You are on page 1of 19

Implementing Six Sigma Quality

at Better Body Manufacturing


D M A I
Define

Measure Analyze

Improve Control

D M A I

Overview

Define

Measure Analyze

Improve Control

ABC Incorporated (ABC) is not achieving Six Sigma quality levels for all critical
Body-Side Sub-Assembly dimensions as requested by their customers.
200000
DPM

150000

DPM

ASM_9Y

3874

ASM_10Y

776

ASM_6Y

ASM_6Y

50000

19786

ASM_10Y

85824

ASM_3Y

ASM_9Y

ASM_8Y

ASM_3Y

172475

ASM_8Y

100000

ASM_7Y

ASM_7Y

Dimension

Ensure that all critical body-side subassembly dimensions are within Six Sigma
quality levels of < 3.4 DPM. Cp 2.0 and Cpk 1.67.
Determined the correlation between body side and assembly dimensions.
Evaluated the significance of Tonnage > 935 for ASM_7Y & ASM_8Y.
Conducted a DOE for Clamp position for ASM_9Y & ASM_10Y.
Change tonnage to > 935 to correct ASM_7Y and ASM_8Y
Set clamp position to location 2 for ASM_9Y and ASM_10Y
Re-machine A-pillar die to correct A_3Y and ASM_3Y
2

Problem Statement & The Goal

D M A I
Define

Measure Analyze

Improve Control

ABC Incorporateds customer wants ABC to apply Six Sigma problem solving
methodology to insure that the body side subassembly is achieving Six Sigma quality
levels of less than 3.4 defects per million for all critical body side subassembly
dimensions.
ABC needs an improvement strategy that minimizes the rework costs while achieving the
desired quality objective. ABCs goal is to produce module subassemblies that meet the

customer requirements and not necessarily to insure that every individual stamped
component within the assembly meets it original print specifications sub-system
optimizations vs. local optimization.
A -Pillar
Reinforcement

B-Pillar
Reinforcement

Body Side Outer

D M A I

Measure Phase

Define

Measure Analyze

Improve Control

Key Variables:
Assembly process variables:

Weld Pattern (density), Clamp Location, and Clamp Weld Pressure


Stamping process variables (body side):
Press Tonnage, Die Cushion Pressure, Material Thickness

Body Assembly Dimensions ASM_1Y through ASM_10Y


Assembly Dimensions with Highest Defects
200000

172475

DPM

150000

100000

85824

50000
19786
3874

776

0
ASM_7Y

ASM_8Y

ASM_3Y

ASM_9Y

ASM_10Y

ASM_6Y

Analyze Phase

D M A I
Define

Measure Analyze

Improve Control

Resolution alternatives (based upon past experience):


1. Make adjustments to assembly process settings

2. Reduce variation of components through better control of stamping


process input variables
3. Rework stamping dies to shift component mean deviation that is off
target and causing assembly defects
Target Performance Level:
All ten critical assembly dimensions at Six Sigma quality level of 3.4 DPM.
Cp 2.0 and Cpk 1.67
Fish Bone and P-Diagrams:
Understanding potential causes of defects. From this we pick the assembly and
component dimensions that require further analysis

D M A I

Analyze Phase

Define

Measure Analyze

Improve Control

Environment
Quality

Component
Variability

Inspection
Process
Clamp Weld
Pressure

Clamp
Location
Methods

For our analysis we will do a DOE to check


for levels that contribute to better quality
product.

