|
Items in cart [0] |
Questions? Call 888-624-8373 |
| Copyright © 2009. National Academy of Sciences. All rights reserved. Terms of Use and Privacy Statement |
Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 47
:Lessons Learned from Commercial
Manufacturing
In recent years, M&S has played a significant role in the development
of a variety of commercial products, including the Boeing 777 aircraft, for
which three-dimensional M&S was used on both the product design and
manufacturing process; jet engine turbine blades at United Technologies,
for which M&S was used to refine blade design; new products at Ford
Motor Company, where M&S is used extensively in vehicle design,
development, trade-off analysis, and verification; the Viper at Daimler-
Chrysler, where M&S was used in design; and wheel rims at John Deere &
Company, where M&S was used to reduce development time. In addition,
M&S was used in the development of new fabrication facilities for
Corning and is also used in the design, fabrication, and assembly of
semiconductors there. Use of M&S in industrial manufacturing is not
without difficulties, however, and significant barriers to pervasive use of
M&S throughout the corporate enterprise remain.
The committee was asked to identify lessons learned from industry
and to identify emerging design, testing, and manufacturing process
technologies that can be enabled by M&S. The committee first examined
the current uses of M&S technologies in commercial manufacturing, using
the automotive industry as an example, and identified barriers to more
widespread use. The committee then analyzed the work of the Integrated
Manufacturing Technology Initiative (IMTI) to further develop the needs
indicated from the commercial manufacturing point of view.
47
OCR for page 48
48
MODELING AND SIMULATION IN MANUFACTURING
MODELING AND SIMULATION IN COMMERCIAL
MANUFACTURING
The Automotive Industry
The automotive industry is one of the world's most competitive
industries because of tight profit margins, the need to get vehicles to
market quickly, and the need for products that are desirable in different
markets worldwide. These factors, added to the complexity of modern
automobile design and the complexity of automobile manufacturing
facilities, have resulted in increased use of M&S within the industry.
The automotive industry needs to reduce the uncertainty involved in
designing and building new products. A recent article in Automotive
Design and Production quotes one expert as claiming that the entire value
of simulation lies in managing risk (Vasilash, 2001~. The article notes that
risk reduction results from the ability to make accurate assessments of the
performance of a system before money is invested in the tooling to build it
Engineering changes can therefore be made at an earlier stage of the
project when they are less costly. The same expert states that all
automobile manufacturers are now aiming for product development cycles
of 18 to 25 months. This results in a reduction of the number of physical
prototypes built and less time for physical testing, at the same time that the
level of technology in vehicles is increasing. In contrast, automotive
product development cycles in the early 1 990s were as long as 5 years
(Eisenstein, 2001~.
Numerous examples show the automotive industry benefiting from
use of models and simulations. Recently, General Motors Corporation was
able to complete its Grand River Assembly plant (in Lansing, Michigan) in
only 21 months from the start of construction. General Motors credits the
use of three-dimensional mathematical modeling with time savings in both
the validation of factory design, including ergonomic issues, and the
integration of equipment, tools, fixtures, and machinery, which was done
before hardware arrived on the factory floor (General Motors Corporation,
2002~. The ability to transfer knowledge developed in the models
throughout the company is seen as a form of technical memory.
Detroit Diesel Corporation was able to design and build a fully
functional prototype V6 diesel engine in 7.5 months. The company credits
rapid prototyping tools with permitting the creation of physical models to
verify designs, and it credits computer engineering tools with permitting
rapid modification of designs as problems were found. The engine was not
derived from previous designs (Vasilash, 1998~.
OCR for page 49
M&S IN COMMERCIAL MANUFACTURING
49
Toyota Corporation made extensive use of simulation in the design
and construction of the 2002 Toyota Campy. Among the benefits cited
were a 65 percent reduction in the number of prototypes needed and a 10-
month reduction in development time. Since the introduction date for the
new model had already been fixed, Toyota used the extra time on
simulating details of the car, such as overforce calculations on fuel lids and
cup holders (Whitfield, 2001~.
