4
The Way Forward: Overcoming Cultural and Organizational Challenges

A lesson learned from the case studies considered by the committee and from other fields that have undertaken major integration efforts is that the cultural issues account for more than half of the effort required for making progress and gaining acceptance. For integrated computational materials engineering (ICME) to gain widespread acceptance, cultural shifts are required within industry, academia, and government. It was clear from several case studies that the cultural changes required to fully benefit from ICME should not be underestimated. In the sections that follow the committee identifies the primary cultural barriers faced by the communities involved in advancing and implementing ICME. In this chapter, the committee considers the cultural barriers faced by the key stakeholders in industry, government, and the materials science and engineering (MSE) community. Its recommendations address barriers to industrial acceptance of ICME within the integrated product development (IPD) process, the acceptance by government of its role as a champion and coordinator of ICME, and acceptance in the MSE community of ICME as an inherent component of its overall education, identity, profession, and practice.

CULTURAL BARRIERS TO ICME IN THE MANUFACTURING INDUSTRY

As described earlier, the IPD process has revolutionized U.S. industry. (For a discussion of the IPD process, see Box 2-1.) However, although the constraints imposed by the choice of materials for a particular product can strongly influence the design and manufacture of that product, materials are not currently part



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 102
4 The Way Forward: Overcoming Cultural and Organizational Challenges A lesson learned from the case studies considered by the committee and from other fields that have undertaken major integration efforts is that the cultural issues account for more than half of the effort required for making progress and gaining acceptance. For integrated computational materials engineering (ICME) to gain widespread acceptance, cultural shifts are required within industry, academia, and government. It was clear from several case studies that the cultural changes required to fully benefit from ICME should not be underestimated. In the sections that follow the committee identifies the primary cultural barriers faced by the communities involved in advancing and implementing ICME. In this chapter, the committee considers the cultural barriers faced by the key stakeholders in indus- try, government, and the materials science and engineering (MSE) community. Its recommendations address barriers to industrial acceptance of ICME within the integrated product development (IPD) process, the acceptance by government of its role as a champion and coordinator of ICME, and acceptance in the MSE community of ICME as an inherent component of its overall education, identity, profession, and practice. CULTURAL BARRIERS TO ICME IN THE MANUFACTURING INDUSTRY As described earlier, the IPD process has revolutionized U.S. industry. (For a discussion of the IPD process, see Box 2-1.) However, although the constraints imposed by the choice of materials for a particular product can strongly influ- ence the design and manufacture of that product, materials are not currently part 0

OCR for page 102
o v e r c o m i n g c u lt u r a l o r g a n i z at i o n a l c h a l l e n g e s 0 and of the IPD computational-tool-based optimization process. The constraints of materials are only considered outside the IPD multidisciplinary design loop, and materials are a fixed and limiting constraint on the overall IPD process rather than a parameter that can be optimized along with other engineering parameters. The allowable list of materials may be taken as fixed or it may include a small subset of materials that are evaluated outside the optimization loop. This approach narrows the design space, resulting in suboptimal product performance.1 Conversely, the development of the optimal materials, currently a lengthy and expensive process, does not benefit from integration of the materials and materials-manufacturing development processes into the product optimization process. Although there have been several successful demonstrations of ICME appli- cations within industry (see the case studies discussed in Chapter 2), as Lesson Learned 1 from that chapter indicates, ICME is still in its infancy and ICME tech- nologies remain immature. Indeed, in the committee’s judgment, gaps between the available and the ideal computational models and methods will persist for some time. While this does not preclude the development of an ICME capability for a particular material or application, it can make the capability much more difficult to achieve. Moreover, and perhaps more important, for many industrial firms the existence of ICME and its benefits are unknown. The engineering orga- nizations that have experimented with the application of ICME have judiciously selected engineering challenges with near-term ICME opportunities. Focused sets of materials tools have been or are being integrated to address a specific engineering component and material system. However, the committee knows of no organiza- tion where ICME is fully institutionalized or represents the norm for the product development process. Inertia in the Engineering Design Community The relatively rapid adoption by industry of the IPD process and computation- ally based multidisciplinary design optimization (MDO) indicates that the integra- tion of engineering tools and computation has become and will remain a critical element in product design and development. However, the insertion of materi- als into this loop to improve product design and optimization may actually be impeded by industry’s prior investment in and commitment to existing integrated engineering tools and processes. The competing business forces that encourage and dissuade ICME implementation are shown in Figure 4-1. Many engineering 1 The design space is the set of possible designs and design parameters that meet a specific product requirement. Exploring the design space means evaluating the various design options possible with a given technology and optimizing the chosen option with respect to specific constraints such as power or cost.

OCR for page 102
i n t e g r at e d c o m P u tat i o na l m at e r i a l s e n g i n e e r i n g 0 • Higher engineering • Cyberinfrastructure productivity Cost Benefit ICME • ICME training • Accelerated insertion Implementation • ICME development • Reduced cost Risk and Issues • ICME maturity • Customer acceptance • Disruption during Transition FIGURE 4-1 Competing business factors that affect the decision to implement ICME technical maturity and validation. organizations, particularly at large companies, have invested considerable human resources and capital in the codification of their engineering practices and prod- uct development processes and timelines. Engineering organizations capture best practices, provide discipline in the design and execution of engineering programs, and build consistent, validated, and approved computational methods. While these formal engineering practices and product development processes enhance the engineering product, they also retard the adoption of progressive, promising new directions such as ICME. As depicted in Figure 4-1, before ICME reaches full maturity and acceptance and before it is validated, engineering managers will take a cautious stance to its implementation where its development costs are high and where they might be subject to liability risks when ICME-based recommended practices do not agree with previously accepted materials engineering practices. For many industries, the design engineer or the production engineer is the leader of the IPD and, in a sense, the primary customer for the materials engineer. The design engineer depends on materials engineers for material selection advice, material property data or constitutive relations (that is, the relation between applied stresses or forces and strains or deformations) for design analyses, and materials evaluation of tested components or components that have already been in use. In general, design engineers who have confidence in the existing support role of mate- rials engineering are likely to resist any extensive, rapid change—such as might be entailed in adopting an ICME approach to materials engineering—because their design methods and inputs have withstood the test of time and gained customer acceptance. For example, a design engineer probably will not want to spend time reconciling the product life predictions from proven, data-driven analyses with

