Chapter 15 Statistical Process

Statistical Control is used to

Process prevent quality
Control problems

MGS3100
Julie Liggett De Jong

Statistical Process Take periodic

Control …. samples from
process
How it works.

1
 Take periodic Take periodic
samples from samples from
process process

Plot sample points Plot sample points

on control chart on control chart

Determine if Determine if
process is within process is within
limits acceptable limits

Variation Variation

1 Common Causes
1. 2 Special Causes
2.
9 Variation inherent in a process 9 Variation due to identifiable factors
9 Eliminated through system 9 Modified through operator or
improvements management action

2
Attribute Attribute
measures measures

Product Product
characteristic characteristics
evaluated with evaluated with
a discrete a discrete
choice: choice:
Good / bad Good / bad
Yes / No Yes / No
Pass / Fail Pass / Fail

Attribute
measures
Variable measures
Product
characteristics Measurable product
evaluated with characteristic:
a discrete
choice: Length, size
Length size, weight,
weight
height, time, velocity
Good / bad
Yes / No
Pass / Fail

3
Variable measures Variable measures

Measurable product Measurable product

characteristics: characteristics

Length, size
Length size, weight,
weight Length, size
Length size, weight,
weight
height, time, velocity height, time, velocity

Hospitals

Timeliness

Responsiveness
SPC Applied to
Services
Accuracy of lab tests

4
Grocery Stores

Check-out time

Stocking

Cleanliness

Luggage handling
Waiting times
Airlines Courtesy

Fast Food Internet

Restaurants Orders
Waiting times Order accuracy
Packaging
Food quality Delivery time
Email
Cleanliness confirmation
Package
tracking

5
 Insurance

Billing accuracy

Timeliness of claims
processing

Agent
g availability
y
Control Charts
Response time

Graphs that establish process control Attribute measures:

limits P-Charts
C-Charts

6
 Variable measures: A Process is in control if:
Mean (x-bar) control charts
Range (R) control charts

A Process is in control if: A Process is in control if:

No sample points are outside Most points are near process

control limits average

7
A Process is in control if: A Process is in control if:

About equal number of points are Points appear to be randomly

above & below centerline distributed

Process Control Chart To develop Control Charts:

Out of control
Upper
control 9Use in-control data
li it
limit

9If non-random causes are present, find

Process
average them and discard data related to them
Lower 9Correct control chart limits
control
limit

1 2 3 4 5 6 7 8 9 10
Sample number
Figure 15.1

8
 Control Charts Control Charts
Control Control
Measures Description Measures Description
Charts Charts
p Chart Attributes Calculates percent defectives p Chart Attributes Calculates percent defectives
in sample
p in sample
p

r Chart Variables Reflects the amount of r Chart Variables Reflects the amount of
(range dispersion in a sample (range dispersion in a sample
chart) chart)
x bar Chart Variables Indicates how sample results x bar Chart Variables Indicates how sample results
(mean chart) relate to the process average (mean chart) relate to the process average

Cp Process Measures the capability

p y of a Cp Process Measures the capability
p y of a
(Process Capability process to meet design (Process Capability process to meet design
Capability specifications Capability specifications
Ratio) Ratio)
Cpk Process Indicates if the process mean Cpk Process Indicates if the process mean
(Process Capability has shifted away from design (Process Capability has shifted away from design
Capability target Capability target
Index) Index)

p-Chart p-Chart

UCL = p + zσp UCL = p + zσp

LCL = p - zσp LCL = p - zσp
where where

p = the sample proportion defective; an estimate of the p = the sample proportion defective; an estimate of the
process average process average

total defectives
p=
total sample observations

9
p-Chart The Normal Distribution

UCL = p + zσp
LCL = p - zσp
where

p = the sample proportion defective; an estimate of the

process average
z = the number of standard deviations from the process
average

95%
99.74%
-3σ -2σ -1σ μ=0 1σ 2σ 3σ

Control Chart Z Values p-Chart

9 Smaller Z values make more narrow control limits UCL = p + zσp

and more sensitive charts LCL = p - zσp
9 Z = 3.00 is standard
where
9 Compromise between sensitivity and errors
p = the sample proportion defective; an estimate of the
process average
z = the number of standard deviations from the process
average
σp = the standard deviation of the sample proportion

