Docstoc

Smart Assembly - Strategy for Establishing a Smart Assembly Initiative

Document Sample
Smart Assembly - Strategy for Establishing a Smart Assembly Initiative Powered By Docstoc
					                    DRAFT




    A Report on “Smart Assembly”


                Prepared for the

National Institute of Standards and Technology
                By IMTI, Inc.
               P.O. Box 5296
            Oak Ridge, TN 37831



               December 1, 2006




                   DRAFT
                            DRAFT




This document was prepared by Integrated Manufacturing Technology
   Initiative, Inc., under NIS T Supplies and Servic es Order Number
      SB134106W0989, Requisition/Reference No. 06 -822-7656.




                           DRAFT
A Report on “Smart Assembly”                                                                                                               Page 3

                                                                   DRAFT

                                                         Table of Contents
Preface----------------------------------------------------------------------------------------------------------------------------------------------5

Introduction --------------------------------------------------------------------------------------------------------------------------------------5

The Business Ca se for Smart Assembly -------------------------------------------------------------------------------------------6

Vision for Smart Assembly ---------------------------------------------------------------------------------------------------------------8

Logical Structure for a Smart Assembly Initiative-----------------------------------------------------------------------------9

Call to Action ---------------------------------------------------------------------------------------------------------------------------------- 13

Appendix A: Workshop Plenary Session Summary ------------------------------------------------------------------------ 14
   Smart Assembly Systems – Bob Tilove ------------------------------------------------------------------------------------------ 14
   Net work Enabled Manufacturing – Bryan G. Dods --------------------------------------------------------------------------- 14
   Electronics Industry Perspective - Tom Babin --------------------------------------------------------------------------------- 15
   Smart Assembly Systems – GM Perspective – Roland Menassa------------------------------------------------------ 16
   Mechanic al Assembly and Researches Needed to Mak e it Smart – S. Jack Hu and Dawn Tilbury ------ 18
   Manufacturing Operations Standards Need to Converge into a Manuf acturing Application Integration
   Framework – Charlie Gifford ---------------------------------------------------------------------------------------------------------- 20
   Production Management – Delivering E nd-to-End PLM – Mitch Vaughn ------------------------------------------- 20
   Smart Assembly – Bob Brown-------------------------------------------------------------------------------------------------------- 23
   Smart Assembly – Jim Caie ----------------------------------------------------------------------------------------------------------- 24

Appendix B: Presentation of Re sults from Workshop Breakout Sessions ------------------------------------- 26

Appendix B1: End User Perspective ----------------------------------------------------------------------------------------------- 27
   Characteristics and Attributes -------------------------------------------------------------------------------------------------------- 27
   Key Themes --------------------------------------------------------------------------------------------------------------------------------- 27
     Simulation and Visualization------------------------------------------------------------------------------------------------------- 27
     Reconfigurable Tools and Systems--------------------------------------------------------------------------------------------- 28
     Assembly System Reuse ----------------------------------------------------------------------------------------------------------- 35
     Business Modeling (PLM/Performance) -------------------------------------------------------------------------------------- 35
     Real-time Actionable Information ------------------------------------------------------------------------------------------------ 35
     Knowledge Capture and Learning----------------------------------------------------------------------------------------------- 35
     Parking Lot-------------------------------------------------------------------------------------------------------------------------------- 36

Appendix B2: Re searcher Perspecti ve ------------------------------------------------------------------------------------------- 41
   Characteristics and Attributes -------------------------------------------------------------------------------------------------------- 41
   Key Themes --------------------------------------------------------------------------------------------------------------------------------- 42
     Automated and Flexible Assembly ---------------------------------------------------------------------------------------------- 42
     Easier Automation Programming ------------------------------------------------------------------------------------------------ 42
     Sensors / Sensing --------------------------------------------------------------------------------------------------------------------- 42
     Alignment of Real Assembly Systems to Their Virtual Counterparts ---------------------------------------------- 43
     Analytical Assembly Tools ---------------------------------------------------------------------------------------------------------- 43


                                                                 DRAFT
A Report on “Smart Assembly”                                                                                                              Page 4

                                                                  DRAFT
     Green Assembly ------------------------------------------------------------------------------------------------------------------------ 43
     Parking Lot-------------------------------------------------------------------------------------------------------------------------------- 43

Appendix B3: Infra structure and Standards Perspective ---------------------------------------------------------------- 54
  Characteristics and Attributes -------------------------------------------------------------------------------------------------------- 54
  Priorit y Themes----------------------------------------------------------------------------------------------------------------------------- 55
    Flexible Manufacturing Systems ------------------------------------------------------------------------------------------------- 55
    Closed-Loop Information Flow ---------------------------------------------------------------------------------------------------- 59
    Common Language ------------------------------------------------------------------------------------------------------------------- 62
    Tracking Systems for Assembly (forward and backward) -------------------------------------------------------------- 65
    Parking Lot-------------------------------------------------------------------------------------------------------------------------------- 65

Appendix B4: Integration and Deployment Perspective ----------------------------------------------------------------- 68
  Characteristics and Attributes -------------------------------------------------------------------------------------------------------- 68
  Key Themes --------------------------------------------------------------------------------------------------------------------------------- 69
    Conc eptual Framework - Demonstration ------------------------------------------------------------------------------------- 69
    Standards – Data Exchange; Hardware/Software ------------------------------------------------------------------------ 69
    OEM/Supplier Collaboration ------------------------------------------------------------------------------------------------------- 70
    Standards – Methodology/Communications/System Engineering -------------------------------------------------- 70
    Skills----------------------------------------------------------------------------------------------------------------------------------------- 70
    Parking Lot-------------------------------------------------------------------------------------------------------------------------------- 70




                                                                DRAFT
A Report on “Smart Assembly”                                                                     Page 5

                                                                  DRAFT
Preface
The purpose of this report is to document and summarize material presented or generated during a Smart
Assembly workshop held at the National Institute of Standards and Technology (NIST) on October 3-4,
2006. This document was prepared by Integrated Manufacturing Technology Initiative, Inc., under NIST
Supplies and Services Order Number SB134106W0989, Requisition/Reference No. 06-822-7656.

Introduction
Manufacturing operations involve the preparation and processing of raw materials, creation of
components, and assembly of components into subassemblies and finished products. The broader scope of
manufacturing embraces innovation, design, engineering, and management of life-cycle performance.
While these basic steps remain unchanged, over the last decade we have experienced a global redefinition
in the distribution of manufacturing functions and operations. What was a trend toward outsourcing has
become a fundamental shift in the way products are manufactured. Twenty years ago if it was said that
“the big three automakers make cars,” the meaning was clear. In their factories and in a small network of
affiliated companies, they made most of the components. They built transmissions and engines and drive
trains and bodies, and the parts and subassemblies came together in their assembly plants. Smaller
companies like Levi Strauss made clothing in small factories across America.
Today, many manufacturing sectors – such as the textile industry – have mostly gone offshore except for
high-end and specialty products. The former vertically integrated manufacturing companies have become
systems integrators, marketers, and distributors, and supply chain management has become a key
competitive discriminator in assuring the quality and timely delivery of all of the components of an
assembly.
There is a great opportunity within this shift. The importance of assembly to our economic well being
remains very high, and to remain strong in the global marketplace, we must maintain our ability to cost-
effectively produce excellent products. This is not an easy challenge. The reports from last year‟s
international auto shows observe that Chinese automobiles are not yet ready for global compet ition, but
are only a few years away. "We're very confident that we will have a five-passenger family sedan ready to
import to the United States, fully in compliance with U.S. emissions and safety regulations, that we can
sell for less than $10,000," says John Harmer, vice president and COO of Geely-USA. The car, probably
to be renamed for the USA, is about the size of a Honda Civic, he says.1 When one considers that the
manufacturing cost of a product produced in China is 30% to 50% lower than the same product produced
in the U.S., the near-term threat to our nation‟s manufacturing base is very real.
The longer-term threat of losing the edge in technological sophistication that is sustaining many U.S.
manufacturers is growing rapidly. Offshore manufacturers have, in their home locations and U.S.-based
operations, brought a willingness and a commitment to automation. This challenge introduces
technological and cultural challenges that require a response.
U.S.-based manufacturers are certainly not being left behind. The quality of the products made by U.S.
manufacturers is excellent. Our manufacturing productivity increased by 94% from 1992 to 2005,
exceeded only by Sweden and South Korea over that time period. 2 Assembly methods have improved
dramatically, aided by automation and systems integration. A notable example is the elaborate integrated
supply chain created by Boeing to enable 3-day final assembly of the 787 aircraft, and there are many
others.



1
    http://forums.autoweek.com/thread.jspa?forumID=10&threadID=24751&messageID=563493
2
    ftp://ftp.bls.gov/pub/special.requests/ForeignLabor/prodsuppt02.txt



                                                                DRAFT
A Report on “Smart Assembly”                                                                         Page 6

                                                  DRAFT
While much progress has been made, there is much more to do. To remain competitive – to offset wage
differentials and counter the increasing technological capability of our offshore competitors – we must
improve assembly operations. The improvement should come from a balance of technologies including
better design for manufacturability, modeling and simulation for process optimization, and effective use
of both automated and human-assisted systems, to assure that every component placed and every product
produced is exactly as required.
In light of the increasing importance of assembly processes, approximately 60 researchers, software and
equipment manufacturers, and end users convened at a Smart Assembly Workshop at NIST October 3-4,
2006. The purpose of the workshop was to assess the current state and to determine needs in smart
assembly processes, thereby establishing a broad industry/academic vision and laying the groundwork for
a Smart Assembly initiative.
Workshop participants were asked to focus on the activities needed to put parts and components together
to make a product, concentrating on process planning, process design, engineering, validation,
construction, installation, launch, and operation of assembly processes and systems. Activities and issues
involving product design and modeling for assembly, supply chain, and enterprise analytics are
recognized as extremely important, but are considered topics for another forum.
The results of that workshop are documented in this volume with the hope that the seeds planted in
October will bear fruit in a collaborative and highly leveraged cooperative relationship, led by industry,
and patterned perhaps along the lines of the NIST Smart Machining Systems program, in which
government agencies, academic and other and research organizations, and industry can work together to
deliver dramatic successes.

The Business Case for Smart Assembly
Assembly represents a large portion of the manufacturing wealth creation for the U.S. economy, and the
importance is growing. There is a systematic shift from manufacturing and assembling products under
one corporate label, to outsourcing of component supply and then managing assembly, marketing, and
distribution to generate revenue and profits. This trend presents an opportunity and a threat. While lower
offshore labor costs make it difficult for U.S. manufacturers to produce cost-competitive components in
the United States, similar arguments are now being made for assembly of increasingly complex and high-
value products. It is alarming to pick up most any product and check its country of origin. The U.S.
content in manufactured goods dropped from 83% in 1973 to 24% in 2004, and there is no sign that this
trend will reverse itself.
The challenge is to sustain the wealth-creation engine of U.S. manufacturing. As the share of
manufacturing content declines, so does the share of the revenue stream and the control of the generated
wealth. Hence, as a nation, we must be aware that while we now live in an increasingly global economy,
control of assembly operations is imperative to our economic well-being. 3
Manufacturing creates wealth, and the manufacturing sector is critical to maintaining our standard of
living. Every dollar invested in manufacturing spawns another $1.43 for the economy – called the
multiplier effect. Private industry is the largest source of R&D funding in the U.S., providing 65.5
percent ($193 billion) of total R&D funding in 2002. U.S. manufacturing R&D accounted for two-thirds
of this investment, or over $125 billion in 2002 4 . Manufactured products also account for two-thirds of
our exports. While the U.S. is still the leading exporter, there are both alarming and encouraging signs on

3
  Presentation by Richard Neal, “Transforming American Manufacturing”, NGMTI Technology Forum, May 8 , 2006,
Andover, Mass.
4
  National Science Foundation, Science and Engineering Indicators, 2004,
http://www.nsf.gov/statistics/seind04/c4/c4s1.htm#c4s1l7.



                                                DRAFT
A Report on “Smart Assembly”                                                                        Page 7

                                                  DRAFT
the horizon. The 2005 deficit in net exports of over $700 billion (up from about $100 billion in 2000)
was to a large extent the result of the imbalance between our exports and imports. 5 On the other hand,
the GDP from manufactured products (as of November 2006) continues to grow each month, increasing
by 9% over the two year period of 2004-2005. Productivity continues to increase, and capacity utilization
is over 81% – higher than the 1972-2005 average.6
It is difficult to assess the impact of the assembly element of manufacturing in the U.S. economy. The
North American Industry Classification System (NAICS) codes against which data is reported do not
group the various categories in such a way that assembly contribution is visible. However, we can make
some informed estimates. The total value of U.S. manufactured products produced in 2004 was slightly
more than $4 trillion, while the cost of raw materials was slightly more than $2 trillion. That means that
the manufacturing value added was approximately $2 trillion.
It is reasonable to postulate that on average, across the range of products manufactured in the U.S. each
year, assembly accounts for more than 25% of manufacturing cost – or roughly $500 billion. One goal of
a smart assembly initiative should be to achieve an improvement of at least 20% in assembly related
manufacturing cost, representing an increase in productivity conservatively estimated at $100 billion
annually. The examples below provide evidence that such a cost reduction is wit hin the realm of
possibility.” Can we have a NIST economist help out here?? While this analysis may vary greatly across
different sectors of manufacturing, it certainly illustrates the point that there is very high value in
improving technologies and processes associated with the assembly of components into final products.
In the narrower perspective, assembly efficiency and capability is a key competitive discriminator in
every product manufacturing sector. Assembly time is a key driver of time-to-market. Companies that can
move an innovative new product from the drawing board to the loading dock before everyone else, gain a
huge advantage in profitability. The ability to quickly retool and reconfigure an assembly facility to
produce the “hot product” is critical to sustaining competitive advantage. Boeing, for example, has
enjoyed a tremendous advantage in obtaining orders for the in-production 7E7 Dreamliner, while chief
competitor Airbus struggles to get its first A350 XWB counterpart out the door – a market swing easily
measured in the tens of billions of dollars. 7
While the value of assembly to the manufacturing sector is clear, the true question is, can we contribute to
improved assembly? The answer to this question is clearly “yes” – numerous case studies show dramatic
results from applying the tools and technologies that would be contained in a smart assembly toolkit.
Toyota’s V-Comm Digital Mockup program va lidates the entire vehicle and the vehicle process through
digital assembly. Before the program, 80% of problems were related to assembly issues. With V-Comm,
the lead time for production has been shortened by 33%, design changes reduced by 33%, and
development costs lowered by 50%.8
Boeing is a pioneer in smart assembly and supply chain integration. In the 777 aircraft program, they
reduced product cycle development time by 91% and reduced labor costs by 71%.9 Using virtual
assembly tools for one defense program, the company:
          Identified and eliminated 160 interferences and “unbuildable” conditions
          Reduced rework flow time by 1 to 2 months

5
    Bureau of Economic Analysis, http://www.bea.gov/bea/dn/nipaweb/SelectTable.asp?Selected=N
6
Industry week, October 2006, http://www.industryweek.com
7
 ThomasNet Industrial Newsroom, Industrial Market Trends,
http://news.thomasnet.com/IMT/archives/2006/10/aircraft_design_king_of_sky_boeing_dreamliner_versus_airbus
_a350xwb.html?t=recent
8
    From materials provided by Delmia, Inc.
9
    From materials provided by Delmia, Inc.


                                                DRAFT
A Report on “Smart Assembly”                                                                         Page 8

                                                    DRAFT
              Reduced final assembly personnel by 40%
Lean Practices are certainly a major component of a smart assembly activity. In a case study reported by
Oracle, a test equipment manufacturer implemented a lean assembly program and realized benefits
including an 80% reduction in unit cycle times, 70% reduction in direct labor costs, and a 90% decrease
in part shortages.10
Maple Landmark Woodcraft is a company of 33 employees that makes high-quality toy trains. By
implementing an assembly cell and optimizing the flow, they were able to reduce the number of times the
product is handled from 15 to 12, reduce processing time from 5 days to 4, slash assembly labor by 50%,
and maintain a 95% on-time delivery rate (shipment within 24 hours). This example in particular shows
that Smart assembly works for any size of operation. 11
NOTE TO PARTICIPANTS/READERS
Can you provide additional success stories for this section that make the business case???
These are just a sampling of cases where implementation of improved assembly technologies and
practices are dramatically reducing the cost of assembly while improving product quality. Companies are
investing in specific solutions that meet their individual needs. However, there are infrastructure tools and
capabilities that are needed across sectors and, in some cases, by all manufacturers. These challenges are
too large for any one organization to undertake alone, and it is impossible for a single organization to
justify the investment. Therefore, delivering success requires a cooperative activity that will define the
priorities for smart assembly and deliver the puzzle pieces to achieve a major success for U.S.
manufacturers.

