Statistical Process Control

(SPC)

Dr. R K Singh

S6 1

Statistical Process Control

(SPC)
Variability is inherent
in every process
Natural or common
causes
Special or assignable causes

Provides a statistical signal when

assignable causes are present
Detect and eliminate assignable
causes of variation
S6 2

Quality Assurance using SPC

Designed Standard

Status of process

Centre of specification
limits (Target)

Centre of the process

(Process Average)

Upper Specification Limit

(USL)

Upper Control Limit (UCL)

Lower Specification Limit

(LSL)

(UCL LCL): Spread of the

process

(USL LSL): Desired

tolerance

This represents the voice

of the customer

Lower Control Limit (LCL)

This represents the voice

of the process

S6 3

Variability
Random

Non-Random

common causes

special causes

inherent in a
process

due to identifiable
factors

can be eliminated
only through
improvements in
the system

can be modified
through operator or
management action

S6 4

Quality Measures
Attribute
a product characteristic that can be
evaluated with a discrete response
good bad; yes - no

Variable
a product characteristic that is
continuous and can be measured

weight - length
S6 5

Applying SPC to Service (cont.)

Hospitals
timeliness and quickness of care, staff responses to
requests, accuracy of lab tests, cleanliness, courtesy,
accuracy of paperwork, speed of admittance and
checkouts

Grocery Stores
waiting time to check out, frequency of out-of-stock
items, quality of food items, cleanliness, customer
complaints, checkout register errors

Airlines
flight delays, lost luggage and luggage handling, waiting
time at ticket counters and check-in, agent and flight
attendant courtesy, accurate flight information,
passenger cabin cleanliness and maintenance

S6 6

Applying SPC to Service (cont.)

Fast-Food Restaurants
waiting time for service, customer complaints,
cleanliness, food quality, order accuracy,
employee courtesy

Insurance Companies
billing accuracy, timeliness of claims processing,
agent availability and response time

S6 7

Control Charts
A graph that
establishes control
limits of a process

Types of charts
Attributes
p-chart
c-chart

Control limits
upper and lower bands
of a control chart

Variables
range (R-chart)
mean (x bar
chart)

S6 8

Setting up a process control

system
Choose the characteristic
for process control

Choose the
Measurement method

Choose an appropriate
Sampling procedure

Choose the type of

Control Chart

Calculate
control limits

Plot the data & Analyse

S6 9

Characteristics for process control

Sl. No.

Some examples

Type of Applications

Component Manufacturing

Characteristic for Measurement

Conformance of physical measurements of

components and sub-assemblies to
specifications
Conformance to operating characteristics of
machines and other resources involved in the
process

Final Assembly

Number of defects in the product

Conformance to test specifications
Number of missing elements

Process Industries

Temperature, Pressure and Heat specifications

Conformance to product specifications
Conformance to equipment specifications
Vibrations and other variations in equipments
and sub-systems
Conformance to specifications of the
automation & control system

Service Systems

Number of defects in various business

processes
Errors in processing documents
Conformance to waiting time/lead time related
specifications
S6 10

Population and Sampling

Distributions
Distribution of
sample means
Mean of sample means = x
Standard
s
deviation of
the sample = sx = n
means

-3sx

-2sx

-1sx

+1sx +2sx +3sx

95.45% fall within 2sx

99.73% of all x
fall within 3sx
S6 11

Control Charts for Variables

For variables that have
continuous dimensions
Weight, speed, length,
strength, etc.

x-charts are to control

the central tendency of the process
R-charts are to control the dispersion of
the process
These two charts must be used together
S6 12

Setting Chart Limits

For x-Charts when we know s
Upper control limit (UCL) = x + zsx
Lower control limit (LCL) = x - zsx
where

x = mean of the sample means or a target

value set for the process
z = number of normal standard deviations
sx = standard deviation of the sample means
= s/ n
s = population standard deviation
n = sample size
S6 13

Setting Control Limits

Hour 1
Sample
Weight of
Number
Oat Flakes
1
17
2
13
3
16
4
18
n=9
5
17
6
16
7
15
8
17
9
16
Mean 16.1
s=
1

