14-1 of 44

Chapter Fourteen
Process Improvement
Using Control Charts

14-2 of 44
McGraw-Hill/Irwin

Copyright 2003 by The McGraw-Hill Companies, Inc. All rights reserved.

Process Improvement Using Control

Charts
14.1
14.2
14.3
14.4
14.5
14.6
14.7
14.8*
14-3 of 44

Quality: Meaning and Historical Perspective

Statistical Process Control and Causes of
Variation
Sampling a Process, Rational Subgrouping,
and Control Charts
and R Charts
x
Pattern Analysis
Comparison of a Process with Specifications:
Capability Studies
Charts for Fraction Nonconforming
Cause and Effect, Defect Concentration
Diagrams

14.1: Quality: Meaning and

Perspective
Quality
Fitness for use
Extent to which customer expectations are met
Types of Quality
Quality of design
Quality of conformance
Quality of Performance

14-4 of 44

History of the Quality Movement

1924
Late 20s
1946
1950
1951
1950s
1980s
1988
1990s

14-5 of 44

Statistical Quality Control/Control Charts, Shewart/Bell

Telephone
Statistical Acceptance Sampling, Bell Telephone
American Society for Quality Control created
W. Edwards Deming introduces statistical quality control in
Japan
Deming Prize established in Japan
Zero Defects programs implemented in the U.S.
Total Quality Management (TQM)
Malcolm Baldrige National Quality Awards established
ISO 9000, international quality standards adopted

ISO 9000
ISO 9000
Series of international series standards
Establishes structures and processes for quality control systems at
every step of the production process design. raw materials, inprocess monitoring, etc.
Imposes quality discipline
Broad acceptance internationally

14-6 of 44

W. Edwards Demings 14 Points

1. Create constancy of purpose toward improvement of
product and service with a plan to become competitive,
stay in business, and provide jobs.
2. Adopt a new philosophy.
3. Cease dependence on mass inspection.
4. End the practice of awarding business on the basis of
price tag.
5. Improve constantly and forever the system of production
and service to improve quality and productivity.
6. Institute training
7. Institute leadership

14-7 of 44

W. Edwards Demings 14 Points

(Continued)
8. Drive out fear, so that everyone may work more
effectively for the company.
9. Break down organizational barriers
10. Eliminate slogans, exhortations, and arbitrary numerical
goals and targets for the workforce which urge the
workers to achieve new levels of productivity and quality
without providing methods.
11. Eliminate work standards and numerical quotas.
12. Remove barriers that rob employees of their pride of
workmanship.
13. Institute vigorous program of education and selfimprovement.
14-8 of 44

14. Take action to accomplish the transformation

14.2Statistical Process Control

Historical Inspection Approach
Inspection of output
Action on output
Scrap, rework, downgrade (expensive!)
Statistical Process Control
Monitor and study process variation
Goal: Continuous process improvement
Preventing by quality through process improvement

14-9 of 44

Causes of Process Variation

Common causes
Typical (random) variation inherent in process design
Process in statistical control
Assignable causes
Unusual process variation
Intermittent or permanent process changes
Not common to all process observations
Process not in statistical control

14-10 of 44

Example: Hole Location Case

Example 14.1:
The Hole Location
Case

14-11 of 44

Time Subgroup
8:00 AM
1
8:20 AM
2
8:40 AM
3
9:00 AM
4
9:20 AM
5
9:40 AM
6
10:00 AM
7
10:20 AM
8
10:40 AM
9
11:00 AM
10
11:20 AM
11
11:40 AM
12
12:00 PM
13
12:20 PM
14
12:40 PM
15
1:00 PM
16
1:20 PM
17
1:40 PM
18
2:00 PM
19
2:20 PM
20
Means

1
3.05
3.00
3.07
3.02
3.01
3.01
3.03
3.05
2.99
3.02
2.97
3.06
2.99
3.00
3.02
3.02
3.01
3.05
3.02
3.00

Measurement
2
3
4
3.02 3.04 3.09
3.04 2.98 2.99
3.06 2.94 2.97
2.96 3.01 2.98
2.98 3.04 3.01
3.02 2.99 2.97
2.98 2.92 3.17
3.03 2.96 3.01
2.96 3.01 3.00
3.02 2.98 3.03
2.96 2.96 3.00
3.04 3.02 3.10
3.00 3.04 2.96
3.01 2.99 3.00
2.96 3.04 2.95
3.02 3.04 2.98
2.87 3.09 3.02
2.96 3.01 2.97
2.99 3.00 2.98
3.00 3.01 3.05