Vision for Smart Assembly
Defining requirements for development of Smart Assembly technologies requires an accepted definition
of the characteristics and attributes of a smart assembly system. The vision suggested below is a
summary/synthesis based on the detailed workshop results contained in the Appendices.
In the extreme futuristic visionary case, all assembly would be done by automated systems with sensory
abilities that enable machines to replace the human element. It is also tempting to define an environment
where modeling and simulation systems would respond to product and business requirements by
automatically developing optimized processes for most efficient assembly. It could be proposed that, with
sufficiently robust process models, automated equipment could be instrumented to enable the execution of
all operations required to ensure the quality of all products – in-process. Finally, this concept could be
extended to include a prognostic capability wherein the health of a machine or system of machines is
monitored and controlled to create a self-diagnosing, self-healing manufacturing environment.
The vision presented below highlights key elements of “smart assembly” that are within the realm of
technical feasibility and provide impetus for collaboration to fill the technical gaps.
A Smart Assembly System is:
               Collaborative: People and automation work collaboratively in a shared environment.
               Re-configurable: The environment can be re-configured/re-programmed (at minimal cost) to
                accommodate new product, equipment, and software variations and to implement corrective
                actions related to fault conditions. The assembly systems consist of modular, plug-and-play
                components that enable flexibility.
               Model and Data Driven:

10
     http://www.oracle.com/industries/high_tech/bringing_lean_ems_oraclewhitepaper.pdf
11
     Success stories from the NIST Manufacturing Extension Partnership



                                                  DRAFT
A Report on “Smart Assembly”                                                                           Page 9

                                                     DRAFT
                  o   Modeling and simulation systems allow all designs to be fully evaluated and all
                      design/engineering changes are made first in the computer (virtual) and are propagated to
                      the plant floor (physical).
                  o In process measurements (of both continuous and discrete variables) are utilized to:
                           Continuously update the virtual representation to match physical reality
                           Control and optimize product quality, throughput and cost during routine
                              operation
                           Predict, diagnose and implement effective corrective actions related to non-
                              routine operations (e.g. failure modes and fault conditions)
               Capable of learning: The same “mistake” is never made twice, and lessons from past
                operations prevent the first occurrence.

Logical Structure for a Smart Assembly Initiative
Workshop attendees were divided into four groups to address smart assembly issues from four different
perspectives:
              End use
              Research
              Infrastructure and standards
              Integration and deployment
In a structured process, each group defined the key attributes and characteristics of smart assembly. After
identifying and discussing key characteristics the participants responded to the question, “what needs to
be done and why?” in the context of the group perspective. In this way, the key ideas for success were
identified and elevated. A consensus was sought concerning the most important topics, and a 1-page
description of the key recommendations was prepared. (See Appendix B for details).
Based on an analysis of the detailed workshop results, we have organized the recommendations into areas
of common interest. Table 1 is an initial attempt at creating such a grouping. Figure 1 depicts a logical
structure for a Smart Assembly Initiative based on this grouping and on the detailed results in Appendix
B: Presentation of Results from Workshop
Breakout Sessions.
The four common themes include:
          Flexible Assembly Systems
          Model-Based Assembly Operations
          Intelligent, Closed-Loop Assembly
          Infrastructure, Standards, and
           Interoperability




                                                        Figure 1: An initi al view of a structure for smart
                                                        assembly around which a program can be devel ope d.




                                                   DRAFT
A Report on “Smart Assembly”                                                                      Page 10

                                                DRAFT
Flexible Assembly Systems:
Advances in robotics, materials, sensors, controls, effectors, and assembly concepts will support the
emergence of assembly systems that enable concurrent production of increasingly disparate and
customized products in a single manufacturing cell or production line. Modular, multifunctional assembly
system components will be readily reconfigurable – ultimately autonomously reconfigurable – allowing
rapid changeover to initiate production of a new product variant or an entirely new member of a product
family. These assembly systems will be self-integrating and self-configuring, negotiating their respective
“roles and responsibilities;” based on digital knowledge of product, process, and business requirements as
defined by applicable product and process models. People and automation will work together safely and
effectively, in a shared environment
Model-Based Assembly Operations:
A virtual environment will enable optimization and error elimination in assembly processes. Product
requirements and manufacturing capabilities/infrastructure will drive the creation of product models for
smart assembly. The product models will support the definition and development of the best assembly
processes, with optimization and evaluation done in “model space” of the virtual environment. The
process models will be proven before operation begins and will be robust enough (accurate and with
sufficient fidelity) to directly transfer to operations. The parameters that ensure satisfaction of
requirements in operation will be included in the models. Hence the result: an environment wherein all
assembly processes are validated before operation and where all operating parameters are included in the
model-based assembly system. The model then becomes the basis for intelligent closed-loop process
control and health monitoring (see below) and provides the foundation for error-free assembly. These
parameters will be continuously updated so that the virtual environment remains an accurate model of the
plant floor throughout the product life cycle.
Intelligent, Closed-Loop Assembly:
The assembly floor will be a sense/analyze/advise-and-respond environment. Sensors will monitor every
parameter that is important to the operation, and control limits will be set for all parameters. The human
in the loop will be aided by excellence in information, instructions on what and how to perform, and
monitor assurance of acceptable completion of tasks. The state of the assembly will be evaluated at all
times, with any deviation made known. The assembly environment will function in a manner that is
similar to the immune system of the human body, wherein anomalies that have no symptoms are
responded to in a very effective manner. This mindset is giving birth to a new discipline called immune
systems engineering. It is an environment wherein there is sufficient intelligence to monitor key
parameters and determine, mandate, and ensure execution of the best response. Self-diagnosing and self-
healing will be attributes of the systems. The intelligent closed-loop assembly environment will be
achieved through advances in control and manufacturing diagnostics/prognosis technologies that embrace
open architecture and modular functionality. The control/diagnostic function will be linked to the model-
based environment to support the application of knowledge with data to enable automated generation of
the necessary information to drive, control, monitor, and maintain assembly operations.
Infrastructure, Standards and Interoperability:
The smart assembly environment will be interoperable at all levels (e.g. tool, cell, zone, line, plant,
enterprise), with plug-and-play systems (both “virtual” and “real-time”) that communicate seamlessly
across domains and different commercial toolsets. The right standards will be in place to assure that data
and information are transferred for easy interpretation and action. Both the sending and the receiving
devices will speak the same language or have integral real-time translation incorporated into the
communications systems. While unified standards are desirable, such total agreement may threaten
competitive environments. Therefore, harmonization of standards for sufficient coverage of all assembly
functions is the minimum requirement of the future state.


                                              DRAFT
                   A Report on “Smart Assembly”                                                                                Page 11
                                                                         DRAFT


Table 1: Recommendations from the workshop are shown grouped by common themes (categories).

       Source                                 Recommendation                                                            Category
    User 1         Reconcile models and physical systems                                       Model-Based Assembly Operations
    User 2         Provide an architecture for modeling                                        Model-Based Assembly Operations
                   Assembly operations and systems
    User 3         Provide "plug-and-play" assembly systems                                    Flexible Assembly Systems
    User 4         Provide intelligent, safe devices (safety engineered in)                    Intelligent, Closed-Loop Assembly
    User 5         Provide modular, low cost, reusable assembly systems                        Flexible Assembly Systems
    User 6         Provide real-time decision making from shop floor data                      Intelligent, Closed-Loop Assembly
    Research 1     Provide additional capability and adaptability in assembly systems          Flexible Assembly Systems

    Research 2     Provide improved, standard programming capabilities                         Model-Based Assembly Operations & Infrastructure,
                                                                                               Interoperability & Standards

    Research 3     Provide improved sensing and control systems for human/machine              Intelligent, Closed-Loop Assembly
                   interaction
    Research 4     Improve integration of modeling systems: fill gaps in modeling capability   Model-Based Assembly Operations

    Research 5     Improve decision support systems for real-time operation                    Intelligent, Closed-Loop Assembly
    I&S 1          Provide a systems engineering approach to create and manage a               Model-Based Assembly Operations
                   persistent capability and requirements model throughout the life cycle

    I&S 2          Develop an assembly taxonomy and populate with associated models            Model-Based Assembly Operations

    I&S 3          Harmonize assembly standards and assure consistent application - at         Infrastructure, Interoperability & Standards
                   least across common applications
    I&S            Provide traceability, forward and backward, for assembly requirements       Model-Based Assembly Operations

    Deployment 1   Mitigate the risk of technology deployment for assembly through
                   collaboration
    Deployment 2   Provide interoperable systems for data exchange                             Infrastructure, Interoperability & Standards

    Deployment 3   Provide tools for collaborative systems engineering                         Model-Based Assembly Operations



                                                                       DRAFT
A Report on “Smart Assembly”                                                                        Page 12

                                                 DRAFT


Conclusions and Next Steps
The workshop provided a substantial first step towards the formulation and launching of a Smart
Assembly initiative. The characteristics and attributes for the future vision state were clearly defined,
which fed the definition of priority recommendations. With this document, the recommendations were
placed in a structure (Figure 1) to support the next steps. The structure addresses the functional areas for a
unified program and recognizes that there must be sector-specific “slices” within a broader Smart
Assembly initiative to develop coordinated “technology” and “business” roadmaps.


The following specific steps are recommended:
    (1) Early in 2007, NIST should convene a larger body to build common technology roadmaps for the
        “horizontal” swimlanes of Figure 1. These roadmaps will provide a much richer view of smart
        assembly technology. Details contained in the Appendices should provide a starting point for the
        technology roadmaps.
    (2) Sector-focused teams should be formed
        around the “vertical” swimlanes in Figure 1.
        These teams would build
        implementation/deployment roadmaps and
        end-user application scenarios for integrating
        smart assembly technologies in their
        businesses. The sector roadmaps will share
        the common foundation of the Smart
        Assembly initiative, and will address
        additional needs specific to that sector.
        Establishment of sectors for initial focus will
        be based on industry leadership and interest.
        It is emphasized that the goal is to create a
        living program to address the challenges of
        smart assembly. It will be a federated
        program with common goals and common            Figure 2: The S mart Assembly Initiati ve will be
        activities, and with interest groups focused    a composite of sector-s pecific programs wi thin
                                                        the context of a highly integrated collaborati on
        on specific sectors. A notional view of a
        smart assembly program is illustrated in Figure 2. The specific mechanism for
                                                        common environment.
        organizing/empowering these sector-focused teams is yet to be determined; however NIST might
        play a facilitating role.
    (3) To facilitate the development of integrated technology and business roadmaps, NIST (possibly
        with additional support from end users) should commission a special study to benchmark current
        technology trends and business oriented application use case scenarios for smart assembly
        systems.
    (4) Assuming successful development of integrated technology and business roadmaps with strong
        industry leadership, NIST should consider the creation of a National Smart Assembly Testbed in
        cooperation with industry sponsors to validate the interoperability and performance of smart
        assembly modules and systems. Use case scenarios identified in the business roadmaps should
        provide the primary requirements for interoperability and performance standards and testing.




                                               DRAFT
A Report on “Smart Assembly”                                                                         Page 13

                                                 DRAFT
In closing this discussion of path forward, let‟s re-visit the vision for smart assembly. In the future, next-
generation assembly equipment and systems will be flexible, agile, and with the appropriate level of
automation. People and automation will work together in a shared and highly collaborative environment.
Flexible systems will respond at minimal cost to changing requirements and will provide processes that
ensure the satisfaction of each assembly process. Automated equipment will embrace modular
construction that supports reconfigurability and provides immunity from obsolescence. Special-purpose
assembly systems will become commonplace for small-lot production. These systems will be designed or
configured for assured satisfaction of requirements and produce dramatic cost savings over conventional
assembly. Safe operation of processes and assurance that the resulting products will operate within a
compliant and safe envelope will be designed and engineered into the products and processes. The
systems will also support intelligent conveyance of components and subassemblies.
Call to Action
The organizers of the workshop call upon the participants to spread the word. Share this work with your
colleagues and urge them to get involved. It is our desire that a fire will be kindled that will grow to a
powerful program to define a new standard of excellence and capability in assembly operations.
This document is released to the participants in the workshop for review and comment. It is a draft
document that will be refined before its final release, before the end of 2006. Comments, corrections, and
additions are welcome. Most importantly, ideas for moving forward are strongly encouraged.
The Smart Assembly WikiSpace website contains copies of all workshop materials and this final report.
This site is a COLLABORATIVE site and all participants/interested parties are encouraged to utilize the
site as a “living forum” to support a Smart Assembly Community of Practice:
                                   http://smartassembly.wikispaces.com/
All comments should be addressed to Dr. Robert Tilove at robert.tilove@nist.gov. His phone number is
(301) 975-4345.
Thanks to all who contributed, and we look forward to moving ahead in this important arena.




                                                DRAFT
A Report on “Smart Assembly”                                                                       Page 14

                                                DRAFT
Appendix A: Workshop Plenary Session Summary
During the first day of the workshop, participants heard presentations from smart assembly stakeholders
representing a cross-section of the manufacturing community – aerospace, electronics, automotive,
research, automation controls, modeling and simulation, product life cycle management, and process
integration. This section gives a synopsis of each presentation. When presenter‟s notes were included with
the slides, they are paraphrased and condensed or, in some cases, used verbatim. The presentation slides
are available to workshop participants at http://smartassembly.wikispaces.com/.

Smart Assembly Systems – Bob Tilove, Technical Fellow, Manufacturing Systems
Research Laboratory, General Motors R&D; Visiting Scientist, Manufacturing
Engineering Laboratory, NIST
This introductory presentation laid out the agenda and established the focus of the two-day workshop.
When we think about what makes assembly systems “smart” or “intelligent” – two questions arise. The
first is “How can we make mechanical systems – like tooling and equipment – smarter, or more adaptive,
or more intelligent?” At least as important as that is “What can we do to improve the effectiveness/quality
of PEOPLE in tasks and decision making?” We can talk about “smart assembly” from either perspective,
but we should also ask the question “What happens as these two worlds start to come together?”
“Intelligence” in agile manufacturing systems implies mechanical systems, sensors, and controls that are
adaptive, automated, and autonomous. Such systems include intelligent robotics and adaptive process
control systems. Just as
improvements are needed in
agile manufacturing systems,
better tools and training are
needed to enable people to be
more agile in their decision
making and their interactions
with intelligent systems. These
include tools for dealing with
unanticipated changes,
reconfiguring systems quickly
as requirements change,
designing and validating
products and processes using
modeling and simulation,
diagnostics and prognostics, and
enterprise analytics that enable    Figure 3. Attri butes of Smart Assembl y include intelligent systems,
better business decisions. Figure   sensors, and controls, combi ned with better decision-making tools for
                                    the people who interact with the systems .
3 shows some of the key
attributes of a smart assembly system.

Network Enabled Manufacturing – Bryan G. Dods, Advanced Manufacturing
Research & Development; AeroStructures, Manufacturing & Support
Technologies, Boeing – St. Louis
This presentation described the dramatic changes that have occurred in aircraft manufacture at Boeing
over the past ten years and forecasted trends over the next ten years. Solid modeling, unitized structure,
assembly automation, and information systems have been the biggest contributors to these changes.
Boeing‟s present focus is on lean manufacturing, common systems, and ergonomics. To prepare for the


                                               DRAFT
A Report on “Smart Assembly”                                                                       Page 15

                                                 DRAFT
next step-function improvement in aircraft manufacturing, the Advanced Manufacturing Research &
Development organization is exploring network-enabled manufacturing.
Boeing‟s Advanced Manufacturing Research and Development organization‟s vision for network-enabled
manufacturing consists of eleven strategies and relies on multiple organizations and a diverse set of skills.
The strategies (which are described in greater detail in the presentation notes) are:
  1)   Data input and collection
  2)   Data management
  3)   Hardware and equipment
  4)   Process variables
  5)   Application
  6)   Model-based environment
  7)   Network
  8)   Architecture
  9)   Business process redesign
 10)   Infrastructure
 11)   People and culture
The first five strategies comprise the network-enabled manufacturing system “front end” – the pieces with
which the operators will interact on the shop floor on a day-to-day basis. The next five strategies are
enabling technologies that will form the “backbone” of the network-enabled manufacturing system. The
final strategy addresses how people and the Boeing culture will adapt to changes in the workplace.
Boeing Advanced Manufacturing Research & Development is developing test beds to demonstrate
various aspects of the network-enabled manufacturing concept. Boeing is relying on close coordination
with industry leaders and universities to create the necessary technology advancements to realize its
manufacturing transformation over the next ten years.

