Automotive Sheet Steel Stamping Process Variation
AK Steel Corporation
Bethlehem Steel Corporation
DaimlerChrysler Corporation
Dofasco Inc.
Ford Motor Company
General Motors Corporation
Ispat/Inland Inc.
LTV Steel Company
National Steel Corporation
Rouge Steel Company
Stelco Inc.
U.S. Steel Group, a Unit of USX Corporation
WCI Steel, Inc.
Weirton Steel Corporation
Auto/Steel
Partnership
This publication was prepared by:
Body Systems Analysis Project Team
An analysis of stamping
The Auto/Steel Partnership Program
process capability and
2000 Town Center, Suite 320
Automotive Sheet Steel
Southfield, Michigan 48075-1123
implications for design,
248.356.8511 fax
http://www.a-sp.org
Stamping Process Variation
die tryout and process
A/SP-9030-3 0100 2M PROG
Printed in U.S.A.
control.
Auto/Steel Partnership
Automotive Sheet Steel
Stamping Process Variation:
An Analysis of Stamping Process Capability and
Implications for Design, Die Tryout and Process Control
Auto/Steel Partnership Program
Body Systems Analysis Project Team
2000 Town Center - Suite 320
Southfield, MI 48075-1123
2000
Auto/Steel Partnership
AK Steel Corporation
Bethlehem Steel Corporation
DaimlerChrysler Corporation
Dofasco Inc.
Ford Motor Company
General Motors Corporation
Ispat Inland Inc.
LTV Steel Company
National Steel Corporation
Rouge Steel Company
Stelco Inc.
U. S. Steel Group, a Unit of USX Corporation
WCI Steel, Inc.
Weirton Steel Corporation
This publication is for general information only. The material contained herein should not be used
without first securing competent advice with respect to its suitability for any given application. This
publication is not intended as a representation or warranty on the part of The Auto/Steel Partnership – or
any other person named herein – that the information is suitable for any general or particular use,
or free from infringement of any patent or patents. Anyone making use of the information assumes
all liability arising from such use.
This publication is intended for use by Auto/Steel Partnership members only. For more information or
additional copies of this publication, please contact the Auto/Steel Partnership, 2000 Town Center, Suite
320, Southfield, MI 48075-1123 or phone: 248-945-7777, fax: 248-356-8511, web site: www.a-sp.org
Copyright 2000 Auto/Steel Partnership. All Rights Reserved.
ii
Table of Contents
Preface
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi
Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.0
Introduction
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1
Motivation for Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2
Study Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.0
Stamping Variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.1
Components of Variation Explained . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2
Calculating Components of Variation Using ANOVA . . . . . . . . . . . . . . . . . . . . . 9
2.3
Description of the Sources of Stamping Variation . . . . . . . . . . . . . . . . . . . . . . . 13
3.0
Analysis of Stamping Variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.1
Mean Conformance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.1.1
Benchmark Comparison - Body Side Outer and Inner Panels . . . . . 14
3.1.2
Mean Bias and Part Tolerances . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.1.3
Benchmark Comparison - Tryout versus Production . . . . . . . . . . . . . 18
3.1.4
Mean Bias Stability over Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.1.5
Impact of Shipping on Mean Bias . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.2
Stamping Process Variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.2.1
Benchmark Comparison - Part-to-Part Variation . . . . . . . . . . . . . . . . 21
3.2.2
Variation Over Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.2.3
Impact of Shipping on Variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.2.4
Components of Variation: Part-to-Part, Run-to-Run,
and Begin-End of Run . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.2.5
Steel Properties and Press Setup Control and Stamping Variation . . 27
3.2.6
Effect of Mean Shifts on Statistical Process Control Techniques . . . . 29
4.0
Tolerance Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
4.1
Tolerances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
4.2
Cp and Cpk (Pp and Ppk) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
4.3
Recommended Tolerances for Sheet Metal . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4.4
Part Tolerances and Functional Build . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
5.0
Conclusions and Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Appendix A - Part Sketches by Company . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
iii
List of Figures
Figure 1.
Body Side Components Chosen for Company C . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Figure 2.
Components of Variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Figure 3.
Potential Sources of Stamping Variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Figure 4.
Total Variation Partitioned into Components . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Figure 5.
Body Side Outer for Company A: 12 Measurement Locations . . . . . . . . . . . . . . . . . 12
Figure 6.
Histogram of Mean Values across 5 Parts for Company C . . . . . . . . . . . . . . . . . . . . 15
Figure 7.
Mean Conformance: Rigid vs. Non-Rigid Panels . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Figure 8.
Mean Conformance: Two-Piece Body Side Panel vs. One-Piece . . . . . . . . . . . . . . . 16
Figure 9.
Correlation of Mean at Part Approval vs. Production . . . . . . . . . . . . . . . . . . . . . . . . . 19
Figure 10.
Effect of Stamping Mean Shift on Body Side Assembly . . . . . . . . . . . . . . . . . . . . . . 20
Figure 11.
Average Variation (Standard Deviation) by Type of Part . . . . . . . . . . . . . . . . . . . . . . 23
Figure 12.
Part-to-Part Variation: Home Line Tryout Approval vs. Production, by Dimension . . . 24
Figure 13.
Components of Variation for Body Side Panel at Company C and Company D . . . . . 26
Figure 14.
Relationship between Press Tonnage and Mean Shift Variation ( mean shift) . . . . . 29
Figure 15.
X-Bar/Range Chart vs. Individuals/ Moving Range Charts . . . . . . . . . . . . . . . . . . . . 32
Figure 16.
Illustration of Cp and Cpk calculations for three scenarios . . . . . . . . . . . . . . . . . . . . 35
Figure 17.
Part Sketches at Company A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Figure 18.
Part Sketches at Company B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
Figure 19.
Part Sketches at Company C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Figure 20.
Part Sketches at Company D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Figure 21.
Part Sketches at Company E . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Figure 22.
Part Sketches at Company F . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
Figure 23.
Part Sketches at Company G . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
iv
List of Tables
Table 1.
Participating Automotive Manufacturers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Table 2.
Components Studied at Each Automotive Manufacturer . . . . . . . . . . . . . . . . . . . . . . 5
Table 3.
Formulae for Calculating Components of Variation . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Table 4.
36-Data Samples for a Stamping Dimension . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Table 5.
SPSS Output Calculations for Mean Squared Errors (all factors) . . . . . . . . . . . . . . . . 11
Table 6.
SPSS Output Calculations for Mean Squared Errors without Begin-End Factor . . . . . 11
Table 7.
Summary of Components of Variation Calculations . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Table 8.
Variance Summary for twelve Body Side Dimensions . . . . . . . . . . . . . . . . . . . . . . . . 12
Table 9.
Mean Conformance by Company . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Table 10.
Mean Bias by Type of Part . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Table 11.
Mean Conformance and Tolerances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Table 12.
Summary of Mean bias: Tryout vs. Production - Case Study Parts . . . . . . . . . . . . . . 18
Table 13.
Comparisons of the Change in Mean Bias from Tryout to Home Line . . . . . . . . . . . . 19
Table 14.
Change in Mean from Home Line to Long-term Production . . . . . . . . . . . . . . . . . . . 20
Table 15.
Summary of Panels Measured Before and After Shipping . . . . . . . . . . . . . . . . . . . . . 21
Table 16.
Part-to-Part Variation for the Body Side Outer Panels . . . . . . . . . . . . . . . . . . . . . . . . 22
Table 17.
Effect of Dimension Location on Variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Table 18.
Part-to-Part Variation: Home Line Approval vs. Production, by Company . . . . . . . . . 24
Table 19.
Summary of Remeasured Data Before and After Shipping via truck . . . . . . . . . . . . . 25
Table 20.
Summary of Part-to-Part and Total Variation for the Body Side Outers . . . . . . . . . . . 25
Table 21.
Sources of Variation by Part for Company A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Table 22.
Sources of Variation by Part for Company C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Table 23.
Summary of Product and Process Variation Compliance . . . . . . . . . . . . . . . . . . . . . 28
Table 24.
Summary of Mean Shift Variation across Companies . . . . . . . . . . . . . . . . . . . . . . . . 30
Table 25.
Process Control Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Table 26.
Effect of Stamping Mean Shifts on Assembly Variation . . . . . . . . . . . . . . . . . . . . . . . 33
Table 27.
General Recommended Tolerances for
Stamped Parts Based upon Process Capability . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
v
Preface
The researchers are indebted to several global
automotive manufacturers for their on-going dedi-
This report is one of a series published by the
cation and participation in this research. They
Auto/Steel Partnership Body Systems Analysis
include DaimlerChrysler Corporation, Ford Motor
Project Team on stamping and assembly variation,
Company, General Motors Corporation, Nissan,
body measurement systems and process valida-
NUMMI (Toyota), Opel and Renault. Each con-
tion. These reports provide a summary of the proj-
ducted experiments under production conditions
ect research and are not intended to be all inclu-
involving hundreds of hours of effort and often
sive of the research effort. Numerous seminars
requiring the commitment of many production
and workshops have been given to individual
workers and engineering personnel. Although it
automotive manufacturers throughout the project
may be impractical to mention each one of these
to aid in implementation and provide direct techni-
people individually, we do offer our sincere appre-
cal support. Proprietary observations and imple-
ciation.
mentation details are omitted from the reports.
These reports represent a culmination of several
This automotive body development report,
years of effort by the Body Systems Analysis
"Stamping Process Variation: An Analysis of
Project Team. Team membership, which has
Stamping Process Capability and Implications for
evolved over the course of this project, includes:
Design, Die Tryout and Process Control," updates
ongoing research activities by the Body Systems
J. Aube, General Motors Corporation
Analysis Team and the Manufacturing Systems
H. Bell, General Motors Corporation
staff at The University of Michigan's Office for the
C. Butche, General Motors Corporation
Study of Automotive Transportation.
G. Crisp, DaimlerChrysler Corporation
T. Diewald, Auto/Steel Partnership
An over-riding goal of this research is to develop
K. Goff, Jr., Ford Motor Company
new paradigms that will drive automotive body-in-
T. Gonzales, National Steel Corporation
white development and production towards a total
R. Haan, General Motors Corporation
optimized processing system. Previous reports
S. Johnson, DaimlerChrysler Corporation
described fundamental research investigating
F. Keith, Ford Motor Company
simultaneous development systems for designing,
T. Mancewicz, General Motors Corporation
tooling and assembling bodies, and flexible body
J. Naysmith, Ronart Industries
assembly. Since the inception of this research pro-
J. Noel, Auto/Steel Partnership
gram, considerable emphasis has been focused
P. Peterson, USX
on benchmarking key world class body develop-
R. Pierson, General Motors Corporation
ment and production processes. These bench-
R. Rekolt, DaimlerChrysler Corporation
marks created foundation elements upon which
M. Rumel, Auto/Steel Partnership
further advances could be researched and devel-
M. Schmidt, Atlas Tool and Die
oped.
The University of Michigan Transportation
This report summarizes recommendations for
Research Institute conducted much of the
moving toward a new "functional build" paradigm
research and wrote the final reports. The principal
by tightly integrating the many individual activities
research team from the Manufacturing Systems
ranging from body design and engineering
Group was:
through process and tooling engineering. Revised
stamping die tryout and buyoff processes receive
Patrick Hammett, Ph.D. (734-936-1121/pham-
special emphasis, as does the launch of stamping
mett@umich.edu)
and assembly tools.
Jay Baron, Ph.D. (734-764-
4704/jaybaron@umich.edu)
Donald Smith, Associate Director (734-764-5262)
vi
Executive Summary
to learn and to apply this knowledge to automotive
body evaluation processes including die buy-off,
The Auto/Steel Partnership (A/SP) is an innovative
production validation and long-term process
international association that includes
capability analysis. Working within the constraints
DaimlerChrysler, Ford, General Motors and eleven
of the production environment, this research eval-
North American sheet steel producers. The
uated stamping variation for several processes
Partnership was formed in 1987 to leverage the
across the seven manufacturers. The research
resources of the automotive and steel industries to
found that stamping variation is related to:
pursue research projects leading to excellence in
the application of sheet steels in the design and
• Check point location on a part: More rigid areas
manufacture of vehicles. The Partnership has
tend to be closer to nominal and have less
established project teams that examine issues
variation.
related to steel properties including strength, dent
• Measurement fixture design: Checking fixtures
resistance, surface texture and coating weights,
with more clamps tend to reflect lower variation.
as well as manufacturing methods including
• Part size, complexity and thickness: Smaller, less
stamping, welding and design improvements.
complex and thicker parts have less variation.
