Common Causes

Quality: n

Statistical ∑x i ∑(x − x)
2

x= i =1 σ= i

Process n n−1

Control

425 Grams

Assignable Causes Assignable Causes

Average Average

Grams Grams
(a) Location (b) Spread

Assignable Causes Effects of Assignable Causes on

Average
Process Control

Grams Out of control

(c) Shape (assignable causes present)
Effects of Assignable Causes on Sample Means and the
Process Control Process Distribution
Mean

Process
distribution

In Control
(no assignable causes) 425 Grams

Sample Means and the The Normal

Process Distribution Distribution
Mean
Distribution of
sample means σ = Standard deviation

Process
distribution
Mean
–3σ –2σ –1σ +1σ +2σ +3σ
68.26%
95.44%
425 Grams 99.74%

Control Charts
Using Control Charts for
UCL Process Improvement
• Measure the process
Nominal • When changes are indicated, find
the assignable cause
LCL • Eliminate problems, incorporate
improvements
Assignable
causes likely • Repeat the cycle
1 2 3
Samples
Control Chart Examples Control Chart Examples

UCL UCL
Variations

Variations
Nominal Nominal
LCL LCL

Sample number Sample number

Control Chart Examples Control Chart Examples

UCL UCL
Variations

Variations

Nominal Nominal
LCL LCL

Sample number Sample number

Control Chart Examples Control Limits and Errors

Type I error: Type II error:

Probability of searching for Probability of concluding
a cause when none exists that nothing has changed
UCL
Variations

Nominal
UCL
Shift in process
LCL average Process
average

Sample number LCL

Two-sigma limits
 Control Charts Control Charts
for Variables for Variables
Special Metal Screw Control Charts – Special Metal Screw
Sample Sample
Number 1 2 3 4
R-Charts R = 0.0021
1 0.5014 0.5022 0.5009 0.5027
2 0.5021 0.5041 0.5024 0.5020 UCLR = D4R
3 0.5018 0.5026 0.5035 0.5023 LCLR = D3R
4 0.5008 0.5034 0.5024 0.5015
5 0.5041 0.5056 0.5034 0.5047

Control Charts Control Charts

for
Control Chart Variables
Factors for Variables
Factor for UCL Factor for Factor
Size of and LCL for LCL for UCL for Control Charts—Special Metal Screw
Sample x-Charts R-Charts R-Charts R-Charts R = 0.0021 D4 = 2.282
(n) (A2) (D3) (D4) D3 = 0
2 1.880 0 3.267 UCLR = D4R
3 1.023 0 2.575 LCLR = D3R
4 0.729 0 2.282
5 0.577 0 2.115
6 0.483 0 2.004
UCLR = 2.282 (0.0021) = 0.00479 in.
7 0.419 0.076 1.924 LCLR = 0 (0.0021) = 0 in.

Range Chart - Control Charts

Special Metal Screw for Variables
Control Charts—Special Metal Screw
x-Charts R = 0.0021 A2 = 0.729
x= = 0.5027

UCLx = x= + A2R
LCLx = x= - A2R
UCLx = 0.5027 + 0.729 (0.0021) = 0.5042 in.
LCLx = 0.5027 – 0.729 (0.0021) = 0.5012 in.
x-Chart— x-Chart—
Special Metal Screw Special Metal Screw

• Measure the process

• Find the assignable cause
• Eliminate the problem
• Repeat the cycle

Control Charts Control Charts

for Variables Using σ for Variables Using σ
UCLx = x= + zσx UCLx = x= + zσx
Sunny Dale Bank Sunny Dale Bank
=
x = 5.0 minutes =
x = 5.0 minutes
LCL = x= – zσ
x x σ = 1.5 minutes LCL = x= – zσ
x x σ = 1.5 minutes
n = 6 customers n = 6 customers
σx = σ / n z = 1.96
σx = σ / n z = 1.96

UCLx = 5.0 + 1.96(1.5)/ 6 = 6.20 min

UCLx = 5.0 – 1.96(1.5)/ 6 = 3.80 min

Control Charts Control Charts

for Attributes for Attributes
Hometown Bank

UCLp = p + zσp
Hometown Bank LCLp = p – zσp

σp = p(1 – p)/n
 Control Charts Control Charts
Sample
Number
Wrong
Account Number for Attributes for Attributes
1 15
2 12 Hometown Bank Hometown Bank
3 19
4 2 n = 2500 n = 2500 p = 0.0049
5 19
UCLp = p + zσp
6
7
4
24 Total defectives
UCLp = p + zσp
8 7 p = Total observations
LCLp = p - zσp
9 10 LCLp = p – zσp
10 17
11 15
σ12
p =
3
p(1 - p)/n σp = p(1 – p)/n
Total 147

Control Charts p-Chart

Wrong Account Numbers
for Attributes
Hometown Bank
n = 2500 p = 0.0049
UCLp = 0.0049 + 3(0.0014)
LCLp = 0.0049 – 3(0.0014)

σp = 0.0014

p-Chart Control Charts

Wrong Account Numbers
for Attributes

Woodland
• Measure the process
Paper
• Find the assignable cause Company
• Eliminate the problem
• Repeat the cycle
 Woodland Control Charts Woodland Control Charts
Paper Paper
Company for Attributes Company for Attributes

c = 20 z=2

UCLc = c + z c

LCLc = c – z c

Woodland Control Charts

Paper
for Attributes Process Capability
Company
• Measure the process
• Find the assignable cause
Nominal
value
• Incorporate the problem Process distribution
• Repeat the cycle Lower Upper
specification specification

800 1000 1200 Hours

(a) Process is capable

Process Capability Process Capability

Nominal value
Six sigma
Nominal
value
Process distribution Four sigma

Lower Upper Two sigma

specification specification
Lower Upper
specification specification

800 1000 1200 Hours

(b) Process is not capable
Mean
Process Capability Process Capability
Lightbulb Production Lightbulb Production
Upper specification = 1200 hours Upper specification = 1200 hours
Lower specification = 800 hours Lower specification = 800 hours
Average life = 900 hours σ = 48 hours Average life = 900 hours σ = 48 hours

Upper specification -
Lower specification
Cp =
6σ
Process Capability Ratio

Process Capability Process Capability

Lightbulb Production Lightbulb Production
Upper specification = 1200 hours Cp = 1.39 Upper specification = 1200 hours Cp = 1.39
Lower specification = 800 hours Lower specification = 800 hours
Average life = 900 hours σ = 48 hours Average life = 900 hours σ = 48 hours

Cpk = Minimum of x= – Lower specification

3σ
,
Process
Upper specification – =
x
Capability
Index 3σ

Process Capability
Lightbulb Production
Upper specification = 1200 hours
Lower specification = 800 hours
Average life = 900 hours σ = 48 hours

Cpk = 0.69 Cp = 1.39

Process Process
Capability Capability
Index Ratio