = 0.10
0.10 shown in Table 1. Construct a control chart for the proportion
0.08 of nonconforming. Revise the control limits, assuming special
0.06 causes for points outside the control limits.
0.04

0.02 LCL = 0.010

2 4 6 8 10 12 14 16 18 20
Sample number
64

c-Chart c-Chart
The number of defects in 15 sample rooms
SAMPLE NUMBER OF DEFECTS
UCL = c + zsc
sc = c 1 12 190
LCL = c - zsc 2 8 c= = 12.67
15
3 16
where UCL = c + zsc
: : = 12.67 + 3 12.67
c = number of defects per sample
: : = 23.35
15 15
LCL = c + zsc
190 = 12.67 - 3 12.67
= 1.99

11
 29-04-2022

c-Chart 24
Problem
UCL = 23.35
21

 The number of scratch marks for a particular piece of furniture

18
Number of defects

c = 12.67
is recorded for sample of size 10. The results are shown in
15
Table 2 for 25 samples. Construct a chart for the number of
12 scratch marks. Revise the control limits, assuming special
9 causes for the out-of-control points.
6

3 LCL = 1.99

2 4 6 8 10 12 14 16
Sample number

68

Control Charts for Variables Range ( R- ) Chart

UCL = D4R LCL = D3R

 Mean chart ( x -Chart )
Uses average of a sample R
R= k
 Range chart ( R-Chart ) where
Uses amount of dispersion in
a sample R = range of each sample
k = number of samples

RangeSAMPLE
(n SIZE
R- FACTOR
) Chart
FOR x-CHART
A 2
FACTORS FOR R-CHART
D3 D4
R-Chart Example
2 1.88 0.00 3.27
3 1.02 0.00 2.57 OBSERVATIONS (SLIP-RING DIAMETER, CM)
4 0.73 0.00 2.28
5 0.58 0.00 2.11 SAMPLE k 1 2 3 4 5 x R
6 0.48 0.00 2.00 1 5.02 5.01 4.94 4.99 4.96 4.98 0.08
7 0.42 0.08 1.92 2 5.01 5.03 5.07 4.95 4.96 5.00 0.12
8 0.37 0.14 1.86 3 4.99 5.00 4.93 4.92 4.99 4.97 0.08
9 0.44 0.18 1.82 4 5.03 4.91 5.01 4.98 4.89 4.96 0.14
10 0.11 0.22 1.78 5 4.95 4.92 5.03 5.05 5.01 4.99 0.13
11 0.99 0.26 1.74 6 4.97 5.06 5.06 4.96 5.03
12 0.77 0.28 1.72
7 5.05 5.01 5.10 4.96 4.99 5.02 0.14
13 0.55 0.31 1.69
14 0.44 0.33 1.67 8 5.09 5.10 5.00 4.99 5.08 5.05 0.11
15 0.22 0.35 1.65 9 5.14 5.10 4.99 5.08 5.09
16 0.11 0.36 1.64 10 5.01 4.98 5.08 5.07 4.99
17 0.00 0.38 1.62
18 0.99 0.39 1.61
19 0.99 0.40 1.61
20 0.88 0.41 1.59

12
 29-04-2022

R-Chart Example R-Chart Example

R 1.15 UCL = D4R = 2.11(0.115) = 0.243
R= = = 0.115
OBSERVATIONS (SLIP-RING DIAMETER, CM) k 10 OBSERVATIONS (SLIP-RING
LCL = D DIAMETER, CM)
3R = 0(0.115) = 0
SAMPLE k 1 2 3 4 5 x R k –
SAMPLE0.28 1 2 3 4 5 x R
1 5.02 5.01 4.94 4.99 4.96 4.98 0.08 1 5.02 5.01 4.94 4.99 4.96 4.98 0.08
0.24 –
2 5.01 5.03 5.07 4.95 4.96 5.00 0.12 2 5.01
UCL 5.03
= 0.2435.07 4.95 4.96 5.00 0.12
3 4.99 5.00 4.93 4.92 4.99 4.97 0.08 3 0.20 – 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 4 5.03 4.91 5.01 4.98 4.89 4.96 0.14
0.16 –