Electronics Industry Perspective - Tom Babin, Fellow of the Technical Staff,
Motorola Labs – Physical Realization Research Center
This presentation gave numerous visual examples that illustrated the current state in the electronics
industry with regard to printed circuit board assembly; test, inspection, and measurement; and final
assembly processes at Motorola. The following technology problems, gaps, and issues were identified for
each of the three areas:

Board assembly
   Lead Free Soldering
   Cost Reduction
   Collaborative Development (OEM-EMS)
   01005 Passive Component Process Technology
   Stencil Technology
   Flexible Substrates
   Fine Pitch Rework

Test, inspection, and measure
    Optical/X-Ray Inspection Improvements
    Access for Test due to Size Reductions
    Reliance on Functional Test (BIST)


                                               DRAFT
A Report on “Smart Assembly”                                                                        Page 16

                                                 DRAFT
Final assembly
    Lack of Standards & Standard Metrics
    Cost Reduction
    Very Little R&D Spending
    Optoelectronics Assembly
Projected trends in the electronics industry are for decreasing cost, time, and pitch; and increasing
functionality, component density, and quality.

Smart Assembly Systems – GM Perspective – Roland Menassa, Laboratory Group
Manager, Agile Equipment and Processes, Manufacturing Systems Research
Laboratory, General Motors R&D
Change is a way of life at GM. It takes about two years of planning to execute a typical new GM vehicle.
GM‟s 55 plants worldwide and over 9-million-vehicle capacity make the launch of a new model a
daunting task with today‟s technology. Much more is needed in terms of equipment intelligence in order
to switch from a hardware-change to a software-change automotive industry. Table 2 shows some of the
major automotive industry challenges along with key technologies that can combat the challenges.
                   Table 2. Today's automoti ve industry faces many challenges – chief among them being
                   affordability.
                    Industry Challenge         Key Technology
                        Energy                    Advanced Propulsion
                        Environment               Advanced Materials
                        Safety                    Vehicle Electronics, Controls and Software
                        Congestion                Telematics
                        Affordability             Smart Manufacturing


About 100 years ago, the auto industry was a craft industry. At that time, Henry Leland, founder of
Cadillac, took an idea from the gun industry that radically altered automotive manufacturing – it was the
idea of having interchangeable parts. The other Henry [Ford] built on the that fundamental idea to create
the assembly line – an idea he took from the meat packing business. Much later, Taiichi Ohno of Toyota,
introduced what we now know as lean manufacturing – an idea adapted from the grocery business – this
is now the dominant paradigm. The next paradigm takes its impetus from the explosion of the World
Wide Web. It will build on “lean” but allow us to go to a whole new level of performance – real-time
optimization of our manufacturing enterprise.
There are technological discontinuities in today‟s manufacturing arena that must be harmonized and
integrated in order to implement smart manufacturing. For example, pervasive sensing and
communications capabilities are needed to support real time manufacturing, pervasive device intelligence
and servo actuation enable agile manufacturing, and pervasive visualization/simulation technologies are
necessary for virtual manufacturing. These must all be tied together.
The manufacturing enterprise (sometimes referred to as “big „M‟ manufacturing”) extends well beyond
the four walls of the plant. The remainder of this presentation will focus on smart assembly as it relates to
those activities that relate to physically assembling parts/components into final products (“small „m‟
manufacturing”).



                                               DRAFT
A Report on “Smart Assembly”                                                                         Page 17

                                                   DRAFT
The vehicle assembly plant
shown in Figure 4 illustrates the
scope/size of a high volume,
highly automated assembly plant
in North America or another
industrialized region. It occupies
2.5-3.5 million square feet,
houses 400-600 robots, 400-600
PLC‟s, and 20-25 miles of
conveyors. There are typically
800-1200 vehicles in the system,
1000-1200 operators per shift,
generating an output of 60
vehicles per hour.
In developing countries, the
manufacturing footprint is
considerably different. There is
much less automation and much
lower volume, but the facility in Figure 4. A typical vehicle assembl y pl ant occupies 2.5-3.5 million s quare
the less-developed country is      feet and may contain 20-25 miles of conveyors.
also usually much more flexible
and capable of building a wider variety of models than its industrialized counterpart.
When we think about smart assembly, we must consider both scenarios. The same business drivers apply
to the high-volume, highly automated plant and to the low-volume, labor-intensive plant. The following
business drivers define some of the most challenging problems that we hope a smart assembly initiative
will address:
       Time to Market
            o     Reduce time to launch and accelerate
       Cost
            o     Reduce hours Per Vehicle (Direct cost)
            o     Improve maintenance efficiency (Indirect cost)
            o     Reduce engineering cost (Structural cost)
       Throughput
            o     Minimize/eliminate downtime
            o     Optimize resource utilization/balancing
       Quality
            o     First time quality
       Responsiveness/Flexibility
            o     Build any product in any plant


Several examples were presented of manufacturing trends in the automotive industry that may become
enabling technologies for smart assembly. The following were presented as key characteristics and
attributes of smart assembly from the automotive industry perspective:


                                                DRAFT
A Report on “Smart Assembly”                                                                      Page 18

                                                DRAFT
Time to market
    The “manufacturing validation” process is eliminated. It becomes continuous improvement of a
        process that is already “in control”.
       A virtual factory is developed that is analogous to the flight simulator for aircraft
       Virtual and real-world environments are fully synchronized.
       The capability is developed to upload real information to a virtual environment. Virtual systems
        “understand” the plant floor.
       100% of the “real” assembly plant is accurately represented mathematically in a “virtual” model.

Cost
       “Intelligent” cooperative robots are used in general assembly.
       Near-zero reactive maintenance is required.
       Centralized remote and highly automated diagnostics/prognostics capabilities exist.
       Minimized disruption to the plant floor during introduction of new products.
       Smart design for reduced hours per vehicle.
       There is no redundant data, and constraints are automatically propagated between systems. All
        systems are converged into one logical “do work” portal. The “engineering factory” is error-
        proofed.

Throughput
    Highly automated “error-recovery”. (e.g.: re-allocation of work to robots).
    Effective plant floor decision support tools. Interoperability from “shop floor” to “top floor”.
    Reconfigurable factory.
    Agent-based, distributed controls and networks.

Quality
   Adaptive dimensional control.

Responsiveness/Flexibility
   “Re-programmable factory.”
   Download virtual to real. Real time systems “understand” virtual.
   Change from hardware to firmware changes for new products.
   Real time “remote control” of the factory.
   Pervasive sensing and wireless connectivity.
   Plug-and-play plant floor devices.
   Reusable plant.

Mechanical Assembly and Researches Needed to Make it Smart – S. Jack Hu and
Dawn Tilbury, Department of Mechanical Engineering, The University of Michigan
This presentation identified and addressed several of the more challenging research problems (related to
the design, engineering, validation, construction, installation, launch, and operation of assembly processes
and systems) that must be solved to make “smart assembly” a reality. Several observations were presented
regarding the role of assembly in product realization:
       A properly defined assembly interface can allow a company to mix and match parts to create
        variety with minimum cost.

                                              DRAFT
A Report on “Smart Assembly”                                                                             Page 19

                                                  DRAFT
         An appropriate assembly sequence can permit a company to customize a product when it adds the
          last few parts.
         Properly defined subassemblies permit a company to design them independently or outsource
          them.
         A well defined assembly-centered product development process can make ramp-up faster.
    The performance of an assembly system can be analyzed and model mathematically in terms of
    quality, cost, and responsiveness.
    Key characteristics and attributes of a smart assembly system from a research perspective include:
         Smart tools, such as feeders, fixtures etc.
         Collaborative robots working with people intelligently and smoothly
         Automatic compensation for quality
         Effective response of maintenance personnel to unexpected change of system to ensure system
          quality and productivity
         Ability to repair and reconfigure itself.
Smart assembly research challenges are presented in Table 3 for five broad areas – quality, reliability and
maintenance, variety and complexity, control design and validation, and the alignment of real and virtual
environments.

Table 3. Research challenges that address key attri butes of a smart assembl y system.
        Topic                                            Research Challenge
Quality                   Developing models incorporating all sources of variation, including operator in fluence
                          Developing models early to aid system design
                          In-line adjustment and compensation
                           o Sensor placement
                           o Controllab ility issues
Reliability and           Interactions of quality and reliability in throughput modeling
Maintenance               Allocation of maintenance resources to ensure fast response
                          Self-reconfiguring of assembly system for exception handling
                           o Task re-allocation
                           o Self repairing
Variety and               What is complexity?
Complexity
                          Understanding the relationship between complexity and performance
                          Balancing between product variety and manufacturing comp lexity
Control Design            Automatic generation of control logic fro m virtual model
and Vali dation            o Developing libraries of control co mponents
                           o Appropriate granularity for the modules in the lib raries, neither too big (entire
                               system) or too small (single axis)
                           o Incorporation of mult iple modes of control (auto, manual)
                           o Interface specificat ions for the modules, across proprietary platforms, to ensure
                               seamless operation
                           o Incorporation of the operator through the HMI panel
                           o Safety criteria, to protect mach ines, parts, and operators
                          Logic verificat ion (i.e. formal mathemat ical p roof of correctness)
                           o Formal representations of logic specification (currently often specified in English)
                           o Formal models of control code, including execution semantics (not adequately
                               addressed by existing standards, e.g. 61131 or 61499)


                                                 DRAFT
A Report on “Smart Assembly”                                                                           Page 20

                                                 DRAFT
                          o Generating mathematical models (e.g. diff. eqns or discrete-event systems) fro m
                             simu lation model for formal verification
Alignment of Real        Validate the fidelity of the simu lation model by co mparing simulated performance with
and Virtual               real performance
Environments
                         Synchronization across mult iple do mains with interacting controllers
                         Integrating models at different levels of abstraction
                          o Machines: force, position, accuracy, CNCs, …
                          o Cells: material handling, maintenance, PLCs,…
                          o System: part flow, scheduling, databases, control & co mmunicat ion networks, …

Manufacturing Operations Standards Need to Converge into a Manufacturing
Application Integration Framework – Charlie Gifford – Director, Lean Production
Management, GE/Fanuc
This presentation addressed the question “What are the tough implementation problems that must be
solved to make „smart assembly‟ a reality?” Emphasis was placed on the infrastructure and standards
necessary to support the manufacturing environment. The opening part of the presentation highlighted the
differences in a 20th century manufacturing system and a 21st century system. Flexibility in operations
and real-time visibility enable the optimization of the profit margin, and are becoming key success
discriminators. In business-to-manufacturing (B2M) and business-to-business (B2B) commerce
environments, there are many applications performing the same functions, and there is no unified
information architecture. This missing link is a key void in a next generation success strategy. The
presentation contrasted the emphasis on information systems and manufacturing operations, highlighting
a study by AMR Research that showed an overwhelming majority of the corporate investment is going
into information and enterprise management systems with very low investments in manufacturing
operations. All of this information was presented to build a case for a manufacturing applications
framework and the coupling of this framework with lean process and six-sigma production improvements.
Smart assembly requires a common schema or architecture that can support lean and six-sigma
transformations. Enterprise Resource Planning (ERP) systems require consistent production data models
and data exchange models. A consistent/standard message structure is imperative for effective B2M and
B2B operations, and success in ERP demands an effective means of data integration. The presentation
closed with a call for a production schema or architecture to assure effective communications.

Production Management – Delivering End-to-End PLM – Mitch Vaughn, Chief
Technology Officer, Production Management, UGS
This presentation discussed how Production Management facilitates innovation by bridging gaps in the
product lifecycle management (PLM) process and delivers the means to measure innovation success. The
process of innovation is about the successful commercialization of great product ideas. As seen in
depiction of the product lifecycle shown in Figure 5, innovation is not just about great product design and
engineering, but also encompasses execution phases including manufacturing and field servicing efforts.
In short, innovation is a complex, interrelated series of processes.




                                                DRAFT
A Report on “Smart Assembly”                                                                    Page 21

                                                 DRAFT




                 Figure 5. Innovation is the successful commercializati on of product i deas.
Production Management is a core PLM component that meaningfully contributes to the process of
innovation in the following ways:
    •   By driving continuous product and process improvement initiatives
    •   Bridging the gaps between “as planned” and “as built”
    •   Improving manufacturing performance with faster time to market
    •   Integrating with enterprise-wide applications to drive the process of innovation
    •   Supporting executive decision making with real-time analytics to drive innovation success.
The innovation challenge is to quickly determine the best innovations and discard those that will be
unsuccessful as early in the PLM process as possible so as not to waste time and resources on poor
innovations. This assures that only good ideas are brought to market much more quickly. In short: fewer
development projects but with greater commercialization success. By establishing repeatable processes
and a global knowledge repository, the process of innovation can be transformed by capturing best
practices.
UGS has identified eight innovation initiatives where PLM contributes to the process of innovation with
the goal of successful delivery of innovative products that enable companies to lead markets in terms of
share, price point, and overall business growth. (These initiatives are new product development, value
chain synchronization, enterprise data management, commonization & reuse, knowledge/IP management,
regulatory compliance, production efficiency, and systems engineering and mechatronics.) However,
promises of innovation can be cut short during manufacturing, when there is a “disconnect” between
product design and process execution. This disconnect can result in missed market opportunities, lowered
price points, and loss of ground to competition, and, possibly, the forfeit of leadership position.
Production Management is a key enabler in terms of integrating PLM capabilities with business needs by
bridging the gaps in innovation execution when it comes to offering true end-to-end product lifecycle
management.
These four initiatives are of particular note in terms of how Production Management can positively
impact any company‟s innovation goals:



                                               DRAFT
A Report on “Smart Assembly”                                                                     Page 22

                                               DRAFT
    •   New Product Development – accelerating production launch and faster delivery to market that
        improves returns on innovation investments.
    •   Enterprise Data Management – facilitate continuous process improvement initiatives with capture
        of “as built” records for reconciliation with product design and manufacturing process planning.
    •   Regulatory Compliance – provide clear proof of compliance from shop-floor both in terms of
        process control and traceability.
    •   Production Efficiency – Maximize manufacturing opportunities thereby assuring that
        manufacturing is executed as planned.
Why hasn‟t Production Management become a mainstream organizational platform, as compared to
Product Data Management and/or Enterprise Resource Planning? There are a number of barriers, but the
following three are the most important:
    1. There is a “disconnect” between business and shop floor applications. There has been insufficient
       integration – meaning that the reconciliation of the two platforms has been nominal, which
       minimizes their effectiveness when it comes to manufacturing operations management.
       Furthermore, there is some confusion about ERP capabilities when it comes to the manufacturing
       shop floor, whereby ERP is deemed to have complete Production Management capabilities.
    2. There has been a historic disconnect between PLM strategies and manufacturing whereby PLM
       efforts have mostly resided around product development.
    3. Shop-floor events are typically reported “after-the-fact” which only allows reactive management.
       A lack of a logical interface means that management does not have the decision support tools nor
       are they able to reconcile analytics with business metrics to effectively determine questions like
       “How am I doing?”; “Will I meet my commitments?”; or “Am I performing within my
       established parameters?”
The UGS vision is to bridge these gaps by facilitating PLM integration via the Teamcenter Manufacturing
Backbone. The backbone will act as a bridge in the current PLM gaps by enabling a collaborative
environment with other PLM applications along with enterprise integration with business applications,
such as ERP.
The foundation of Production Management is being able to connect to the shop floor in a meaningful way
both in terms of interacting with the shop floor as well as parsing data from the shop floor to facilitate
real-time monitoring and control. By taking this path, the shop floor becomes an integral part of the PLM
process and removes the barriers to achieving greater levels of innovation.
In summary, Production Management is a core PLM component that, in the past, has been overlooked as
a result of integration barriers which are now resolved by the Teamcenter Manufacturing Backbone.
Production Management transforms PLM and helps drive the process of innovation by:
    •   Being a strategic enabler of innovation initiatives
    •   Filling the gaps between product and process design by facilitating continuous process
        improvement
    •   Driving the enterprise-wide process of innovation through the Teamcenter manufacturing
        backbone
    •   Providing reconciled analytics to drive innovation success.