• Press process control: Different press lines
Automotive manufacturers face the challenge of
demonstrate higher die set to die set mean shift
identifying when a process is capable of produc-
control which often is reflected in the control of
ing dimensionally acceptable stamped panels.
process variables such as draw press tonnage.
The non-rigid nature of many stamped parts has
• Shipping and handling: The shipping and han-
always made them difficult to measure. Often
dling of parts tends to increase variation and
parts do not meet the dimensional quality objec-
shift dimensions on the parts.
tives, as measured by C
• Changes in stamping presses: Some dimension-
pk, seen in many other
vehicle components. In fact, no manufacturer has
al shifts occur as dies are moved from a tryout
successfully achieved a C
press line to the home production press line.
pk of 1.33 on all part
dimensions using the original specifications. This
is particularly true for the larger, lighter gauge
Different automotive manufacturers manage varia-
body panels. Furthermore, achieving a high C
tion, in part, by how they manage these factors and
pk
value alone is not necessarily a good predictor of
several examples are cited in this report. Although
final dimensional quality. Factors, such as the
the effects of steel material properties such as
rigidity of the mating panels, the assembly locat-
gauge, yield strength, percent elongation and n-
ing process and the clamp and welding effects,
value were investigated, this factor is not included
influence how body panels build into an assembly.
in the list above because it had minimal influence
Consequently, a number of automotive manufac-
on variation. All of the manufacturers that supplied
turers have opted not to use C
steel coupons in this study had material properties
pk as the principal
measure of panel quality.
sufficiently controlled to virtually eliminate any influ-
ence on stamping variation.
This stamping report analyzes dimensional data to
characterize stamping variation by short-term
One of the objectives of this research is to under-
(part-to-part), long-term (die set to die set), and
stand the amount of variation experienced by dif-
mean bias (long-term deviation from design nomi-
ferent manufacturers and how they manage varia-
nal) to better understand process capability.
tion issues. Together, this information may be used
Numerous factors affect the observed variation in
to improve the overall validation process for
a stamping process, making stamping one of the
stamping and sheet metal assembly. The uncer-
more difficult processes to control. The complexity
tainty of sheet metal assembly clearly supports a
of stamping makes it extremely difficult to conduct
functional build approach where component qual-
rigorous experimental studies that can be general-
ity is determined by how it influences the assem-
ized beyond a given part and process configura-
bly. These methods are outlined in other reports by
tion. Thus, the knowledge base of stamping varia-
the Body Systems Analysis Project Team.
tion is very sparse, and a great opportunity exists
1
1.0 Introduction
additional clamps beyond 3-2-1 requirements.
These additional clamps in the fixtures over-con-
1.1 Motivation for Research
strain parts, thereby shifting mean dimensions.
• Stamping processes have so many input vari-
Leading global automotive manufacturers have
ables affecting variation, with some estimates at
been challenged with applying traditional design
well over 100, that even world-class stamping
practices to sheet metal design and assembly.
operations routinely operate outside of statistical
The goal behind this effort is to help achieve high
control, with non-stable process means between
quality car bodies with minimal lead-time and
die sets, especially on larger flimsy parts.
development costs. These practices include geo-
Consequently, measuring several parts from a
metric dimensioning and tolerancing (GD&T), vari-
single die set or run does not provide sufficient
ation simulation analysis, tolerance stack-up
information about the expected long-term varia-
analysis and setting quality standard targets for
tion of the process.
process capability, such as Cp and Cpk. Most man-
• Assembly processes often distort parts - some-
ufacturers have expressed concerns over the lim-
times closer to and sometimes further away from
ited success these methods have had on sheet
nominal - during assembly because of clamping,
metal processes including the assignment of
spot welding, and inconsistencies of part locat-
dimensional part tolerances, translating compo-
ing schemes. These distortions can shift panel
nent designs into tools that can make them and
mean dimensions and affect process variation,
predicting assembly conformance based on
resulting in a low correlation between stamping
stamping capability. Several important observa-
dimensions and assembly dimensions.
tions account for the limited success of applying
traditional design principles to these processes:
A purpose of this report is to provide a basic
understanding of stamping variation. The data in
• Manufacturers experience difficulties estimating
this report are intended to illustrate general char-
mean part dimensions, relative to nominal and
acteristics of stamping variation, and are not
process variation because these attributes are
intended to be a comprehensive data base to sup-
product and process co-dependent. Potential
port design. World-class automotive manufactur-
attributes affecting variation include material
ers that are most adept at designing, producing
properties (steel variations in gauge, grade, and
and assembling sheet metal are those who have
coatings), part geometry (size and shape), die
effectively learned from past designs, while man-
engineering and construction, and stamping
aging the variation in new parts and processes as
press variables. The infinite number of design
they become known. By researching a number of
and process possibilities make it nearly impossi-
stamping and assembly processes across several
ble to accumulate sufficient historical knowledge
manufacturers, this report begins to establish
for a designer to accurately assign tolerances
boundaries for the limits of variation that can be
that consistently meet future process capability.
expected under different situations. This report
• The lack of component rigidity allows less stable
examines the implications of this inherent stamp-
panels to conform to more rigid ones, making it
ing variation on several design and validation
difficult to predict final assembly dimensions
activities including:
based on component quality.
• Component dimensions that deviate from their
• Tolerance assignment,
design nominal cannot always be predictably
• Check point selection,
centered or shifted to the desired nominal with-
• Stamping process control limits,
out excessive rework costs. Moreover, this
• Process validation - die tryout,
rework may correct one particular deviation but
• Production part approval process - stamping,
adversely affect correlated points on the same
• Part measurement systems and measurement
part.
strategies, and
• Part measurement systems often have limited
• Assembly strategies with respect to part locating
capability to measure non-rigid parts without
and clamping.
2
The majority of the data in this report was collect-
design options, controlled experiments in stamp-
ed under production conditions, resulting in sever-
ing and assembly often have a limited value in
al advantages and disadvantages over a more
generalizing results.
controlled experiment approach. The advantages
are that the data actually reflect what can be
expected in production at normal line rates
1.2 Study Background
and with typical levels of process control.
The seven automotive manufacturers noted partic-
Generalizations about process variation are made
ipated in this study by providing data about their
where similar observations are seen over several
stamping and assembly processes. These manu-
case studies. The disadvantages are that the data
facturers, and the respective vehicles studied are
cannot be used, in most instances, to support
shown in Table 1 below. Note that companies
direct cause-and-effect conclusions, thus often
are referred to as A, B, C, D, E, F, and G in this
limiting observations to hypotheses. However,
report and they do not correspond to the order
given the infinite number of part and process
presented.
Data
Stamping Assembly
Company
Model
Collection
Location
Location
GM
Grand Am
1996
Lansing, MI
Lansing, MI
NUMMI (Toyota)
Corolla
1996
Fremont, CA
Fremont, CA
DaimlerChrysler
Neon
1997
Twinsburg, OH; Belvidere, IL
Belvidere, IL
Nissan
Altima
1997
Smyrna, TN
Smyrna, TN
Ford
Taurus
1997
Chicago, IL
Chicago, IL
Opel
Vectra
1998
Ruesselsheim, Germany
Ruesselsheim, Germany
Renault
Clio II
1998
Flins, France
Flins, France
Table 1. Participating Automotive Manufacturers
3
The scope of body panels included from all the
Dimensional studies at each manufacturer are
automotive manufacturers are:
based on the body side assembly and its major
stamped components. One participant provided
• One-piece body side outer,
data for panels on both the right and left body
• Two-piece body side outer,
side, resulting in a total of eight body-side assem-
• Center pillar reinforcement (B-pillar),
bly studies. The difference among manufacturers
• Front pillar reinforcement (A-pillar),
was the type of body side outer. Three used body
• Quarter outer panel,
side outers with an integrated quarter panel while
• Quarter inner panel,
the remaining manufacturers used two-piece body
• Roof rail outer,
sides. The other panels chosen in each body side
• Wheelhouse outer, and
assembly study, typically 4-5 mating parts,
• Windshield frame reinforcement.
depended on the design, but with the goal to
include critical rigid structural reinforcements
Figure 1 below illustrates a typical body side case
with thicker gauges greater than 1.25 mm.
study for a two-piece body side.
Windshield Frame
Reinforcement
Roof Rail
Center Pillar
Body Side
Front Pillar
Figure 1. Body Side Components Chosen for Company C
4
Table 2 below lists the body side components cho-
identified consistently by the same letter, A
sen from each of the automotive manufacturers for
through G, throughout the report. (See Appendix
this study. Each of the seven manufacturers is
for sketches of components in study.)
Company
Part
Number of
Steel
Steel
Identifier
Description
Dimensions
Gauge
Coupons
Body Side – One Piece
39
0.69
Quarter Inner
76
0.90
A
Wheelhouse Outer
38
0.61
Yes
Front Pillar Reinf
69
1.70
Center Pillar Reinf
60
1.44
Body Side – One Piece
104
0.90
Body Side Inner
54
0.80
Wheelhouse Outer
42
0.75
B
Center Pillar Reinf
17
1.00
Yes
Cowl Side
24
1.10
Roof Rail Inner
8
0.80
Roof Rail Outer
13
1.00
Body Side – Two Piece
60
1.10
Roof Rail Outer
22
0.90
C
Front Pillar Upper
9
1.85
Yes
Front Pillar Lower
8
1.85
Center Pillar Reinf
14
1.87
Windshield Side Inner
30
2.70
Body Side – Two Piece
17
0.73
Quarter Otr
14
0.82
D
Front Pillar Lower
6
Yes
Center Pillar Lower
6
Front Pillar Upper
15
Center Pillar Upper
4
Body Side – Two Piece
35
E
Front Pillar Lower
2
No
Center Pillar Lower
2
Roof Rail
2
Body Side – One Piece
38
0.90
F
Quarter Outer
11
No
Body Side Inner
6
Center Pillar
6
Body Side – Two Piece
54
Frt. 1.17;
(tailor welded blank)
Rr: 0.77
Body Side Inner
13
0.67
G
Windshield Side Inner
5
1.17
Yes
Front Pillar Reinf
6
0.97
Center Pillar Reinf
10
1.17
Table 2. Components Studied at Each Automotive Manufacturer
5
A consistent sampling plan was applied to each of
The measurement of the parts was conducted in a
the stamped panels. This plan was designed to
manner to reduce potential error. The 36 samples
ascertain both short-term and longer-term varia-
were collected over a period of several weeks and
tion under production conditions. Six panels were
set aside for measurement at one time. This
taken during each die set or production run for a
approach was intended to reduce measurement
given part: three consecutive panels near the
errors by using a single operator and a standard
beginning of the run and three consecutive panels
measurement protocol of a loading, clamping and
near the end of the run. This six panel sampling
measurement sequence. To measure the body
plan was then repeated over six separate die sets,
side outer and many of the other inner panels, five
thus producing a total case study of 36 panels (6
manufacturers used CMMs, while E and F used
per die set x 6 die sets = 36 panels). This sam-
hard fixtures. A few of the smaller parts were
pling plan was executed on all of the major panels
measured on hard checking fixtures using
in each case study. A few smaller reinforcements
datamyte collection devices and measurement
had less than 6 die sets. The length of each die set
probes. In all cases, part locating was based on
varied by participant, but tended to be greater
the standard checking fixtures used by each man-
than four hours. The time between each die setup
ufacturer for internal quality monitoring. A few
also varied and was typically between 2 and 7
chose to modify their CMM measurement routines
days. This sampling plan allowed the calculations
to include additional dimensions to provide a more
of short-term variation, or variation across consec-
comprehensive geometric database.
utive panels, and long-term variation, or variation
both within and between die sets.