95%
99.74%
-3σ -2σ -1σ μ=0 1σ 2σ 3σ

10
p-Chart
p(1 - p)
UCL = p + zσp σp = n
LCL = p - zσp

total defectives
p = total sample observations

p-Chart Example ~
Western Jeans Company p337

p-Chart Example ~ p-Chart Example ~

Western Jeans Company Western Jeans Company
0.20

20 samples of 100 pairs of jeans (n = 100) 0.18 UCL = 0.190

NUMBER OF PROPORTION 0 16
0.16
SAMPLE DEFECTIVES DEFECTIVE
0.14
1 6 .06
Proportion defective
0.12
2 0 .00
p = 0.10
0.10
3 4 .04
: : : 0.08
P

: : : 0.06

20 18 .18 0.04

200 0.02 LCL = 0.010

2 4 6 8 10 12 14 16 18 20
Ex 1, P337 Sample number Ex 1, P337

11
 Control Charts Range ( R ) Chart
Control
Measures Description
Charts
p Chart Attributes Calculates percent defectives in
sample
UCL = D4R LCL = D3R
r Chart Variables Reflects the amount of
(range chart) dispersion in a sample
∑R
x bar Chart
(mean chart)
Variables Indicates how sample results
relate to the process average
R= k
Cp Process Measures the capability
y of a where:
(Process Capability process to meet design
Capability specifications
Ratio)
Cpk Process Indicates if the process mean R = range of each sample
(Process Capability has shifted away from design
Capability target
k = number of samples
Index)

Factors for R-Chart: D3 & D4

SAMPLE SIZE FACTOR FOR x-CHART FACTORS FOR R-CHART
n A2 D3 D4

2 1.88 0.00 3.27

3 1.02 0.00 2.57
4 0 73
0.73 0 00
0.00 2 28
2.28
5 0.58 0.00 2.11
6 0.48 0.00 2.00
7 0.42 0.08 1.92
8 0.37 0.14 1.86
9 0.44 0.18 1.82
10 0.11 0.22 1.78
11 0.99 0.26 1.74
12 0.77 0.28 1.72
13 0.55 0.31 1.69
14 0.44 0.33 1.67
15 0.22 0.35 1.65
16 0.11 0.36 1.64
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 R-Chart Example ~
Table 1, P343 Goliath Tool Company (p345)

12
R-Chart Example ~
Goliath Tool Company

OBSERVATIONS (SLIP-RING DIAMETER, CM)

SAMPLE k 1 2 3 4 5 x R
1 5.02 5.01 4.94 4.99 4.96 4.98 0.08
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
R = max5.03
– 4.91
min = 5.01
5.024.98 4.89
– 4.94 = 4.96
0.08 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
total 50.09 1.15

Ex 3, P344

Factors for R-Chart: D3 & D4 R-Chart Example ~

Goliath Tool Company
SAMPLE SIZE FACTOR FOR x-CHART FACTORS FOR R-CHART
n A2 D3 D4

2 1.88 0.00 3.27

3 1.02 0.00 2.57
4 0 73
0.73 0 00
0.00 2 28
2.28
5 0.58 0.00 2.11
6 0.48 0.00 2.00
7 0.42 0.08 1.92
0.28 –
8 0.37 0.14 1.86 0.24 –
9 0.44 0.18 1.82 UCL = 0.243
10 0.11 0.22 1.78 0.20 –
11 0.99 0.26 1.74

ange
12 0.77 0.28 1.72 0.16 – R = 0.115
13 0.55 0.31 1.69

Ra
14 0.44 0.33 1.67 0.12 –
15 0.22 0.35 1.65
16 0.11 0.36 1.64 0.08 –
17 0.00 0.38 1.62
18 0.99 0.39 1.61 0.04 – LCL = 0
19 0.99 0.40 1.61 | | | | | | | | | |
20 0.88 0.41 1.59
0–
1 2 3 4 5 6 7 8 9 10
Table 1, P343 Sample number Example 15.3

13
 x-Chart Example ~
x-Chart Calculations
Goliath Tool Company
= =
UCL = x + A2R LCL = x - A2R
OBSERVATIONS (SLIP-RING DIAMETER, CM)
SAMPLE k 1 2 3 4 5 x R
1 5.02 5.01 4.94 4.99 4.96 4.98 0.08
2 5.01 5.03 5.07 4.95 4.96 5.00 0.12
x1 + x2 + ... xk
x= =
3 4.99 5.00 4.93 4.92 4.99 4.97 0.08
4 5.03 4.91 5.01 4.98
k (5.02 + 5.01 + 4.95 + 4.99 +4.89 4.96
4.96)/5 =0.14
4.98
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
where 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
x= = the average of the sample means 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
R bar = the average range values total 50.09 1.15