Hour
1
2
3
4
5
6

Mean
16.1
16.8
15.5
16.5
16.5
16.4

Hour
7
8
9
10
11
12

Mean
15.2
16.4
16.3
14.8
14.2
17.3

For 99.73% control limits, z = 3

UCLx = x + zsx = 16 + 3(1/3) = 17
LCLx = x - zsx = 16 - 3(1/3) = 15
S6 14

Setting Control Limits

Control Chart
for sample of
9 boxes

Variation due
to assignable
causes

Out of
control

17 = UCL
Variation due to
natural causes

16 = Mean
15 = LCL
| | | | | | | | | | | |
1 2 3 4 5 6 7 8 9 10 11 12

Sample number

Out of
control

Variation due
to assignable
causes

S6 15

Setting Chart Limits

For x-Charts when we dont know s
Upper control limit (UCL) = x + A2R
Lower control limit (LCL) = x - A2R
where

R = average range of the samples

A2 = control chart factor found in Table.1
x = mean of the sample means

S6 16

Control Chart Factors

Sample Size
n

Mean Factor
A2

Upper Range
D4

Lower Range
D3

2
3
4
5
6
7
8
9
10
12

1.880
1.023
.729
.577
.483
.419
.373
.337
.308
.266

3.268
2.574
2.282
2.115
2.004
1.924
1.864
1.816
1.777
1.716

0
0
0
0
0
0.076
0.136
0.184
0.223
0.284
Table 1
S6 17

Setting Control Limits

Process average x = 12 ounces
Average range R = .25
Sample size n = 5

S6 18

Setting Control Limits

Process average x = 12 ounces
Average range R = .25
Sample size n = 5
UCLx

= x + A2R
= 12 + (.577)(.25)
= 12 + .144
= 12.144 ounces
From
Table.1

S6 19

Setting Control Limits

Process average x = 12 ounces
Average range R = .25
Sample size n = 5
UCLx

LCLx

= x + A2R
= 12 + (.577)(.25)
= 12 + .144
= 12.144 ounces

UCL = 12.144

= x - A2R
= 12 - .144
= 11.857 ounces

LCL = 11.857

Mean = 12

S6 20

R Chart
Type of variables control chart
Shows sample ranges over time
Difference between smallest and
largest values in sample

Monitors process variability

Independent from process mean

S6 21

R Chart
For R-Charts
Upper control limit (UCLR) = D4R
Lower control limit (LCLR) = D3R
where
R = average range of the samples
D3 and D4 = control chart factors from Table.1

S6 22

Setting Control Limits

Average range R = 5.3 pounds
Sample size n = 5
From Table S6.1 D4 = 2.115, D3 = 0
UCLR = D4R
= (2.115)(5.3)
= 11.2 pounds

UCL = 11.2

LCLR

LCL = 0

= D3R
= (0)(5.3)
= 0 pounds

Mean = 5.3

S6 23

Mean and Range Charts

(a)

(Sampling mean is
shifting upward but
range is consistent)

These
sampling
distributions
result in the
charts below
UCL

(x-chart detects
shift in central
tendency)

x-chart
LCL
UCL

(R-chart does not

detect change in
mean)

R-chart
LCL

S6 24

Mean and Range Charts

(b)
These
sampling
distributions
result in the
charts below

(Sampling mean
is constant but
dispersion is
increasing)
UCL

(x-chart does not

detect the increase
in dispersion)

x-chart
LCL
UCL

(R-chart detects
increase in
dispersion)

R-chart
LCL

S6 25

Control Charts for Attributes

For variables that are categorical
Good/bad, yes/no,
acceptable/unacceptable

Measurement is typically counting

defectives
Charts may measure
Percent defective (p-chart)
Number of defects (c-chart)
S6 26