One challenge with comparing manufacturers in
this study was the significant differences in meas-
Several of the manufacturers also collected panel
urement systems. These differences relate prima-
coupons and process variable data to see if rela-
rily to the locating and clamping of parts in the fix-
tionships could be found between the material or
tures. Some manufacturers attempt to minimize
equipment setup and stamped panel variation.
the influence of the fixture on the part by minimiz-
They collected a steel blank at the destacking side
ing the number of clamps and clamping pressure,
of the press while a contiguous sample of 3 pan-
while others intentionally over-constrain their parts
els was collected. At the same time, several com-
for measurement. Manufacturers attempting to
panies collected data on the process, such as ton-
reduce the influence of the measurement system
nage and cushion pressure in the draw die. The
use a minimal number of clamps and locators to
actual variables collected at each manufacturer
obtain an adequate measurement system repeata-
varied by part and die design. The steel coupons
bility and reproducibility, or gage R&R. Other man-
were collected and tested for several properties,
ufacturers more readily obtain high gage repeata-
including R-value, n-value and blank gauge varia-
bility and reproducibility by adding a larger num-
tion.
ber of clamps. This approach, masks variation in
the panel, making the measurement system less
able to detect variation. The difference in meas-
urement systems requires caution when generaliz-
ing variation across manufacturers.
6
2.0 Stamping Variation
out, a mean dimension is -0.65 mm or 0.65 mm,
then its mean bias is 0.65 mm. If the mean from
2.1 Components of Variation Explained
two die sets are -0.40 mm and 0.10 mm, assum-
ing equal sample size from each die set, then the
Dimensional variation from the stamping process
grand mean is -0.15 and the mean bias of those
may be categorized into a number of components.
two die sets is 0.15 mm ([-0.40 + 0.10] ÷ 2 =
Generally, different variation components are
0.15).
attributable to different sources and often have a
different impact on downstream operations. The
• Part-to-part variation is also referred to as the
following are the general components of variation
short-term or inherent variation. It is the amount
that will be used throughout this report. Figure 2
of variation that can be expected across con-
below illustrates each variation component.
secutive parts produced by the process during a
given run. The assumption is that the variation is
• Mean bias deviation is the process bias relative
a reflection of numerous incidental random vari-
to the design nominal. Mean bias is the absolute
ables over a short-term and is not affected by
value of the average deviation from nominal.
any special causes of variation such as a
When a process is centered exactly at its nomi-
change in the steel coil or process settings. This
nal dimension, its mean bias is zero. If, after a
part-to-part variation is denoted as part-part.
single die set, for example at the die source try-
Tryout
Regular Production
Part-to-Part
Total Variation
Die Source to Home Line
Mean Shift
Upper
1.5
Specification
1
0.5
Mean Shift
Nominal
0
Mean Bias
-0.5
Die Source Tryout
Mean Bias
-1
Lower
Specification
Legend:
Individual Measurements
Mean of the Stamping Run
Figure 2. Components of Variation
7
Estimates for part-to-part variation for the
end, which may be several hours later. This
36-panel study are based upon the 12 sub-
change in a mean dimension may occur due to
groups of 3 consecutive panels. Again, the
process changes during a run such as a steel coil
assumption is that the process is stable during
change, changes in operating speeds or tonnage,
three consecutive parts.
or adjustments to draw lubrication. If a mean
dimension significantly shifts during a run due to
• Run-to-run variation is commonly referred to as
some special cause, the stable mean assumption
mean-shift variation. It is the measure of the
is violated and begin-end variation is greater than
repeatability of the die setting process, and its
part-to-part variation. We denote this variation as
derivation is based on the variation of the mean
begin-end. Estimates for begin-end of run variation
dimension across two or more die sets. Run-to-
are based on the variation of the mean dimension
run variation is denoted as run-to-run. Estimates
from the beginning to the end of each die set.
for run-to-run variation are based on the varia-
tions in mean dimensions between die sets.
Figure 3 below illustrates a run chart for a single
stamping dimension with unusually large variation.
• Begin-end of run variation is another type of
Each of the three variation components, part-to-
mean-shift variation in that it is a measure of the
part, run-to-run, and begin-end variation, is illus-
stability of the process mean within a run. Since
trated in the plot.
stamping production runs can be long, the mean
of the run can change from the beginning to the
Run Chart of a Stamping Check Point
run 1
run 2
run 3
run 4
run 5
run 6
4
within run
3
(mm)
mean shift
run-to-run
alue
2
mean shift
Part -
1
to- part
0
Measurement V
-1
-2
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35
Measurement Values
Mean of Group
Figure 3. Potential Sources of Stamping Variation
8
• Mean shift variation is the sum of the run-to-run
2.2 Calculating Components of Variation
variation and begin-end of run variation. Since
Using ANOVA
run-to-run variation and begin-end variation are
An efficient method for estimating the components
both forms of mean instability, they can be com-
of variation is through Analysis Of Variance, or
bined into one variation number that is called the
ANOVA. Briefly, the parameters for an ANOVA
mean shift variation. The mean shift variation is
model for this sampling plan were defined in the
denoted by mean shift, where 2 mean shift = 2 run-
following manner:
to-run + 2 begin-end. In most cases with stamping,
the run-to-run variation dominates the within run
d = number of die sets
= 6
variation ( 2 run-to-run >> 2 begin-end), so rather
s = number of groups
= 2 samples of 3
than separate the two, the total mean shift varia-
per die set
per batch
tion ( 2 mean shift) is used.
n = sample size per group
= 3 consecutive
panels
• Total variation is the sum of part-to-part variation
and mean shift variation. This represents the
The total number of panels sampled is equal to
total variation that the downstream assembly
dsn, or 6x2x3 = 36. This ANOVA model estimates
process is subject to over the long-term. The
part-to-part, run-to-run and begin-to-end of run
total variation is denoted total. Equation 1 and
variation using the expected Mean Squares (MS).
Figure 4 below summarize variation partitioned
The equations shown in Table 3 on page 10 are
into components.
used to estimate the sources of variation. If
both factors of run-run and begin-end of run are
Total variation is equal to the sum of the compo-
statistically significant, then a begin-end of run
nents of variation:
and a run-run variance may be calculated. If only
one of the two factors is significant, then only that
2 total = 2 part-to-part + 2 mean shift, or
variable will have a variance estimate. Finally, if
neither of the two factors is significant, then all of
Equation 1
the total variance may be attributed to part-part
2
variation.
total = 2 part-to-part + 2 run-to-run + 2 begin-end
2
total
2
2
part - to - part
mean shift
(inherent process
variation)
2
2
run - to - run
begin-end
Figure 4. Total Variation Partitioned into Components
9
Variation
Formula
Description
2
Equation 2
=
MSE
mean squared error
part-to-part
Equation 3
2
=
(MSBE – MSE)
(mean squared – mean
begin-end run
begin-end
n
squared error) – sample size
Equation 4a*
2
=
(MSRR – MSBE)
(mean squared – mean
run-to-run
run-to-run
sn
begin-end
squared error ) – (number
of samples x sample size)
Equation 4b*
2
=
(MSRR – MSE)
run-to-run
run-to-run
(mean squared – mean
sn
squared error) – (number of
samples x sample size)
*Note: If all variation sources are significant, use Equation 4a. If begin-end factor is not significant, use Equation 4b.
Table 3. Formulas for Calculating Components of Variation
The following table illustrates an application of an
ANOVA analysis for a stamping dimension.
Die Set
Group
Panel 1
Panel 2
Panel 3
Sample Average
1
begin run
0.13
0.16
0.17
0.15
1
end run
0.10
0.03
0.03
0.05
2
begin run
0.06
0.16
0.08
0.10
2
end run
0.13
0.08
0.23
0.15
3
begin run
0.18
0.72
0.14
0.34
3
end run
0.19
0.49
0.16
0.28
4
begin run
-0.35
-0.43
-0.47
-0.42
4
end run
-0.41
-0.39
-0.38
-0.40
5
begin run
-0.32
-0.31
-0.35
-0.33
5
end run
-0.32
-0.33
-0.29
-0.31
6
begin run
-0.16
-0.10
-0.21
-0.16
6
end run
-0.20
-0.17
-0.20
-0.19
Grand Mean (Mean Bias)
-0.06 (.06)
Table 4. 36-Data Samples for a Stamping Dimension
10
The ANOVA output for this data is summarized in
than . For this data set, the Mean Squared Error
Table 5 below, or using the Statistical Software
for the begin-end factor is not significant. That is,
Package SPSS based on Type I error, = 0.05.
the mean does not significantly change from the
Note that a significant variable has a value less
beginning to the end of the stamping run.
Source
df
Mean Square
F
Significance
Run-Run
Hypothesis 5
.479a
103.81
.000
Error
6
4.6E-03b
Run-Run x
Hypothesis
6
4.6E-03b
.352
.901
Begin-End Run
Error
24
0.013c
a.
Mean square run-to-run
b.
Mean square begin-run of run
c.
Mean squared error
Table 5. SPSS Output Calculations for Mean Squared Errors (all factors)
Since the begin-end factor is not significant, the
put is shown in Table 6, where again significance
ANOVA model must be revised and the Mean
is based on a Type I error of 0.05.
Square Errors recalculated. The revised SPSS out-
Source
df
Mean Square
F
Significance
Run-Run
Hypothesis 5
.479
25.836
.000
Error
30
0.011
Table 6. SPSS Output Calculations for Mean Squared Errors without Begin-End Factor
The mean squared errors in Table 6 above may be
through 4 (note: begin-end = 0 because this factor
used to estimate the variation for each of the com-
is not significant for this dimension). The variation
ponents of variation present using Equations 2
estimates are shown in Table 7 below.
Variation Source
Mean Squared Error Calculation
Variance (mm2)
Standard Deviation (mm)
Part-to part
MSE
0.011
0.11
Begin-end run
Not significant
0.00
0.00
Run-to-run
(0.479 – 0.011) ÷ (2)(3)
0.078
0.28
Total Process
(.078+.011)
0.089
0.30
Table 7. Summary of Components of Variation Calculations
11
Table 8 below summarizes the components of vari-
clamps. It will be shown in the following section
ation and the mean bias at 12 measurement loca-
that as the number of clamps increases on a
tions for the Company A Body Side Outer as,
checking fixture, the amount of observed variation
shown in Figure 5. One noteworthy finding is the
decreases due to the masking of variation by the
large range in variation for part-to-part, 0.01 to
clamps. Table 8 also indicates that part-to-part
0.48, and run-to-run, 0.00 to 0.18, across different
variation, 65.4%, and run-to-run variation, 30.3%,
measurement locations. This contrast is attributa-
are much greater than begin-end run variation of
ble to the differences in location/axis on the part
4.3% for these dimensions.
and also to the proximity of measurement system
#10 & #11
#1
#12
#9
#5
#4
#7
#3
#2
#8
#6
Figure 5. Body Side Outer for Company A: 12 Measurement Locations
Measurement
Component of Variation (mm2)
Part-to-Part
Begin-end
Run-to-run
Total Process
Mean Bias (mm)
Location
Direction
( 2 part-to-part)
( 2 begin-end)
( 2 run-to-run)
( 2 total)
1
Y
0.09
0.00
0.06
0.16
0.313
2
Y
0.04
0.00
0.00
0.04
0.342
3
Z
0.01
0.01
0.00
0.02
0.786
4
Y
0.05
0.03
0.00
0.08
1.187
5
X
0.48
0.00
0.00
0.48
0.851
6
Z
0.05
0.05
0.00
0.10
3.673
7
Z
0.06
0.00
0.03
0.09
1.609
8
Y
0.14
0.00
0.18
0.32
2.530
9
X
0.34
0.00
0.17
0.51
1.139
10
Y
0.01
0.00
0.00
0.01
0.618
11
Z
0.02
0.00
0.03
0.05
0.837
12
Y
0.14
0.00
0.18
0.32
0.675
Average
0.12
0.10
0.05
0.18
1.130
Percent of Total
65.4%
4.3%
30.3%
100.0
––
Table 8. Variance Summary for 12 Body Side Dimensions
12
2.3 Description of the Sources of Stamping
One area of this research that is often not exam-
Variation
ined is the effect of process variables on mean
Extensive research has been conducted regard-
conformance. Most of the research on reducing
ing dentification and elimination of the sources of
stamping mean biases has been directed toward
variation associated with stamping sheet metal.
metal forming and die design. Little research
The stamping process is complex, with many
exists on eliminating mean biases once a die has
variables that can influence variation. One related
been made and the actual mean biases become
research effort, by John Siekirk,(1) identified 30
known. Even less attention has been given to non-
major factors, and then classified them into the
die related influences on mean bias such as the
following seven categories:
measurement system effects. Among the factors
that influence mean bias include:
• Blank condition,
• Blank lubrication,
• Measurement System:
• Stamping press variables,
• Clamping sequence
• Metal properties,
• Clamping forces
• Die condition,
• Part locating (datum)
• Miscellaneous and
• Product Design:
• Interactive variables.