Material Thickness

Humidity

Gage R&R

Weld Pattern
(density)

Inputs

Temperature

Operator

Training
Body
Assembly

Yield
Strength

Material
Thickness

Die Cushion
Pressure

Elastic
Limit
Materials

Press
Tonnage
Machine

Control Variables
Clamp Location Press Tonnage
Weld Density
Die Pressure
Clamp Pressure

Body Side Sub-Assembly


Stamping Process

Yield Strength

Outputs
Body Side Sub-Assemblies at
Six Sigma quality levels

Noise Variables
Environment
Inherent Variation

Error
States
Dimensional
defects
6

D M A I

Analyze Phase

Define

Measure Analyze

Improve Control

Analysis of ASM_7Y and ASM_8Y


Run Chart for ASM_7Y

Run Chart for ASM_8Y


1.0

ASM_8Y

ASM_7Y

1.0

0.5

0.5

0.0

0.0

12

4.00000
7.00000
5.00000
0.03464
0.96536

Number of runs up or down:


Expected number of runs:
Longest run up or down:
Approx P-Value for Trends:
Approx P-Value for Oscillation:

12

Subgroup Number

Subgroup Number
Number of runs about median:
Expected number of runs:
Longest run about median:
Approx P-Value for Clustering:
Approx P-Value for Mixtures:

6.00000
7.66667
3.00000
0.10778
0.89222

Number of runs about median:


Expected number of runs:
Longest run about median:
Approx P-Value for Clustering:
Approx P-Value for Mixtures:

4.00000
7.00000
5.00000
0.03464
0.96536

Number of runs up or down:


Expected number of runs:
Longest run up or down:
Approx P-Value for Trends:
Approx P-Value for Oscillation:

8.00000
7.66667
2.00000
0.59781
0.40219

XY Plot of ASM_8Y and ASM_7Y

1.0

ASM_7Y

Conclusion: BS_7Y and ASM_7Y are following a similar trend.


A correlation chart to study this further shows high correlation.
(Pearson correlation, R of 0.701).

0.5

0.0

0.0

0.5

1.0

ASM_8Y

D M A I

Analyze Phase

Define

Capability Analysis of B_7Y


Capability of B_7Y

USL
Target

0.70
*

LSL
Mean

-0.70
0.11

USL

0 DPM

Within
Overall

LSL
Mean

Potential (Within) Capability


Cp
2.96
CPU
2.50

Potential (Within) Capability


Cp
1.56
CPU
-0.44

CPL
Cpk

3.43
2.50

CPL

3.56

Cpk

-0.44

Cpm

Cpm

Overall Capability

-0.4

-0.2

Observed Performance

0.0

0.2

0.4

Exp. "Within" Performance

0.6

0.8

Exp. "Overall" Performance

USL

698416 DPM

Within

-0.700000
0.899444

Sample N
36
StDev (Within) 0.149640
StDev (Overall) 0.383691

-0.6

LSL

Process Data
USL
0.700000
Target
*

Sample N
36
StDev (Within) 0.0788122
StDev (Overall) 0.0791215

-0.8

Improve Control

Capability Analysis of BS_7Y


Capability of BS_7Y

LSL

Process Data

Measure Analyze

Overall Capability

Overall

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

0.61
-0.17

Observed Performance
PPM < LSL
0.00
PPM > USL
666666.67

Exp. "Within" Performance


PPM < LSL
0.00
PPM > USL
908706.09

Exp. "Overall" Performance


PPM < LSL
15.33
PPM > USL
698400.06

PPL

1.39

PPM Total

PPM Total

PPM Total

Ppk

-0.17

Pp
PPU

2.95
2.49

PPM < LSL


PPM > USL

0.00
0.00

PPM < LSL


PPM > USL

0.00
0.00

PPM < LSL


PPM > USL

0.00
0.00

Pp
PPU

PPL
Ppk

3.41
2.49

PPM Total

0.00

PPM Total

0.00

PPM Total

0.00

666666.67

908706.09

698415.39

XY Plot of ASM_7Y and BS_7Y


1.3

Furthermore, BS_7Y shows strong correlation on


dimension ASM_7Y. (Pearson correlation, R of
0.786).

1.2
1.1
1.0

BS_7Y

Conclusion: B_7Y has 0 ppm compared to ~700K


DPM in BS_7Y.