Like General Motors, Toyota has also used simulation tools to study
and resolve ergonomic issues. Toyota has used digital assembly software
to characterize the difficulty of motions made by employees in production
as green, yellow, or red. A pilot assembly line was used before production
began to improve the ergonomics of processes deemed red and to achieve a
large reduction in those rated yellow (Whitfield, 2001~.
Barriers to Widespread Use of M&S Technologies
Despite the successful examples described above, M&S technologies
are not yet deeply ingrained in most corporations or industrial sectors. On
the basis of a literature review and the experience of its own members, the
committee identified a number of barriers, both technological and
nontechnological, to the widespread, systemic use of M&S. These barriers
include the lack of reusability of existing successful applications, the lack
of model reliability and robustness, limitations on integration of systems,
and barriers caused by management and process structures.
Lack of Reusability
Most successful M&S applications have been solutions to specific
problems at the level of a single project or a single part. Few applications
of M&S at higher levels, such as supply chain integration, have been
successful. The applications that have succeeded at higher levels have
involved a single product line or a single process, such as continuous
materials processing. No examples of successful enterprise-level M&S
exist, although there is a trend toward making M&S a part of continuous
scheduling, production analysis, and troubleshooting (Gould, 20011.
Because of their specificity, it is difficult to integrate existing product
solutions into larger systems or to reuse M&S elements in solving new
problems. In part, this is due to limitations in the use of computer-aided
design (CAD) software. For example, unless all parts are designed using
the same CAD software, data sets from several product parts cannot be
merged into an overall system design. One solution would be to require all
designers to use the same software, but this is not optimal because different
OCR for page 50
50
MODELING AND SIMULA TION IN A1ANUFAC TURING
software packages work best for modeling different types of systems. In
addition, because CAD designs are geometric and static, it is not possible
to simulate parts or systems under dynamic use conditions. A better
understanding of the fundamental physics underlying product performance
and manufacturing processes could improve the specificity problem, but
such an understanding is lacking.
The design of the Boeing 777 aircraft is an example of the restriction
regarding simulation under use conditions. A 1997 NRC report discussed
this problem as follows:
While the Boeing 777 experience is exciting for the VE [virtual
enterprise], we should recognize just how limited the existing CAD
tools are. They deal only with static solid modeling and static
interconnection, and not~r at least not systematically with
dynamics, nonlinearities, or heterogeneity. The virtual parts in the
CATIA [computer-aided three-dimensional interactive application]
system are simply three-dimensional solids with no dynamics and none
of the dynamic attributes of the physical parts. For example, all the
electronics and hydraulics had to be separately simulated, and while
these too benefited from CAD tools, they were not integrated with the
three-dimensional solid modeling tools. A complete working physical
prototype of the internal dynamics of the vehicle was still constructed,
a so-called "iron-bird" including essentially everything in the full 777.