• Part geometry (size and complexity)
• Part rigidity (shape and gage)
Because this body side research project investi-
• Check point location
gated process variation under production condi-
• Process:
tions, only a limited number of process and mate-
• Press setup and control of process
rial variables could be collected. More important-
variables (see above)
ly, process variables were not purposely altered.
• Changes in stamping presses
Therefore, only inferences can be made between
(e.g., tryout to production presses)
stamping variation and the observed variability in
• Material handling and storage
process variables. In many of these case studies,
the process and material variables were under
control. As a result, our findings do not necessar-
ily identify those variables that could affect part
variation, but rather which variables explain the
dimensional mean shifts in these case studies.
1 Process Variable Effects on Sheet Metal Quality, Journal of Applied Metalworking, American Society for Metals,, July 1986.
13
3.0 Analysis of Stamping Variation
does as well, sometimes in an unpredictable way.
Another difficulty is trying to rework dimensions
3.1 Mean Conformance
exactly to their design nominal. Basically, there is
a limited ability to hit the nominal dimension even
One of the greatest challenges in die making and
after rework. A final difficulty concerns the ability to
stamping is minimizing mean biases for dimen-
measure a part and to know precisely what the
sions on stamped parts. As defined earlier, the
mean bias really is. In addition to die processing,
mean bias is the absolute value of the average
mean dimensions also are affected by the number
deviation from nominal. Ideally, manufacturers
and positioning of clamps in measurement fix-
would produce every stamped component such
tures. Because of the many variables in forming a
that each dimension is, on the average, at the
part, such as changes in stamping press variables
specification nominal. By doing so, design capa-
and steel properties, and the limited ability to
bility (Cpk) would be maximized for a given level of
measure sheet metal, ascertaining the precise
process variation. Achieving minimal mean biases
mean bias can be very difficult - both before and
in stamping also facilitates the “tune-in” of assem-
after a die change. Manufacturers often face a
bly tooling, which is initially designed for parts at
complicated decision in determining when to
nominal, and increases the likelihood of producing
rework a die or when to allow a mean bias to
dimensionally acceptable assemblies within the
remain (see Body System Analysis Project Team
shortest possible lead-time. The problem is that no
report “Event-Based Functional Build: An
manufacturer in the world has demonstrated the
Integrated Approach to Automotive Body
capability to produce stamped body parts without
Development”).
mean biases.
3.1.1 Benchmark Comparison - Body Side
Manufacturers who have minimized their mean
Outer and Inner Panels
biases relative to their competition appear to
maintain a competitive advantage in terms of cost,
Figure 6 on page 15 shows a histogram for 143
quality and lead-time. To achieve lower mean
mean dimensions across 5 parts at Company C.
biases, manufacturers employ a combination of
Several observations may be made from these
technology and applied learning and limit the evo-
data:
lution of product design to reduce uncertainty.
Future product designs with uncertain forming
• The distribution of mean dimensions is approxi-
challenges might be subject to soft tool evaluation
mately normal. Assuming the measurement sys-
in order to evaluate metal forming and die design
tem does not unfairly influence mean deviations,
before production tools are machined.
this finding suggests an inherent variation in the
ability to design and construct dies to produce
Modifying hard dies, or die rework, after they have
part dimensions at nominal.
been machined to reduce mean biases represents
• The distribution of mean dimensions is centered
one of the most difficult tasks in getting dies
approximately at zero (i.e., average mean bias
approved for production. Manufacturers attempt-
is near zero). This is as expected since the
ing to rework dimensions to reduce mean biases
distribution is normal and the die maker's target
face several challenges. First, since a stamped
is to have zero bias.
part has a continuous surface, reworking a die to
• Approximately 10% of mean values have a bias
shift one dimension may affect other areas of the
greater than 1.0 mm, and about 35% have a
part. Many areas of a part are interdependent, so
bias greater than 0.5 mm.
that when one dimension changes another area
14
45%
40%
65%: [Mean] <0.5 mm
35%
30%
25%
20%
15%
10%
5%
% of Dimensions (143 total)
0%
<-1.25
-1.25~-.75
-0.75~-.25
-0.25~+.25
0.25~.75
0.75~1.25
>1.25
Range of Mean Deviations
Figure 6. Histogram of Mean Values across 5 Parts for Company C
In general, the amount of spread in the mean dis-
less rigid panels, such as one-piece body side
tribution will vary significantly by type of part.
outer and quarter inner. The less rigid panels have
Larger, less rigid panels like the body side outer
larger mean biases and also a greater dispersion
often have significantly more dimensions with
in mean deviations than the rigid panels. This is
large mean biases than rigid panels, or panels
evident in comparing reinforcements to a one-
with blank thickness greater than 1.5 mm.
piece body side, and it also occurs in comparing
a one-piece body side of 0.69 mm gauge to a two-
Figure 7 below compares the mean deviations for
piece body side of 1.10 mm gauge as shown in
smaller rigid reinforcement panels at Company A,
Figure 8 on page 16.
such as A and B pillar reinforcements, to larger,
100%
Non-rigid: 44% [Mean] <0.5
80%
Rigid:
83% [Mean] <0.5
60%
40%
20%
% of Dimensions
0%
<-1.5
-1.5 ~ -0.5
-0.5 ~ 0.5
0.5 ~ 1.5
> 1.5
Range of Mean Deviations
Body Side Otr/Qtr Inr (Non-rigid)
Front/Center Pillar Reinforcements (Rigid)
Figure 7. Mean Conformance: Rigid vs. Non-Rigid Panels
15
80%
60%
One-Piece: 31% [Mean] <0.5
Two-Piece: 65% [Mean] <0.5
40%
20%
% of Dimensions
0%
<-1.5
-1.5 ~ -0.5
-0.5 ~ 0.5
0.5 ~ 1.5
> 1.5
Range of Mean Deviations
One-Piece Body Side
Two-Piece Body Side
Figure 8. Mean Conformance: Two-Piece Body Side Panel vs. One-Piece
Table 9 below summarizes the mean bias for the
G use constrained measurement systems, and all
body side outer panels at each of the automotive
have lower biases for the body side outer panel. In
manufacturers. These data suggest several gener-
addition, the same body side panels at Company
alizations. First, larger one-piece body side outers
B exhibited less mean bias when measuring the
with integrated quarters tend to exhibit greater
parts in a more constrained fixture. A major influ-
biases than two-piece body sides. Second, manu-
ence of the constrained checking system is that
facturers using constrained measurement systems
extreme mean biases, those greater than 1.0 mm,
such as excess part locating clamps have signifi-
are greatly reduced.
cantly less mean deviations. Companies E, F, and
Body Side
# cross car
Average
% Dimensions % Dimensions
% Dimensions
Company
Type
clamps in fixture
[Mean]
[Mean] <.5
[Mean] >1
[Mean] > tol (t)
A
Integrated Quarter
11
1.10
34%
56%
66%
B (remeasured)
Integrated Quarter
14
0.73
49%
29%
39%
C
Two-piece
7
0.51
65%
15%
5%
D
Two-piece
8
0.88
42%
39%
39%
E
Two-piece
22
0.36
74%
3%
14%
F
Two-piece
16
0.31
84%
3%
39%
G*
Integrated Quarter
17
0.37
69%
2%
28%
* Over-constrained (excess clamps) during measuring
Table 9. Mean Conformance by Company
16
The effect of a constrained measurement system
across several part types. Although the clamping
is limited to larger, less rigid panels since addi-
strategy may be correlated with mean bias in the
tional clamps beyond 3-2-1 on rigid parts with
body outer panels, the same cannot be done for
gauge greater than 1.5 mm have little or no effect.
rigid panels.
Table 10 below compares mean conformance
Body Side
Non-Rigid
Rigid Inner
Body Side
Non-Rigid
Rigid Inner
Outer
Inner Panels
Panels
Outer
Inner Panels
Panels
Average
Average
Average
% Dimensions % Dimensions
% Dimensions
Company
[Mean]
[Mean]
[Mean]
[Mean]>1
[Mean]>1
[Mean]>1
A
1.10
0.79
0.22
56%
30%
1%
B
0.90
0.56
no data
33%
16%
no data
C
0.51
0.31
0.34
15%
0%
3%
D
0.88
0.32
0.27
39%
0%
0%
E*
0.36
0.35
no data
3%
0%
no data
F*
0.31
0.38
no data
3%
17%
no data
G*
0.37
0.39
no data
2%
6%
no data
* Over-constrained (excess clamps) during measuring
Table 10. Mean Bias by Type of Part
3.1.2 Mean Bias and Part Tolerances
Another observation across study participants is
Another contrast across manufacturers is the
that although Companies A through C use Cpk as
assignment of part tolerances. When comparing
their principal buyoff criteria, they do not achieve
manufacturers, the same physical dimension on a
greater mean conformance. In fact, it might be
body side may have a tolerance of +/- 0.3 mm at
argued that the use of Cpk at Company C has led
one manufacturer and +/- 1.25 mm at another.
primarily to wider tolerances to achieve greater
Table 11 on page 18 shows the typical tolerance
Cpk conformance, not greater mean conformance.
for the body side panel and the percentage of
Another finding is that only those manufacturers
dimensions whose mean biases exceeds the tol-
using constrained measurement systems
erance limit. On average, more than 30% of the
assigned tolerances less than ±0.70 mm.
dimensions in companies A through G have their
Company F assigns the tightest tolerance at ±0.3,
mean bias outside the tolerance. It is important to
but uses a constrained measurement system and
note that whenever the mean bias exceeds the tol-
also has a two-piece body side which tends to
erance limit, at least 50% of the panels have that
have lower mean bias than the larger one-piece
dimension outside of tolerance. It is clear that a
design.
significant number of vehicles are being produced
with acceptable final body quality, but with a sig-
nificant number of body panel dimensions out of
tolerance.
17
Body Side
Typical
# cross car
Average
% Dimensions
% Dimensions
Company
Type
tolerance
clamps in fixture
[Mean]
[Mean] >tol (t)
Cpk > 1.33
A
Integrated Quarter
+/- 0.7
11
1.10
66%
15%
B
Integrated Quarter
+/- 0.7
14
0.73
39%
80%
C
Two-piece
+/- 1.25
7
0.51
5%
75%
D
Two-piece
+/- 1.0
8
0.88
39%
23%
E
Two-piece
+/- 0.5
22
0.36
14%
43%
F
Two-piece
+/- 0.3
16
0.31
39%
29%
G
Integrated Quarter
+/- 0.5
17
0.37
28%
37%
Table 11. Mean Conformance and Tolerances
3.1.3 Benchmark Comparison - Tryout versus
these findings with mean biases experienced at
Production
production buyoff, the data are consistent.
Many manufacturers apply common dimensional
Although the other four manufacturers did not pro-
validation procedures and criteria to all body pan-
vide tryout data, discussions with their personnel
els even though the expected mean bias differs by
suggest that their mean conformance distributions
type of panel, whether rigid versus non-rigid or
in production also corresponded to die tryout. The
small/simple form versus large/complex form.
main point is that even though manufacturers may
Table 12 below depicts mean conformance across
adjust some mean biases to correct build con-
multiple parts during regular production at
cerns, the overall ability to produce mean dimen-
Companies A, B and C. These data suggest that
sions at nominal does not significantly change
manufacturers produce stamped parts with 50-
from die tryout.