OCR for page 102
o v e r c o m i n g c u lt u r a l o r g a n i z at i o n a l c h a l l e n g e s 0 and predictions from ICME when he/she feels confident that existing methods work and when there is skepticism about the value of materials computational methods and tools. Also, design engineers focus on the demands of their own discipline and generally have less understanding of and interest in material mechanisms and models, and they may be constrained by formal design practices and regulatory requirements that cannot be modified without rigorous validation and approval by a chief engineer or regulatory agency. The industrial product development community must have confidence that ICME has sufficient maturity and validity to provide, at acceptable risk levels, tangible offsetting benefits in terms of product improvement, cost savings, and/or reductions in the development cycle time. Given the option to invest in the new and “unproven” (to their way of thinking) capability provided by ICME, many engineering and R&D managers will choose the traditional route to a new product in the absence of some incentive. Inertia in the Industrial Materials Engineering Community Although the role of the materials engineering discipline’s support of IPD varies from industry to industry and firm to firm, it can undermine confidence in ICME and prevent its adoption. Given the current state of MSE curricula and com- putational materials tools, experienced materials engineers are often ill equipped to develop or use ICME tools to perform their job. They may be skeptical of ICME capability and will insist on acquiring their own data to support their engineering decisions. A further cultural barrier is that materials engineering managers may resist adopting radically new engineering processes that have not demonstrated the ability to deliver the required materials technologies within established product development and investment cycles. For the foreseeable future, ICME tool sets will be developed by specialists, either in industry or academia. In large and medium companies this role may fall to either in-house materials researchers or materials engineers with specialized training. Materials engineering managers, cognizant of a gap in ICME skills, may avoid large-scale, rapid ICME deployment, especially if they believe that ICME requires the addition of new employees with a higher level of education. In addition, managers who have traditionally imposed restrictions on the external disclosure of materials information will need to redefine what information is truly sensitive and what information can safely be disseminated among the com- munity for mutual benefit. This will be more readily accomplished in industries where specific materials do not represent a strategic advantage—for example, the automotive, consumer product, and packaging industries. In industries where the materials of construction are not themselves propri- etary, the primary resistance to sharing data is the high cost to develop a mate-

OCR for page 102
i n t e g r at e d c o m P u tat i o na l m at e r i a l s e n g i n e e r i n g 0 rial. Mechanical engineering managers do not worry about who among their competitors is using a particular finite-element analysis (FEA) software package, and—similarly—progressive materials engineering managers will take the same view of ICME-enabling infrastructures. Rather they will emphasize the benefits of skillful use of ICME tools. Their support for open-access databases will accelerate the further development and adoption of ICME. Materials engineering leaders and decision makers must also identify a pathway by which to transition gracefully from current practice to ICME without disrupting the supply of materials to design, manufacturing, and product support operations. A successful pathway must account for the needs imposed by ICME on the skills and training of materials engineers within an organization, the handling of pro- prietary information and information controlled by regulation, the expansion of computational infrastructure, and the risks perceived by customers. Inertia in the Manufacturing Engineering Community Resistance to adopting ICME may also come from outside the design and mate- rials communities—namely, from manufacturing engineers. One of the primary objectives of ICME is the integration of manufacturing and product design into a holistic computational system that includes the materials developer, leading to an increased role for manufacturing engineers in providing new and quantitative outputs to the materials and product development process. In most industries, the main focus of computer-aided engineering (CAE) for manufacturing is to make sure the part being designed can be manufactured. In contrast, the main focus of design CAE is to design the “perfect” part, without considering manufacturing constraints or variability. Even in IPD, design CAE analysts often work in isolation from manufacturing simulation, and the integration of manufacturing CAE and design CAE is rarely considered. Integrating design, materials, and manufacturing engineering using computational tools and methods will require overcoming orga- nizational chimneys, rising above the resistance to change established processes, and adjusting product and process development cycles to provide or receive proper and interconnecting inputs and outputs. A significant barrier to achieving this kind of integration is that manufacturing CAE is often conducted by suppliers, and in general there is no paradigm for passing manufacturing CAE information to original equipment manufacturers (OEMs). This results in barriers associated with purchasing policies and concerns about proprietary information. Overcoming Inertia in the Manufacturing Industry Even if there were no other barriers to the implementation of ICME, many engineering organizations would opt to either delay it or phase it in owing to the