0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.5

1.0

1.5

ASM_7Y

D M A I

Analyze Phase

Define

Measure Analyze

Improve Control

XY Plot of Tonnage vs. BS_7Y

XY Plot of Tonnage vs. BS_7Y

BS_7Y

1.5

1.0

0.5

905

915

925

935

945

Tonnage

Conclusion: Tonnage values above 935 greatly improves BS_7Y and brings it closer
to the mean. Lets see what impact this has on ASM dimensions 7Y, 8Y, 9Y, and
10Y by creating a subset of the data looking only at Tonnage > 935.
9

D M A I

Analyze Phase

Define

Measure Analyze

Improve Control

Impact this has on ASM dimensions 7Y, 8Y, 9Y & 10Y on Tonnage
Capability Analysis of ASM_8Y at Tonnage > 935

Capability Analysis of ASM_7Y at Tonnage > 935


LSL

Process Data
USL

USL

1.00

Target
LSL

*
-1.00

Mean

0.09

Sample N

Within
Overall

Process Data
1.00000

USL
Target
LSL

*
-1.00000

Mean

-0.12833

Sample N

12
0.163174

StDev (Within)

0.101825

StDev (Overall)

0.147855

StDev (Overall)

0.089161

Potential (Within) Capability

Potential (Within) Capability

Cp

2.04

Cp

3.27

CPU

1.86

CPU

3.69

CPL

2.23

CPL

2.85

Cpk

1.86

Cpk

2.85

-1.0

Overall Capability

-0.5

0.0

Observed Performance

0.5

Exp. "Within" Performance

1.0

Cpm

Within
Overall

-1.0

-0.5

Overall Capability

Exp. "Overall" Performance

USL

12

StDev (Within)

Cpm

LSL

0.0

Observed Performance

0.5

Exp. "Within" Performance

1.0
Exp. "Overall" Performance

Pp

2.25

PPM < LSL

0.00

PPM < LSL

0.00

PPM < LSL

0.00

Pp

3.74

PPM < LSL

0.00

PPM < LSL

0.00

PPM < LSL

0.00

PPU
PPL

2.05
2.46

PPM > USL


PPM Total

0.00
0.00

PPM > USL


PPM Total

0.01
0.01

PPM > USL


PPM Total

0.00
0.00

PPU
PPL

4.22
3.26

PPM > USL


PPM Total

0.00
0.00

PPM > USL


PPM Total

0.00
0.00

PPM > USL


PPM Total

0.00
0.00

Ppk

3.26

Ppk

2.05

Capability Analysis of ASM_9Y at Tonnage > 935


Process Data
1.00000

USL
Target
LSL

*
-1.00000

Mean

0.52083

Sample N

LSL

Capability Analysis of ASM_10Y at Tonnage > 935

USL

LSL

Process Data

Within
Overall

12

USL

*
-1.00

Mean

0.39

Sample N

0.206010

StDev (Within)

0.215541

StDev (Overall)

0.177098

StDev (Overall)

0.187663

Potential (Within) Capability

Potential (Within) Capability

Cp

1.62

Cp

1.55

CPU

0.78

CPU

0.94

CPL

2.46

CPL

2.15

Cpk

0.78

Cpk

0.94

Cpm

*
Overall Capability

-0.5

0.0

Observed Performance

0.5

Exp. "Within" Performance

1.0

Cpm

Exp. "Overall" Performance

Pp

1.88

PPM < LSL

0.00

PPM < LSL

0.00

PPM < LSL

0.00

PPU
PPL

0.90
2.86

PPM > USL


PPM Total

0.00
0.00

PPM > USL


PPM Total

10010.77
10010.77

PPM > USL


PPM Total

3408.51
3408.51

Ppk

0.90

Within
Overall

12

StDev (Within)

