Pittsburgh, PA—Oct 9, 2008
NASA, Westinghouse Electric Company lead Pittsburgh discussions
Granta hosted a open seminar entitled "Managing and Using Materials Information" at the MS&T conference in Pittsburgh, PA. The event was attended by representatives from industries including aerospace, defense, energy, medical devices, home and construction products, and architecture. Presentations, software demonstrations, and case studies established the need for materials information management in engineering enterprises, and explained how an integrated approach can help organizations to get and manage the information that they need, deploy it to the people that require it, and use it to make better business and design decisions.
The highlight of the event was two case studies from users of Granta software—Dr Steven Arnold of NASA Glenn Research Center, and Paula Freyer of Westinghouse Electric Company. NASA are long-standing users of the GRANTA MI system. Westinghouse Electric Company is currently initiating pilot evaluations of the system. NASA's requirements include management of large quantities of complex test data and its use in modeling and simulation. The Westinghouse project is focused on access to key reference data and capture of in-house materials knowledge. The case studies thus provided two quite different perspectives and applications.
Dr Steven Arnold is Chief of the Mechanics and Life Prediction Branch in the Structures and Material Division at NASA Glenn Research Center. He also chairs the Material Data Management Consortium, a collaboration of leading aerospace, energy, and defense enterprises that has developed a best practice approach to materials data management, including guiding the development of the GRANTA MI software system.
He began by outlining the work of the Mechanics and Life Prediction Branch, and the intense materials information requirements that result. This work includes a broad range of material characterization and experimental mechanics. Results then need to be analyzed to generate mathematical characterizations of material behavior. These are used as input to in-house modeling tools and to commercial software such as Abaqus—for example, for FEA analysis of a component.
A key point in Dr Arnold's talk was a paradigm shift in modeling approaches that is dramatically increasing materials data management challenges. Historically, the need was to gather, store, and use the 'point-wise' materials property data required for conventional design and analysis methods. Now, the focus is on capturing the complete fundamental response spectrum of a material—for example, not simply a property such as Young's Modulus, but the full stress-strain curve for a material at a range of different temperatures. Not only is there more (and more complex) data to store—but the level of effort required to generate this data (particularly when testing at high temperatures) means that it is a much more valuable commodity. However, if such complete data can be captured, not only does it enable more sophisticated constitutive modeling, it also creates an information resource that can support new or refined models in the future, without the need to repeat expensive tests because the correct data was not captured first time around.
NASA Glenn has focused on the likely use—and the likely users—of materials data, now and in the future, in developing an informatics infrastructure for this data. This focus drove the organization to take a lead in the foundation, in 2002, of the Material Data Management Consortium (MDMC), along with ASM International and Granta. The MDMC's philosophy, now supported by 17 leading engineering enterprises, is to analyze the full 'lifecycle' of materials data—its capture, analysis, deployment, and maintenance—and to provide tools within the GRANTA MI system that support each stage.
Sample NASA homepage within the GRANTA MI system. Materials data can be navigated in the left panel. The right panel provides quick access to specific data and tools in language tailored to the user community.
NASA Glenn has implemented GRANTA MI to capture, analyze, and share both test data and external reference data. The database contains thousands of records and continues to grow. GRANTA MI's flexible web browser user interface has enabled development of a variety of 'homepages' (above) through which different users access and apply the data and tools that are relevant to them. These homepages use pictures and text tailored to the specific user community, and also provide links enabling access and export to third-party software and NASA in-house systems. NASA has also been instrumental in specifying underlying database features such as Access Control, which is now in use to ensure that users only see data appropriate to their role and authorization status. Current effort is focused on developing a model table within the system, and on meeting the particular needs of composite data.
NASA Glenn provides an excellent example of the need for effective materials data management—and how that need can be met through a comprehensive, integrated approach. It also shows the benefits of the MDMC collaboration in allowing one member to influence the development of the GRANTA MI system and benefit from that development, while sharing the costs with other partners.
Paula Freyer is a Fellow Engineer with Westinghouse Electric Company in Pittsburgh. She has been leading a project to optimize the management of the organization's materials data.
She began her talk with an overview of the history of Westinghouse Electric and the nuclear industry. This provided important context for materials information management requirements in the sector. The nuclear industry underwent a 30 year decline from its heyday in the early 1970s. It is now in the midst of a renaissance, as countries adopt their energy policies in the light of climate change, volatile oil prices, and the need for energy independence. Demand for new nuclear power stations is rising, and the industry is seeking to extend the life of existing facilities—many of which are coming to the end of their original 40-year lifespan.
This is good news for the industry, but also poses a challenge. It needs to scale up quickly. But, with low recruitment levels for the last several decades, it is also set to lose much of its workforce and expertise. 35% of the engineers in the nuclear industry are eligible for retirement in the next 5 years. These facts are making organizations re-evaluate how they capture, store, and access knowledge in their design and maintenance processes—including materials knowledge. In fact, a major corporate asset at Westinghouse is its long history of materials research, and the resulting rich materials data resource, comprising tens of thousands of laboratory reports, test results, and documents. The challenge is that, in common with most engineering enterprises, this resource is spread across filing cabinets, discs, computer files, proprietary databases, and handbooks. This makes it hard to find and share materials information across the materials team and the hundreds of engineers who may require data, for example as input to their analysis codes. And knowledge about what data is available, and where to access it, will dissipate as materials experts retire or move on.
The risk of losing data or duplicating tests that have already been done represents a major cost in the nuclear industry. Samples with sufficient irradiation exposure to represent, for example, a component in a 40-year old nuclear reactor, are extremely rare. Tests must evaluate a wide range of parameters, and these tests are logistically complex—requiring highly specialized hot cell laboratory testing facilities and highly skilled radiologically trained personnel.
These are the reasons behind a new project to give Westinghouse Electric a '21st century materials data management system'. The goal is to create a server-based system, accessible only on the company's intranet, which becomes the single, central place to store materials data. The system will enable anyone with a web browser to intuitively find key materials data and to compare materials, plot curves, and prepare reports. Data will be exportable to Excel and to key CAD and FEA codes. Key reference information—for example, the ASME Boiler and Pressure Vessel Code data, will be included. Security and traceability features are vital.
Westinghouse has defined a four-stage process to implement such a system corporation-wide. The first stage is a data management pilot project, which is currently being implemented using the GRANTA MI system—selected because it meets all of the requirements outlined above.
Westinghouse Electric Company, and the nuclear industry in general, offers a particularly compelling example of the need for effective materials information management to help preserve materials knowledge and fully leverage investment in expensive materials testing.
Acknowledgements and further information
Granta would like to thank Steven Arnold and Paula Freyer for speaking at the seminar and for permission to distribute presentation materials to meeting attendees.