                                              DRAFT
A Report on “Smart Assembly”                                                                      Page 23

                                                DRAFT
Smart Assembly – Bob Brown, CEO Advisor, DELMIA Corporation
Smart assembly is a next-generation capability in assembly systems and technologies, which integrate
virtual and real time methods (as depicted in Figure 6) in order to achieve dramatic improvements in
productivity, lead time, and agility. A smart assembly system uses smart product and resource data to
drive the assembly process, and is able to quickly respond to engineering changes and new equipment
configurations.
Our vision for a smart assembly
environment should include paperless, 3D-
only models that are able to completely
capture design geometry, design intent, and
engineering specifications, completely
eliminating the need for 2D drawings.
Information from the models will be
directly usable by and reconcilable with
geometry-based manufacturing process
planning software. Knowledge contained in
the design and manufacturing planning
definition can leveraged to automatically
generate shop floor work instructions.
Knowledge contained in design and            Figure 6. S mart assembly systems integrate virtual and
manufacturing planning definition can also real time methods.
be used to automatically generate robot and
PLC programs, and will provide a concurrent engineering environment that is appropriate for both the
production engineer and the control engineer. Such programs would be capable of self-adapting to design
changes, and provisions would be made for defining multiple models and configurations in the same
program, thus reducing reprogramming time and effort.
Challenges of Smart Assembly – What are the tough implementation problems (infrastructure, standards)
that must be solved to make smart assembly a reality? In order to understand the challenges, it is
necessary to examine two manufacturing paradigms – the traditional paradigm in which the OEM‟s
internal and external suppliers build to the OEM‟s print; and the emerging paradigm in which the OEM
becomes a systems integrator, and partners/suppliers design and build components based on requirements
communicated by the OEM. Table 4 lists some of the major challenges to achieving the vision of smart
assembly for both paradigms.
In the first paradigm (build-to-print), all product specifications that drive manufacturing and assembly
must be communicated and maintained constantly. In most cases, these specifications are still drawing
based, even if the drawing is communicated electronically. The supplier must review each drawing in
detail to understand the requirements. Drawing-based specifications are not easily accessible to a smart
assembly solution. In some cases, model-based production specifications are available to the supplier.
Although these digital specifications are more readily available to a smart assembly so lution, there are
currently no widely used standards for the exchange of such information. A single supplier may provide
manufacturing services to multiple OEM‟s. Because of the lack of standards, suppliers must maintain
interface capabilities to the CAD, PDM/PLM and Knowledge systems of each of their customers.
In the outsourcing, or design and build paradigm, the OEM is a system integrator who communicates
requirements to its partner/supplier, who designs and builds components. This approach eliminates many
of the data transmittal problems associated with the build-to-print business model. It minimizes problems
of disparate CAD, PDM, and Knowledge systems between the OEM and the supplier. However, a
standard model is still needed for communicating form, fit, function and interface requirements to. Such a


                                              DRAFT
A Report on “Smart Assembly”                                                                             Page 24

                                                      DRAFT
standard model should facilitate communications with suppliers and provide the ability to obtain
competitive quotes on a well understood set of requirements.
Table 4. A few of the major challenges that must be addressed in order to achieve the smart assembly vision.
Build-to-pri nt                                              Design and build

         CAD t ranslations including features, attributes         Ability to co mmunicate and manage change of
          and requirements data                                     requirements for
         PDM / PLM interoperability for product                         o Form
          structure, attributes, requirements                            o Fit
         Interoperability between knowledge                             o Function
                                                                         o Interface
          management systems
                                                                   Methodology and standards for OEM to validate
         Change and configuration management of the
                                                                    integration
          above
                                                                   Core Manufacturing Simu lation Data
                                                                    Information Model (CMSDIM)


Smart Assembly – Jim Caie – ARC Advisory Group
Smart assembly can be defined as a next-generation capability in assembly systems and technology,
which integrates assembly processes, people, equipment, and information using both virtual and real-time
methods to achieve dramatic improvements in productivity, lead time, and agility.
         ARC‟s Collaborative Discrete Automation System (CDAS), which was introduced three years
          ago to define a vision for the factory of the future. At the time it was introduced, interoperability
          and complexity were two significant issues that manufacturers had to deal with in order to
          establish a collaborative environment across the manufacturing enterprise. As Figure 7 illustrates,
          these and many other challenges are driving manufacturing operations today.
One of the biggest challenges involves connecting the shop floor with the rest of the manufacturing
enterprise. One of the main themes of CDAS is how to address the P2B challenge. The ARC CDAS
conceptual model continues to evolve based on technology, markets, business drivers, standards, and the
                                                                     needs of the manufacturing
                                                                     community. This model represents a
                                                                     high level contextual view of the
                                                                     functional relationships and domains
                                                                     that comprise a discrete
                                                                     manufacturing enterprise. In the
                                                                     middle we have manufacturing
                                                                     operations management, where the
                                                                     production processes and factory
                                                                     operations have to interoperate and
                                                                     collaborate. Each of these high level
                                                                     functional domains can be
                                                                     decomposed into multiple and more
                                                                     complex sets of processes that
                                                                     define the overall manufacturing
                                                                     process. Modeling these processes
                                                                     and workflows will be essential.
                                                                                Next-generation assembly drivers
        Figure 7. Today's manufacturing operations face unprecedented
                                                                                will require flexible plants that
        challenges, including connecting the production floor wi th the
        rest of the manufacturing enterprise.                                   offer more speed and agility,

                                                     DRAFT
A Report on “Smart Assembly”                                                                  Page 25

                                              DRAFT
lower investment costs, and support for continuous product launches. Smart assembly can address these
challenges by adhering to a vision that addresses:
      PEOPLE – knowledgeable, empowered work teams,
      INFORMATION PROCESSING – event-driven, real-time and actionable
      PROCESSES – strategic, lean, agile, with flawless execution
      TECHNOLOGY – common, safe, and smart environment for man and machine.




                                            DRAFT
A Report on “Smart Assembly”                                                                    Page 26

                                               DRAFT


Appendix B: Presentation of Results from Workshop Breakout Sessions
Workshop attendees were divided into four groups to address smart assembly issues from four different
perspectives – end users, researchers, infrastructure and standards, and integration and deployment. In a
very structured process, each group was asked to define the key attributes and characteristics of smart
assembly. After identifying and discussing key characteristics and attributes of smart assembly, each
group was asked to brainstorm and generate ideas for achieving these characteristics and attributes. The
ideas were stated in the format
“What needs to be done and
why?” Each group was then
asked to organize and group its
ideas into key themes, which the
group then prioritized. The top
three or four ideas from each
group were documented as
recommendations on one-page
summary sheets. The
recommendations were
supported by a problem
statement, identification of root
cause(s), benefits that would be
realized, and a brief action plan.
The basic process followed by
each breakout group is shown in
Figure 8.
The four groups then reconvened
as a single large group and each
group presented its prioritized
ideas in the workshop wrap-up
session. Because the amount of
time available for the groups to
generate their information was   Figure 8. During the workshop, the four breakout groups brai nstormed
less than four hours, most ideas to devel op recommendations for addressing smart assembly problems
were described at a very high    and issues.
level.




                                             DRAFT
A Report on “Smart Assembly”                                                                     Page 27

                                                DRAFT
Appendix B1: End User Perspective
The following attendees participated in the group and process:
        Rikard Soderberg - Chalmers University of Technology
        Swee Leong - NIST
        Johan Carlson - Fraunhofer Chalmers Centre
        Jack Hu - University of Michigan
        Jim Caie - ARC Advisory Group
        Keith Ridgway - AMRC with Boeing, University of Sheffield
        Tom Babin - Motorola
        Bob Owens - Honeywell FM&T
        Roland Menassa - General Motors
        Alkan Donmes - NIST, Scribe
        Kathy Thomas - IMTI, Facilitator

Characteristics and Attributes from the End User Perspective
Based on the knowledge, experience, and interests of the participants, and with an end user perspective,
the following key characteristics and attributes of smart assembly were identified:
       Next paradigm of manufacturing systems, look at the assembly process and drive the intelligence
       Integration of knowledge workers, lean processes, intelligent equipment, and real-time
        information
       Designed and operated in an intelligent way to insure robust performance
       Reconfigurable with interchangeable parts
       Visualization and virtual reality used for assembly process design and operator training
       Flexible automation/robotics, lower cost, business driven, accuracy, cost, and speed
       Optimized geometry and motions together
       System designed around product family and has capability to track quality (geometric variation)
       Data requirements fully defined and supported

Key Themes from the End User Perspective
The group process built on the characteristics and attributes to answer the question, “What needs to be
done and why?” The group identified and prioritized the following six themes for smart assembly:
       Simulation and Visualization
       Reconfigurable Tools and Systems
       Assembly System Reuse
       Business Modeling (PLM/Performance)
       Real-time Actionable Information
       Knowledge Capture and Learning
The group then discussed the needs and issues associated with each theme and generated six
recommendations to address them. Although the recommendations clearly support the needs, further
discussion, input and clarity may be needed to fully define them due to the limited time available at the
workshop. The six themes, along with prioritized recommendations that address the needs of the themes,
are described below.

Theme #1 – Simulation and Visualization
The top priority theme from an end user perspective for achieving smart assembly goals was simulation
and visualization, which includes the application of modeling and virtual reality. Simulation and

                                              DRAFT
A Report on “Smart Assembly”                                                                       Page 28

                                                DRAFT
visualization is used for multiple aspects of Smart Assembly, including, but not limited to, the modeling
of manual and automated assembly process flows and steps, equipment functionality and safety, space
allocation, and operator training. When applied effectively and efficiently with the right data, simulation
and visualization has been proven to significantly reduce cost, reduce cycle time and increase quality
associated with the assembly processes and the delivered product.
The use of modeling and virtual reality allows the manufacturer to analyze, visualize and optimize an
assembly process for making critical business decisions from the initial investment through changes and
final disposition. Essentially, they provide the capability to “try before you buy” before making a
significant investment in equipment, space and other resources. Once created, the models can be used to
quickly analyze failures or potential changes and provide solutions or impacts to the assembly process.
Although the group recognized the value of simulation and visualization, they were quick to identify the
need for real plant data to develop and validate the models, for better integration between the models and
the physical system (real world), for reducing the time to create models, and for more efficient methods of
keeping the models up to date. As such, the two recommendations defined by the group were focused on
the accuracy of models reflecting the physical world and the time to create models. The key needs and
issues associated with the simulation and visualization theme are shown in Table 5, and the two
recommendations are summarized in Table 6 and Table 7.

Theme #2 – Reconfigurable Tools and Systems
The second priority theme from an end user perspective for achieving smart assembly goals was
reconfigurable tools and systems. With the continuous changes in product designs and manufacturing
technologies, it is imperative the assembly process be flexible and quickly reconfigurable for companies
to be responsive and competitive.
Smart assembly processes require both intelligence and functionality that delivers the right quality
product at the right time in an efficient and safe environment. Employing reconfigurable tools and
systems is critical in achieving that goal. The group clearly agreed that it takes entirely too long to
reconfigure manufacturing systems, keeping them from being responsive to their customers and meeting
the competitive challenges. The need for “plug & play” capabilities in plant floor systems must be
satisfied to significantly reduce the cost and time for delivering new product designs, implementing new
technologies, and running smaller batches.
Although manufacturers have devised methods to assure the safety of operators running machines, the
group recognized a significant need to reducing the cost of intelligent, safe assembly processes.
Understanding that potentially new equipment designs may require changes in regulations, the group felt
it was critical to improve the intelligence and safety of equipment for making the factory safer, improving
labor efficiencies, lowering capital costs, optimizing floor space and optimizing the synergy between the
equipment and the human.
After much discussion, the group defined two recommendations for reconfigurable tools and systems
focused on developing “plug & play” capabilities and intelligent and safe devices. The key needs and
issues associated with the reconfigurable tools and systems theme are shown in Table 8, and the two
recommendations are summarized in Table 9 and Table 10.




                                               DRAFT
                     A Strategy for Establishing a Smart Assembly Initiative                                               Page 29
                                                                       DRAFT


Table 5. The top priori ty theme from an end user perspecti ve for achieving smart assembl y goals was simulation and visualization.
What is Needed?                                     Why?                                                                               Participant
Visualize the assembly line. Build the                Save time and money by working out the details before installing                Bob Owens
assembly line and optimize it in virtual               machines into the assembly line
reality before building it.                           More efficient use of space and other resource s
Virtual modeling to run from real plant data          Eliminate field checks and allow remote monitoring of plants                    Roland Menassa
to assi st in the design and operations of            More realistic modeling
plants.
Modeling and simulation of manufacturing              Reduce new product introduction cycle time, optimize                            Tom Babin
operations and linkage to shop floor                   manufacturing throughput
systems.
Need to integrate virtual models with the               Virtual models become rather useless if they do not reflect reality           Jim Caie
real world and keep them up to date.                    Be able to optimize and validate changes in design quickly
Integration of virtual models, physical                 Performance                                                                   Jack Hu
system and data in design and operations.               Quality
                                                        Producti vity
                                                        Lead time
Try out new ideas on a virtual reality                  Weeds out the good ideas from bad ideas without a large                       Bob Owens
assembly line. For example a worm robot                  investment
handling tools or parts to operator.
Non-nominal simulation models for                     Better accuracy between virtual and real world                                  Rikard
assembly system s.                                                                                                                     Soderberg
Assembly system feedback to virtual                   Reliability                                                                     Johan Carlson
models.
Optimized master location scheme s,                     Optimize geometry calculations                                                Rikard
consi stently used in all steps.                        Robustne ss in assembly system                                                Soderberg
Fast virtual tool s for optimization of                 To find motions that withstands va riation                                    Johan Carlson
geometry and motions.                                   To generate back-up plans and re-allocation of tasks
                                                        To assure diagnostic ability
Understanding of cost of tolerances by                  Fundamental understanding and vision                                          Keith Ridgway
using the appropriate tools.




                                                                      DRAFT
                    A Report on “Smart Assembly”                                                                   Page 30
                                                                   DRAFT




Table 6. Recommendation #1 – Simulation Models and Physical System Differences .
Problem/Issue            Manufacturing process designs are not optimized or validated before deployment or for any changes
                         (include the entire assembly system).
Root Cause                Models are not kept up to date
                          Cannot get plant floor data for models
                          Models take too long to create, too costly, lack accuracy
Recommendation           Understand and prioritize differences and contributors to differences between simulat ion models and physical system.
                         Focus on accuracy of models.
Benefit                   Understand accuracy of models
                          Develop calibration guidelines
                          Ties to build dat a
Action Plan                                                                                      Owner/Time Frame
Define scope, select an assembly system(s)

Select the type of simulation models to study

Research and develop links (communication protocol, data interfaces, etc.) between virtual
model and physical system

Research the differences between the simulation and physical system

Document results and recommend next steps




                                                                  DRAFT
                    A Report on “Smart Assembly”                                                                     Page 31
                                                                    DRAFT




Table 7. Recommendation #2 – Generic Model Archi tecture.
Problem/Issue            Manufacturing process designs are not optimized or validated before deployment or for any changes
                         (include the entire assembly system).
Root Cause                Models are not kept up to date
                          Cannot get plant floor data for models
                          Models take too long to create, too costly, lack accuracy
Recommendation           Develop generic model architecture to allow a significant time reduction in creating and using models for as sembly
                         systems. Focus on time to develop models.

Benefit                   Less time to develop models
                          Increase utilization of models
Action Plan                                                                                        Owner/Time Frame
Define scope, select an assembly system(s)

Select the type of simulation models to study

Benchmark model cycle time and find opportunities

Investigat e model architecture options

Develop generic model architecture (library, data, equipment, standard models, modularity, etc.)

Test model architecture – verification and validation

Document results and recommend next steps




                                                                  DRAFT
                     A Report on “Smart Assembly”                                                                         Page 32
                                                                       DRAFT




Table 8. The second priori ty theme from an end user perspecti ve for achieving smart assembl y goals was reconfigurable tools and systems.
What is Needed?                                     Why?                                                                             Participant
“Plug & Play” assembly technologies.                 Ability to react to new product features and changes                           Tom Babin

Smart, reconfigurable tools, machines and            To ensure responsiveness                                                       Jack Hu
systems.

Standardize communication protocols for              Plug and play functionality reduces engineering cost, accelerates launch       Roland Menassa
robots, PLC, and any other plant floor device.        and assures first time quality
Make intelligent equipment (like robots) safe.       Need robots and people to interact without expensive, restrictive safety       Jim Caie
                                                      fences
Develop advanced robotics solutions that can         Eliminate non-value added time and increase what humans are good at            Roland Menassa
work with humans.                                     – dexterity
                                                     Augment people not replace them
Common product and process data                      To enable data-driven intelligent/aut omated creation of models to             Swee Leong
requirements and data definition.                     support smart assembly operations and control
                                                     Efficient data exchange bet ween manufacturing systems




                                                                     DRAFT
                    A Report on “Smart Assembly”                                                      Page 33
                                                                    DRAFT




Table 9. Recommendation #1 – Plug and Play Capabilities.
Problem/Issue             Manufacturing systems take too long to reconfigure.
Root Cause                 No plug and play capability
                           Lacks standard prot ocols

Recommendation            Develop plug and play capabilities for plant floor systems.