-1.0

USL

1.00

Target
LSL

*
Overall Capability

-1.0

-0.5

0.0

Observed Performance

0.5

Exp. "Within" Performance

1.0
Exp. "Overall" Performance

Pp

1.78

PPM < LSL

0.00

PPM < LSL

0.00

PPM < LSL

0.00

PPU
PPL

1.08
2.47

PPM > USL


PPM Total

0.00
0.00

PPM > USL


PPM Total

2326.72
2326.72

PPM > USL


PPM Total

576.00
576.00

Ppk

1.08

Conclusion: Setting Tonnage to greater than 935 resulted in ASM_7Y and ASM_8Y
meeting the goal of <3.4 DPM. ASM_9Y and ASM_10Y require further analysis.
10

D M A I

Analyze Phase

Define

Measure Analyze

Improve Control

DOE for Response Variable ASM_9Y


DOE factorial analysis shows Clamp Position is the only significant factor in
determining ASM_9Y dimension
Input Variable

Proposed ASM_9Y Setting

Proposed ASM_10Y Setting

DOE Response Optimization for ASM_9Y


Clamp Location
Location 2
Location 2
Weld Density (welds per X inches)
1.33
1.33
Set Clamp Position to Location 2Clamp
(level
Pressure1)
2100 psi
2100 psi
Optimizer recommends setting Weld Density to 1.33 weld per inch (level 1),
but this appears to be a robust parameter, which could be changed for the benefit

of process without reducing quality if processing time or cost shows a benefit.


Optimizer recommends setting Clamp Pressure to 2100 psi (level 1), but this
appears to be a robust parameter, which could be changed for the benefit of process
without reducing quality if processing time or cost shows a benefit.
Run additional tests at recommended settings to confirm results
Weld Density and Clamp Pressure are robust parameters and can be set to optimize
the process capability to maximum level and lowest cost.

11

Analyze Phase

D M A I
Define

Measure Analyze

Improve Control

DOE for Response Variable ASM_10Y


DOE factorial analysis shows Clamp Position is also the only significant

factor in determining ASM_10Y dimension


DOE Response Optimization for ASM_10Y
Setting clamp to location 2 also improves ASM_10Y

Recommend same settings used to improve ASM_9Y to improve process


capability which also allows for no changes to machine setup and helps reduce
possible process concerns
Run additional tests at recommended settings to confirm results
Weld Density and Clamp Pressure are robust parameters and can be set to optimize
the process capability to maximum level and lowest cost.

12

Analyze Phase

D M A I
Define

Measure Analyze

Improve Control

DOE for Response Variable ASM_3Y


DOE factorial analysis shows that no factors are significant

Response Optimization shows no solution for response optimizer


Observe Process Capability of A_3Y and BS_3Y
ASM_3Y and A_3Y have a similar mean shift in the -Y direction
Correlation of Output Variables
No dimensional correlations appear to exist between ASM_3Y and
A_3Y or BS_3Y
Stepwise Regression Analysis of BS_3Y
Tonnage and Die Pressure appear to be significant in determining
dimension BS_3Y
Tonnage values < 920 may improve BS_3Y
Die Pressure appears to have no clear correlation to BS_3Y
13

D M A I

Analyze Phase

Define

Measure Analyze

Improve Control

Process Capability of BS_ 3Y and ASM_3Y at Tonnage < 920


Created subset of body data looking only at dimensions with Tonnage < 935

Tonnage < 920 appears to improve the mean of BS_3Y slightly, but has no
impact on improving the mean of ASM_3Y.
Capability Analysis of ASM_3Y
Die remachined to move mean +0.80