OCR for page 102
o v e r c o m i n g c u lt u r a l o r g a n i z at i o n a l c h a l l e n g e s 0 and uncertainty surrounding its completeness and the lack of maturity for the full suite of ICME models and tools. Prudent organizations will critically identify, assess, and prioritize those applications where ICME methods add value to their engineering function and then judge which are feasible with the existing state of the art. Most firms will judge the technology readiness by comparing ICME’s results with the results of existing, proven, data-driven methods. Organizations will encounter lower introduction barriers if they implement ICME gradually and initially learn how to use it to supplement existing methods or provide results that fuse traditional and ICME methods. Conversely, once ICME achieves greater acceptance, current data-driven methods can be applied to establish empirical relationships that can substitute for missing physically based models in an ICME system. Regardless of the ICME implementation strategy, imple- menters must emphasize validation and uncertainty measurement to gauge and demonstrate progress in the maturation of ICME and provide a sound basis for its application. It is important to recognize that such cultural barriers can be and have been overcome in certain sectors where ICME’s value is recognized. One prominent example of such a cultural shift occurred in the auto industry. Government stan- dards require crash testing of new vehicle designs, which has led to the costly and time-consuming building of vehicle prototypes and their crash testing during the design process. A complicating factor is that the response of prototypes is not necessarily the same as that of production cars. Thus a sufficiently accurate com- putational model offers large payoffs. For example, in the mid-1980s, a standard vehicle development testing program would have required approximately 30 crash tests per year for 5 years, each at a cost of approximately $250,000. For the Ford Mondeo program, finite element modeling enabled a reduction in test costs of 30 percent, or $12.5 million, over 5 years.2 Since that time, despite the imposition of more stringent safety requirements, the number of crash tests has been cut in half through the use of CAE. Recognizing the utility and validity of computational crash testing, automakers have shifted their cultures and rely increasingly on CAE to cost-effectively engineer safe vehicles.3,4 A critical challenge is providing the human and financial resources required to fully develop a firm’s ICME capability. In the near term, overcoming this challenge 2 David Bensen, University of California, San Diego (UCSD), “Integrating FEM with materials models for crash tests,” Presentation to the committee on March 13, 2007. Available at http://www7. nationalacademies.org/nmab/CICME_Mtg_Presentations.html. Accessed February 2008. 3 P. Prasad, Ford Motor Co., private communication (2007). 4 David Bensen, UCSD, “Integrating FEM with materials models for crash tests,” Presentation to the committee on March 13, 2007. Available at http://www7.nationalacademies.org/nmab/CICME_ Mtg_Presentations.html. Accessed February 2008.

OCR for page 102
i n t e g r at e d c o m P u tat i o na l m at e r i a l s e n g i n e e r i n g 0 will require investment in ICME development across different product develop- ment cycles. The term “industry” implies a monolithic entity with common interests and motivations. This is clearly not the case. So the ability of “industry” to organize a coordinated development of ICME is limited, although the establishment of con- sortia of companies with common interests within industrial segments is one way to foster development of ICME, particularly with augmented governmental sup- port. Such consortia could be funded by industry alone or jointly by industry and government agencies. To focus resources and demonstrate the capability of ICME, an appropriate foundational engineering problem could be selected, as described in Lesson Learned 10 in Chapter 2. Generally a foundational engineering problem would consist of a manufacturing process or set of processes; a material system; and an application or set of applications that define the critical properties and geometries. Examples of such foundational engineering problems are the ICME capabilities described in Chapter 2 for forged nickel-based superalloy turbine disks (aerospace sector) and the cast aluminum power train components (automobile sector). As mentioned in Chapter 1, the committee envisions that ICME could impact a range of industries. Some possible foundational engineering problems are these: • High-dielectric materials and processes for improving the performance of microelectronic devices, • Low-cost organics for robotics sensors, • Thermal protection materials for hypersonic vehicle surfaces, • Catalysts for optimizing the performance of hydrogen-fueled systems, • Reliable and rapid recertification of components in aging structures, • Materials for ballistic and blast survivability of ship hulls, • Thermoplastic injection-molded materials for automotive structures, • Materials and electrochemical processes for advanced batteries, • Nanoparticles for magnetic storage devices, and • Composite or advanced metallic materials for aeroengine components. A good model of a consortium in this area is the U.S. Automotive Materials Partnership (USAMP), an ICME project in the U.S. auto industry. Ford, Gen- eral Motors, and Chrysler have selected “Magnesium for Body Applications” as a foundational engineering problem for the consortium, and together they are developing an ICME infrastructure and knowledge base for these materials and the manufacturing processes that are used to fabricate engineering components from them. This 5-year international program was initiated in 2007 and is jointly sponsored with the U.S. Department of Energy (DOE), the China Ministry of Sci-

OCR for page 102
o v e r c o m i n g c u lt u r a l o r g a n i z at i o n a l c h a l l e n g e s 0 and ence and Technology, and Natural Resources Canada.5 The approximate funding over 5 years is between $6 million and $7 million. It involves participation from researchers at Chrysler, Ford, and General Motors, and more than 15 universities and government laboratories. This level of funding is acknowledged to be small relative to the needs for this particular foundational engineering problem, but it is an important first step. Similar consortia would appear to be appropriate in aircraft engines, airframes, and electronic materials to name a few obvious areas. Despite the clear vision and persuasive logic of ICME, the cultural challenges outlined above will impede the widespread implementation of ICME in industry. There is a reluctance to fix a process that is generally not thought to be broken, particularly one that was expensive to build. In summary, the committee concludes as follows: Because of ICME’s relative immaturity, the remaining computational gaps, and the potential for ICME to disrupt current IPD processes, the industrial product development community is skeptical of or unaware of the benefits of ICME. Sustained funding across multiple product development cycles will be required to advance ICME and build confidence in it to the point where it is fully accepted. There are two main cultural challenges impeding the widespread industrial adoption of ICME: • PD engineers are not aware of ICME (including tools and suppliers) and I ICME has not been accepted into the IPD process because its value in specific problems of interest to industry has not been proven. • esources for things such as R&D investments and personnel must be R committed to an ICME project for more than 1 year, which is longer than the typical 1-year investment cycle and outside the typical product- oriented R&D cycles. Establishing consortia within industrial segments having common interests offers a means to foster development of ICME, build awareness and confi- dence, and augment governmental support. It may be that government will have to offer incentives to enable such consortia and collaborations, and the terms of the consortia may need to explicitly state that the results will be open to the wider ICME community, as appropriate. 5 Joseph Carpenter, DOE, “DOE’s work on extruded long-fiber-reinforced polymer-matrix compos- ites,” Presentation to the committee on March 13, 2007. Available at http://www7.nationalacademies. org/nmab/CICME_Mtg_Presentations.html. Accessed February 2008.