Benefit                    Reduced time and cost
                           Reduced down time for smaller batches
                           Reduced lot sizes

Action Plan                                                                             Owner / Time Frame
Understand building blocks of assembly systems

Define scope, select devices for standardization

Identify device components requiring standards

Study other industry methods for plug and play (e. g. personal computers )

Develop standards for manufacturing devices to pass on to equipment manufacturers

Document results and recommend next steps




                                                                   DRAFT
                     A Report on “Smart Assembly”                                                               Page 34
                                                                     DRAFT




Table 10. Recommendati on #2 – Intelligent, Safe Devices
Problem/Issue             Inability to integrate people and intelligent devices cost effectively.
Root Cause                 No optimization of models between machines and people
                           Technology not available for sensing and control
                           Safety fences are costly
                           Intelligent devices and people cannot share the same space
Recommendation            Develop technologies to enable intelligent, safe devices.

Benefit                      Labor efficiencies (time and cost)
                             Robot/human synergy
                             Safer processes and factory
                             Lower capital cost
                             Optimized floor space
                             Significant paradigm shift in assembly processing
Action Plan                                                                                        Owner/Time Frame
Define/target essential assembly processes to be studied and develop business case

Investigat e current autonomous devices (e.g. robot) capabilities, sensing technology related to
safety

Define technologies/devices capitalizing on technologies available

Develop any new technologies or required devices

Build test bed and demonstrate safety of new technology or device

Review regulations for potential changes to support implementation

Document results and recommend next steps




                                                                     DRAFT
A Strategy for Establishing a Smart Assembly Initiative                                            Page 35

                                                 DRAFT
Theme #3 – Assembly System Reuse
The third priority theme from an end user perspective for achieving smart assembly goals was assembly
system reuse. Although assembly system reuse is very closely associated with reconfigurable tools and
systems, it is focused on reusing or reapplying processes or equipment after the assembly process is no
longer needed for its current use.
Manufacturers have made significant investments in assembly equipment and processes and when
possible, would expect to reuse them rather than invest in new. Reusing assembly equipment and
processes increases the equipment life, lowers the equipment life cycle costs, reduces the ramp up time
for new product, and overall lowers the capital costs. Collectively, the group recognized there was no
clear process (methodology and tools) for designing assembly systems for reuse. Thus, they defined a
recommendation to address that need. The key needs and issues associated with the Assembly Systems
Reuse theme are shown in Table 11, and the recommendation is summarized in Table 12.

Theme #4 – Business Modeling (PLM/Performance)
The fourth priority theme from an end user perspective for achieving smart assembly goals was business
modeling. Smart assembly systems must be planned and integrated into the company business model and
provide a clear benefit and payback for the business. For manufacturer‟s to effectively and efficiently
plan, implement and use smart assembly systems may require a paradigm shift in the methodology they
employ to design their product and assembly processes.
Although the group did not identify and define any recommendations for this priority theme, they did
recognize the need for manufacturers to assess their business model and understand the financial impact
when investing in smart assembly systems. The key needs and issues associated with the Bueiness
Modeling theme are shown in Table 13.

Theme #5 – Real-Time Actionable Information
The fifth priority theme from an end user perspective for achieving smart assembly goals was real-time
actionable information. Smart assembly systems require manufacturers to have accurate and timely data
for executing the systems and making timely business decisions. Today manufacturing systems provide
extensive data, but lacks the needed interpretation or assimilation to transform it into the information or
knowledge required for making business decisions.
Although this theme was a lower priority, the group felt that the need to develop an information and
control architecture for real-time decision making was critical to achieving the smart assembly goals.
Such an architecture would provide more informed decision making, improved quality of product and
processes, reduced process downtime, and overall, improved throughput. The recommendation below
addresses this need. The key needs and issues associated with the Real-Time Actionable Information
theme are shown in Table 14, and the recommendation is summarized in Table 15.

Theme #6 – Knowledge Capture and Learning
The final priority theme from an end user perspective for achieving smart assembly goals was knowledge
capture and learning. The knowledge capture and learning theme is from the worker perspective and the
intelligence they require to effectively interact with smart systems. This knowledge is extremely valuable
in designing, implementing, executing and maintaining assembly systems. It is imperative that the
manufacturing teams have the knowledge needed to do their job as well as the business capturing the
knowledge they generate while doing their job. The key needs and issues associated with the Knowledge
Capture and Learning theme are shown in Table 14.




                                               DRAFT
A Report on “Smart Assembly”                                                                   Page 36

                                               DRAFT
Parking Lot
Because the workshop was focused on assembly processes and not product, the group identified two
product related needs for smart assembly on the parking lot. These issues are documented in Table 17.




                                             DRAFT
                     A Strategy for Establishing a Smart Assembly Initiative                                                Page 37
                                                                          DRAFT
Table 11. The third priori ty theme fr om an end user perspecti ve for achieving smart assembl y goals was assembly system reuse .
              What is Needed?                                                            Why?                                        Participant
Design assembly system together with product             Flexible to product variants                                         Rikard Soderberg
family (platforms).

Improved recognition of assembly at designed             Assembly is firefighting                                             Keith Ridgway
stage.                                                   Fundamental understanding and vision


Table 12. Recommendati on #1 – Designing Assembly Systems for Reuse.
Problem/Issue              Legacy assembly systems are difficult to upgrade for future products and technologies.
Root Cause                  Systems are not modular
                            Too long and costly to reconfigure
Recommendation             Develop a process (methodologies and tools) for designing assembly systems for reuse (support product family and
                           generations of products as well as technology upgrades).

Benefit                       Longer equipment life
                              Lower life cycle costs
                              Less down time
                              Faster ramp up time
                              Lower capital costs
Action Plan                                                                                             Owner/Time Frame
Define/target assembly proc esses to be studied and develop business case

Investigat e/identify what technology or device you want to reuse

Establish methodology to assess reuse for fut ure products

Determine opportunities to modularize technology/devices

Develop design process for reuse

Validate design process through simulation

Document results and recommend next steps



                                                                        DRAFT
                     A Report on “Smart Assembly”                                                                          Page 38
                                                                       DRAFT




Table 13. The fourth priority theme from an end user pers pecti ve for achieving s mart assembly g oals was business modeling.
What is Needed?                                     Why?                                                                      Participant
Integration through business.                        Needs to be repositioned                                                Keith Ridgway
                                                     PLM
Fundamental appraisal of business model.               We may be designing the wrong thing                                   Keith Ridgway
                                                       Business drivers – vision
Need clear benefit and payback for a smart             Optimize investment                                                   Jim Caie
assembly initiative.                                   Must have a payback
Simplify the manufacturing system – e.g. by          Reduce fixturing complexion                                             Roland Menassa
increasing the use of vision/sensors.                Allow line tracking
                                                     Improve labor efficiency
Micro factory retail center                          Redefine the business model                                             Keith Ridgway
                                                     Cost


Table 14. The fifth priority theme from an end user pers pecti ve for achievi ng smart assembly goals was real-time actionable information.
What is Needed?                                     Why?                                                                      Participant
Need to provide real -time information to            Needed to optimize the operation of an assembly system                  Jim Caie
knowledge workers and intelligent equipment.




                                                                      DRAFT
                     A Report on “Smart Assembly”                                                                    Page 39
                                                                      DRAFT



Table 15. Recommendati on #1 – Real-time Information and Control Architecture.
Problem/Issue             Real-time operation data not used for decision making for asse mbly operations.
Root Cause                 Too much data and not enough information (data is fragmented)
                           Lack of mobility (data is not where it is needed)
                           Data is at station level rather than system level
                           Data is not in the needed context
Recommendation            Develop information and control architecture for real-time decision making for smart assembly.

Benefit                      People more informed
                             Better informed for decision making
                             Better quality
                             Improved throughput
                             Less downtime
Action Plan                                                                                          Owner/Time Frame
Identify the decisions that are needed for assembly operations

Establish baseline of data available today and what new data are needed (gap analysis)

Establish a method/ proc ess for real -time decision making for assembly operations based on
real-time dat a (how to turn data int o relevant information in the needed context)

Develop information control architecture and required tools to support real -time dat a collection
and decision making

Validate architecture and tools

Document results and recommend next steps




                                                                     DRAFT
                     A Report on “Smart Assembly”                                                                         Page 40
                                                                       DRAFT




Table 16. The final pri ority theme from an end user perspecti ve for achieving smart assembl y goals was knowledge capture and learning.
What is Needed?                                     Why?                                                                    Participant
Need to find ways to capture knowledge in the        Teams need to use collective knowledge                                Jim Caie
company.                                             Lost knowledge represents waste
Let assembly line workers have a say in the          Including them makes them a stakeholder                               Bob Owens
layout of the assembly line.                         Fresh ideas may surface from their perspective
Effective training                                   Performance oriented                                                  Jim Caie
                                                     Virtual training

Table 17. These product-rel ated needs were placed in the "Parking Lot" section for future considerati on.
What is Needed?                                     Why?                                                                    Participant
Identify next generation product trends.               Product characteristics will drive process R&D                      Tom Babin
                                                       Anticipate significant changes in process technologies
Design the product for Smart Assembly.                 Address process needs upfront                                       Bob Owens
                                                       Maximize efficiencies in assembly processes




                                                                      DRAFT
A Strategy for Establishing a Smart Assembly Initiative                                                Page 41

                                                  DRAFT

Appendix B2: Researcher Perspective
The following attendees participated in the group and process:
        Gene Coffman – Ford Motor Company
        Mark Dausch - GE
        Sergio Durante - DIAD
        Roger Eastman – NIST/Loyola
        Jack Harris – Rockwell Collins
        Sanjay Jain – GWU/NIST
        Matt Mason – Carnegie Mellon Robotics Institute
        Wyatt Newman – Case Western
        Detlef Pauly – Siemens AG
        Ray Puffer - RPI
        Nick Dagalakis – NIST, Scribe
        Sam McSpadden – IMTI, Facilitator

Characteristics and Attributes
Based on the knowledge, experience, and interests of the participants, and with a researcher perspective,
the following key characteristics and attributes of smart assembly were identified:
       Soft Technology, Better Assembly and Testing
       3D-vision sensing to integrate human motions with assembly process virtualizations
       Flexible fixturing technology; ability to handle components of different shape
       Informatics, virtual reality, new technologies for reducing assembly cost
       Ability to identify and better understand barriers to Smart Assembly
       Integration and inclusion of analytical tools into manufacturing tools with simplified user
        interface at the plant (shop floor) level
       Goal-oriented, self-tuning assembly processes and systems
       Standards to drive factory systems and process, plus modeling and simulation; (i.e., optimized
        simulation output drives process).
                                                                                      Application of design for
       Integrate tools and develop tools to fill gaps in support of modeling         manufacturing modeling
        and simulation to address the intersection of model-based                     and simulation
        engineering and model-based manufacturing.
       Flexible and adaptive assembly, plug-and-play
        technology, smart solenoids, part recognition,
        active compensation and damping
       Real and virtual world alignment and a                  Model-based              Model-based
        mechanism for maintaining alignment.                    Engineering              Manufacturing
       Interaction between the real and virtual
        world.
       Ability to sense and accommodate natural
        variations and properties of materials and
        components in order to accomplish a successful assembly.
       Goal-oriented, self-tuning, creature-like intelligent systems.
       New technologies for reducing assembly costs.
       Easy linking of dozens of tools in a seamless way (CAD / CAM / PLM / MES / ERP / SIM /
        Automation Engineering Tools / …).



                                                DRAFT
A Report on “Smart Assembly”                                                                      Page 42

                                                DRAFT
       Ability to deal with large amounts of data in a SMART assembly system, to convert that data into
        useful information, and to communicate it

Key Themes from the Researcher Perspective
The group built on the characteristics and attributes to answer the question, “What needs to be done and
why?” The group identified and prioritized the following six themes for smart assembly:
       Automated and Flexible Assembly
       Easier Automation Programming
       Sensors/Sensing
       Alignment of Real Assembly Systems to Their Virtual Counterparts
       Analytical Assembly Tools
       Green Assembly
The group then discussed the needs and issues associated with each theme and generated five
recommendations to address them. Because the concept of “smart assembly” is still relatively new, many
research opportunities exist, but a comprehensive roadmap that identifies and prioritizes the research
topics does not yet exist. Therefore, several of the group‟s initial recommendations focused on a need to
identify current best practices, evaluate the available technologies, and prioritize research. Although the
recommendations clearly support the needs, further discussion, input and clarity may be needed to fully
define them due to the limited time available at the workshop.

Theme #1 – Automated and Flexible Assembly
The top priority theme from a researcher perspective for achieving smart assembly goals was automated
and flexible assembly. This is a very broad theme with numerous interrelated issues and needs. For
example, existing assembly systems do not support easy and fast reconfiguration for new products or new
versions of existing products, nor do they take advantage of plug-and-play technology to simplify the
integration of fixtures and machines. In general, existing assembly systems do not currently take
advantage of proven and emerging manufacturing technologies that enhance performance and simplify
use and maintenance. Inexpensive and flexible machines are needed for production and assembly. The
key needs and issues associated with the automated and flexible assembly theme are shown in Table 18,
and the recommendation to address these needs is summarized in Table 19.

Theme #2 – Easier Automation Programming
The second priority theme from a researcher perspective for achieving smart assembly goals was easier
automation programming. Currently, automation programming is slow, expensive, and somewhat
cumbersome. Capabilities for error detection are often lacking, and recovery from errors is not well
supported if at all. Furthermore, small changes in the automated assembly process can cause the software
to work incorrectly or stop working completely. The group made one recommendation that addresses
automation programming issues. The key needs and issues associated with the easier automation
programming theme are shown in Table 20, and the recommendation to address these needs is
summarized in Table 21.

Theme #3 – Sensors / Sensing
The third priority theme from a researcher perspective for achieving smart assembly goals was the broad
topic of sensors and sensing. Better sensor technologies are needed for part recognition, location, and
orientation. Technology is needed to sense attributes of the product during assembly that correlate with
product performance in operation. Advances are needed in fail-safe sensor technology that guarantees
worker safety. Sensor technologies are needed to accurately recognize and capture human activity in the
factory during interactions with automated assembly systems. These and other key needs and issues

                                               DRAFT
A Report on “Smart Assembly”                                                                      Page 43

                                                DRAFT
associated with the sensors / sensing theme are shown in Table 22, and the recommendation to address
these needs is summarized in Table 23.

Theme #4 – Alignment of Real Assembly Systems to Their Virtual Counterparts
The fourth priority theme from a researcher perspective for achieving smart assembly goals was the
alignment of real-world assembly systems to their virtual counterparts. Currently, accurate virtualization
of factory assembly operations is difficult to achieve for a number of reasons. For example, installed
equipment locations tend to vary somewhat from design specifications. Equipment modifications are not
communicated to the virtual system designer, and variations in real-world assembly process are
sometimes difficult to predict in a virtual environment. The group‟s idea-generation exercise identified
problems/solution ideas that tended to fall into two broad categories – those that pertain to assembly
processes at the localized, or cell, level and those that are more global in scope. No distinction between
the two categories is made in Table 24, which shows the key needs for the theme. The recommendation
for this theme is shown in Table 25.

Theme #5 – Analytical Assembly Tools
The fifth priority theme from a researcher perspective for achieving smart assembly goals was analytical
assembly tools. This is a very broad theme that generated a great deal of discussion and a multitude of
ideas that could become exciting research areas for smart assembly. The key needs and issues associated
with the analytical assembly tools theme are shown in Table 26, and the recommendation to address these
needs is summarized in Table 27.

Theme #6 – Green Assembly
The sixth and final priority theme from a researcher perspective for achieving smart assembly goals was
Green Assembly. Although this theme is no less important than other themes discussed, the group did not
prepare a recommended action, primarily because of insufficient time. Rather than relegate the two
issues/ideas to the “parking lot” category, they were assigned to the Green Assembly theme. The key
needs and issues associated with the green assembly theme are shown in Table 28.

Parking Lot
The research group did not place any ideas in the “parking lot”.