Control Limits for p-Charts

Population will be a binomial distribution,
but applying the Central Limit Theorem
allows us to assume a normal distribution
for the sample statistics
UCLp = p + zsp^

sp =
^

LCLp = p - zsp^
where

p
z
sp^
n

=
=
=
=

p(1 - p)
n

mean fraction defective in the sample

number of standard deviations
standard deviation of the sampling distribution
sample size
S6 27

p-Chart for Data Entry

Sample
Number

1
2
3
4
5
6
7
8
9
10

Number
of Errors

Fraction
Defective

6
5
0
1
4
2
5
3
3
2

.06
.05
.00
.01
.04
.02
.05
.03
.03
.02

80
p = (100)(20) = .04

Sample
Number

Number
of Errors

11
6
12
1
13
8
14
7
15
5
16
4
17
11
18
3
19
0
20
4
Total = 80

sp^ =

Fraction
Defective

.06
.01
.08
.07
.05
.04
.11
.03
.00
.04

(.04)(1 - .04)
= .02
100
S6 28

p-Chart for Data Entry

UCLp = p + zsp^ = .04 + 3(.02) = .10

Fraction defective

LCLp = p - zsp^ = .04 - 3(.02) = 0

.11
.10
.09
.08
.07
.06
.05
.04
.03
.02
.01
.00

UCLp = 0.10

p = 0.04

10

12

14

16

18

20

LCLp = 0.00

Sample number
S6 29

p-Chart for Data Entry

UCLp = p + zsp^ = .04 + 3(.02) = .10

Fraction defective

Possible
LCLp = p - zsp^ = .04 - 3(.02) =
0
assignable
causes present

.11
.10
.09
.08
.07
.06
.05
.04
.03
.02
.01
.00

UCLp = 0.10

p = 0.04

10

12

14

16

18

20

LCLp = 0.00

Sample number
S6 30

Control Limits for c-Charts

Population will be a Poisson distribution,
but applying the Central Limit Theorem
allows us to assume a normal distribution
for the sample statistics
UCLc = c + 3 c
where

LCLc = c - 3 c

c = mean number defective in the sample

S6 31

c-Chart for Cab Company

c = 54 complaints/9 days = 6 complaints/day

LCLc = c - 3 c
=6-3 6
=0

14
12
10
8
6
4
2
0

Number defective

UCLc = c + 3 c
=6+3 6
= 13.35

|
1 2

UCLc = 13.35

c= 6

|
3

|
4

|
5

|
6

|
7

|
8

LCLc = 0
|
9

Day
S6 32

Which Control Chart to Use

Variables Data
Using an x-chart and R-chart:
Observations are variables
Collect 20 - 25 samples of n = 4, or n =
5, or more, each from a stable process
and compute the mean for the x-chart
and range for the R-chart

Track samples of n observations each

S6 33

Which Control Chart to Use

Attribute Data
Using the p-chart:
Observations are attributes that can
be categorized in two states
We deal with fraction, proportion, or
percent defectives
Have several samples, each with
many observations

S6 34

Which Control Chart to Use

Attribute Data
Using a c-Chart:
Observations are attributes whose
defects per unit of output can be
counted
The number counted is a small part of
the possible occurrences
Defects such as number of blemishes
on a desk, number of typos in a page
of text, flaws in a bolt of cloth
S6 35

Process Capability
The natural variation of a process
should be small enough to produce
products that meet the standards
required
A process in statistical control does not
necessarily meet the design
specifications
Process capability is a measure of the
relationship between the natural
variation of the process and the design
specifications
S6 36

Process Capability Ratio

Upper Specification - Lower Specification
Cp =
6s
A capable process must have a Cp of at
least 1.0
Does not look at how well the process
is centered in the specification range

Often a target value of Cp = 1.33 is used

to allow for off-center processes
Six Sigma quality requires a Cp = 2.0
S6 37

Process Capability Ratio

Insurance claims process
Process mean x = 210.0 minutes
Process standard deviation s = .516 minutes
Design specification = 210 3 minutes

Upper Specification - Lower Specification

Cp =
6s

S6 38

Process Capability Ratio

Insurance claims process
Process mean x = 210.0 minutes
Process standard deviation s = .516 minutes
Design specification = 210 3 minutes