OCR for page 102
i n t e g r at e d c o m P u tat i o na l m at e r i a l s e n g i n e e r i n g 0 CULTURAL BARRIERS TO ICME IN THE MSE COMMUNITY Advancing ICME will also require adaptation by the materials community in the United States and abroad. There must be a shift in the MSE community’s mind-set so that the materials discipline becomes focused on problem solving alongside the scientific endeavor—that is, making materials engineering a more quantitative discipline and positioning MSE as an integral part of the engineering process. In addition, the MSE acceptance of ICME can be accomplished by shifts in education and research and by advocacy and communication as well as support and encouragement by government agencies and industry. The changes needed in the MSE community can be assessed by looking at the traditional roles of MSE practitioners and how they might have to change if ICME is to proceed successfully. Need for Change in the Roles of MSE Professionals An environment characterized by multidisciplinary, collaborative teamwork is integral to the future success of ICME. The modern materials expert must work across disciplinary boundaries in multidisciplinary teams throughout the design process, sharing information and communicating outside the original company, country, or discipline. This requires an ability to work with others and collaborate across disciplinary boundaries, a recognition that materials are not the only design parameter or constraint, and a mind-set for sharing data and information. While the MSE community will be a primary contributor to the development of ICME, experts from a variety of other fields, including engineering, physics, mechanics, information sciences, systems engineering, mathematics, and computer sciences, will also play important roles in product design and manufacturing and in MSE research. Product Design and Manufacturing Materials engineers have adopted an approach to their profession that often focuses on acquiring material property data sets from which they calibrate aver- age (nominal) materials properties, sometimes with statistical information, and, in collaboration with CAE analysts, reduce these data to conventional constitutive models for use in performance CAE. But for ICME to be truly successful, materials engineers will need to think of ICME as a key enabler capable of providing quan- titative information in the product development process. This will be a significant cultural shift for materials engineers. Materials designers, by contrast, often optimize an alloy or process for a single property without regard to other design constraints—for example, other properties,

OCR for page 102
o v e r c o m i n g c u lt u r a l o r g a n i z at i o n a l c h a l l e n g e s  and product geometries, and so on—or manufacturing constraints. Alloy development projects tend to be Edisonian and experimental rather than driven by model predic- tions. While designers make use of the relationships between structure, underlying physical mechanisms, and macroscopic properties, they do not use quantitative analytical or computational tools. For ICME to progress, materials designers will need to look at the material development process as an integrated computational process involving optimization of multiple attributes (microstructure, properties, geometries, and manufacturing cost) and its role in overall product design. Materials scientists traditionally work as single investigators to gain narrow but deep technical knowledge and insights, generally not part of a larger inte- grated effort. Individual materials scientists can, however, provide other ICME practitioners with important inputs, and they can also be beneficiaries of inputs from those same practitioners. For such benefits to accrue, the model that focuses on individual investigators gaining deep scientific insight must be expanded to include the integration of these insights into quantitative tools that combine col- lective insights from many different experts. A concomitant increase in the role of materials knowledge in computational system analysis is paramount, leading to quantitative tools used by product engineers and materials engineers working together in integrated teams. MSE Research Within the MSE community there is a tension between the materials-science- based, fundamental efforts and the materials-engineering-based, applied efforts. ICME provides an important linkage between these two activities and the two categories of experts—scientists and engineers—and can have a profoundly impor- tant synergistic benefit. MSE academic research is playing a key role in develop- ing ICME computational tools and experimental data sets. Although academic research is typically fundamental and not geared toward immediate industrial application, some transformational steps could allow the research to more readily translate into industrial needs by means of ICME. Protocols could be developed to facilitate a viable and easy translation. As discussed above, small businesses can play a vital role in this translation. For example, in an IPD process, a new compu- tational modeling capability developed in academia could be readily incorporated into an IPD framework if there was a known interface format for the model or for the experimental data passing through the model. Such standards or user inter- faces could be part of the IPD framework, enabling and encouraging researchers to formulate their model interfaces for easier incorporation into IPD processes, speeding up their use by others in a win-win situation for the researcher and for the ICME/IPD users. In addition, various forms of experimental data could have specific reporting formats depending on data type and could be deposited and

OCR for page 102
i n t e g r at e d c o m P u tat i o na l m at e r i a l s e n g i n e e r i n g  shared, enabling the more ready use of data in model validation, data mining, and materials informatics. Chapter 3 discusses in greater depth the database needs for ICME. Such an approach would require establishing database depositories and shifting the mind-set of the researcher to include preparation of data in a standard form for inclusion on a curated, shared database. Finally, ICME capabilities can enrich fundamental materials research by providing new insights at the interfaces between different subfields and by providing a focus that enables identifying important new or underexplored areas of research. Materials professional societies have a unique opportunity to foster ICME and make the MSE culture more collaborative and fertile for it. This support can come from assisting the materials research community in the development of standards and taxonomies, encouraging and providing incentives for open access to research results, and providing ICME collaborative Web sites, which can serve as repositories for high-quality data and models that are important parts of these research results. By hosting the ICME cyberinfrastructures described in Chapter 3, materials professional societies can enable the materials community to contribute to and in some cases to self-organize in the establishment of a broad ICME infra- structure. Professional societies can also communicate the results and successes of ICME developments in their publications and programming. The ICME efforts of The Minerals, Metals & Materials Society (TMS) stand out in this regard. TMS has established an ICME coordination committee and an ICME Web site and is actively programming and publishing in this area.6 In summary, the committee concludes as follows: For ICME to succeed, it must be embraced as a discipline by the interna- tional materials science and engineering community, leading to changes in education, research, and information sharing. This would transform the role of materials science and engineering to one of uniting engineering and scientific endeavors into more holistic and integrated activities. Education and Workforce Readiness Implementing cultural change in the materials discipline will require the inte- gration of ICME into the MSE curriculum if ICME is to become part of the iden- tity of an MSE professional. With the recent reforms in engineering accreditation, the role of materials in design and the importance of computation in materials engineering undergraduate curricula are now recognized, and graduates must demonstrate the following: 6 For more information, see http://materialstechnology.tms.org/icme/ICMEhome.asp. Accessed February 2007.