                                              DRAFT
                     A Strategy for Establishing a Smart Assembly Initiative                                               Page 44
                                                                       DRAFT




Table 18. The top priority theme from a researcher pers pecti ve for achieving s mart assembly g oals was automated and flexible assembly.
What is Needed?                                     Why?                                                                     Participant
Inexpensive and flexible machines for                Existing assembly equipment does not support easy and fast             Detlef Pauley
production and assembly                               reconfiguration for new products or new versions of existing
                                                      products

Tools and guidelines for the identification and      Not all assembly processes are ideal or even appropriate               Matt Mason
understanding of barriers to adoption of              candidates for automation.
assembly automation
Flexible and adaptive Fixturing: Development         There are currently no widely used interfac e standards that           Sergio Durante
of plug-and-play technology for assembly lines        support the integration of machines and fixtures in assembly
to simply the integration of fixtures and             lines.
machines
Active compens ation and damping                     Dynamic performance characteristics of equipment need to be            Sergio Durante
technologies for assembly systems                     improved

Application of smart mat erials for actuation and    Advanced materials offer the potential for Increasing the              Sergio Durante
sensing components                                    flexibility, adapt ability, and performanc e of assembly
                                                      components

Intelligent assembly systems that                    Existing assembly systems do not currently take advantage of           Wyatt Newman
      automatically generate detailed code           proven and emerging manufacturing technologies that enhance
        from an abstract plan                         performance and simplify use and maintenance
      respond automatically to disruptions
      support “novelty” detection for
        diagnostics of predictive maintenance
      are flexible/rapidly reprogrammable




                                                                     DRAFT
                    A Report on “Smart Assembly”                                                                    Page 45
                                                                    DRAFT




Table 19. Recommendati on #1 – Develop and Deploy Cost-Effective Assembly Solutions.
Problem/Issue             Current assembly solutions are not cost effective, and they are difficult to deploy and maintain.
Root Cause                Current automation systems do not have the required capability and adaptability.

Recommendation            Identify & develop new capabilities (or enhance existing capabilities) needed to ensure deployment of assembly
                          solutions where appropriate.

Benefit                      Lower cost
                             higher quality
                             improved productivity
                             increased flexibility
                             mass customization
Action Plan                                                                                       Owner/Time Frame
Learn from other smart assembly programs and document best practices e.g. LFMA, programs
in Germany and Japan

Conduct workshops to address research areas and generate more specific ideas including
application of smart materials for actuation and sensing, plug & play for assembly l ines,
intelligent systems, flexible and adaptive Fixturing

Understand barriers to adoption of assembly automation
Investigat e opportunities for forming a research consortium




                                                                  DRAFT
                     A Report on “Smart Assembly”                                                                        Page 46
                                                                      DRAFT


Table 20. The second priority theme from a researcher pers pecti ve for achieving s mart assembly g oals was easier automation programming.
What is Needed?                                    Why?                                                                    Participant
3D displays to support robot programming on         Current displays do not accurately and realistically depict the       Matt Mason
the plant floor                                      location and functionally of assembly equipment on the shop
                                                     floor

Simplify the reconfiguration of assembly            The skill level required to reconfigure automation systems is too     Matt Mason
systems so that the reconfiguration can be           high.
done by factory floor workers who do not have
specialized programming skills. The underlying
complexity of programming should be
transparent to the worker who reconfigures the
assembly system.

Robotic assembly with the following                 Current robotic assembly systems do not embody these                  Wyatt Newman
characteristics:                                     characteristics and are not adequate for complex assembly
                                                     systems
o   Fast, gentle (safe for personnel and
    product), and reliable
o Easy to program
o Able to replace manual assembly
o Fixtureless
o Able to recover from errors
o Capable of autonomous learning and self
    tuning
Standardization in robot /sensor                    There is currently a lack of standardization in robot / sensor        Roger Eastman
communication                                        communication, requiring extra cost in vendor programming

A descriptive programming language is               It is not feasible to support programming of aut omated “smart        Detlef Pauley
needed to address machine, kinematics,               assembly” systems with today’s manual programming tools.
function, and safety issues in automat ed            Current tools for automatic generation of code also lack the
assembly systems.                                    needed features.




                                                                    DRAFT
                    A Report on “Smart Assembly”                                                             Page 47
                                                                   DRAFT




Table 21. Recommendati on #1 – A New Paradigm for Automation Programming.
Problem/Issue            Automation programming is slow, expensive, and fragile (error detection, recovery, software is
                         sensitive to small changes)
Root Cause                Proprietary architectures and languages
                          Lack of modularity granularity
                          Safety requirement for LOTO
Recommendation           Develop a new paradigm for aut omation programming to overcome these problems.

Benefit                   Reduced time and cost
                          Improved reliability

Action Plan                                                                                     Owner/Time Frame
Conduct a workshop to identify best practices in programming languages, tools, methodologies
joint workshop with academia, industries, and funding sources

Conduct workshops to address the research areas and generate more specific ideas including
automat ed code generation, user interfaces for plant floor programming, and machine learning




                                                                 DRAFT
                     A Report on “Smart Assembly”                                                                          Page 48
                                                                       DRAFT



Table 22. The third priori ty theme from a researcher perspecti ve for achievi ng smart assembly goals was the broad topi c of sensors and sensing.
What is Needed?                                     Why?                                                                     Participant
There is a lack of effective 3D vision / sensing      Current products are eit her ineffective or complet ely               Roger Eastman
products for assembly tasks in less structured         unavailable.
environments.                                         Current products require a fairly
A fast, low cost method to map the thickness of       Currently available methods are eit her slow, expensive,              Ray Puffer
flexible materials                                     unreliable, or inaccurate

Better technologies are needed for part               Flexible assembly systems must be able to automatically sense         Sergio Durante
recognition                                            the geometries and locations of components without error
Changes are needed in robot safety standards,         Standards that fence in workcells preclude the use of advanced        Roger Eastman
coupled with advances in fail-safe technology          cooperative assembly methods
that guarant ee worker safety.                        They consume valuable space
                                                      They limit flexibility of assembly systems
                                                      They require additional employee time during maintenance
                                                       activities
Technology incorporating vision sensors and           It is extremely difficult to realistically depict in an assembly      Matt Mason
pattern recognition is needed to accurately            simulation the random and sometimes unpredictable behavior
capture human activity in the factory during           of humans as they interact with assembly equipment.
interactions with automated assembly systems.

Computer vision and machine learning                  Existing simulation software uses simplified and sometimes            Matt Mason
technology needs to be used to synchronize             unrealistic models of human activity.
real-world and virtual-reality assembly

Technology is needed to sense attribut es of the      Adaptive process control based on sensed product                      Ray Puffer
product during assembly that correlate with            characteristics would improve process yield and reduce
performance in operation.                              production costs.




                                                                      DRAFT
                     A Report on “Smart Assembly”                                                                          Page 49
                                                                       DRAFT




Table 23. Recommendati on #1 – Improve Sensor / Sensi ng Technologies to Capture Data and Convert it to Useful Informa tio n on Plant Floor Activities.
Problem/Issue             Lack of data and information on plant floor activities
Root Cause                 Current solutions are inadequat e e.g. slow, costly, fragile, too long to implement

Recommendation            Develop technologies for human/machine monitoring for safety and activity analysis; Overcome reliability and
                          acceptance barriers; Integrate data with virtual models.

Benefit                    Safer systems
                           higher fidelity models
                           better performance, quality, and yield
Action Plan                                                                                            Owner/Time Frame
Learn from other advance sensor technology programs and document best practices e.g. Home
Land Security technologies, ballistic missile defense initiative

Conduct workshops to address research areas and generate more specific ideas including
sensing for process control, safety, system characterization and integration.




                                                                     DRAFT
                      A Report on “Smart Assembly”                                                                         Page 50
                                                                       DRAFT




Table 24. The fourth priority theme from a researcher pers pecti ve for achieving s mart assembl y goals was the alignment of real-world assembly systems
to their virtual counterparts.
What is Needed?                                          Why?                                                                     Participant
Human models need to be integrated into the virtual       Models that do not mimic human behavior and do not consider            Jack Harris
reality environment to help optimize factory and           human limitations can not be accurately portray the physical
sustainability processes                                   response

Real and virtual world not aligned – extend the           Digital/ virtual world bas ed on “as designed” data                    Gene Coffman
calibration concept used in offline programming           Real equipment varies from nominal or design intent
(OLP ) to all aspects of manufacturing

Combined / hybrid real -world with virtual-world          Similar to the ideas above, the models must bring together the         Sanjay Jain
model                                                      reality of the operations world with the models, simulations,
                                                           and visualizations.
Correct representation of real world in virtual world.    The process is effort intensive                                        Sanjay Jain

Real-time updat e of virtual world                        Current technology is limited.                                         Sanjay Jain
                                                          Resource/Effort intensive
Integration of mathematical models with virtual           Virtual reality is often focused on visualization and loses the link   Sanjay Jain
world                                                      to the science and math foundation. The data, not just the
                                                           “view” is important for incorporation in operations information
                                                           and for trans fer to ot her models.
The ability to create digital data directly from          Often programming on the shop floor is based on training for           Gene Coffman
operating information.                                     operations. The digital data may not be captured for product
                                                           history or for future use.
Develop tools that provide value to plants that are       Plant focus is on production and continuous improvement.               Gene Coffman
usable by plants to support their objectives, but          There is no time to make updates that don’t provide “value
dependent on up-to-date digital data. (P rovide an         added”
incentive for plants to keep digital information
current.)

Improved simulation tools                                 Creating simulation models for all the different                       Detlef Pauly
                                                           questions/conditions that need to be simulated is very time
                                                           consuming today


                                                                     DRAFT
                     A Report on “Smart Assembly”                                                                        Page 51
                                                                      DRAFT




Table 25. Recommendati on #1 – Improve simulati on of the factory floor, and develop methods to keep virtual and real worl ds aligned.
Problem/Issue             Lack of integration between modeling/simulation and factory application (virtual and real worlds).

Root Cause                Failure to adopt interfac e protocols between modeling/simulation systems and faculty automation systems.

Recommendation            Identify gaps in tools and techniques for integration, improve simulation of the factory floor, and develop methods to
                          keep virtual and real worlds aligned.
Benefit                    More rapid configuration of processes
                           Better monitoring of processes
                           Easier problem resolution
Action Plan                                                                                          Owner/Time Frame

E valuate existing standards for their applicability and based on that analysis, identify gaps where
new standards should be developed.

Hold workshops to identify technologies for plant floor monitoring and virtual/real world
coordination, including bring humans into the simulation.

Hold workshops on the improvement of modeling and simulation tools, including improving
simulation modeling to use Monte Carlo and ot her statistical rather than simply discrete
simulation, and identifying gaps in simulation tools to better cover all plant floor assembly
processes, and develop technologies for developing models by reverse engineering from
existing floor dat a sources.




                                                                    DRAFT
                     A Strategy for Establishing a Smart Assembly Initiative                                                  Page 52
                                                                        DRAFT



Table 26. The fifth priority theme from a researcher pers pecti ve for achieving s mart assembly g oals was analytical assembly tools.
What is Needed?                                               Why?                                                                  Participant
Next-generation analytics that incorporate                     Such tools are not widely used in current assembly                  Mark Dausch
    sensor fusion                                              systems.
    process informatics (the application of                   Analytical tools are the “building blocks” needed to enable
       computer and statistical techniques to the               the development and deployment of smart assembly
       management of information)                               systems
    identification of signatures in the process data
       for proactive maintenance of shop floor
       equipment and reduction of unplanned outages

Smart tools for capturing process data (e.g. time,             Easier and more aut omated capture and storage of                   Mark Dausch
operator, cost, part ID).                                       assembly process history
                                                               Improved capability to linking process variation to specific
                                                                events, people and components.
Better tools for the aggregation of real-world data for        Current methodology and technology limitations have                 Sanjay Jain
analysis                                                        hindered the development of such tools
Analytical tools for alternative generation and modeling       Intelligent analysis capability are not currently available         Sanjay Jain
in the virt ual world

Decision making in the real world based on virtual world       Both technology and human factors are limitations                   Sanjay Jain
/ model recommendations

A common theory of assembly methods – a science or             Assembly planning is viewed by some as an art, but art              Detlef Pauly
set of rules that defi ne best assembly practices               doesn’t’ enable standardization and optimization.
Automatically generate assembly ordering paths for             No standard mechanism currently exists for including such           Mark Dausch
shop floor processes from CAD models                            information or extracting it from CA D models.
Analytical tools are needed for assessing the feasibility      Such tools either don’t exist at a functional level today, or       Mark Dausch
and cost of manual assembly versus automated                    are in the early stages of development.
assembly, including life-cycle costs




                                                                       DRAFT
                     A Report on “Smart Assembly”                                                                          Page 53
                                                                       DRAFT




Table 27. Recommendati on #1 – Analytical Tools that Support Decision Making at the Factory Floor Level .
Problem/Issue              Decision support tools are not being used effectively at the shop floor level.
Root Cause                  The level of expertise required to use existing analytical tools is too high for effective use by shop floor personnel.

Recommendation             Enable shop floor personnel wit h tools that are better focused on their needs and level of expertise

Benefit                    Group did not have time to list benefits.

Action Plan                                                                                            Owner/Time Frame
No action plan was developed for this recommendation due to lack of time.




Table 28. The sixth and final priority theme from a researcher pers pecti ve for achieving s mart assembly g oals was Green Assembly.
What is Needed?                             Why?                                                                             Participant
Application of clean processes               The laws dictate it, and it is simply good business. The total cost of         Sergio Durante
(actuation oils, energy saving                operation and product must be revised to include environmental
strategies, etc.)                             sustainability. Often responsibility is also more economical.

Address the sustainability aspects of        The society and the laws are changing to make life-cycle responsibility a      Jack Harris
life cycle support of products                necessity.




                                                                       DRAFT
A Strategy for Establishing a Smart Assembly Initiative                                           Page 54

                                                DRAFT

Appendix B3: Infrastructure and Standards Perspective
The following attendees participated in the group.
        Mahesh Mani - NIST
        Richard Schmidt - Lockheed Martin
        Mitch Vaughn - UGS
        Charile Gifford - G.E. Fanuc
        Charles Mclean - NIST
        Eric Berg - Procter & Gamble
        Matt Hobson-Rohrer – NIST
        Sudarsan Rachuri - Scribe
        Richard Neal - Facilitator

Characteristics and Attributes
Based on the knowledge, experience, and interests of the participants, and with an infrastructure and
standards perspective, the following key characteristics and attributes of smart assembly were identified:
       Intelligence between systems
       Interoperability between related domains
       Operating within a common (standard) framework
       Agile, intelligent man-machine interactions
       Intelligent automation with collaboration between machines and humans
       Interoperability based on semantic understanding
       Real-time actionable information for automated decisions
       Self-diagnosing and self healing
       Capable of optimization across an enterprise
       Easily useful
             Flexible in operation
             Easy to program
             Easy to train
             Easy to operate
             Self organizing
             Immune from human error
       Model based – working from a library of state models
       Integrated production across organizational boundaries and supply chains
       Integral to product and process lifecycle and connected to the product lifecycle
       Integrated with other systems outside the plant floor
       Bridges the physical and virtual worlds without differences
       Adapts virtual applications and technologies for direct shop floor use
       Adapts feedback from the shop floor (shop floor data) for use in design
       Common language between business areas, supporting timely decisions
       Information flow with standard forms and presentation
             Configuration managed
             Standardized and modularized fixtures and sub-fixtures with multiple uses
       Effective communication and integration of lessons learned in a closed-loop system\
       The ability the “synthesize” a system based on requirements
             Model-based systems engineering

                                              DRAFT
A Report on “Smart Assembly”                                                                        Page 55

                                                 DRAFT
              Semantic linkage
              Object-based technologies
              Standards for managing lifecycle performance by integrating hardware and
               software (IEC 61346)
            Structuring reference designations
            Allocating functions to the best equipment
       Common language for representing needs including crossing domains without loss of
        content of context

Key Themes from an Infrastructure and Standards Perspective
The group built on the characteristics and attributes to answer the question, “What needs to be done and
why?” The group identified and prioritized the following four high-priority themes for smart assembly:
       Flexible Manufacturing Systems
       Closed Loop Information Flow
       Common Languages
       Tracking Systems for Assembly

Priority Theme #1 – Flexible Manufacturing Systems
The priority theme of flexible manufacturing systems gave rise to a broad range of ideas that all support
the objectives of quickly responding to changing requirements and reducing cost, slashing the time for
integration of new products and new versions of products into the production stream, enabling mu lti-
product assembly, and optimizing capital utilization.
The most compelling idea, and the one under which the others can well be aligned, is model-based
systems engineering. The implementation of a model-based approach would separate the management of
the physical manufacturing system from the requirements. This would enable a requirements-driven
system with the management of the assembly functions performed in a real-time decision process.
       High level product, process, and business requirements drive the definition of the functional roles
        (what will be made where and how).
     Functional roles are allocated to physical systems based on the design parameters.
     Functional roles are traceable to high level requirements allowing the roles to be reallocated over
        the lifecycle.
This model-based systems engineering approach would build on emerging standards for engineering tools
that support a model-based approach. It is important to note that this approach is a commitment to an
environment that is supported by a rich modeling and simulation capability and allows the functionality of
the requirements driven system to be fully tested and all assignments to be optimized. The requirements
drive the planning, the planning directly drives creation of all information for production, and the loop is
closed with tracking of performance to requirements. In its full extension, the requirements based
modeling environment can, become the factory controller and smart assembly can be realized.
Other important ideas support the priority theme of flexible manufacturing systems. Statistical process
control and decision support at the system level, incorporating tolerance information in the decision
process, can improve product quality and reduce rejects. Matching of in-tolerance parts to assure that
systems are in-tolerance can greatly improve systems level performance. Continuous tracking of
components and systems, using RFID or other systems, enables a continuous decision process supporting
smart assembly. Standards are needed to support the use of RFID in continuous tacking mode for
manufacturing applications.
The ultimate achievement in flexible assembly systems might be a smart assembler in the virtual world
that directly tracks and transfers to the physical world (and controls the physical world). Intelligent agents

                                               DRAFT
A Report on “Smart Assembly”                                                                  Page 56

                                              DRAFT
are seen as a promising technology for achieving this goal. The smart assembler would support
knowledge-based representation of all components of the system and interoperability between
components and domains. The agent-based system would support the evaluation and optimization of
interactions between the functions of the assembly system, cooperative management of resource
allocation, and reconfiguration of factory operations based on real-time operations requirements.
The key needs and issues associated with the flexible manufacturing systems theme are shown in Table
29, and the recommendation to address these needs is summarized in Table 30.