Ex 4, P345

Factors for R-Chart: D3 & D4

SAMPLE SIZE FACTOR FOR x-CHART FACTORS FOR R-CHART
n A2 D3 D4

2 1.88 0.00 3.27

3 1.02 0.00 2.57
4 0 73
0.73 0 00
0.00 2 28
2.28
5 0.58 0.00 2.11
6 0.48 0.00 2.00
7 0.42 0.08 1.92
8 0.37 0.14 1.86
9 0.44 0.18 1.82
10 0.11 0.22 1.78
11 0.99 0.26 1.74
12 0.77 0.28 1.72
13 0.55 0.31 1.69
14 0.44 0.33 1.67
15 0.22 0.35 1.65
16 0.11 0.36 1.64
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

Table 1, P343

14
x-Chart Example ~
Goliath Tool Company
5.10 – Using x- and R-charts together
5.08 –
UCL = 5.08
5 06 –
5.06

5.04 – 9 Each measures the process

5.02 –
x= = 5.01
differently
Mean

5.00 –
9 Both process average (x bar chart)
4.98 –
and variability
y ((R chart)) must be in
4.96 –
LCL = 4.94
control
4.94 –

4.92 –
| | | | | | | | | |
1 2 3 4 5 6 7 8 9 10
Sample number
Example 15.4

Sample Size Determination Sample Size Determination

9 Attribute control charts (p chart) 9 Variable control charts (R- & x bar- charts)
• 50 to 100 parts in a sample • 2 to 10 parts in a sample

15
Process Capability Process Capability
9 Range of natural variability in process
• Control limits (the “Voice of the Process”
or the “Voice of the Data”): based on • Measured with control charts.
natural variations (common causes) 9 Process cannot meet specifications if
natural variability exceeds tolerances
• Tolerance limits (the “Voice of the 9 3-sigma quality
Customer”): customer requirements • Specifications equal the process control
limits.
• Process Capability: A measure of how 9 6-sigma
6 i quality
li
“capable” the process is to meet customer • Specifications twice as large as control
requirements; compares process limits to limits
tolerance limits

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

Figure 15.5 Figure 15.5

16
Process Capability Measures Process Capability Measures
Process Capability Ratio (Cp ) Process Capability Ratio ( Cp )

a) Cp < 1.0 Design

Specifications
ttolerance
l range
Design
Cp = process range c) Cp > 1.0 Specifications

upper specification limit - Process

lower specification limit
= b) Cp = 1.0 Design
6σ
Process
Specifications
S ifi i

Process
Figure 15.5

Computing Cp Process Capability Measures

Munchies Snack Food Company Process Capability Index ( Cpk )

Net weight specification = 9.0 oz ± 0.5 oz

Process mean = 8
P 8.80
80 oz =
x - lower specification limit
Process standard deviation = 0.12 oz ,
3σ
Cpk = minimum =
upper specification limit - upper specification limit - x
lower specification limit 3σ
Cp =
6σ
D i
Design
Specifications
9.5 - 8.5
= = 1.39
6(0.12)

Ex 6, P 354 Process

17
Process Capability Measures Computing Cpk
Process Capability Index ( Cpk ) Munchies Snack Food Company

Net weight specification = 9.0 oz ± 0.5 oz

Cpk > 1.00:
1 00: Process is capable of meeting design Process mean = 8.80 oz
specifications Process standard deviation = 0.12 oz

Cpk < 1.00: Process mean has moved closer to one of the =
x - lower specification limit
upper or lower design specifications and will ,
3σ
generate defects Cpk = minimum =
upper specification limit - x
Cpk = 1.00: The process mean is centered on the design 3σ
target.
8.80 - 8.50 9.50 - 8.80
= minimum , = 0.83
3(0.12) 3(0.12)

Ex 7, P354

The Process Capability Index

Cpk < 1 Not Capable

p
Cpk > 1 Capable at 3σ
Cpk > 1.33 Capable at 4σ
Cpk > 1.67 Capable at 5σ
Cpk > 2 Capable at 6σ

18