OCR for page 102
i n t e g r at e d c o m P u tat i o na l m at e r i a l s e n g i n e e r i n g  of ICME and how incorporating it into industrial processes can impact product design and manufacturing. As described above, materials professional societies have an important role to play in establishing an ICME development network and infrastructure and in communicating the progress and success of ICME. Materials societies also have an important role to play in including ICME in the continuing education of professionals. The committee considers that the UMC is in a unique position to influence curricula and change the culture of MSE academic institutions and that it could take an active role in promoting ICME and the curricular changes that support improvements in the computational ability of the students who graduate from their departments. Materials professional societies can share best practices on the development and introduction of computational tools into MSE curricula. The committee also believes that materials professional societies can meet the need for workforce training by offering workshops and tutorials in ICME. Finally, the committee believes that alliances between the small businesses who are developers of software tools and MSE teaching institutions can be particularly effective for propagating ICME tools, particularly as students make the transition to engineer- ing practice.8 ROLE OF SMALL BUSINESS IN ICME DEVELOPMENT During the course of the study, the committee heard from a number of small science and engineering companies that have played a key role in developing, advo- cating, and maturing ICME technologies.9 They have acted at times as the scouts for the OEMs in identifying and integrating MSE research into viable commercial products. In the course of these briefings, the committee learned that one signifi- cant concern of the end user of the products discussed is the long-term support and sustainability of the products. The committee observed a variety of business models employed in small ICME firms that may serve as models for emerging small businesses in the ICME sector. In one approach, user requirements are reviewed through consortia, focus groups, and special interest groups, with consortium members paying for a seat at the table, giving them access to the decision-making and priority-setting process for new model development. The resultant software applications are tailored to specific needs in industries that produce products rang- ing from pharmaceuticals to consumer goods and are licensed by the industries that 8 Nuno Rebelo, Simulia, “CAE: Past, Present, and Future,” Presentation to the committee on May 30, 2007. Available at http://www7.nationalacademies.org/nmab/CICME_Mtg_Presentations.html. Accessed February 2008. 9 The committee received briefings from Accelrys (www.accelrys.com), Automation Creations (www. aciwebs.com), Engenious Systems (www.engenious.com), Materials Design (www.materialsdesign. com), Phoenix Integration (www.phoenix-int.com), and QuesTek (www.questek.com).

OCR for page 102
o v e r c o m i n g c u lt u r a l o r g a n i z at i o n a l c h a l l e n g e s  and specified development goals.10 Other companies are building materials and process modeling tools and offer ICME as a service by applying systems integration tools to solve customer problems.11,12 Another well-established business model for small firms is the development of databases; an example of success is the CALculation of PHAse Diagram (CALPHAD) software and the thermodynamic databases that have become key elements of ICME. There are several candidate business models for building and marketing databases, ranging from fully proprietary to government funded, including the following: • Commercially available proprietary databases (Thermotech, CompuTherm, Granta, Thermo-Calc), • Consortium (Sematech), • Advertising-supported (MatWeb), • Community/emergent (Wikipedia), • Government support (National Institute for Standards and Technology, National Science Foundation, National Science Digital Library, and Center for Computational Materials Design), and • Hybrid government/industry or government/university collaboration (NIST Solder Database13). Each business model has advantages and disadvantages in its ability to meet goals for ICME, such as relevance, lack of bias, verification, innovation, efficiency, and broad availability. While any of the above models could emerge as the preferred structure(s) for ICME, each will face challenges to being accepted by the current industrial community. For U.S.-based firms, SBIR and STTR funds are generally critical for the success of these firms. That said, it must be recognized that not all the small businesses that are developing or are capable of developing ICME tools and models are U.S. firms. 10 The committee notes that ICME will challenge how users want to use existing and new software tools, and this in turn will challenge how software vendors currently license their software; these issues will also need to be explored and addressed in due course. 11 Frank Brown, Accelrys, “Scientific business intelligence,” Presentation to the committee on May 30, 2007. Available at http://www7.nationalacademies.org/nmab/CICME_Mtg_Presentations.html. Accessed February 2008. 12 Charles Kuehmann, QuesTek, “Experiences integrating theory & experiment & industrial prac- tice/culture,” Presentation to the committee on March 13, 2007. Available at http://www7.national academies.org/nmab/CICME_Mtg_Presentations.html. Accessed February 2008. 13 Ursula Kattner, NIST, “Thermodynamic databases in support of materials,” Presentation to the committee on May 29, 2007. Available at http://www7.nationalacademies.org/nmab/CICME_Mtg_ Presentations.html. Accessed February 2008.

OCR for page 102
i n t e g r at e d c o m P u tat i o na l m at e r i a l s e n g i n e e r i n g  Another business approach that has been adopted14,15 involves a small busi- ness partnering with a larger company to share the cost of developing, say, a new material model or a new computational feature that is needed to simulate their particular problem. In return, each partner company has access to the advances as soon as they have been completed. Commercial integration software tools are available that are designed to link a variety of disparate methods into an integrated package that can then be used to optimize some multistep or multiphysics process.16,17 Software tools such as these have frequently been developed by small to medium-sized software companies using government (for example, SBIR) funding. As a result of these efforts, de facto standards are emerging for wrapping models, running parallel parametric simulations, sensitivity analysis, and reducing the complexity (order) of systems. While these codes are becoming widely used for IPD and MDO, they are also well suited for ICME. Such companies market and apply systems integration tools that will solve specific engineering problems, tools for interoperability across organiza- tions, and tools that are experiencing a period of sustained business growth. The rapid adoption of IPD by industry is further evidenced by the large international participation in IPD conferences; in particular, although IPD was developed in the United States, it has been widely adopted by competitors from abroad, and now licenses for this MDO integration software and attendance at IPD conferences in Japan are reported to be twice that in the United States.18 Based on information from a representative group of small ICME businesses, the committee concludes as follows: Small science and engineering companies are playing a key role in develop- ing, advocating, and maturing ICME technologies. They act as the scouts for the OEMs in identifying and integrating MSE research into viable commer- cial products. These innovative firms develop just the kinds of technologies 14 David Hibbit, Abaqus, Inc., “A perspective from a commercial finite element software vendor,” Presentation to the committee on May 29, 2007. Available at http://www7.nationalacademies.org/ nmab/CICME_Mtg_Presentations.html. Accessed February 2008. 15 Nuno Rebelo, Simulia, “CAE: Past, present, and future,” Presentation to the committee on May 30, 2007. Available at http://www7.nationalacademies.org/nmab/CICME_Mtg_presentations.html. Accessed February 2008. 16 Brett Malone, Phoenix, “Phoenix integration,” Presentation to the committee on March 13, 2007. Available at http://www7.nationalacademies.org/nmab/CICME_Mtg_Presentations.html. Accessed February 2008. 17Alex Van der Velden, Engineous, “Use of process integration and design optimization tools for product design incorporating materials as a design variable,” Presentation to the committee on March 14, 2007. Available at http://www7.nationalacademies.org/nmab/CICME_Mtg_Presentations.html. Accessed February 2008. 18 Ibid.