Upper Specification - Lower Specification

Cp =
6s
213 - 207
=
= 1.938
6(.516)

S6 39

Process Capability Ratio

Insurance claims process
Process mean x = 210.0 minutes
Process standard deviation s = .516 minutes
Design specification = 210 3 minutes

Upper Specification - Lower Specification

Cp =
6s
213 - 207
=
= 1.938
6(.516)

Process is
capable
S6 40

Process Capability Index

Upper
Lower
Cpk = minimum of Specification - x , x - Specification
Limit
Limit
3s
3s

A capable process must have a Cpk of at

least 1.0
A capable process is not necessarily in the
center of the specification, but it falls within
the specification limit at both extremes
S6 41

Process Capability Index

New Cutting Machine
New process mean x = .250 inches
Process standard deviation s = .0005 inches
Upper Specification Limit = .251 inches
Lower Specification Limit = .249 inches

S6 42

Process Capability Index

New Cutting Machine
New process mean x = .250 inches
Process standard deviation s = .0005 inches
Upper Specification Limit = .251 inches
Lower Specification Limit = .249 inches
(.251) - .250
Cpk = minimum of
,
(3).0005

S6 43

Process Capability Index

New Cutting Machine
New process mean x = .250 inches
Process standard deviation s = .0005 inches
Upper Specification Limit = .251 inches
Lower Specification Limit = .249 inches
(.251) - .250
.250 - (.249)
Cpk = minimum of
,
(3).0005
(3).0005
Both calculations result in
.001
Cpk =
= 0.67
.0015

New machine is
NOT capable
S6 44

Interpreting Cpk
Cpk = negative number
Cpk = zero
Cpk = between 0 and 1
Cpk = 1
Cpk > 1
S6 45

Acceptance Sampling
Form of quality testing used for
incoming materials or finished goods
Take samples at random from a lot
(shipment) of items
Inspect each of the items in the sample
Decide whether to reject the whole lot
based on the inspection results

Only screens lots; does not drive

quality improvement efforts
S6 46

Acceptance Sampling
Form of quality testing used for
incoming materials or finished goods
Take samples at random from a lot
Rejected lots can be:
(shipment) of items
Returned
to the
Inspect each of the
items in the
sample
supplier
Decide whether to reject the whole lot
Culledresults
for
based on the inspection
defectives
Only screens lots; does
drive
(100%not
inspection)

quality improvement efforts

S6 47

Operating Characteristic
Curve
Shows how well a sampling plan
discriminates between good and
bad lots (shipments)
Shows the relationship between
the probability of accepting a lot
and its quality level

S6 48

P(Accept Whole Shipment)

The Perfect OC Curve

Keep whole
shipment
100
75

Return whole
shipment

50
25
|
0 | |
0 10 20

Cut-Off
|

30 40

50 60

70 80

90 100

% Defective in Lot
S6 49

An OC Curve
Figure S6.9
100
95

= 0.05 producers risk for AQL

75

Probability
of
50
Acceptance
25
10
= 0.10
0 |
0

Consumers
risk for LTPD

|
1

|
2

|
3

|
4

|
5

AQL
Good
lots

|
6

|
7

|
8

Percent
defective

LTPD
Indifference
zone

Bad lots
S6 50

AQL and LTPD

Acceptable Quality Level (AQL)
Poorest level of quality we are
willing to accept

Lot Tolerance Percent Defective

(LTPD)
Quality level we consider bad
Consumer (buyer) does not want to
accept lots with more defects than
LTPD
S6 51

Producers and Consumers

Risks
Producer's risk ()
Probability of rejecting a good lot
Probability of rejecting a lot when the
fraction defective is at or above the
AQL

Consumer's risk ()
Probability of accepting a bad lot
Probability of accepting a lot when
fraction defective is below the LTPD
S6 52

SPC and Process Variability

Lower
specification
limit

Upper
specification
limit

(a) Acceptance
sampling (Some
bad units accepted)
(b) Statistical process
control (Keep the
process in control)

(c) Cpk >1 (Design

a process that
is in control)
Process mean, m
S6 53