70% of dimensions within 0.5 mm. Comparing
Tryout
Production
Tryout
Production
Company
% Dimensions
% Dimensions
% Dimensions
% Dimensions
[Mean]<0.5
[Mean]<0.5
[Mean]>1
[Mean]>1
A
59%
63%
26%
22%
B
51%
53%
15%
23%
C
64%
66%
13%
10%
Table 12. Summary of Mean Bias: Tryout vs. Production (Case Study Parts)
3.1.4 Mean Bias Stability over Time
each die setup provides another opportunity to
Another important consideration regarding mean
estimate the mean bias. Most of the data in this
bias concerns its stability over time. Most automo-
study was collected during production, a year or
tive manufacturers first evaluate mean bias during
more after the dies were initially brought to the
die source tryout. A decision is eventually made to
home line. An important question affecting dimen-
move the die to the production press, often
sional validation is how does the estimate of mean
referred to as the “home line”, where another esti-
bias change from tryout to the home line and then
mate of the mean bias is made. Finally, as the dies
to future production.
are repeatedly run on the home line for production,
18
Table 13 below examines changes in part dimen-
tainty of change is one reason manufacturers rec-
sional means between die source tryout and home
ognize that it is necessary to re-evaluate the
line tryout. The first two data sets are based on the
dimensions on a part when the dies are trans-
case study parts at two of the manufacturers. Two
ferred to the home line. Interestingly, a similar
more extensive studies of die source to home line
number of dimensions shift toward nominal as
mean shifts are also included. These data suggest
opposed to away from nominal, or the shift in
that approximately 30% of dimensions shift at least
mean bias from tryout to the home line appears
0.5 mm when the dies are moved from the tryout
random.
presses to the home line. The amount and uncer-
Die Source to Home Line
Company
# of Parts/
Median Shift [Die source
% Dimensions
of the Dimensions
# Dimensions
mean. Home line Mean]
[Mean Shift] >0.5
with shift >0.5
% closer % away
B-1
4/104
0.30
30%
40%
60%
C-1
5/86
0.20
25%
68%
32%
C-2
47/652
0.23
30%
63%
37%
C-3
26/182
0.27
28%
50%
50%
Overall
0.25
Table 13. Comparisons of the Change in Mean Bias from Tryout to Home Line
Figure 9 below compares the mean dimension at
program. Table 14 on page 20 shows further that
time of part approval versus the production mean
of the dimensions with significant mean differ-
approximately one year later at Company C. For
ences, a similar number shifted closer to nominal
the parts in this study, nearly 50% of the dimen-
than away from nominal.
sions shifted more than 0.5 mm over the life of the
Correlation, R = .12
2.00
1.50
1.00
0.50
0.00
-0.50
oduction Mean
Pr
-1.00
-1.50
-2.00
-2.00
-1.00
0.00
1.00
2.00
Home Line Mean at Part Approval
Figure 9. Correlation of Mean at Part Approval vs. Production Mean
19
Home Line Approval Mean to Production Mean
Company
Median Shift
% Dimensions
% of the Dimensions
[Home line mean- [Mean Shift] >0.5
with shift >0.5
Production]
% closer % away
A
0.48
48%
41%
59%
B
0.57
53%
45%
55%
C
0.35
40%
44%
56%
Table 14. Change in Mean from Home Line to Long-term Production
These findings suggest that automotive body
variation observed in assembly actually becomes
parts continue to evolve from die source tryout,
higher.
through home line tryout, and even through regu-
lar production. Although some of these dimension-
In this example, maintaining a stable mean over
al changes are intentional based on die rework to
time appears more important than the magnitude
correct a problem, the majority are not. They shift
of the mean deviation. Similar to stamping,
because of lack of process control or die rework or
assembly processes evolve over time to match
maintenance in another related dimensional area.
stamping mean deviations. If these mean devia-
tions change significantly, assembly processes
Interestingly, some dimensions shift away from
will likely experience problems. Thus, manufactur-
nominal with no apparent impact on the assembly
ers must develop a better understanding of how to
process. In addition, some dimensions may shift
minimize mean instability. Fortunately, mean insta-
significantly closer to nominal, greater than 1 mm,
bility is not inherent to a process like part-to-part
but with an adverse affect on the final assembly.
variation, rather it is caused by some special influ-
Figure 10 below depicts a stamping dimension
ence such as a process variable change or die
that shifts 1.3 mm between stamping runs four and
rework. Thus, a potential exists to control these
five. Even though this shift is toward nominal, the
special causes.
run 1-4
run5-6
4.5
Stamping
.30
.19
4
Assembly
.24
.80
3.5
mm
3
2.5
2
1.5
1
Measurement,
1.3 mm mean shift
0.5
0
1
2
3
4
5
6
Stamping Run #
Stamping
Assembly
Figure 10. Effect of Stamping Mean Shift on Body Side Assembly
(Note: above dimension is coordinated between stamping and assembly)
20
3.1.5 Impact of Shipping on Mean Bias
An experiment was performed at Company A
One final investigation into the factors influencing
where four inner panels were measured both
mean bias looked at the impact of material han-
before and after shipment. The measurement sys-
dling, including:
tem used the same locating fixtures and CMM pro-
grams at both the production and assembly
• Racking of parts and container design for con-
plants; however, different operators performed the
sistency and impact resistance,
actual measurements. The panels were shipped in
• Time lag as stressed parts become stress
their specified containers, via truck over several
relieved and
hundred miles. Two small parts were dropped into
• Movement of parts - including manual, forklifts,
bins, one larger wheelhouse outer was stacked
and truck and rail mass transit, all of which
and the fourth part, the quarter inner, was shipped
impact the distortion of parts through vibration.
in a special rack. The results are shown in Table 15
below.
Number of
Average Mean
% of Dimensions With
Part
Check Points
Bias Shift (mm)
Mean Bias Shift>0.2mm
Wheelhouse Outer
69
0.89
76%
Quarter Inner
91
0.10
15%
B-pillar Reinforcement
59
0.16
19%
A-pillar Reinforcement
70
0.08
6%
Table 15. Summary of Panels Measured Before and After Shipping
This experiment indicates a potentially significant
3.2 Stamping Process Variation
impact of shipping on mean bias. It should be
3.2.1 Benchmark Comparison - Part-to-Part
noted, however, that the impact of shipping is con-
Variation
founded by different measurement operators.
Reproducibility is a potential source of gage error
Part-to-part or short-term variation is a measure of
in this study because the same operator did not
the inherent variation for a particular product, or
measure the panels before and after shipping.
set of dies, and process, or stamping press line.
However, reproducibility of a CMM based on an
Key variable set-up parameters, such as shut
automatic program is generally insignificant.
height, lubrication, cushion pressure, etc, and
incoming steel coils or blanks are presumed to be
The wheelhouse outer suffered the greatest mean
constant or consistent. Several variables may
shift with an average change of 0.89 mm. The
explain differences in part-to-part variation across
wheelhouse panels at the bottom of the stack had
companies. Some of these differences were inves-
the largest dimensional differences, reflecting the
tigated, including:
accumulated weight effect. For the quarter inner,
special racks are used which clearly help reduce
• measurement and clamping system,
the shipping effect. The more rigid panels also
• check point location/axis on the part, and
experienced less shipping impact, with most
• part rigidity, size and material thickness.
mean dimensions shifting less than 0.2 mm.
Similar to the die source tryout to home line analy-
Table 16 on page 22 summarizes part-to-part vari-
sis, the direction of the mean shift appears ran-
ation for the body side outer panels for each of the
dom, or equally likely to get closer to or further
manufacturers studied. A comparsion of variation
away from nominal.
across manufacturers is again difficult because of
21
the different measurement strategies as demon-
ment clamps will likely reduce the observed part
strated by the number of clamps. The three manu-
to part variation for large, non-rigid parts. Both
facturers using the most clamps (E, F, and G) have
the average and the extreme variation points, or 6
the lowest part to part variation. In addition, part to
part-to-part greater than 1.5 mm, appear to be sig-
part variation at Company B is significantly lower
nificantly reduced by the additional secondary
when using a more constrained measurement sys-
locating clamps.
tem. These data suggest that adding measure-
Body Side
# cross car
Average
95th Percentile
% Dimensions
Company
Type
clamps in fixture
6 part-part
6 part-part
6 part-part >1.5
A
Integrated Quarter
11
1.14
2.90
20%
B (remeasured)
Integrated Quarter
14
1.09
2.35
23%
C
Two-piece
7
0.99
1.89
18%
D
Two-piece
8
0.99
1.57
10%
E*
Two-piece
22
0.48
0.81
0%
F*
Two-piece
16
0.32
0.50
0%
G*
Integrated Quarter
17
0.40
1.08
0%
* Over-constrained (excess clamps) during measuring
Table 16. Part-to-Part Variation for Body Side Outer Panels
Note: 95th percentile is the level of variation where 95% of the dimensions on the part are less than this amount.
The type of body side, one-piece versus two-
and rigid body side inner panels. The body side
piece, also appears to affect variation. Companies
outer panel is the largest and one of the lightest
A through D use roughly the same number of
gauge panels, varying from 0.69 mm to 0.90 mm
clamps, but have two different body side styles,
thickness. The non-rigid body side inner panels
integrated quarter panel and two-piece. It appears
are arbitrarily limited to 1.5 mm thickness and are
that the two-piece body side results in lower aver-
smaller than the outer panel. These panels include
age part-to-part variation than the larger and more
the quarter inner, wheelhouse outer and roof rails.
complex integrated quarter body side by about
The third category consists of small, heavy-gauge
10%. The same relationship is seen among com-
parts, including the A- and B-pillar reinforcements.
panies E, F, and G using the more constrained
measurement approach. Of these, company G
Figure 11 on page 23 plots the average standard
with the larger body side has the highest 95th per-
deviation, or sigma, for all parts studied at the
centile part to part variation. Follow-up analysis at
seven manufacturers. The changes in these
company G indicates that most of their high varia-
groupings from large and flimsy to smaller and/or
tion dimensions are in non-stable measurement
more rigid can be seen to correlate with the aver-
areas in the tail area of the body side outer panel.
age amount of part-to-part variation. As panels
become smaller and more rigid, their part to part
Since part-to-part variation differs according to
variation decreases. In addition, Figure 11 sug-
body side style, it would be expected to vary
gests that the body side panels with the lowest
according to part size and rigidity for non-body
variation are from manufacturers using more
side outer panels. The panels in these case stud-
measurement clamps, thus masking some of the
ies may be grouped into three categories: body
actual process variation.
side outer panel, non-rigid body side inner panels
22
0.30
integrated
0.25
quarters
t
6 =1
t-par
0.20
par
e
0.15
g
0.10
vera
A
0.05
small,
E, F, G
simple
0.00
Body Side
Non-Rigid
Rigid (guage>1.5)
Figure 11. Average Variation (Standard Deviation) by Type of Part
Another difference among manufacturers is the
illustrate the impact of dimension location, Table
number and location of dimensions measured.
17 below shows the body side variation for com-
Two manufacturers, companies D and E, collect
panies D and C. At company D, 24% of their body
less data on their stamped panels than the other
side dimensions have an average standard devia-
manufacturers, and primarily collect data from
tion greater than 0.2 mm. Company D measures
points located in more rigid localized part areas.
near the A- and B-pillars and on the flanges in the
Although a body side outer panel tends to be flim-
door openings. In contrast, company C measures
sy, certain areas in highly formed sections of the
dimension throughout the body side and has 73%
part, such as the door openings, are typically
of their dimensions exceeding 0.2 mm. However,
more rigid than the tail or wheelhouse areas.
when comparing dimensions in similar locations,
Control of these more rigid areas often is more
the variability at company C more closely resem-
important than other areas because they are less
bles company D. Thus, the expected variation on
likely to conform to reinforcements during assem-
a stamped panel appears dependent upon where
bly. As has been shown, dimensions on less rigid
the dimension is located and how rigid the part is
parts tend to have greater variation. In order to
at that location.