OCR for page 102
o v e r c o m i n g c u lt u r a l o r g a n i z at i o n a l c h a l l e n g e s  and and human expertise that are required to mature ICME and that the SBIR and STTR programs were designed to support. PROPOSED APPROACH TO ICME FOR THE GOVERNMENT As described above, significant effort will need to be expended in industry and within the MSE community to support the widespread development and adoption of ICME. But there is another key stakeholder. The federal government supports materials research with two goals in mind: (1) to further the fundamental under- standing of materials science and materials engineering and (2) to promote the adoption of materials technologies in industries of national importance—that is, in priority areas such as national defense, the energy sector, automobiles, aerospace, and health care. Government support of materials engineering R&D often focuses on technologies that can be successfully matured and implemented in a specific product. There are numerous examples of successful research translating into new capabilities that underpin the competitiveness of the country and its economic and national security. The committee believes that government agencies play a critical role in cham- pioning, developing, and implementing ICME in the United States. Bringing ICME from its infancy today and turning it into a mature, widely adopted element of the materials profession in academia and in the industries that rely on materials engi- neering will require the resources, organizational capability, and the collective long- range vision that are best supported by government. The committee is convinced that just as government has undertaken coordinated activities that supported the development of nanotechnology from its nascent state 10 years ago; the mapping of the human genome; and the development of computing, networking, and software technologies, it can clearly also play a crucial role in the development and valida- tion of an ICME methodology calling on the capabilities of its research agencies. Although not strictly a cultural challenge, insufficient coordinated government support for ICME would constitute a nontechnical barrier to ICME’s emergence as a mature discipline. Government has the opportunity to underpin the cultural changes described in this chapter and the technical challenges described in Chapter 3. Government agencies are uniquely able to organize, advocate, and sustain the long-range programs needed to make ICME a reality. Absent a significant level of government organization and funding progress, development of ICME will take many decades and might never fully materialize. So far federal research investment in tools for computational materials science has focused mainly on supporting individuals or small teams to develop tools for specific materials technologies. There have been successful government-supported programs for ICME-related activities—for example, the Advanced Insertion of

OCR for page 102
i n t e g r at e d c o m P u tat i o na l m at e r i a l s e n g i n e e r i n g 0 Materials (AIM) program supported by the Defense Advanced Research Projects Agency (DARPA) and the Advanced Simulation and Computing (ASC) program supported by the National Nuclear Security Administration (NNSA)—and new efforts such as DOE support of ICME development in the automobile sector. The success of these efforts has made ICME ready today to benefit from a systematic, sustained program to develop the tools and infrastructures needed for ICME. The right government action now can promote the development of the technical tools required and the sharing of the data and models developed by MSE R&D with the engineering community. Integrating the disjointed activities of today will require a change in philosophy and a coordinated, substantial, long-term effort but will result in less duplication of effort and less wasting of precious R&D resources. If properly captured and designed, much of the current government R&D portfolio could be integrated into ICME efforts, with potentially a very modest amount of new fund- ing. The committee speculates that this could be accomplished by targeting existing research funds and eliminating redundant research. Incremental funding would clearly be required for activities such as material informatics, cyberinfrastructure, and database development and curation, which are not currently being funded in any meaningful way within the materials community. As was discussed in Chapter 3, while some of the gaps in the ICME tool set have been bridged, the full set of tools for ICME remains to be developed. The time is ripe to develop concerted, coordinated efforts to calibrate existing models for particular materials systems and applications and integrate them into usable integrated models that include manufacturing, materials, and design inputs and outputs. Where modeling gaps exist, the experience summarized in Chapter 2 has shown they can generally be filled by the empirical relationships developed for the particular material system under consideration. The experience to date has shown that making progress in ICME capabilities is generally an organizational challenge and that the rate of progress in ICME capability will be directly related to the degree of organization and level of funds directed at the problem. ICME is an emerging discipline. In most cases it is a precompetitive activity, although, as with nanotechnology, niche commercial applications are likely to emerge. Sustained funding as part of longer-term R&D efforts will be required given the need for the discipline’s maturation and the need to train researchers who can effectively contribute and educate engineers who are critical to its implementa- tion. Firms can also reap financial benefit from ICME capabilities, suggesting to the committee, as mentioned above, that a useful mechanism for funding ICME pro- grams will be public-private partnerships in the form of consortia aimed at solving particularly challenging problems of high priority to the nation. Establishment of international cooperative efforts will be important for accelerating progress and reducing costs. This is a classic cooperation/competition situation in which resources must be leveraged to develop a basic capability while ensuring a com-