While there was finite element modeling of static stresses and
loads, all dynamical modeling of actual flight, including aerodynamics
and structures, was done with "conventional" CFD [computational fluid
dynamics] and flight simulation, again with essentially no connection
to the three-dimensional solid modeling. Thus while each of these
separate modeling efforts benefited from the separate CAD tools
available in their specialized domains, this is far from the highly
integrated VE environment that is envisioned for the future, and is
indeed far from even some of the popular images of the current
practice. Thus while a deeper understanding of the 777 does nothing to
reduce our respect for the enormous achievements in advancing VE
technology or dampen enthusiasm for the trends the 777 represents, it
does make clear the even greater challenges that lie ahead. (NRC,
1997b, p. 138)
Lack of Model Reliability and Robusiness
Increased acceptance and use of models and simulations in
manufacturing and defense systems acquisition will depend on increasing
the credibility of the models (Lucas, 1997~. Increasing credibility depends
on performing appropriate verification, validation, and testing activities
throughout the simulation life cycle (Balci, 1998; Robinson, 19991;
OCR for page 51
M&S IN COMMERCIAL MANUFACTURING
51
examining the engineering processes used to develop the simulation
(Ketcham and Muessig, 20001; and understanding the intended use of the
simulation (Muessig et al., 20009. Further development of these practices
and of how to integrate them with model development is needed (Balci,
19983. Although some tools exist to support the activities that lead to
credibility, knowledge of the use of these tools and the related techniques
may not be as widespread as it should be (Pace and Glasgow, 1999~. More
research is needed to increase the automation of verification, validation,
and testing (Balci 1998~. Some modeling methods are less robust than
desired; for example, the results of finite element modeling can differ if
different meshes are used (Xu and Liao, 20011. Theoretical and practical
development is required to improve the reliability and robustness of
models. Development is needed as well in dealing with model data
uncertainty (Doyle, 1997; Tolk, 1999) and in quantifying the effect
uncertainty has on the validity of models (Pace, 2002~.
Lack of System Integration Capabilities
Systems engineering is the flow-down process of determining needs,
exploring concepts for product systems that fulfill those needs, selecting a
concept, developing a design, and setting product specifications. The
integration of systems, such as weapons containing software and hardware
that are both complex, is hindered by the limitations of systems
engineering. For example, it is not possible to directly model the actual
outcome of a system in response to its inputs (Sage and Olson, 2001~.
Rather, the processes that the system will use to produce outputs can be
modeled and then the system can be simulated using a variety of inputs to
characterize the output behaviors with respect to the inputs. Systems
engineering is limited by the fact that the individual parts of a system, as
well as subsystems, influence each other. They adapt to their environment
and in so doing change the environment of other parts and subsystems.
Only M&S can shed light on this process, but exploration of system
behavior through simulation response to random inputs is time-consuming.
Existing Management and Process Structures
Existing management and process structures are outdated and
therefore represent barriers to the widespread use of M&S technologies in
manufacturing. Designers are skilled tradespeople who produce and
release detailed part drawings, usually with the aid of CAD and CAM
software tools. Degreed engineers have an impressive array of M&S tools,
known as computer-aided engineering (CAE), available to analyze designs.
These tools are often bypassed, however, because analysis takes time and
OCR for page 52
52
MODELING AND SIMULATION IN MANUFACTURING
designers are rated on the number of drawings released rather than on
designs that are certified to meet product requirements. Designs are
therefore open pushed forward before analysis is complete, and CAE
analysis and simulation of product reliability remains divorced from the
critical design path (Versprille, 2001~.
Project engineers are still rated on the speed at which they can
produce and test prototypes (the "build and break" philosophy). Prototype
construction therefore begins early in the product development cycle,
before up-front modeling and simulation are able to provide guidance.
Since the modeling and simulation of an entire product from concept
to disposal crosses the boundaries of many disciplines, systemic use of
M&S in manufacturing faces large cultural resistance. In addition,
although the "build-test-fix" product development cycle is recognized as
being inefficient, particularly for large and complex projects, it is still in
wide use. Systemic use of M&S requires substantial up-front investment
in personnel, training, and software tools. Change is hindered by the
significant investment needed to develop the infrastructure necessary for
incorporating M&S. In today's business climate, return on investment is
evaluated quarterly, and it is difficult to justify the overhead dollars needed
to build substantial M&S capabilities. In addition, many corporations are
organized into business units, manufacturing units, and support units, each
seeking to look like a profit center. The enterprise-level thinking needed to
achieve pervasive M&S use even within a product line, much less at the
enterprise level itself, is difficult to achieve.