5
3.05
2.99
3.01
3.02
3.01
2.96
2.96
2.97
2.95
3.02
3.04
3.05
3.02
3.01
2.97
3.03
3.00
2.98
3.00
3.01

Mean Range
3.05
0.07
3.00
0.06
3.01
0.13
3.00
0.06
3.01
0.06
2.99
0.06
3.01
0.25
3.00
0.09
2.98
0.06
3.01
0.05
2.99
0.08
3.05
0.08
3.00
0.08
3.00
0.02
2.99
0.09
3.02
0.06
3.00
0.22
2.99
0.09
3.00
0.04
3.01
0.05
3.0062 0.0850

Example: Graph of Process

Performance
Example 14.1: The Hole Location Case

14-12 of 44

Control Charts
1020

Abnormal variation
due to assignable sources

Out of
control

UCL

1010
1000

Mean
Normal variation
due to chance

990
980

LCL

Abnormal variation
due to assignable sources

970
0

9 10 11 12 13 14 15

Sample number

14-13 of 44

14.4 x Chart

14-14 of 44

xand R Chart: Control Limits

UCLx

LCLx

3( / n )

3( / n )

R / d2

x 3
n

3
R
x

d2 n

R / d2

x 3
n

3
R
x

d2 n

x A2 R

x A2 R

UCLR

LCLR

D4 R

D3 R

Control Limits for x Charts

Estimate by x, by R / d 2

A2

3
d2 n

Center line is x
Control Limits for R Charts
Center line is R

Values of d2, A2, D3 and D4 depend on subgroup size n and are found in Table 1

14-15 of 44

Control Chart Constants

14-16 of 44

Subgroup Size
n
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

Charts for Averages

Charts for Ranges
Limit Factor Std Dev Divisor Control Limit Factors
A2
d2
D3
D4
1.880
1.128
0.000
3.267
1.023
1.693
0.000
2.574
0.729
2.059
0.000
2.282
0.577
2.326
0.000
2.114
0.483
2.534
0.000
2.004
0.419
2.704
0.076
1.924
0.373
2.847
0.136
1.864
0.337
2.970
0.184
1.816
0.308
3.078
0.223
1.777
0.285
3.173
0.256
1.744
0.266
3.258
0.283
1.717
0.249
3.336
0.307
1.693
0.235
3.407
0.328
1.672
0.223
3.472
0.347
1.653
0.212
3.532
0.363
1.637
0.203
3.588
0.378
1.622
0.194
3.640
0.391
1.608
0.187
3.689
0.403
1.597
0.180
3.735
0.415
1.585
0.173
3.778
0.425
1.575
0.167
3.819
0.434
1.566
0.162
3.858
0.443
1.557
0.157
3.895
0.451
1.548
0.153
3.931
0.459
1.541

Table 14.3

Example:

xand R Charts
Example 14.3:
The Hole Location
Case

Center linex x 3.0062

UCLx x A2 R 3.0062 (0.577)(0.0850) 3.0552
LCLx x A2 R 3.0062 (0.577)(0.0850) 2.9572
Center lineR R 0.0850
UCLR D4 R 2.114 (0.0850) 0.1797
14-17 of 44

LCLR D3 R 0.000(0.0850) 0.0000

Detecting a Shift in Process Mean

Shift in process mean shows up on x-bar chart

14-18 of 44

Detecting an Increase in Process

Variation

Increase in process variation shows up on both x-bar and R char

14-19 of 44

14.5 Pattern Analysis

An observation
beyond the control
limits indicates the
presence of an
assignable cause.
Other types of
patterns can also
indicate the
presence of an
assignable cause.
These patterns are
more easily
described in terms
of control chart
zones A, B, C.
14-20 of 44

Pattern Analysis for Control Charts

Any of the following conditions or patterns is evidence of the likely
presence of an assignable cause of variation.