OCR for page 102
o v e r c o m i n g c u lt u r a l o r g a n i z at i o n a l c h a l l e n g e s  and petitive and security advantage by being proficient in the application of the tools and by being early adopters. While national security interests may dictate restrict- ing some information on the building blocks of ICME, the committee believes that ensuring as much open access to ICME as possible and making information available to the widest possible audience are also in the best interest of the nation. Such information would include all the elements of the ICME cyberinfrastruc- ture, including collaborative Web sites and repositories of data and models. While export control laws have important purposes, the unnecessary expansion of export control to include ICME could substantially increase the time and cost to develop a widespread ICME capability and limit the ability of multinational corporations to obtain maximum value from it. Solving the so-called foundational engineering problems in specific high-pri- ority materials systems could provide a focus for the development of ICME and demonstrate its capability. An ICME solution to a key set of engineering problems critical to national security or competitiveness would provide the impetus needed for widespread development and utilization of IMCE. Some possible examples of foundational engineering problems are listed in the report. The committee speculates that based on the case studies identified in Chapter 2, development of an ICME capability for a given material system to solve a particular foundational engineering problem will require an investment of $10 million to $40 million over 3 to 10 years, depending on the completeness and complexity desired. Inter- estingly, these funding levels are similar to those devoted to genomic challenge problems such as the NIH-funded Rhesus monkey genome project described in Chapter 2. The NNSA effort to develop an ICME capability for nuclear warheads represented a unique situation in that the required experimental efforts were understandably difficult and expensive given the complexities of the materials and systems involved, complexities that led to total costs of more than $150 million. It is important to note that having made an initial investment to develop an ICME capability for a particular foundational engineering problem, this capability can be extended at much less expense. An important feature of these foundational engineering problems for ICME would be the establishment of permanent Web- based cyberinfrastructures to serve as repositories for data and models that would enable future extensions. The committee concludes that a number of government agencies could play a role. Specific recommendations for these agencies are discussed below and in Chapter 1. Department of Defense Since DOD was the sponsor and primary catalyst for one of the central ICME demonstration projects (DARPA’s AIM), there is growing and broad awareness of

OCR for page 102
i n t e g r at e d c o m P u tat i o na l m at e r i a l s e n g i n e e r i n g  the potential for ICME within DOD. DOD develops and deploys advanced weapons systems of extreme complexity, systems that depend critically on the development of new materials systems and their incorporation into new products.19,20,21 By providing focused product development objectives as well as long-term sustainable investments, DOD has a unique opportunity to champion the development of a broad ICME capability to enhance and ensure national security. While there will most likely be a slight increase in near-term resource requirements to develop an ICME capability, there is potential for long-term efficiency improvements. In a time of increasingly constrained funding for DOD materials research, ICME provides a means to improve the efficiency of the development of new materials systems, in terms of both cost and time. The committee has concluded that DOD would benefit from establishing a defense ICME coordination group to champion development of ICME within the defense services and in the industries that supply the services. Because of the many overlaps between the materials needs of the DOD services, a coordinated effort would reap particularly attractive benefits, maximize the value of DOD investments, and minimize the duplication of effort. The tasks for the coordina- tion group would include • Coordinating and monitoring of DOD ICME efforts that are currently focused on single-agency needs. • Defining a long-range (15-20 year) strategy and roadmap for the coordi- nated development of an ICME capability that is specific to the needs of DOD. The strategy would include these: —Identifying DOD’s ICME needs. —Identifying initial high-priority foundational engineering problems to demonstrate the promise of ICME. —Establishing and curating a cross-service ICME cyberinfrastructure to include data repositories and libraries of online tools to support the networking and collaboration needed to advance a collaborative ICME culture. 19 Defense Science Board, Defense Science and Technology, 2002. Available at http://www.acq.osd. mil/dsb/reports/sandt.pdf. Accessed October 2007. 20 National Research Council (NRC), Materials Research to Meet st Century Defense Needs, Wash- ington, D.C.: The National Academies Press (2003). 21 NRC, Accelerating Technology Transition: Bridging the Valley of Death for Materials and Processes in Defense Systems, Washington, D.C.: The National Academies Press (2004).

OCR for page 102
o v e r c o m i n g c u lt u r a l o r g a n i z at i o n a l c h a l l e n g e s  and • Setting policies and establishing procedures to promote public access to data and tools developed from DOD-supported ICME tools, subject to national security concerns. Department of Energy ICME has significant potential to provide for the design of new materials for use in energy production and improving the efficiency of its storage and use. The DOE’s Office of Energy Efficiency and Renewable Energy (EERE) has championed the ICME discipline, funding the first ICME consortium in the U.S. automotive industry. The consortium was a pilot project for development of an ICME knowledge base and cyberinfrastructure, in this case for magne- sium in automotive body applications. The committee concludes that there are likely to be many other areas where EERE can support ICME efforts to develop materials that will impact energy production and energy efficiency. These areas could include efforts in priority areas such as the development of the hydrogen economy, civil infrastructure technologies, solar energy, and vehicle technologies beyond the automobile sector. Moving technology out of the laboratory and into the marketplace is one of EERE’s major hurdles, and ICME is a way to achieve that goal. DOE’s Office of Science Basic Energy Sciences (BES) supports a suite of national facilities that could participate in the development of ICME by providing the core fundamental science, computational models, and theory. The committee concludes that there is a particular opportunity for BES to leverage its efforts in computational materials science by linking these theory-based resources to new and yet-to-be-developed rapid characterization techniques for materials, bridging the gaps in theory and computational technique. BES could also accelerate the development of ICME by promoting broad public access to data and tools coming out of federally supported ICME development programs. Here, BES could support coordinated collaborative ICME Web sites which will be used as repositories for the information generated. DOE’s NNSA has played a key role in developing computational materials science tools that predict long-term behavior of selected materials of importance in nuclear weapons systems. NNSA’s program on the plutonium life cycle, for example, successfully integrated modeling and experiment across scales, employ- ing a materials modeling approach that ranged from calculations of the electronic structure of materials to large-scale atomistic simulations to determine the long- term properties of microstructural features. While successful in meeting its objec- tives, the program focused on a single system, and no wider applicability of the ICME process was demonstrated.