INTEGRATED MANUFACTURING TECHNOLOGY
INITIATIVE
The Integrated Manufacturing Technology Initiative (IMTI)' was
launched in 1998 to develop a research and development (R&D) agenda
for integrated manufacturing technology in the 21 st century. In this
context, "integrated manufacturing" was defined as the effective
integration of production, design, supply, and marketing functions to
enable improved control, management, and planning for the enterprise. The
R&D agenda that was developed addressed key technology goals cutting
across all manufacturing sectors and recognized M&S as a critical enabler
to support future manufacturing. Indeed, the IMTI report concluded that no
' The initiative, formerly known as the Integrated Manufacturing Technology Roadmapping
Initiative, was sponsored by the National Institute of Standards and Technology, the U.S.
Department of Energy, the National Science Foundation, and the Defense Advanced Research
Projects Agency.
OCR for page 53
M&S IN COMMERCIAL MANUFACTURING
53
other technology offers more potential for improving products, perfecting
processes, reducing design-to-manufacturing cycle time, and reducing
product realization costs.
The IMTI road map for M&S distinguishes between product and
process applications of M&S. Product applications include the following
functions: representation of the physical attributes of a product, the
effectiveness with which the product performs its advertised functions, cost
and affordability, producibility, and requirements related to different
phases of the product life cycle. Process applications include the following
functions: the material operations performed in manufacturing processes,
such as preparation, treatment, forming, removal and addition; the
assembly, disassembly, and reassembly of components to form the overall
product; the testing and evaluation of product quality; and packaging and
remanufacture.
The first two columns of Table 3-1, "IMTI Vision for Product
Functions," and of Table 3-2, "IMTI Vision for Process Functions,"
summarize IMTI conclusions regarding the current state of practice and the
ideal state of product and process functions, respectively. The study
committee developed the material in the remaining two columns regarding
real-world limitations on each function and the requirements needed to
achieve the ideal state. The limitations place realistic constraints on what
can be achieved using M&S. The requirements point to R&D needed to put
the prerequisites in place before the desired capabilities can be attained.
The committee also partitioned the aspects of M&S addressed in
Tables 3-1 and 3-2 into two categories those relating to "in the small"
and "in the large" considerations. Modeling and simulation "in the small"
refers to aspects of M&S that concern one or, at most, a limited number of
modelks) addressed in isolation from the range of all other models. For
example, development of a product model and concern for its validation
are an "in the small" aspect. On the other hand, "in the large"
considerations address problems and issues that cover M&S technologies
across the board. For example, integration of models into a common
framework is "in the large" concern.
As indicated in Table 3-1, the IMTI vision for a future ideal state of
M&S use in product design applications includes models that capture all
product attributes; interoperability between product and performance
models; more accurate cost estimating; manufacturing process
requirements included in an integrated design system; all life-cycle
considerations included in the product model; and a situation in which
analysis leads design, rather than supporting it. Limitations on this vision
include those on bandwidth, computation speed, memory, and other
communication and computation resources. R&D is required in the areas
of model standards and integration; modularity between different M&S
OCR for page 54
54
MODELING AND SIMULA TION IN MANUFACTURING
components; continuity of modeling information across life-cycle phases
and between manufacturing facility and site of product use; development
of improved search methodologies; and advances in parametric modeling,
variational analysis, and probabilistic design.
As indicated in Table 3-2, the ideal future state of M&S process
applications would include production processes generated from design
and enterprise models; reliable models for materials and materials
development; micro to macro continuum modeling; automated
optimization of complex process models; quality engineered into every
manufacturing process via virtual testing; packaging integrated into
product and process design; modeling for disassembly, remanufacture, and
reuse integrated into product life-cycle model; integration of stochastic and
deterministic models to optimize manufacturing processes; and controller
simulations that evolve into optimum operations controllers.
Real-world limitations to the achievement of the ideal state shown in
Table 3-2 include limitations on model content and available knowledge.
R&D required to reach this state includes continuity of modeling
information across life-cycle phases, standards for product models,
improved interoperability, improved composability, use of families of
multi-resolution models, integrated verification and validation, placing
M&S tools and systems under knowledge-based control, and a universal
framework for model construction.