Selected
Company
Dimension
<0.2
>0.2
D
14
76%
24%
C
40
27%
73%
C
14 (common with D)
60%
40%
Table 17. Effect of Dimension Location on Variation
23
3.2.2 Variation Over Time
to-part variation typically increases from part
In theory, part to part variation produced from a set
approval runs to regular production. These data
of dies on the same press line should remain con-
suggest that the average six sigma increases from
stant over time. In practice, part-to-part variation
0.8 mm to 1.2 mm after more than a year in pro-
does vary for some dimensions. Variables that
duction. The most likely explanation for this differ-
may affect part-to-part variation over time include:
ence is that operating conditions at buyoff are
substantially more controlled than in regular pro-
• The condition of the press line, a function of the
duction. Although the overall variation increases,
level of maintenance of the presses,
not every dimension exhibits an increase. Figure
• The condition of the dies, a function of die main-
12 below compares the observed part-to-part
tenance and engineering change rework, and
standard deviation at buyoff versus regular pro-
• Processing variables, such as the control of
duction. This illustration indicates a general lack of
cushion pressure, material handling, automation
correlation between part approval variation and
between presses, etc.
regular production. For some dimensions, the vari-
ation increases and for others it decreases,
Although many of these changes often are associ-
although more dimensions have higher part to part
ated with mean shifts, part-to-part variation can be
variation in production.
affected as well. Table 18 below shows that part-
Home Line
Production
Home Line
Production
Company
# Parts
Average
Average
% Dimensions
% Dimensions
(# Dimensions)
6 part-part
6 part-part
6 part-part>1
6 part-part>1
A
1 (37)
0.79
1.16
14%
48%
B
5 (132)
0.96
1.32
26%
48%
C
39 (327)
0.84
1.14
23%
38%
Table 18. Part-to-Part Variation: Home Line Approval vs. Production by Company
Note: production data 1 year + after home line buyoff
Correlation, R = .21
0.80
0.60
oduction
0.40
Pr
–
t
t-par
0.20
par
0.00
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
part-part – Home Line Tryout Approval
Figure 12. Part-to-Part Variation: Home Line Tryout Approval vs. Production by Dimension
24
3.2.3 Impact of Shipping on Variation
were measured using the same fixtures and auto-
As previously mentioned, part shipping from the
mated CMM programs.
stamping plant to the assembly plant caused sev-
eral mean dimensions to shift, particularly for the
Table 19 below indicates that part-to-part variation
non-rigid wheelhouse outer panels. This section
increased on 87% of the dimensions for the four
examines the effects of shipping on variation. As
parts: wheelhouse outer, quarter inner, A-pillar
noted earlier, some potential operator noise exists
reinforcement, and B-pillar reinforcement, with a
because different operators measured the panels
lesser increase on the more rigid components, the
before and after shipping. However, this operator
A- and B-pillar reinforcements. Clearly, part ship-
effect is unlikely to be significant, as the panels
ment increases part-to-part variation.
Panel
Measurement Points
Variation Increased
Wheelhouse Outer
69
91%
Quarter Inner
91
92%
B-Pillar Reinforcement
59
86%
A-Pillar Reinforcement
70
76%
Table 19. Summary of Remeasured Data Before and After Shipping (via truck)
3.2.4 Components of Variation: Part-to-Part,
variation components. One reason for looking at
Run-to-Run, and Begin-End of Run
the components of variation individually is that the
Stamping variation may be broken down into three
each is a reflection of different root causes. Table
components of variation: part-to-part, run-to-run,
20 below shows the part-to-part and total variation
and begin-end of run (see Section 2.0). Total vari-
for each auto company's body side outer panel.
ation, total, is a statistical summation of these three
Body Side
# cross car
Average
Average
% Dimensions
Company
Type
clamps in fixture
6 part-part
6 total
6 total >1.5
A
Integrated Quarter
11
1.14
1.41
29%
B
Integrated Quarter
14
1.09
1.93
57%
C
Two-piece
7
0.99
1.88
42%
D
Two-piece
8
0.99
1.21
23%
E
Two-piece
22
0.48
0.52
0%
F
Two-piece
16
0.32
0.49
0%
G
Integrated Quarter
17
0.40
0.77
3%
Table 20. Summary of Part-to-Part and Total Variation for the Body Side Outers
25
Companies E, F, and G, which used the most con-
company D. Among companies E, F, and G, the
strained measurement systems at 16, 17, and 22
constrained measurement companies, company
clamps respectively on the body side outer panel,
G appears to have less control over their mean
have the lowest part-to-part and total variation.
shift variation. In general, manufacturers with sim-
Comparing companies C and D, excluding the
ilar panels and similar checking systems should
clamping effect, showed that although the two
have similar levels of dimensional variation. When
manufacturers exhibit similar part-to-part variation,
they do not have similar levels of variation, the dif-
company C has much higher total variation. Figure
ference typically is not related to inherent part-to-
13 below shows that company C has significantly
part variation, but rather to how well one manufac-
more run-run and begin-end of run mean shifts.
turer controls its process over time.
Thus, company C does not appear to control their
process as well as company D. A similar finding is
observed in comparing companies A and B with
90th Percentile part-part C: 0.27 D: 0.24
90th Percentile total
C: 0.47 D: 0.29
100%
ariation
V
80%
20%
ved
60%
46%
40%
71%
20%
35%
otal Obser
T
0%
Company C
Company D
% of
Part-part
Run-run
Begin-end of run
Figure 13. Components of Variation for Body Side Panel at Company C and D
(Note:
total is greater at Company C due to mean shifts not part-part variation)
Table 21 on page 27 shows the amount of variation
the same for right and left mirror image parts. At
for each of the parts studied at company A, by
company A, the right hand body side outer
source of variation. The sample size for each type
exhibits significantly less variation than the left
of panel is 36 right and 36 left, or 72 total for each
side. Overall, variation at company A is relatively
type. The numbers expressed in Table 21 are
low with the exception of the left body side.
averages across all the dimensions on a part and
Although part-to-part variation is typically larger
therefore are non-additive. These data indicate
for a one-piece body side, the principal reason
that less rigid panels exhibited the largest part-to-
that the left side has significantly higher variation
part and mean shift variation. Interestingly, the
than the right side is due to mean shifts between
variation for a particular component is not always
stamping runs.
26
Average
Average
Average
Average
% of Variation
Part
run-run
begin-end
part-part
total
Explained by
Mean Shifts
Body Side - RH
–
0.15
0.19
0.24
31%
Body Side - LH
0.26
0.15
0.26
0.34
43%
Quarter Inner
0.07
0.07
0.08
0.10
43%
Wheelhouse Outer
0.11
0.09
0.09
0.14
50%
B-Pillar Reinforcement
0.04
0.07
0.05
0.08
59%
A-Pillar Reinforcement
0.06
0.05
0.07
0.08
28%
Table 21. Sources of Variation by Part for Company A
Table 22 below shows the percentage of total vari-
assessing mean shifts. Because analysis of vari-
ation at company C according to variation source:
ance methods are used to estimate mean shift
part-to-part and mean-shift (run-run and/or begin-
variation, higher part-to-part variation will mask
end). The effects of mean shifts at company C are
mean shift variation. In other words, the true mean
more significant than company A. The variation of
shift variation cannot be effectively evaluated if the
the body side, front pillar and center pillar rein-
inherent variation is unstable, a violation of the
forcements are approximately doubled due to
homogeneity of variance assumption used in
mean shifts. An analysis of the roof rail and wind-
ANOVA models.
shield frame suggests one potential challenge in
Average
Average
Average
% of Variation
Part
mean-shift
part-part
total
Explained by
Mean Shifts
Body Side - RH
0.26
0.17
0.31
79%
Roof Rail
0.23
0.28
0.34
32%
Front Pillar Upper
0.16
0.09
0.18
76%
Front Pillar Lower
0.15
0.09
0.18
76%
Center Pillar
0.21
0.08
0.23
92%
Windshield Frame
0.15
0.20
0.22
22%
Table 22. Sources of Variation by Part for Company C
3.2.5 Steel Properties and Press Setup
to-run and within run, however, is generally related
Control and Stamping Variation
to changes in the process over time, such as the
These case studies under production conditions
repeatability of press setup or changes to material
provide an opportunity to investigate possible root
properties. Although this study did not provide an
causes of mean shift variation. Short-term or part-
opportunity to rigorously control variables to
to-part variation is assumed to result from several
ascertain direct cause and effect relationships
factors related to product design, part size and
between process input variables and variation, it
rigidity, die design, stamping press condition or
does allow for some general conclusions regard-
the measurement system. Mean shift variation run-
ing the causes of mean shifts.
27
Five manufacturers collected input data for both
account for only 20% of the total observed varia-
process and material variables across thirty parts.
tion, then the most variation that can be explained
(companies E and F did not participate). They col-
with the input variables collected is 20%. This
lected this data for each sampling of three panels,
analysis only identifies relationships between
or, in some cases, once per run. The material
control of input variables and mean shifts.
coupons were analyzed later, either at an inde-
pendent test laboratory, three participants, or in-
Of the thirty parts with process input data, approx-
house, two participants. The following variables
imately 33% of the dimensions, 330 out of 1135,
were collected when possible:
had at least one large mean shift greater than 0.5
mm over the data collection period. Thus, prior to
• Process data (at each setup)
any mean shift analysis, over two-thirds of the
- Draw press shut height
dimensions studied were found robust to the vari-
- Draw Tonnage
ability of their respective process and material
- Die cushion pressure (if applicable)
input variables.
- Outer ram tonnage (if double-action
press used)
The next step was to examine the relationship
between process variable control and mean-shift
• Material data (a steel coupon was sampled
variation. Table 23 below compares mean shift
when a sample of parts was taken from the
variation with process input variation using
production run)
allowed ranges. Allowed ranges are essentially the
- Gauge
tolerances of the process and material input vari-
- Yield strength
ables. Thus, if manufacturers control their
- Ultimate strength
process-input variables within these ranges, they
- n-value
should not observe significant mean shifts related
- Percent elongation
to these variables. Generic allowed ranges are
used instead of tolerances to permit comparison
Due to data collection limitations, it was not possi-
among manufacturers with different process and
ble to match process and material variable data
material variable specifications. Furthermore,
directly to a particular panel. For example, the
since this analysis only looks for relative variation
material properties of the steel for each individual
differences, the nominal or average value of each
panel are unknown. Thus, the analysis is limited to
variable is not important.
trying to explain mean shift variation and not part-
to-part variation. For instance, if mean shifts
% Parts within
Correlation, R,
Variable
Robust Range
Robust Range
to mean-shift
Material Gauge
0.06 mm
96%
0.23
Yield Strength
6 ksi
95%
0.22
Ultimate Strength
6 ksi
92%
0.24
% elongation
9%
100%
0.19
n-value
0.04
100%
0.09
Inner Tonnage
60 tons
45%
0.69
Outer Tonnage/Cushion
50 tons/+/- 10%
48%
0.47
Table 23. Summary of Product and Process Variation Compliance
28
Table 23 shows that most steel variables, 92% to
as the limit to the allowable range of variation for
100%, fall well within their expected ranges of
controlling dimensional mean shifts. Note that
variation. Consequently, it is not surprising to see
these observed ranges relate only to dimensional
that their correlation with mean shifts is relatively
mean shifts and do not consider potential impacts
low with values ranging from 0.09 to 0.24, where 0
on formability issues such as splits or wrinkles.
has no correlation, 1.0 is a perfect correlation and
a value greater than 0.6 is considered correlated.