Range
5 4.95 4.92 5.03 5.05 5.01 4.99 0.13 5 4.95R =4.92
0.115 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 6 0.12 – 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 7 5.05 5.01 5.10 4.96 4.99 5.02 0.14
0.08 –
8 5.09 5.10 5.00 4.99 5.08 5.05 0.11 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 9 0.04 – 5.14 5.10 4.99 5.08 5.09 5.08 0.15
LCL = 0
10 5.01 4.98 5.08 5.07 4.99 5.03 0.10 10 5.01
| | 4.98| 5.08
| 5.07
| | 4.99| 5.03
| | 0.10 |
0–
50.09 1.15 1 2 3 4 5 6 8
7 50.09 91.1510
Sample number

x-Chart Calculations x-Chart Example

OBSERVATIONS (SLIP-RING DIAMETER, CM)
x1 + x2 + ... xk SAMPLE k 1 2 3 4 5 x R
x= = = x 50.09
k x1 = = 5.02 5.01= 5.01
4.94 4.99
cm 4.96 4.98 0.08
2 k 10 5.03 5.07 4.95
5.01 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
UCL = x + A2R LCL = x - A2R UCL5 = x= + A2R4.95
= 5.01
4.92+ (0.58)(0.115)
5.03 5.05 5.01= 5.08
4.99 0.13
6 4.97 5.06 5.06 4.96 5.03 5.01 0.10
where LCL7 = x= - A2R5.05
= 5.01 - (0.58)(0.115)
5.01 = 4.94
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
x = the average of the sample means 10 5.01 4.98 5.08 5.07 4.99 5.03 0.10
50.09 1.15

x-Chart Example 5.10 – Using x- and R-Charts

5.08 –
UCL = 5.08
OBSERVATIONS (SLIP-RING DIAMETER, CM)
Together
5.06 –
SAMPLE k 1 2 3 4 5 x R  Each measures the process differently. It is
= x 5.04 – 50.09
possible for sample averages to be in
x1 = = 5.02 5.01= 5.01
4.94 4.99
cm 4.96 4.98 0.08
2 k5.02 – 5.0110=5.03 5.07 4.95 4.96 5.00 0.12
control, but ranges might be very large
Mean

3 4.99 x5.00
= 5.01 4.93 4.92 4.99 4.97 0.08
4 5.00 – 5.03 4.91 5.01 4.98 4.89 4.96 0.14
=
UCL5 = x 4.98
+ A–2R4.95
= 5.01
4.92+ (0.58)(0.115) = 5.08
5.03 5.05 5.01 4.99 0.13
 It is possible for samples to have very
6 4.97 5.06 5.06 4.96 5.03 5.01 0.10 narrow ranges, but their averages is
LCL7 = x= -4.96
A2–R5.05
= 5.01
5.01- (0.58)(0.115) = 4.94
5.10 4.96 4.99 5.02 0.14
beyond control limits
8 5.09 LCL
5.10 5.00 4.99 5.08 5.05
= 4.94 0.11
4.94 –
9 5.14 5.10 4.99 5.08 5.09 5.08 0.15
 Both process average and variability must
10 4.92 – 5.01 4.98 5.08 5.07 4.99 5.03 0.10
| | | | | | | | 50.09
| |1.15
be in control
1 2 3 4 5 6 7 8 9 10
Sample number