Knowledge management refers to a deliberate approach to
recognizing knowledge as a resource to be managed in a corporate
environment (House and Bell, 2001~. Its advent is an important
development for M&S in the enterprise context, since models are an
important form of corporate knowledge. Moreover, knowledge
management can provide a broader framework in which M&S is fed
knowledge from other sources and, in turn, generates new knowledge as an
output. For example, knowledge management could help couple
functionality that is specified at a high level of abstraction to detailed
design. Basic research is needed here, since it could significantly reduce
modeling time and ensure consistency in system acquisition.
OCR for page 55
M&S IN COMMERCIAL MANUFACTURING
a
C~
C~
o
._
C)
C~
~o
~o
o
._
C~
._
~4
E~
~
_.
m
_ o~,
E r: ~ ~ O E ' 0
o =' 2 '~ E ~ E co
cn cn
C~ _
.
o E
~
~c
_
._
C
o
._
CD
E ~~ c~
'~ o E ~ tn j ~ ~ ~ E ~ S ~ ~ ~ '
E ~ ~e ~ 0: ~ =- =, ~ ~ ~ E
cx5 ~ ~ ~ E a' ~ 'C Q ~ _ ~ CC
° a~ a~ (D
au : ~c E ° o ' a,
c E E ~0 0 <~ =
(~ G' ~ a, —~ a' ~ Q
E co E `° E ~
55
OCR for page 56
56
.=
o
Cal
_'
Cal
o
._
Cal
C)
s°-
s~
to
o
._
._
EM
m
¢
MODELING AND SIMULA TION IN MANUFACTURING
CD
a'
. _
fir
a)
To ~
~ ~ ~ a) ~ a'
o ._ _
~ ~ o ~ ~ o ~ ~
(~' ~ C ~ ° ~ C ~ o in
~ '0 Is ~ 'I
~ ~ ~~ ; ~ ~ ~ ~ ~ ~ O
E— ~~ ~ Q
il ~ ~ ~ -= ~ r j ~ e,
gel j ~52( ,,~j
as i,, a) ,, g [D o
C,0 · ~ ~ ~ ~ ~ ~ ~ ~
OCR for page 57
M&S IN COMMERCIAL MANUFACTURING
In
c
a)
a)
._
a)
try
an
. _
,_ ~ C5) ~ ~
a . r 8
~ it| ~ 1 ~ ~ ~ 1 E
=- ~
~ (I _ D ~ O
o E b 3 ~ · E
en a)
m~ I, , ° E of E E
o ~ ~ a' ~ =~m ~ 3 o
~ ~ E ~ A' C' E a) E
v ·, E e
a ~ ~ ,, 2. ~ t 1 ~ ¢~ E
57
OCR for page 58
58
U'
o
._
C)
a
V'
C~
a~
o
S~
~0
o
.
C~
.