A few additional comments with respect to this
In general, the steel manufacturers studied had
analysis are appropriate. First, tonnage readings
control of their variation, and even when they did
may be affected by several setup variables such
not, material property variability could not be
as lubrication, die placement in press, shut height,
correlated with dimensional mean shifts.
etc., and thus the correlation to mean shifts should
be viewed principally as an indicator of lack of
The process variables of inner tonnage, outer ton-
process control. Second, the relationship between
nage, and cushion pressure, had considerably
tonnage and mean shifts over a continuous range
more variation and operated within the allowed
of tonnage settings was not analyzed scientifically
range only 45% to 48% of the time. The result was
for every part. Therefore, these data should not be
a much higher correlation to mean shifts.
used to identify tonnage specifications for a par-
Presumably, the opportunity for variation reduction
ticular part. Rather, simply recognize that those
for these part dimensions is significant if the press
parts in this study exhibiting large mean shifts
setup variable of tonnage and cushion pressure
tended to have relatively poor control of the
can be controlled more tightly. Figure 14 below
process variables but good control of the material
suggests an observed threshold of around 75 tons
variables.
0.35
Range 90-300
0.30
0.25
0.20
(<75)
mean-shift
0.15
0.10
0.05
0.00
0
100
200
300
400
Range of Draw Die Tonnage
Figure 14. Relationship between Press Tonnage and Mean Shift Variation ( mean shift)
3.2.6 Effect of Mean Shifts on Statistical
group plot out-of-control on an X-bar chart.
Process Control Techniques
However, only 20% of the dimensions had a mean
All manufacturers in the benchmark study exhibit-
shift greater than 0.5 mm. Note that the majority of
ed some level of mean shift variation for the major-
these mean shifts occurred on the parts at
ity of their part dimensions as shown in Table 24 on
companies B and C. Again, these mean shifts
page 30. Of the 1287 dimensions examined,
largely explain why certain manufacturers have
approximately 80% would have at least one sub-
more variation in their process than others.
29
# of
Average
% Dimensions w/
% Dimensions
Company
Dimensions
total
Significant Mean Shift
[mean shift> .5]
A-RH
329
0.12
80%
3%
A-LH
282
0.15
88%
12%
B
262
0.36
80%
51%
C
143
0.28
84%
31%
D
62
0.19
85%
3%
E
41
0.09
34%
0%
F
61
0.10
82%
3%
G
107
0.15
82%
14%
Total
1287
0.18
81%
19%
Table 24. Summary of Mean Shift Variation across Companies
The fact that such a large percentage of dimen-
are observed. The main concern with X-bar/
sions would plot out-of-control on an X-bar chart
Range charts for stamping is that they do not
has serious implications for process control. One
effectively separate problems from insignificant
interpretation is that stamping and die processes
process changes. One approach to desensitize
by nature are not stable enough to produce parts
charts is to replace X-bar/ Range charts with
with stable mean dimensions, even at world-class
Individual and Moving Range charts.
facilities. Another interpretation is that the part-to-
part variation of a stamping process often is so low
Individual and Moving Range charts are based on
that even well maintained processes will exhibit
subgroup sizes of one. Control limits to assess
some process drifts over time. Assuming, for
mean stability are then based on moving ranges.
example, that the inherent standard deviation of a
Because moving range values are based on con-
stamping process is 0.10 mm, a process will be
secutive subgroups, variation estimates reflect the
deemed statistically out-of-control if a mean shifts
part-to-part variation and some mean shift varia-
by more than 0.15 mm(2). Most manufacturers
tion. Table 25 on page 31 presents process control
would not want to adjust a process for a 0.15 mm
data for a stamping dimension. Using traditional X-
mean shift.
bar charts, this process would be considered
unstable or out-of-control as shown in Figure 15,
Assuming that small mean shifts are inevitable
on page 32. Interestingly, if only the first observa-
with the die changeover process, the traditional
tion in each subgroup is measured and Individual
use of X-bar charts to assess mean stability may
and Moving Range charts are used, this same
be unnecessarily stringent. The small part-to-part
process would be deemed in control. The reason
variation results in tight control limits, and this in
is that Individual charts based on moving ranges
turn results in many out-of-control dimensions.
are less sensitive than X-bar charts if small mean
Since small mean shifts rarely affect assembly
shifts are inherent to the process. Of course, with
builds, manufacturers using control charts often
individual and moving range charts, large signifi-
ignore the results. This is true even if larger shifts
cant mean shifts may still be identified.
2 The control limit for an X-bar chart is equal to A2(n) x d2(n) x part-part, where A2 and d2 are functions of subgroup size. If the sub-
group size, n, is equal to 4, then the control limits are +/-0.729 x 2.059 x 0.1 or +/- 0.15mm.
30
Subgroup
Sample
Sample
Sample
Sample
X-bar
Range
X
Rm
(i)
1
2
3
4
(i)
(I)
(I=2)
(I)
1
0.40
0.30
0.20
0.50
0.35
0.30
0.30
0.00
2
0.25
0.50
0.40
0.30
0.36
0.25
0.50
0.20
3
0.25
0.25
0.05
0.15
0.18
0.20
0.25
0.25
4
0.50
0.20
0.10
0.20
0.25
0.40
0.20
0.05
5
0.90
0.75
0.85
0.70
0.80
0.20
0.75
0.55
6
0.65
0.40
0.50
0.90
0.61
0.50
0.40
0.35
7
0.20
0.40
0.25
0.25
0.28
0.20
0.40
0.00
8
-0.10
0.10
0.25
0.20
0.11
0.35
0.10
0.30
9
0.25
0.30
0.30
0.25
0.28
0.05
0.30
0.20
10
0.40
0.25
0.10
0.20
0.24
0.30
0.25
0.05
11
0.40
0.65
0.50
0.30
0.46
0.35
0.65
0.40
12
0.30
0.25
0.20
0.25
0.25
0.10
0.25
0.40
13
0.10
0.10
0.00
0.10
0.08
0.10
0.10
0.15
14
0.40
0.30
0.70
0.50
0.48
0.40
0.30
0.20
15
0.30
0.25
0.30
0.30
0.29
0.05
0.25
0.05
16
0.35
0.60
0.50
0.40
0.46
0.25
0.60
0.35
17
0.15
0.15
-0.05
0.05
0.08
0.20
0.15
0.45
18
0.60
0.30
0.20
0.30
0.35
0.40
0.30
0.15
19
0.70
0.55
0.65
0.50
0.60
0.20
0.55
0.25
20
0.75
0.60
0.90
1.00
0.81
0.40
0.60
0.05
21
0.15
0.20
0.35
0.40
0.28
0.25
0.20
0.40
22
0.30
0.50
0.25
0.60
0.41
0.35
0.50
0.30
23
0.15
0.20
0.20
0.15
0.18
0.05
0.20
0.30
24
0.30
0.55
0.40
0.50
0.44
0.25
0.55
0.35
25
0.75
1.00
0.85
0.65
0.81
0.35
1.00
0.45
Average
0.38
0.26
0.39
0.25
Table 25. Process Control Data
31
x bar Chart
Range Chart
0.90
0.70
0.80
UCL
0.60
R
0.70
0.50
0.60
UCL
0.50
0.40
0.40
CL
xbar (i) 0.30
xbar (i)
CL Rbar
0.30
0.20
0.20
LCL
0.10
0.10
0.00
0.00
1
3
5
7
9
1
3
5
7
9
11
13
15
17
19
21
23
25 LCL
11
13
15
17
19
21
23
25
R
subgroup #
subgroup #
Individuals Chart
Moving Range Chart
1.2
0.9
1
0.8
UCL Rm
0.8
UCLx
0.7
, xi
0.6
0.6
0.4
0.5
0.4
0.2
CL
ving Range 0.3
Individuals
0
Mo 0.2
-0.2
CL
0.1
-0.4
LCLx
0
1
3
5
7
9
11
13
15
17
19
21
23
25
1
3
5
7
9
LCL
11
13
15
17
19
21
23
25
Rm
subgroup #
subgroup #
Figure 15. X-Bar/Range Chart vs. Individuals/ Moving Range Charts
(Note: charts based on the same process data)
The use of Individual and Moving Range charts for
then need to operate their processes within these
stamping processes solves the problem of over-
windows. If they can meet this objective, there is
sensitive control charts; however, it does not nec-
little need to measure stamped parts during regu-
essarily result in better process control. The fun-
lar production. However, many manufacturers
damental problem with statistical process control
either have insufficient knowledge of the robust-
charts for stamping is that they merely expose
ness of their processes to input variables or are
mean shifts. Effective process control requires an
not consistent in monitoring them.
understanding of the robustness of dimensional
measurements to input variables and then the dis-
Ultimately, whether a non-stable mean is accept-
cipline to control the variation within these robust
able depends on the influence that the variation
levels. For example, manufacturers should identify
will have on the assembly. In these case studies,
safe operating windows for draw tonnage, cushion
most assembly dimensions were robust to the vari-
pressure, shut height, counterbalance pressure,
ability of their coordinated stamping dimensions.
air pressure, n-value, material thickness etc. They
32
Table 26 below indicates that relatively few dimen-
dimensions demonstrating higher variation. Thus,
sions, less than 5%, exhibited strong correlation.
the elimination of large stamping mean shifts
Although stamping-to-assembly correlation is low,
would likely lead to a reduction in some assembly
some stamping dimensions with mean shifts
variation.
greater than 0.5 mm corresponded with assembly
Effect of Stamping Mean Shifts
# Dimensions with # Dimensions with
Company
# Coordinated
Significant
Stamping Mean
Median assembly
Median assembly
Dimensions
Correlation
Shift > 0.5
if shift < .5
if shift < .5
A
33
1
1
0.18
0.30
B
104
8
62
0.16
0.23
C
32
2
14
0.19
0.38
D
31
0
1
0.22
0.21
E
32
0
0
0.16
none
F
8
2
0
0.20
none
G
77
1
9
0.13
0.13
Totals
317
14 (4%)
87 (27%)
Average=0.16
Average=0.25
Table 26. Effect of Stamping Mean Shifts on Assembly Variation
33
4.0 Tolerance Considerations
system design. It was shown earlier that meas-
urement fixtures with more clamps tended to
4.1 Tolerances
have parts with tighter tolerances than those
measured with fixtures using fewer clamps. The
Two objectives for assigning sheet metal toler-
amount of observed variation with constrained
ances are to help insure that final assembly quali-
checking fixtures is less than that of less-con-
ty will be met and to minimize productivity losses
strained fixtures and therefore, tighter tolerances
during assembly because of large stamping varia-
can be achieved.
tion. Assigning tight tolerances help achieve this
goal. The tradeoff to assigning overly tight toler-
ances, however, is that die and stamping costs
4.2 Cp and Cpk (Pp and Ppk)
may become excessive trying to meet them. In
some respects, the tolerance has the effect of
The predominant tolerance strategy used by auto-
shifting costs from stamping to assembly or from
motive manufacturers is to assign tolerances
assembly to stamping, depending on the toler-
which may be difficult to achieve but are believed
ance assigned. A reasonable and meaningful
to help final assembly quality while reducing
sheet metal tolerance needs to consider the fol-
assembly problems. In some cases, overly tight
lowing three factors:
tolerances are assigned; if not readily achieved,
they can be re-evaluated during development. An
• Stamping process capability:
advantage of this strategy is that certain parts
The tolerance must reflect what a stamping
where all the tolerances are met are approved
process is capable of achieving, otherwise
without special intervention. One concern with this
unnecessarily high stamping costs will accrue.
strategy, however, is that many dies are unneces-
There are many current instances where
sarily reworked to meet the original tolerances
stamped parts are out of tolerance, but are being
even though they may not impact assembly build.
assembled successfully. All of the benchmark
This unnecessary rework leads to delays. Since
automotive manufacturers had body side outer
these manufacturers often use process capability
panels with a significant number of points out of
indices to measure conformance to tolerance,
tolerance. This is evidence that manufacturers
they will be discussed next.
tend to assign unnecessarily tight tolerances on
stamped parts, particularly for less-rigid outer
Two process capability indices often used to com-
panels. When stamping plants have difficulty
pare how well a process is achieving the design
meeting assigned tolerances, there is a tenden-
tolerances are Cp, process capability, and Cpk,
cy to overlook the tolerance and wait to hear if
design capability. These indices are a function of
assembly generates build problems. This wait-
the tolerances, part-to-part variation and mean
and-see approach would be improved upon if
bias, and were developed to measure the capabil-
the tolerances were known to be meaningful.
ity of a process relative to design intent. The for-
• Impact on assembly:
mula for Cp is:
Unlike many other rigid assembly processes, the
assembly of sheet metal affects final part geom-
Equation 5
Cp = USL - Nominal
etry. The assembly process has the ability to add
3 part-part
or reduce variation depending upon the compo-
nents and the assembly process. Many assem-
bly processes are robust to a wide range of
The Cp index is determined by dividing one half of
stamping variation showing virtually no impact
the tolerance, where one half the tolerance equals
on assembly quality due to stamping variation. In
the upper specification limit (USL) minus the nom-
these instances, it would benefit manufacturers
inal, by three standard deviations of part-to-part
to widen stamping tolerances, at least to the
variation. The formula for Cpk is:
point where they begin to impact assembly.