                                            DRAFT
                     A Strategy for Establishing a Smart Assembly Initiative                                             Page 57
                                                                      DRAFT




Table 29. The highest priority theme from the Infrastructure and Standards perspecti ve was flexible manufacturing syst ems.
     What is Needed?                                                      Why?                                                   Participant
Model-based Systems                We need to be able to connect requirements with solutions throughout the                Eric Berg
Engineering                        lifecycle of the product. Model-based engineering is the key to doing that.

Flexible Manufacturing             To quickly adjust to changing requirements                                              Richard Schmidt
systems
                                     Reduce long-term non-recurring costs
                                     Reduce time to market
                                     Optimized capital utilization
Direct execution of “as-           Efficient communication of exactly what is needed to conversion to product              Mitch Vaughn
planned” process model by a
                                      Reduces errors and iterative development
shop floor execution system
                                      Saves money
Manufacturing standards for        Present resourc e tracking systems apply a nodal mode, not a continuous tracking        Charlie Gifford
RFID                               mode. Continuous tracking supports decision making. RFID standards are needed
                                   for tracking manufacturing work flow and are essential for continuous tracking.

Statistical Process Control at     Good components don’t always make good systems. Tolerance stack -up or                  Sudarsan Rachuri
the systems level including        marginal compliance, when coupled wit h multiple parts, multiplies the variability
tolerance information flow         and leads to unacceptable product. SPC, based on full tolerance awareness, can
throughout the assembly            support selection and decision proc esses for optimized system performanc e
process

Agent-based Smart Assembler        Knowledge representation and interoperability are essential if we are to have           Mahesh Mani
supporting cooperative             flexible assembly systems. Cooperative management of assembly operations –
management                         across systems, domains, and products – will enable maximum flexibility with all of
                                   the associated advantages




                                                                    DRAFT
                    A Report on “Smart Assembly”                                                                      Page 58
                                                                       DRAFT




Table 30. Recommendati on #1 – Persistent Requirements and Capability Model.
Problem or Issue         Present systems lack a clear two way mapping from requirements to models and to solutions.
                         (solutions includes tooling, fixtures, etc.)
Root Cause               Organizational structures inhibits such a smooth flow of information. The model based engineering toolset is still under
                         development and has evolved.

Recommendation           Use model based systems engineering principles to maintain a persistent requirements and capability model that is
                         used to manage a distributed assembly environment throughout its lifecycle.
                         Attributes: Trade space, Traceability of requirements (forward/backward)

Benefit                  Flexibility, reduced rework (lean manufacturing), reduced time-to-market, improved quality, improved capital utilization,
                         increased reuse of components, knowledge and design elements, reduced number of unique systems, improved
                         capability assurance

Action Plan                                                                                         Owner/Time Frame
Select a SE approach to implementing MBSE (ICTT, S TEP (AP 233), SysML, ISA95)

Define a test case with feat ures of common interest and establish a testbed for implementation

Establish and maintain a persistent model for the selected test case




                                                                  DRAFT
A Strategy for Establishing a Smart Assembly Initiative                                               Page 59

                                                  DRAFT
Priority Theme #2 – Closed-Loop Information Flow
Closed-loop information flow has a broad context. It couples the as-designed to as-built and, in smart
assembly, it embraces the toolset that enables and assures this capability. Closed-loop operation couples
the information from the process and assembly plans to the travelers and procedures that enable the
human and the machines to execute the process. It closes the loop as the information from the shop floor
is analyzed and verified – even to the point of intelligent real-time closed loop operation. Intelligent, real-
time operation simply means that the shop floor systems are tuned, monitored and empowered to accept
performance commands, execute those commands, assure and verify the result, and take appropriate
action in case of deviant results.
A strong emphasis in closed-loop information flow is placed on the need for excellence in preparation and
presentation of materials. The materials produced to support assembly should clearly define the tasks,
provide instruction for performance, and support collection and analysis of data to assure that the product
is within spec and in tolerance at every step. This capability calls for a model-based information creation
system and the ability to present the correct views to the shop floor. From a functionality perspective, the
requirement can be quickly linked to multiple views in 3-D CAD. From a futuristic infrastructure
perspective, the creation of the right information and the provision of that information to the point-of-use
calls for a richer set of product and process models that can automatically support generation of “views”
or abstractions. The capability to support composable and decomposable models is similar to the topic of
systems-of-systems that is so pervasive in the Department of Defense. Perhaps the model-based
manufacturing equivalent could be viewed as a system-of-models.
The ideas gathered under this category support the realization of the vision of close-loop information
flow. The key enabler is a rich set of data models that fully define the capabilities of the assembly
environment and the requirements that fall within the domain. The data models make possible the
generation of information models, operational models, and process models. The models link directly to
the requirements that flow from the product definition. In other words, the data models enable the model-
based systems engineering approach, defined in the first priority theme, to flow to the shop floor for
intelligent, closed-loop assembly.
The key needs and issues associated with the closed-loop information flow theme are shown in Table 31,
and the recommendation to address these needs is summarized in Table 32.




                                                DRAFT
                     A Strategy for Establishing a Smart Assembly Initiative                                               Page 60
                                                                     DRAFT




Table 31. The second priority theme from the Infrastructure and Standards perspecti ve was closed-loop information flow.
What is Needed?                   Why?                                                                                      Participant
Closed-Loop information flow      Understanding, in real-time or near real -time, pres ent conditions and operations        Richard Schmidt
                                  supports informed decisions about future operations.

Comprehensive assembly            A rich set of data models will support the creation of operational models and             Matt Hobson-Rohrer
system data models                predictive analytics. The data m odels will also support capturing and sharing
                                                                                                                            Mahesh Mani
                                  knowledge.

Interface definitions and,        Will support the connection and interoperability of systems and the seamless              Matt Hobson-Rohrer
longer term, standards            flow/trans fer of information.

Embedded 3-D product data in      To assure that the human has the information that is needed, in the form that it is       Mitch Vaughn
shop floor manufacturing          best presented. This information, coupled with data collection and validation can
instructions with two way         lead to closed-loop assured compliance in assembly proc esses.
communications. It is noted
that various levels of
abstraction from the CA D
models is required top support
this functionality.

Decision support systems          Decision support is needed to turn data and information into actionable knowledge.        Mahesh Mani
using standard knowledge          Prognostics and intelligent assembly/automation will be the result.
representation

Standardized “engineering         Enables intelligent decision making in a closed-loop operating environment.                Sudarsan Rachuri
analysis interfaces”




                                                                    DRAFT
                    A Report on “Smart Assembly”                                                                  Page 61
                                                                     DRAFT




Table 32. Recommendati on #1 – Data Models for Smart Assembly.
Problem or Issue        We do not capture common aspects between different data models for assembly. A cost-effective
                        methodology does not exist for the verification and validation of models.
Root Cause              Vertical industry silos and lack of collaboration

Recommendation          Develop a taxonomy for assembly and consistent set of assembly scenario (role or activity) based d ata models that
                        support assembly functions

Benefit                 Deploying defined finite sets of familiar data models:
                               Supports activity based work flow,
                               Optimizes operations, supports seamless interoperability
                               Saves transaction cost.
Action Plan                                                                                     Owner/Time Frame
Survey of assembly model taxonomies and extend the taxonomy. (take a look at VCOR,
DoDAF, STEP)

Identify the functions and rules for assembly scenarios and identify common and unique
aspects.

Select a high frequency and high value use cases and build a set of data models

Apply these data models to test cases




                                                                   DRAFT
A Strategy for Establishing a Smart Assembly Initiative                                          Page 62

                                               DRAFT
Priority Theme #3 – Common Language
The case for a common language for assembly can best be made by looking at a factual scenario. 12 A
typical company has approximately 3000 IT applications in its engineering functions and across a
manufacturing supply chain. All of these applications have different data models and custom interfaces.
With 5 transactions per interface, the complexity grows to 15,000 interfaces per application. With 20
transactions per interface, the number of custom transactions that a company must manage is an
astounding 300,000. A common language supporting a common integration baseline would dramatically
reduce the complexity factor – saving money and paving a pathway to better and more efficient assembly
systems.
There are several components to the common language. To address the needs highlighted in the example
above, a common semantic understanding supporting configurable interfaces with common work
instructions is essential. The semantic understanding also must be supported by rule sets for specific
applications. Foundational to this achievement is the development of a lexicon of terms and their
meaning. Starting with a taxonomy of assembly operations and moving to ontological unity seems to be
in a right direction.
Perhaps not as dramatic as a standard common language for the vision of the future, there is a short-term
need for consistency in documentation standards. There are multiple standards available, but unless
standards are uniformly applied, the benefits are often lost. The suggestion from the infrastructure and
standards group is that invested parties come together and select the best available standards for
application now e.g. IEC 61346, as we move to common languages.
From an infrastructure perspective, all of these arguments for a common language call for standards. The
harmonization of existing standards to allow unity in documenting, defining, and executing assembly
operations is seen as the highest near term objective. The call for action is for the harmonization of
existing models, on the pathway to semantic understanding.
The key needs and issues associated with the common language theme are shown in Table 33, and the
recommendation to address these needs is summarized in Table 34.




12
   Information provided by Charlie Gifford of General Electric. There is research to
substantiate these numbers.


                                              DRAFT
                    A Strategy for Establishing a Smart Assembly Initiative                                             Page 63
                                                                     DRAFT




Table 33. The third priori ty theme from the Infrastructure and Standards pers pecti ve was common language.
     What is Needed?                                                     Why?                                                  Participant
Common language through           We must be able to relate the disparate components and documentation of a              Eric Berg
documentation standards           design throughout the lifecycle in a distributed environment.

Common language across            Enterprise-wide business decisions making is hampered by an inability to               Richard Schmidt
business areas and domains        communicate efficiently.

Harmonized Standards              The number of trans actions that support manufacturing functions is huge. There        Charlie Gifford
                                  are many “standards” and protocols that support these transactions. We need a
                                                                                                                         Sudarsan Rachuri
                                  common set of standards that simplify these processes

Semantic Understanding of         The hardware and software that make up the assembly systems lack a common              Sudarsan Rachuri
assembly related processes        language. Short of total compliance with a single standard or set of standards, the
                                  best pathway to efficiency and unity is through a semantic capability that adapts
                                  everyone’s language to achieve common understanding




                                                                   DRAFT
                    A Report on “Smart Assembly”                                                                      Page 64
                                                                    DRAFT




Table 34. Recommendati on #1 – Harmonized Standards for Assembly.
Problem or Issue          Lack of harmonized standards prevents effective information exchange limiting the ability to deliver
                          the right information to the point of use
Root Cause                Lack of a big picture and lifecycle pers pective leads individuals to select standards in an ad hoc manner.

Recommendation            Require that all standards used in a given set of assembly operations are harmonized this establishing a canonical
                          corporate schema.

Benefit                   Serves as a basis for knowledge management systems and business analytics. Enables mapping operational metric to
                          business metrics

Action Plan                                                                                        Owner/Time Frame
Create a “Smart Assembly” consortium

Find the gaps and overlaps bas ed on standards typology

Resolve the gaps and overlaps to harmonize for specific application (near term)

Define a standards landscape




                                                                  DRAFT
A Strategy for Establishing a Smart Assembly Initiative                                           Page 65

                                                DRAFT
Priority Theme #4 – Tracking Systems for Assembly (forward and backward)
The final priority theme that was addressed by the Infrastructure and Standards working group was
tracking systems for assembly operations. It is noted that the call for tracking systems is inclusive. It
encompasses the state and allocation of components, fixtures, equipment, subassemblies, information
systems… all elements of the assembly system. Further, tracking is more than knowing where the pieces
are. It addresses traceability of “as built” to “as designed” for product data, and to “as planned” for
process data. The final element is the tracking of requirements to design, to production, and throughout
the product lifecycle – forward and backward. These capabilities support the total connection of all
aspects of product development into an integrated product realization and support system.
Attention is called to the parenthetical addition to the title – forward and backward. Often tracking is
viewed as understanding where the pieces are and being sure that the pieces come together. Throughout
this discussion, you will note the continuing themes of providing traceability from requirements to design,
from design to process plans and execution, and from execution back to design and planning in a
continuous, closed-loop process.
 Standards are very important for success in advanced assembly tracking. Technologies like RFID and
other identification methods have become commonplace in assembly. However, there is a lack of
standardization in their application. Standards are needed for the collection and management of data,
across the product lifecycle.
The key needs and issues associated with the common language theme are shown in Table 35, and the
recommendation to address these needs is summarized in Table 36.

Parking Lot
The following extremely important ideas and suggestions did not fit into any of the previously identified
compelling themes. They were assigned to the “parking lot” category so that the information would be
preserved for later use.
    1. Conduct a requirements analysis for smart assembly.
    2. Develop an integration framework for smart assembly.
    3. Develop a roadmap for technology standards.
    4. Develop a roadmap for smart assembly.
    5. Create an “industrial strength use case” as an example of smart assembly and build a
       test-bed for smart assembly technologies.
    6. Create a plan to promote and market smart assembly and to grow a large initiative.




                                              DRAFT
                    A Strategy for Establishing a Smart Assembly Initiative                                            Page 66
                                                                     DRAFT




Table 35. The fourth and fi nal priority theme that was addressed by the Infrastructure and Standards working group was tracking systems for assembly
operations.
     What is Needed?                                                     Why?                                                   Participant
Feedback from shop floor to        Mistakes are repeat ed unless the opportunity for error is removed. Mistake            Eric Berg
design                             opportunities can be designed out and engineered out of the assembly process
Tracking standards for manual      Without a standard, inefficiencies abound. The standard would support the              Matt Hobson-Rohrer
assembly                           efficient collection of the right data to enable:
                                         Collection of the right information to support lean assembly
                                         The definition of aut omation opport unities
Shop floor data captured wit h     Smart assembly will only be achieved with the full understanding of the important      Mitch Vaughn
associativity back to “as-         parameters and their state coupled wit h continuous process improvement. This
designed” for quality and “as-     requires that shop-floor data be int egrated with the design process.
planned” for process data
Cons umer/Producer Resource        As assemblies are aggregated into larger systems, common resources can be              Eric Berg
Models                             combined e.g. electrical power. This applies to the product being produced and
                                   the equipment producing it. Also, understanding the state of a resource and its
                                   availability is critical for utilization decisions.




                                                                   DRAFT
                     A Report on “Smart Assembly”                                                                        Page 67
                                                                      DRAFT




Table 36. Recommendati on #1 – Associativity of Shop Floor Data and Design Data.
Problem or Issue            Problems persist on the shop floor that could and should be corrected and eliminated. The solution
                            includes real-time closed loop operation, and feedback to design for elimination of the cause.
Root Cause                  Often assembly operations are open-loop. Problems occur and recur with no identification of the problem and no
                            systemic solution

Recommendation              Provide systems that are based on tracking standards, which enable the collection of data on the shop floor, analysis
                            of that data, and feedback to design.