m
MODELING AND SIMULATION IN MANUFACTURING
cn
-
a)
._
<:r
CY
a) ,
~) ~ cn .5L~ — E
~ _ _, — ~ ~
e 0, ~ <, ~ 2 o'
_ ~ 0 a) 0 a) cn ° a) ~ O
::E ~ C o O ~ c: w -o
c~cn ~ S='
.c~,o a) C\5 C<5,0
`° E _ °)_ E tn
~ ,~ E "o
_ _-3 " o
C~
,
co E: <,, ~, ~ e ~ ~ ~ ~ e
~ E ~ L i C ~ ~ ~ ~ ~ ~ U
a)
-
cn
-
CD
CO
' a
o o
_ ~ . _
E o~ o E a) Q co o C'
~ ~ 9 ~ 5i o ~ j 5 ~
OCR for page 59
M&S IN COMMERCIAL MANUFACTURING
CO
c
0
~0
._
0
1Y
_ -
=
Co~ 0
' CO
~ ~n
o 0
C)
C~
o
~o
o
._
Cq
._
E~
m
0
r
cn
-
~0
0
C~
-
0
c
._
,O
-
~ CT5 ~
E =m
cn 0 c
~c . - ' E
C~ C~ 0
~ _ 0
_ cn ~
0 0 -=
° E 3
a.) ._
.> cn
.— ~
C c
CD C~
o o
C~
a
CD
=5,
co cn
CD ~
a) O
o ~
,c~
cn .C
C~
c
_ —0
-
._
~ C~
a), '
C~ 0
o, o
~ C~ C
_ _ _
'c~ ~ 3
_ 0 0 ~
~ 0 ~ 3
O Q 0 0
~C 0
o
E ~'
-
~,
— CTS
Q
O C~
(TS ~S
C
~ ,,0
c cn
0
CD 0
C~ C
C~
CO
Q
0
~0
o
-
0
-
Q
U'
59
~ _
0 0
C C
~ O
CD ~
0 0
~ C~
C ~
~0 0
~ C5)
0 0
cn 3
oes Co
CO
-
0
-
o
c~
- ~
0 0
o ~
c 0
0 o
~ 0
a~ co
- -
~ - ~ -
c ~
~— ~
J C(5
o
0 - cD
C,, ~ o E
[= —
0 C C
tn 3 .~° 0
cn ~ 0 ~
0 —c =,
2' '0 ° o
G) (0 ~ Q
—3 cnQ 0 _
y Ct5 0 a
0
_ C~:5
O ccn
O ~ O O
_ C 0 (~
0 0 ~ ~
0 cD 0 0 Q
Q 0 ~0 -—
— c _
~ t5 c,, ~ E ~'
o
o
o
. .
C~
V:
OCR for page 60
60
.
MODELING AND SIMULATIONIN MANUFACTURING
CONCLUSIONS
On the basis of the barriers to widespread use of M&S in industry
identified above and the analysis of the IMTI vision of future M&S
product and process applications, this NRC committee identified the
following needs for improvement in M&S technologies for product and
manufacturing process design applications:
.
Increased capabilities to reuse successful product design
applications for other problems or to integrate successful product
design applications into larger systems; product model standards
modularity of M&S components, and improved comparability;
Improved integration of models; improved CAD software that
enables use of product models in performance simulations of
dynamic-use conditions; improved interoperability;
Improved model validation and verification methods to increase
reliability and robustness to uncertainty of product models;
integrated verification and validation of models and simulations;
Improved parametric modeling, variational analysis, and
probabilistic design to increase use of M&S analysis in design
process;
Universal framework for model construction that incorporates
both stochastic and deterministic models to optimize
manufacturing parameters.
The committee identified needs for improvement in M&S
technologies for process applications, including the following:
Improved capabilities for integrating systems, such as improved
methods for understanding systems behavior and improved
integration of performance modeling and effectiveness
simulations with product modeling and engineering simulations;
. Continuity of models across life-cycle phases;
· Improved heuristic search methods to decrease simulation times
and to support an integrated design system of business, product,
and process models;
· Knowledge-based control of M&S environments to improve
testing and evaluation.
OCR for page 61
M&S IN COMMERCIAL MANUFACTURING
TABLE 3-3 M&S Needs for Commercial Manufacturing
61
Category of Need
.
Product and manufacturing
Specific Needs
process design
Process applications
Product development process
Increased reuse capabilities
Improved integration of models
Improved model validation and verification
Improved design modeling methods
Universal framework for model construction
Improved system integration
Continuity of models across life cycle
Improved heuristic search methods
Improved testing and evaluation
Encourage use of M&S in product design, testing,
and evaluation
Finally, the committee identified the need for nontechnical
improvements in the product development process to encourage, rather
than discourage, full use of M&S analysis capabilities in design and full
use of M&S capabilities in product testing and evaluation (see Table 3-3~.
OCR for page 62
Representative terms from entire chapter:
process applications