• Measurement system limitations:
Equation 6
Cpk = USL - Mean Bias
Because of the impact the measurement system
3 part-part
has on the ability to measure stamped panels,
(Note: mean bias = process mean - nominal)
part tolerances need to reflect the measurement
34
The Cpk index is determined similarly to Cp, except
If the sample standard deviation is used to esti-
that any mean bias is first subtracted from the
mate part-to-part rather than the above formula, then
numerator. If there is no mean bias and the
the Cp and Cpk indices are referred to as Pp and
process is operating exactly at the design nomi-
Ppk. Their interpretation, however, is the same
nal, then Cp = Cpk. For the purpose of these cal-
regardless of the method used to estimate part-to-
culations, part-to-part is estimated using statistical
part. Figure 16 below illustrates differences in Cp
tables and the formula:
and Cpk for three different scenarios.
Equation 7
part-part = R
d2
Mean
Mean
0.5
Mean
nominal
nominal
nominal
tolerance
tolerance
tolerance
Tol
±1.0
±1.0
±1.0
0.25
0.33
0.25
Cp
1.33
1.0
1.33
Cpk
1.33
1.0
0.67
Figure 16. Illustration of Cp and Cpk calculations for three scenarios
4.3 Recommended Tolerances for Sheet Metal
than 1.5 mm, have greater process capability, or
The tolerance guidelines shown in Table 27 on
smaller variation, exhibit more influence on the
page 36 are based on these empirical benchmark
assembly, and therefore warrant smaller toler-
studies. These guidelines allow consideration for
ances. Rigid components also tend to exhibit
process capability, or achieving a C
greater repeatability from die set to die set, so both
p = 1.33, influ-
ence on assembly dimensions and measurement
short-term and long-term tolerances are smaller
system limitations. They also assume that the data
than other components. Dimensions for non-rigid
is obtained without over-constrained measurement
panels are divided into two groups; mating sur-
systems. Furthermore, these tolerances only
faces and non-mating surfaces. Mating surfaces
reflect manufacturing variation about the long-term
often are more critical for assembly, and thus may
process mean, and do not consider the ability to
have tighter tolerances than non-mating surfaces.
hit the design nominal. Since dimensions routinely
In all cases represented in Table 27, a tolerance
deviate from design nominal, initial specifications
range is shown because the ability to control vari-
may account for both mean bias and process
ation may differ around the part. These general tol-
variation, resulting in wider tolerances than those
erances are a function of the inherent sigma and
shown in Table 27.
assume that a Cp of 1.67 is desired. For all three
categories, manufacturers should be able to at
These case studies also suggest that rigid com-
least meet the high end of the tolerance guideline
ponents, typically with material gauges greater
based on the capability of stamping processes.
35
Part
Location of
Inherent
Tolerance to Achieve Cp > 1.67
Rigidity
Dimension
Sigma
(tol > 3Cp or +/- 5sigma)
rigid dimensions
(~gauge > 1.5 mm)
all
.06 ~ .15
0.3 to 0.75
Mating Surface
.10 ~ .20
0.5 to 1.0
non-rigid dimensions
(~gauge > 1.5 mm)
Non-Mating
.10 ~ .25
0.5 to 1.25
Surface
Table 27. General Recommended Tolerances for Stamped Parts Based on Process Capability
(Note: data based on measurements systems without over-constrained clamping)
4.4 Part Tolerances and Functional Build
• Development lead-time is saved because less
The assignment of part tolerances often hinges on
die rework is required.
whether to allow for mean bias, or deviation from
• Lower overall process variation is achieved both
nominal. The previous section recommended part
in stamping and in assembly. Many engineers
tolerances based on manufacturing variation with-
believe that as the amount of die rework increas-
out consideration of mean bias. Since mean bias
es from shifting many dimensions toward nomi-
is not considered, the Cp index may be used to
nal, the less robust the die becomes.
measure conformance to design, but Cpk is not
used. This development strategy relies on two
This functional build strategy may also help
steps: minimize variation to an acceptable level
improve process control because the final specifi-
and evaluate the impact of mean bias on the
cations for mean bias and process variation are
assembly to determine which points, if any, require
determined during tryout and thus better reflect
rework. Here, the assembly build is used to identi-
process capability and the influence on assembly.
fy dimensional shifts and not product specifica-
The consequence of not meeting the final toler-
tions. Several manufacturers use this functional
ances is better understood without waiting to hear
build strategy and the advantages include:
from assembly.
• Less die rework is needed because only dimen-
sions that adversely affect the final assembly are
identified for rework.
36
5.0 Conclusions and Summary
ments. Larger parts experience from 20% to
500% more mean bias on the average, from
The following conclusions are based on the analy-
0.5 mm to more than 1.0 mm for unconstrained
sis contains in this report and from observations
measured parts. The amount of mean bias
made throughout the study. Since much of the
varies considerably across manufacturers
data collection was obtained under production
depending on many factors, including meas-
conditions in a non-statistically structured manner,
urement strategy, panel size and die buyoff
the analysis is not sufficiently rigorous to establish
strategy. Large parts also demonstrate up to
conclusive results in many areas. Due to the num-
twice as much variation as small parts, and the
ber of product and process variables seen at a
variation is distributed across part-to-part and
single manufacturer, rigorous experimentation
mean shift variation.
would have severely limited the breadth of analy-
sis. The following general conclusions provide
3. Short-term variation is relatively small with the
insight from several manufacturers, and reflect dif-
95% 6-sigma less than 1.0 mm for rigid parts
fering design and manufacturing strategies struc-
and less than 2.0 mm for the body side outer,
tured around common operating principles of
using unconstrained measuring. If the mean
sheet metal design, die construction and metal
bias could be eliminated, many parts would
forming. These conclusions also provide guide-
readily achieve a Cpk = 1.33. A significant
lines to developing more rigorous research
challenge during dimensional validation is
deemed necessary at particular manufacturers
eliminating mean bias, particularly for large or
choosing to develop a more scientific approach to
small complex panels.
stamping variation and measurement.
4. Large, less-rigid panels also are more suscepti-
1. An important distinction across companies was
ble to changes in variation due to transferring
the type of panel measuring system used on
the dies from the tryout source to the home line
large, non-rigid parts like the body side and
and from home line tryout to future production.
wheelhouse outer panels. The greater the num-
In both cases, both the mean bias and the
ber of clamps, the less observed variation and
amount of variation are likely to increase. Small,
mean biases were seen in the measurement
rigid panels have smaller changes in variation
data. Constrained measurement systems had
when transferring from tryout presses to the
between 16 and 22 in/out clamps, whereas the
home line. In some cases during production,
lesser-constrained systems used from 5 to 11
they show a decrease in mean bias from the
clamps. The use of clamps and their location is
home line tryout. It is likely that die rework has
indicative of different dimensional validation
taken place during the production life to reduce
and process control strategies not discussed in
mean bias, and attention may have been
this report. It is important to note the difference
focused more on the rigid panels than on the
because of the impact on the measurement
larger ones. Small, rigid panels are also less
data for large panels. Manufacturers using con-
susceptible to increased variation and mean
strained measurement systems also assigned
bias due to shipping influences than are larger
tighter tolerances to the body side. The con-
panels. Several small panels experienced from
strained tolerances varied from 0.3 mm to
6% to 19% of the dimensions shifting at least
0.50 mm, where the unconstrained tolerances
0.2 mm due to shipping, whereas the wheel-
varied from 0.70 mm to 1.25 mm.
house outer had 76% of the dimensions shift at
least 0.2 mm. The difference in the variation
2. There are significant differences in the amount
increase was not as significant, where the small
of variation seen in larger, less-rigid parts such
panels averaged 85% of their dimensions
as a body side outer panel versus smaller rein-
increasing in variation and the wheelhouse
forcements such as A and B pillar reinforce-
increasing 91%.
37
5. Manufacturers with similar part design and
sions. For this reason, conventional X-bar and R
measurement systems, but with different levels
charts are inappropriate for process control
of total variation, usually experience varying
because they would routinely indicate that the
degrees of run-to-run mean shifts. As expected,
processes are out of control, despite the capa-
part-to-part variation is similar for manufacturers
bility to be assembled into acceptable bodies.
with similar measuring strategies and product
Some manufacturers are better than others at
designs. The two setup-related variables inves-
minimizing mean shift variation, but all manu-
tigated in this study, tonnage and cushion pres-
facturers in this study are producing a signifi-
sure, showed a correlation with dimensional
cant percentage of parts with dimensions out-
mean shifts. No significant relationship could be
side of their assigned tolerances. The meaning-
found between material property variation and
fulness of currently assigned tolerances to
mean shift variation.
sheet metal part dimensions is suspect, partic-
ularly for less rigid panels.
6. All stamping processes in this study operated
out of statistical control. Stamping processes
have inherent complexity making it difficult or
impossible in production to set up repeatedly
with a constant mean value on all panel dimen-
38
Appendix
39
Appendix A - Part Sketches by Company
Locating Pin
Bodyside Outer
Front
U/D & F/A
Pillar
Center
Pillar
Quarter
Inner
Clamps
Wheelhouse
Detail Fixture
Outer
Figure 17. Part Sketches at Company A
Locating Pin
U/D & F/A
Front
Pillar
Pin U/D
Center
Pillar
Quarter
Inner
Clamps
Detail Fixture
Figure 18. Part Sketches at Company B
41
Windshield Frame
Reinforcement
Roof Rail
Locating Pin
U/D & F/A
Bodyside
Center
Pillar
Front Pillar
Clamps
Detail Fixture
Figure 19. Part Sketches at Company C
Bodyside Outer
Quarter Outer
Center
Pillar
Not included:
A-Pillar Upper and Lower Reinforcement
B-Pillar Upper and Lower Reinforcement
Figure 20. Part Sketches at Company D
42
Roof Rail
Locating Pin
Quarter Outer
U/D & F/A
Pin U/D
Bodyside Panel
Front Pillar
Center Pillar
Assembly
Clamps
Assembly
Detail Fixture
Figure 21. Part Sketches at Company
Bodyside Inner
Center Pillar
Reinforcement
Quarter Outer
Bodyside Outer
Clamps
Detail Fixture
Figure 22. Part Sketches at Company F
43
Quarter Inner
Bodyside
Outer
Locating Pin
U/D & F/A
Front Pillar
Center Pillar
Clamps
Detail Fixture
Figure 23. Part Sketches at Company G
44
AK Steel Corporation
Bethlehem Steel Corporation
DaimlerChrysler Corporation
Dofasco Inc.
Ford Motor Company
General Motors Corporation
Ispat/Inland Inc.
LTV Steel Company
National Steel Corporation
Rouge Steel Company
Stelco Inc.
U.S. Steel Group, a Unit of USX Corporation
WCI Steel, Inc.
Weirton Steel Corporation
Auto/Steel
Partnership
This publication was prepared by:
Body Systems Analysis Project Team
An analysis of stamping
The Auto/Steel Partnership Program
process capability and
2000 Town Center, Suite 320
Automotive Sheet Steel
Southfield, Michigan 48075-1123
implications for design,
248.356.8511 fax
http://www.a-sp.org
Stamping Process Variation
die tryout and process
A/SP-9030-3 0100 2M PROG
Printed in U.S.A.
control.
Auto/Steel Partnership