Benefit                     Supports intelligent assembly operations: assures availability and status of resourc es: systemati cally assures quality
                            of product

Action Plan                                                                                           Owner/Time Frame
Under the umbrella of the smart assembly work that was started with the workshop, form ad -hoc        NIS T
groups that continue the work

Socialize the recommendations across the groups to create a unified set of agreed to priorities

Be sure that the topic defined here is included in the priorities

Flesh out the work needed, build a project plan, and seek a collaborative activity to move
forward and provide the needed solution




                                                                     DRAFT
A Strategy for Establishing a Smart Assembly Initiative                                           Page 68

                                                DRAFT

Appendix B4: Integration and Deployment Perspective
The following attendees participated in the group and process.
        Bob Brown – DELMIA Corporation
        Bryan Dods – The Boeing Company
        Tom Kurfess – Clemson University
        Allen Martel – O3neida, Incorporated
        Glen Oliver – Lockheed Martin Missiles & Fire Control
        Gordon Shao – NIST
        Michael Wynblatt – Siemens Technology-To-Business (TTB) Center
        Fred Proctor – NIST (Scribe)
        Charlie Neal - IMTI, Incorporated (Facilitator)

Characteristics and Attributes
Based on the knowledge, experience, and interests of the participants, and with an integration and
deployment perspective, the group identified the following key characteristics and attributes of smart
assembly:
       Quality – improvements to the assembly process that increase quality
       Real time information – reduction of data exchange barriers that impeded downstream flow of
        design intent, upstream flow of “production actuals”
       System - product and production
       Hardware – electrical (in autos) and mechanical
       Software – interface and integration
       Traceability - birth certificates; product/process history
       Networks to collaborate – need to build networks and nodes to foster effective collaboration;
        deployment
       Leverage existing entities – draw on existing entities; capitalize on current/future projects in EU
        and Asia; international perspective
       Integration – loosely integration of people, equipment, processes, and IT into an assembly system
       Tailor – assemble is about complexity and scale; we need the ability to tailor our assembly
        systems to our needs
       Flexibility versus Complexity – emphasis on “smart” to enable sweet spot on the curve
       “Smart” – automation without adding undue noise, power consumption, complexity; reduce
        complexity on the floor as holistic approach
       Interoperability – between systems: software & hardware; control systems; virtual & real
       Systems Integration in an Outsourcing Business Model – communicate requirements for form, fit,
        function, interface, technologies, from OEM to Supplier; in reverse direction Supplier should
        evaluate components against requirements and ability to integrate BEFORE going into finished
        product/assembly
       Interchangeability – of part data/simulation data between programs and platforms
       Re-configurability and Re-use – how to get early systems engineering viewpoint to carry on into
        manufacturing ; getting manufacturing engineering and system engineering to embrace virtual
        manufacturing as a way of business for trade analysis on common platforms
       Make sure great ideas make their way into the tools people rely on day-to-day
       Reduce the ripple effects of changes throughout the enterprise



                                              DRAFT
A Report on “Smart Assembly”                                                                       Page 69

                                               DRAFT
       User front end sensors and integration; integration (loose) of people, equipment and IT into a
        system; tailor plug and play; critical skills/knowledge diversity

Key Themes from an Integration and Deployment Perspective
The group built on the characteristics and attributes to answer the question, “What needs to be done and
why?” The group identified and prioritized the following five themes for smart assembly:
       Conceptual Framework - Demonstration
       Standards – Data Exchange; Hardware/Software
       OEM/Supplier Collaboration
       Standards – Methodology/Communications/System Engineering
       Skills
The group discussed the needs and issues associated with each theme and generated three
recommendations to address the key needs for the three highest-priority themes. Although the
recommendations clearly support the needs, further discussion, input and clarity may be needed to fully
define them due to the limited time available at the workshop. The five themes, along with prioritized
recommendations that address the needs of the themes, are described below.

Key Theme #1 – Conceptual Framework - Demonstration
The top priority theme from an integration & deployment perspective for achieving smart assembly goals
was establishing a solid conceptual framework for the full “smart assembly” concept. Too many
stovepipes of “local only”/proprietary solutions at the manufacturing and vendor levels have resulted in
no cohesive approach to either the integration or deployment aspects of smart assembly.
A huge barrier to pursing integration and deployment of that integration is the willingness of companies
to invest capital – both human and $ - in smart assembly for the future when faced with near term,
bottom-line, and risk-averse environments. The current
approach is to make cautious, incremental advances building                  NEW
on legacy systems and concepts. Putting existing technology
(including hardware and software) into an existing
environment is much easier to sell and support than boldly
attempting new technology into a new environment. The four                      NE/ET    NE/NT
quadrants of “risk” shown in Figure 9 were identified to           Environment
characterize the “risk/difficulty” of integration and
                                                                                EE/ET    EE/NT
deployment:
                                                                                                       NEW
                                                                      Existing    Technology
EE/ET = Existing Environment and Existing Technology                  Legacy
EE/NT = Existing Environment and New Technology
NE/ET = New Environment and Existing Technology
NE/NT = New Environment and New Technology                       Figure 9. The risk/difficulty of integration
                                                                 and depl oyment will fall into one of the
Currently, companies – to varying degrees of                     four quadrants shown.
aggressiveness – are independently pursing projects
within this framework, but mostly in the EE/ET quadrant.
The key needs and issues associated with the conceptual framework demonstration theme are shown in
Table 37, and the recommendation to address these needs is summarized in Table 38.

Key Theme #2 – Standards – Data Exchange; Hardware/Software
The second priority theme from an integration &deployment perspective for achieving smart assembly
goals was the need for standards, to include:


                                              DRAFT
A Report on “Smart Assembly”                                                                      Page 70

                                                DRAFT
   Data exchange formats, terminology, within and across industries, and within and across vendors –
    need for a common language of “smart” assembly.
   Specific hardware and software interface protocols, terminology, and design specifications for
    integration
The lack of standards perpetuates complex interface solutions which are costly to initially build and
continually maintain as a change on either side requires modifications. The task of executing smart
assembly is difficult enough without the additional burden of system-to-system “fixes”.
The key needs and issues associated with the standards – data exchange, hardware/software theme are
shown in Table 39, and the recommendation to address these needs is summarized in Table 40.

Key Theme #3 – OEM/Supplier Collaboration
The third priority theme from an integration & deployment perspective for achieving smart assembly
goals was improvements to OEM and Supplier Collaboration.
Suppliers support multiple OEMs while OEMS utilize multiple suppliers. All incur the costs of
interacting/interfacing with different systems to meet requirements, evaluate solutions, and communicate
with one another.
The key needs and issues associated with the OEM/Supplier Collaboration theme are shown in Table 41,
and the recommendation to address these needs is summarized in Table 42.

Key Theme #4 – Standards – Methodology/Communications/System Engineering
The fourth priority theme from an end user perspective for achieving smart assembly goals was the need
for standards methods to improve communications and to leverage existing system engineering efforts.
The key needs and issues associated with the Standards – Methodology/Communications/System
Engineering theme are shown in Table 43. No specific recommendations were developed for this area.

Key Theme #5 – Skills
The fifth and final priority theme from an end user perspective for achieving smart assembly goals was
skills specifically attuned to assembly “engineering.” The task of manufacturing and specifically
assembly is not adequately addressed in the current educational curricula for engineers.
The key needs and issues associated with the Skills theme are shown in Table 44. No specific
recommendations were developed for this area.

Parking Lot
No additional items outside of the integration & deployment perspective were identified for the parking
lot.




                                              DRAFT
                     A Strategy for Establishing a Smart Assembly Initiative                                               Page 71
                                                                       DRAFT




Table 37. The top priority theme from an integrati on & depl oyment pers pecti ve for achievi ng smart assembly goals was establishing a solid conceptual
framework for the full “smart assembl y” concept.
What is Needed?                    Why?                                                                                       Participant
Make it easier to pilot and        Assembly involves coordination between many parties; getting sign-off on pilots or         Fred Proctor
deploy assembly technology –       deployment is naturally harder
develop scenarios that show
smart assembly in action

Recognize there are two types      Lack of knowledge about new technology is a barrier and must be accommodated               Allan Martel
of deployment: existing            in planning
technology and new
technology

Be able to simultaneously put      To make rapid advances wit hout being encumbered by legacies                               Tom Kurfess
together new production lines
and new products




                                                                      DRAFT
                     A Report on “Smart Assembly”                                                                        Page 72
                                                                        DRAFT




Table 38. Recommendati on #1 – Technology Deployment Demonstrations.
Problem/Issue              It’s difficult to deploy technology or process concepts because of the risk (see Four Quadrant EE-
                           NN, Figure 9).

Root Cause                 Compens ation policies and personal interests give little incentive to take risks, e.g., production managers are asked to
                           take all the risks, and get no rewards.

Recommendation             Reduce risk through convincing pilot demonstration projects, phased deployment approach. Develop performance
                           metrics for success. Demonstrate technologies within a representative environment to show they can work within
                           practical constraints.

Benefit                    Increased deployment

Action Plan                                                                                          Owner/Time Frame
Establish a group wit h broader participation, focus ed on the narrower problem of deployment of     Allan Martel (ISA ) others TB D, 1 month
smart assembly. Candidates include ISA, SME, SAE. Then,

Benchmark current state of the art in concept ual frameworks, internationally, e.g., Fraunhofer,     The aforementioned group
MITI, KAIS T.

Establish conceptual framework for supporting, developing or exploiting smart assembly               Ditto
testbed environments that service multiple industries, including success metrics.

Initiate pilot demonstration projects in US; link with international projects.                       All: involve end user, vendor, system
                                                                                                     integrator




                                                                       DRAFT
                     A Report on “Smart Assembly”                                                                          Page 73
                                                                       DRAFT




Table 39. The second priority theme from an integrati on &depl oyment pers pecti ve for achieving s mart assembl y g oals was the need for standards.
     What is Needed?                                                       Why?                                                      Participant
Reduce barriers to information      So that investments in technology can persist                                            Fred Proctor
exchange                            So that new technology can be easily deployed
Development of standard            To support design and production updates plus provide traceability                         Tom Kurfess
databases for easy information
access and data exchange;
Development of standard            To enable systems integration and update capability (e.g. CA N bus on vehicle OK;          Tom Kurfess
hardware and software              higher data rates needed, but lots of proprietary solutions
interfaces
Interface standards                   To solve interoperability problems                                                     Gordon S hao
                                      Ensure the smooth integration among systems - easy and reliable
                                      Eliminate those applications not actually needed
                                      By using interface standards, can exchange information both within internal
                                       and with external applications
                                      To reduce point-to-point data exchange solutions
Software extensibility –              To collect solutions and make them work together more easily                           Michael Wynblatt
industrial software needs to be       To break the labor bottleneck for software compatibility, standards compliance
extensible by the user or third       To jump start “community software” movement
party, not just the vendor
Develop a new generation of        Need to extend the “network” as close to the product as possible – even t the point        Bryan Dods
tools and equipment that           of incorporating into the product.
incorporates connectivity




                                                                      DRAFT
                     A Report on “Smart Assembly”                                                                       Page 74
                                                                      DRAFT




Table 40. Recommendati on #1 – Data Exchange Pathways Analysis
Problem/Issue             It is difficult to exchange data that would enable smart assembly, e.g., design torques transmitted to
                          torquing equipment, CAD-to-MES
Root Cause                Lack of standards, or lack of compliance, for a variety of reasons. Following eac h could lead to deeper root causes.

Recommendation            Identify those data exchange paths that are causing the most pain to end users, or that should exist, determine how to
                          improve those data exchanges, and follow through wit h improvement

Benefit                   None were developed by the team

Action Plan                                                                                           Owner/Time Frame
Identify industry sector champions for actions below. Output is list of champions.                    Bob Tilove, now to Dec

Identify areas of pain or opportunity that most benefit industry, and candidate painkillers. Output   Bob and champions, Mar 07
is Bob’s final report.

Organize industry teams around specific common interests who will conduct case studies that           Bryan Dods for Boeing, other champions, 2 yr
include needs analyses, solution development, pilot testing and reporting (lessons learned, gap
analyses), e.g., Boeing-led industry team for the aerospace sector. Out puts are reports. First
interim report is Boeing’s, Mar 07.

Communicate approach, results, lessons learned at first champion’s conference shortly after           Parties in previous step, ongoing
Mar 07. Publicize more broadly as other reports become available at conferences, ongoing.




                                                                    DRAFT
                    A Report on “Smart Assembly”                                                                       Page 75
                                                                     DRAFT


Table 41. The third priori ty theme from an integration & depl oyment perspecti ve for achieving smart assembl y goals was improvements to OEM and
Supplier Collaboration.
     What is Needed?                                                      Why?                                                   Participant
Provide a standards-based         Suppliers do work for multiple OEMs, all with different systems for providing           Bob Brown
solution for communicating        requirements and evaluating solutions
between OEM and supplier:
  Requirements to supplier
  Solutions to OEM



Table 42. Recommendati on #1 – Collaborative Systems Integration.
Problem/Issue             Data exchange supporting the customer’s systems integration process with suppliers is difficult, in
                          both directions, particularly with small suppliers.
Root Cause                Complexity and scale: global supply chain, different levels of sophistication. Component suppliers don’t share
                          ownership of systems integration with OEMs.
Recommendation            Develop tools and skills for collaborative systems integration, including day-to-day shop floor collaborative operations.

Benefit                   None were developed by the team

Action Plan                                                                                          Owner/Time Frame
Identify industry sector champions to develop industry-specific requirements specifications and      Bob Tilove, as before
definitions for collaborative systems integration. Output is list of champions.
Survey and determine best practices in particular industry. Identify areas of pain or opportunity    Bob and champions, Mar 07
that most benefit industry, and candidate painkillers. Output is Bob’s final report.
Organize industry teams around specific common interests who will conduct case studies that          Tom Kurfess for Clemson/BMW, other
include needs analyses, solution development, pilot testing and reporting (lessons learned, gap      champions, 2 yr
analyses), e.g., Clemson-led industry team for the automotive sector (BMW ). Outputs are
reports. First interim report is Clemson’s, Mar 07.
Communicate approach, results, lessons learned at first champion’s conference shortly after          Parties in previous step, ongoing
Mar 07. Publicize more broadly as other reports become available at conferences, ongoing.




                                                                    DRAFT
                     A Report on “Smart Assembly”                                                                          Page 76
                                                                       DRAFT




Table 43. The fourth priority theme from an end user pers pecti ve for achieving s mart assembly g oals was the need for st andards methods to improve
communications and to leverage existing system engineering efforts.
What is Needed?                                                                       Why?                                    Participant
Need a standardized methodology to determine if something is producible -             Software, organizations,                Glen Oliver
able to be assembled.                                                                 companies do not have a
                                                                                      standardized way of describing
Need a standard means of describing process data across programs, divisions,
                                                                                      producability, process, assembly
and industries - Standard data format for proc ess/simulation data exchange
                                                                                      information (e.g. Production
between packages.
                                                                                      readiness levels, TRLs, MRLs not
Work with InCOSE to create/embrace manufacturing engineering produceability           specific enough).
trades well defined and embedded in the systems engineering process.
Require Manufacturing/Assembly trades as part of concept design review.

Systems need to be rec onfigurable and re-useable in atmosphere of rapid and          Better response time for customer       Allan Martel
pervasive change; Recommend examining and joining like -minded efforts like           design specification changes.
MPOSE: proposed by INTE L as an Intelligent Manufacturing Systems (IMS)
                                                                                      To avoid duplication of effort.
R&D projects named MPOSE (www.ims.org) Purpose of the projects is to
produce an IT technology infrastructure underpinning a service orient ed
architecture (SOA) focused on international supply network management.


Recommendation: No specific recommendations were developed for this area.




                                                                     DRAFT
                     A Strategy for Establishing a Smart Assembly Initiative                                                 Page 77
                                                                        DRAFT


Table 44. The fifth and final priority theme from an end user perspecti ve for achievi ng smart assembly goals was skills specifically attuned to assembly
“engineering.”
What is Needed?                    Why?                                                                                        Participant
Develop assembly engineering       Engineering programs are currently stove piped.                                             Bryan Dods
curriculum at universities akin
                                   Assembly engineering requires a diverse set of skills to integrate each engineering
to manufacturing engineer
                                   specialty – i.e., current curricula don’t emphasize systems integration needed for
                                   assembly


Recommendation: No specific recommendations were developed for this area.




                                                                      DRAFT

				
DOCUMENT INFO
Jun Wang Jun Wang Dr
About Some of Those documents come from internet for research purpose,if you have the copyrights of one of them,tell me by mail vixychina@gmail.com.Thank you!