OCR for page 102
i n t e g r at e d c o m P u tat i o na l m at e r i a l s e n g i n e e r i n g  The three NNSA laboratories are able to more widely apply an ICME framework to meet their materials development and assessment needs. They are leaders in the development and application of computational materials science, have remarkable computational facilities, and operate a wide-ranging, high-quality experimental program on materials development and characterization. The committee concludes that NNSA laboratories have a unique opportunity to develop such capabilities and to share the ICME framework with the wider materials community. National Science Foundation To derive significant benefit from ICME as a new venue for transformational collaborative cross-functional science in materials, physics, and mechanics, as well as to exploit ICME as an efficient mechanism for providing the outputs of fundamental materials research to the engineering community, the committee has concluded that NSF has a significant opportunity to accelerate considerably the development of ICME.22 The ICME cyberinfrastructure described in Chapter 3 is critical for advancing ICME, and the NSF cyberinfrastructure initiative could be an important source of funding for collaborative, cross-functional international net- works in pursuit of NSF’s goal of supporting research leading to the development and/or demonstration of innovative cyberinfrastructure services for science and engineering research and education. The protocols and approaches required for 22 NSF has a number of ICME-related programs. For instance, the Center for Computational Materials Design is a collaborative effort between the Pennsylvania State University and the Georgia Institute of Technology and a number of industrial and government sponsors, including the Air Force Research Laboratory, the Army Research Laboratory, Corning, the Ford Motor Company, the General Electric Global Research Center, General Motors, Knolls Atomic Power Laboratory, Lawrence Livermore National Laboratory, Procter & Gamble, Thermo-Calc, and Timken. For more informa- tion, see http://www.ccmd.psu.edu/. Accessed February 2008. The nanoHUB was created by the NSF-funded Network for Computational Nanotechnology (NCN), a network of universities with a vision to pioneer the development of nanotechnology from science to manufacturing through inno- vative theory, exploratory simulation, and novel cyberinfrastructure. The nanoHUB hosts over 790 resources, including online presentations, courses, learning modules, podcasts, animations, teaching materials, and more. Most importantly, the nanoHUB offers simulation tools accessible from a Web browser. For more information, see http://www.nanohub.org/about. Accessed February 2008. NSF’s Office of Cyberinfrastructure coordinates and supports the acquisition, development, and provision of state-of-the-art cyberinfrastructure resources. It supports cyberinfrastructure resources, tools, and related services such as supercomputers, high-capacity mass-storage systems, system software suites and programming environments, scalable interactive visualization tools, productivity software libraries and tools, large-scale data repositories and digitized scientific data management systems, networks of various reach and granularity, and an array of software tools and services that hide the complexities and heterogeneity of contemporary cyberinfrastructure while seeking to provide ubiq- uitous access and enhanced usability. For more information, see http://www.nsf.gov/od/oci/about. jsp. Accessed February 2008.

OCR for page 102
o v e r c o m i n g c u lt u r a l o r g a n i z at i o n a l c h a l l e n g e s  and ICME cyberinfrastructure have yet to be defined, and NSF could play an essential role in helping the ICME community explore this important new area and identify best practices. NSF could also require that all data and models developed during NSF-funded materials research be placed in publicly available ICME cyberinfra- structures. Development of innovative curricula and education for future profes- sionals trained in ICME can also be an important and essential NSF role. National Institute of Standards and Technology NIST has a mission to ensure the competitiveness of U.S. industry. ICME holds enormous promise for maintaining and enhancing this competitiveness by provid- ing a highly efficient means to capture materials knowledge and provide it to U.S. manufacturers with a view to optimizing products and manufacturing processes to produce high-quality goods at the lowest possible cost. Material properties in engineered components depend on the manufacturing processes by which they are produced and the manner in which they are used. Thus the data challenges in materials science and engineering are formidable. No single fixed database can be created from which materials engineers can derive the information they need to incorporate materials into a design; instead, there are related databases that can be cross-referenced. This is similar to the situation in bioinformatics, where a diversity of databases must be cross-linked. For example, in the Entrez Genome Project database at the National Center for Biotechnology Information (NCBI), a gene database can be linked to databases on proteins, nucleotides, taxonomy, molecular abundance, three-dimensional structure, and the PubMed database on life sciences journals. Given its unique mission and traditional role as a developer of standardized test techniques and curator of databases, NIST could establish and curate materials informatics databases that can be integrated into ICME tools and collaborative Web sites. SUMMARY As discussed in this chapter the committee is convinced that many of the primary barriers to advancing and implementing ICME are cultural and organiza- tional. The time it takes to develop ICME and gain acceptance for it will be directly proportional to the efforts expended in overcoming these cultural barriers. Doing so will remove the major constraint on new industrial products: the absence of a computational materials engineering capability in the product development and optimization cycle. ICME is the means to integrate materials into the broader computational engineering. Industrial acceptance of ICME is hindered, however, by the slow conversion of science-based materials computational tools to engineering-based

OCR for page 102
i n t e g r at e d c o m P u tat i o na l m at e r i a l s e n g i n e e r i n g  tools, by inertia in industry’s current product development processes, and by a lack of trained computational materials engineers. To overcome these challenges, industry has to develop an understanding of ICME and its capabilities, and gov- ernment agencies have a critical key role to play in championing the development of ICME. The MSE community also has a major role to play. While some aspects of ICME have been successfully implemented, ICME does not exist as a subdiscipline within MSE. For ICME to succeed, it must be embraced as a discipline by the MSE community, the community from which this committee is drawn, and changes in education, research, and information sharing must be brought about. The rate of progress in development of ICME will be proportional as well to the participation of academic researchers in information sharing, model integration, and develop- ment of an ICME infrastructure. Materials professional societies have their role to play, too, in establishing the ICME infrastructure, in the continuing education of professionals, and in communicating the progress and successes of ICME through programming and publications.