Capability of A_3Y and ASM_3Y with +0.80


mm mean offset

LSL

Process Data
USL

Target
LSL

*
-1

Mean
Sample N

USL
Within
Overall

0
36

StDev (Within) 0.0851436

Manipulate data for A_3Y and ASM_3Y


by +0.80 mm to simulate re-machining

StDev (Overall) 0.0971725

Potential (Within) Capability


Cp

3.91

CPU

3.91

CPL

3.91

Cpk

3.91

Cpm

Process capability shows 0 defects for


A_3Y and ASM_3Y with this mean offset

*
Overall Capability

-1.0

-0.5

0.0

Observed Performance

0.5

Exp. "Within" Performance

1.0
Exp. "Overall" Performance

Pp

3.43

PPM < LSL

0.00

PPM < LSL

0.00

PPM < LSL

0.00

PPU
PPL

3.43
3.43

PPM > USL


PPM Total

0.00
0.00

PPM > USL


PPM Total

0.00
0.00

PPM > USL


PPM Total

0.00
0.00

Ppk

3.43

14

Analyze Phase

D M A I
Define

Measure Analyze

Improve Control

Conclusions
From the analysis of ASM_7Y and ASM_8Y we can conclude that:

Setting tonnage > 935 results in ASM_7Y and ASM_8Y meeting the goal

Analyzing ASM_9Y and ASM_10Y helps determine that:

Setting clamp position to location 2, weld density to 1 weld every 1.33


and clamp pressure to 2000 psi helps with dimensions ASM_9Y and
ASM_10Y

Analyzing ASM_3Y helps us conclude that:

Re-machine A-Pillar die to move A_3Y to nominal which could cause


BS_3Y to shift towards nominal effectively shifting ASM_3Y to nominal

15

D M A I

Analyze Phase

Define

Measure Analyze

Improve Control

With the recommended changes the process performance will improve significantly
Dimension Mean

StDev
Overall

DPM_Obsv DPM_Within DPM_Exp

Pp

Ppk

Cp

Cpk

ASM_1Y

-0.035

0.165

2.01

1.94

2.47

2.39

ASM_2Y

0.259

0.152

2.20

1.63

2.31

1.71

ASM_3Y

0.000

0.097

ASM_4Y

0.009

0.115

2.90

2.87

3.53

3.50

ASM_5Y

-0.330

0.145

2.30

1.54

3.72

2.50

ASM_6Y

-0.284

0.160

2.08

1.49

2.24

1.60

ASM_7Y

0.090

0.148

2.25

2.05

2.04

1.86

ASM_8Y

-0.128

0.089

3.74

3.26

3.27

2.85

ASM_9Y

0.521

0.180

ASM_10Y

0.395

0.191

16

Improve Phase

D M A I
Define

Measure Analyze

Improve Control

Recommendations for improving the process:


Set Tonnage to above 935 to improve ASM_7Y & ASM_8Y
Set Clamp to Location 2 to improve ASM_9Y & ASM_10Y
Re-machine the A-Pillar die to move the mean of A_3Y to nominal which in turn

will move ASM_3Y to nominal

Implement the above recommendations and run additional samples to verify results.

17

Control Phase

D M A I
Define

Measure Analyze

Improve Control

Recommended controls :
Implement a gauge on the body side component press to monitor tonnage
Implement an alarm and shut-off feature on the body side press if tonnage

falls below 935 tons


Implement poke-yoke clamping fixture that ensures clamp is always in

Position 2
Establish an affordable control plan for ongoing monitoring of the 10

critical assembly dimensions.

18

Summary

D M A I
Define

Measure Analyze

Improve Control

ABC Incorporated is not achieving Six Sigma quality levels for all critical BodySide Sub-Assembly dimensions as requested by their customers. BBM needs to
apply Six Sigma problem solving methodology to establish an improvement strategy
that minimizes rework costs, yet achieves the desired quality objective.
Bring the key process output variables within Six Sigma quality level of < 3.4 DPM.
Cp 2.0 and Cpk 1.67
Set Tonnage to above 935 to improve ASM_7Y & ASM_8Y
Set Clamp to Location 2 to improve ASM_9Y & ASM_10Y
Re-machine the A-Pillar die to move the mean of A_3Y to nominal

Implement a gauge on the body side component press to monitor tonnage


Implement an alarm & shut-off feature on body side press if tonnage falls below 935
Implement poke-yoke clamping fixture that ensures clamp is always in Position 2
Establish control plan for ongoing monitoring of the 10 critical assembly dimensions.

19

You might also like