13
 29-04-2022

Subgroup X1 X2 X3 X4 13 6.4 6.38 6.47 6.35

Control Chart Patterns
1 6.35 6.4 6.32 6.37 14 6.38 6.39 6.45 6.42
2 6.46 6.37 6.36 6.41 15 6.5 6.42 6.43 6.45 UCL
3 6.34 6.4 6.34 6.36 16 6.33 6.35 6.29 6.39
4 6.69 6.64 6.68 6.59 17 6.41 6.4 6.29 6.34
5 6.38 6.34 6.44 6.4 18 6.38 6.44 6.28 6.58 UCL
6 6.42 6.41 6.43 6.34 19 6.35 6.41 6.37 6.38
7 6.44 6.41 6.41 6.46 20 6.56 6.55 6.45 6.48 LCL
8 6.33 6.41 6.38 6.36 21 6.38 6.4 6.45 6.37
9 6.48 6.44 6.47 6.45 22 6.39 6.42 6.35 6.4 Sample observations
10 6.47 6.43 6.36 6.42 23 6.42 6.39 6.39 6.36 consistently below the LCL
11 6.38 6.41 6.39 6.38 24 6.43 6.36 6.35 6.38 center line
12 6.37 6.37 6.41 6.37 25 6.39 6.38 6.43 6.44 Sample observations
consistently above the
center line

Control Chart Patterns Zones for Pattern Tests

UCL =
3 sigma = x + A2R
UCL Zone A
= 2
2 sigma = x + 3 (A2R)

Zone B
= 1
1 sigma = x + 3 (A2R)
UCL
Zone C
Process =
x
LCL average
Zone C
=
1 sigma = x - 1 (A2R)
3
Sample observations Zone B
consistently increasing LCL =
2 sigma = x - 2 (A2R)
3
Zone A
=
Sample observations LCL 3 sigma = x - A2R
consistently decreasing | | | | | | | | | | | | |
1 2 3 4 5 6 7 8 9 10 11 12 13
Sample number

Control Chart Patterns Process Capability

1. 8 consecutive points on one side of the  Range of natural variability in process

center line.  Measured with control charts.
2. 8 consecutive points up or down across  Process cannot meet specifications if
Zones. natural variability exceeds tolerances
3. 14 points alternating up or down.  3-sigma quality
4. 2 out of 3 consecutive points in Zone A  Specifications equal the process control
limits.
but still inside the control limits.
 6-sigma quality
5. 4 out of 5 consecutive points in Zone A
 Specifications twice as large as control
or B.
limits

14
 29-04-2022

Process Capability Process Capability

Design Design
Specifications Specifications

(a) Natural variation (c) Design specifications

exceeds design greater than natural
specifications; process variation; process is
is not capable of capable of always
meeting specifications conforming to
all the time. specifications.
Process Process
Design Design
Specifications Specifications

(b) Design specifications (d) Specifications greater

and natural variation the than natural variation,
same; process is capable but process off center;
of meeting specifications capable but some output
most the time. will not meet upper
specification.
Process Process

Process Capability Measures Process Capability Measures

 A capable process must have a Cp of at least 1.0

Process Capability Ratio
 Does not look at how well the process is centered in
the specification range
tolerance range
Cp = process range  Often a target value of Cp = 1.33 is used to allow for
off-center processes
upper specification limit -  Six Sigma quality requires a Cp = 2.0
lower specification limit
=
6s

Computing Cp Process Capability

Measures
Net weight specification = 9.0 oz  0.5 oz Process Capability Index
Process mean = 8.80 oz
Process standard deviation = 0.12 oz =
x - lower specification limit
,
3s
upper specification limit - Cpk = minimum =
lower specification limit upper specification limit - x
Cp = 3s
6s
• A capable process must have a Cpk of at
9.5 - 8.5 least 1.0
= = 1.39 Process is
6(0.12) capable • A capable process is not necessarily in the
center of the specification, but it falls within
the specification limit at both extremes

15
 29-04-2022

Computing Cpk

Net weight specification = 9.0 oz  0.5 oz

Process mean = 8.80 oz
Process standard deviation = 0.12 oz
=
x - lower specification limit
,
Cpk = minimum 3s
=
upper specification limit - x
3s

8.80 - 8.50 9.50 - 8.80

= minimum , = 0.83
3(0.12) 3(0.12) New machine
is NOT
capable

16