One point beyond zone A (three standard deviation limits. )

Two of three consecutive points in zone A (the two standard
deviation warning limits, or beyond) on one side of the center
line.
Four of five consecutive points in zone B (the one standard
deviation limits, or beyond) on one side of the center line.
A run of eight consecutive points (runs up, down or on the
same side of center line).
Any nonrandom pattern trend, fanning out, cycle or
alternating pattern.

Otherwise, the process is said to be in statistical control.

14-21 of 44

Out-of-Control Conditions
1 beyond limits

2 of 3 in A or beyond

Other out-of-control patterns

14-22 of 44

4 of 5 in B or beyond

14.6 Process Capability Studies

Natural tolerance limits for a normally distributed process that is in
statistical control will contain approximately 99.73 percent of the process
observations and is given by

R
R
R
x 3
, x 3

x 3
d2
d2
d2

The value of d2 depend on subgroup size n and is found in Table 14.3

If the natural tolerance limits are inside the process specification limits,
we say that the process is capable of meeting specifications.

14-23 of 44

Example: Process Capability

Analysis
Example 14.9: The Hole Location Case
Specification limits (given)

3.00 0.05 2.95, 3.05

Natural tolerance limits

R

x 3
d
2

0.0675
3.0006 3

2
.
326

3.0006 0.0871
[2.9135, 3.0877]

14-24 of 44

The natural tolerance limits fall outside the process specification

limits. Thus the process is not capable of meeting
specifications.

Example: Calculating Fraction out of

Specification
Example 14.9: The Hole Location Case

Specification limits
are [30.000.05]

14-25 of 44

Example: A Capable Process

Natural tolerance
limits are within
specification limits.

14-26 of 44

Sigma Level Capability and Process

Leeway

14-27 of 44

14.7 Charts for Fraction

Nonconforming
Control Limits and Center Line for a p Chart

UCL p
p3

LCL p
p(1 p )
n

p 3

Estimate p by p

p3
14-28 of 44

p (1 p )
n

p (1 p )
n

total nonconforming, all subgroups

total inspected, all subgroups

p 3

p (1 p )
n

Center line is p

Example: Sales Invoice Data

Example 14.10: Sales Invoice Data

Week
1
2
3
4
5
6
7
8
9
10

14-29 of 44

Incorrect
Invoice
1
5
4
0
3
2
1
3
0
6

Fraction
NonCon
0.01
0.05
0.04
0.00
0.03
0.02
0.01
0.03
0.00
0.06

Week
11
12
13
14
15
16
17
18
19
20

Incorrect
Invoice
4
3
2
0
2
3
3
2
1
0

Fraction
NonCon
0.04
0.03
0.02
0.00
0.02
0.03
0.03
0.02
0.01
0.00

Week
21
22
23
24
25
26
27
28
29
30

Incorrect
Invoice
4
2
1
2
5
2
3
4
1
0

Fraction
NonCon
0.04
0.02
0.01
0.02
0.05
0.02
0.03
0.04
0.01
0.00
0.0230

Example: p Chart Control Limits and

Center Line
Example 14.10: Sales Invoice Data

Center line is
total nonconforming, all subgroups 69
p

0.023
total inspected, all subgroups
3000
p (1 p )
0.023(1 0.023)
UCL p p 3
0.023 3
n
100
0.023 0.04497 0.06797
p (1 p )
LCL p p 3
n
0.023 0.04497 0.02197
14-30 of 44

Example: p Chart for Fraction

Nonconforming
Example 14.10: Sales Invoice Data
Minitab output of p
chart for sales
invoice data.

14-31 of 44

14.8* Cause-and-Effect Diagram

A cause-and-effect diagram for why tables are not
cleared quickly in a restaurant. (also known as
Ishikawa diagrams or fishbone charts.

14-32 of 44

Defect Concentration Diagram

A defect concentration diagram showing the
locations of enamel chips on kitchen ranges.

14-33 of 44

Process Improvement Using

Summary Control Charts
:14.1
14.2
14.3
14.4
14.5
14.6
14.7
14.8*
14-34 of 44

Quality: Meaning and Historical Perspective

Statistical Process Control and Causes of
Variation
Sampling a Process, Rational Subgrouping,
and Control Charts
and R Charts
x
Pattern Analysis
Comparison of a Process with Specifications:
Capability Studies
Charts for Fraction Nonconforming
Cause and Effect, Defect Concentration
Diagrams