|
Divisions
PED
MMD
ISD
MSID
MEL
Postdoc
page |
Gaithersburg, Maryland
U.S. citizenship is required
New Methods for the Development of Manufacturing Simulation Models 50.82.61.B5202
Chuck Mclean. (301) 975-3511
charles.mclean@nist.gov
Simulation technology holds tremendous promise for reducing costs, improving quality, and shortening the time-to-market for manufactured goods. Unfortunately, this technology still remains largely underutilized today. The development of new simulation methods and supporting interface standards has been repeatedly identified by industry studies as a top research priority that promises high payback.
The simulation model development process is often time-consuming and costly. How could the modeling process be improved? The solution would appear to be to simplify the model development process through modularization and the creation of re-usable simulation-model code and data. Neutral, vendor-independent data formats for storing simulation models could greatly improve the accessibility of simulation technology by industry by enabling the sharing and re-use of models. Neutral models and interface standards could improve the accessibility of this technology to helping to reduce the expenses associated with acquisition and deployment, minimize model development time and costs, and provide new types of simulation functionality that are not available today. Individual companies, simulation vendors, equipment and resource manufacturers, consultants, and service providers could use these format to develop sharable models. Neutral formats would help enlarge the market for simulation models and make their development a viable business enterprise.
Our program is focusing on the establishment of modeling techniques, standard interfaces and test methods to support the rapid construction of distributed manufacturing simulation systems. We are investigating the development of distributed manufacturing simulations based upon extensions to the Defense Modeling and Simulation Office's (DMSO) High Level Architecture (HLA). We are interested in exploring solutions that would simplify and accelerate the development process for virtual reality and discrete even-based simulation models. Solutions must be commercially-viable so that they may be adopted by simulation software vendors in future product offerings.
Our research interests include: 1) development of a taxonomy simulation study types to better understand the different kinds of models that are needed by industry, 2) identification of study templates for addressing different classes of simulation problems, 3) development of new methods for defining building block modules of manufacturing system components to e used in he templates, 4) creation of new techniques for automatically generating specific simulation solutions from templates and building block modules, 5) development of neutral interfaces for interconnecting simulation component modules and distributed simulations together, and 6) development of libraries of simulation reference data sets for industrial users.
Return to top
Ontology-based Manufacturing Standards 50.82.61.B4787
Jones, Albert (301) 975-3554
jonesa@cme.nist.gov
As the use of information technology in manufacturing has matured, the necessity for software applications to inter-operate has become crucial to the conduct of business and operations in organizations. To be competitive and maintain good economic performance, manufacturing organizations need to employ increasingly effective and efficient systems. Such systems should result in the seamless integration of manufacturing applications and exchange of manufacturing processes between applications. Organizations should also be able to conserve and retrieve on demand the knowledge contained in their business and operational processes, regardless of the applications used to produce and handle these processes.
Existing approaches to interoperability lack an adequate specification of the semantics of the terminology of the process models, which leads to inconsistent interpretations and uses of knowledge. Within enterprise operations, obstacles to interoperability arise from the fact that the legacy systems that support the functions in many enterprises, were created independently, and do not share the same semantics for the terminology of their enterprise models.
The proposed research focuses on the development of ontologies for existing and emerging manufacturing standards to facilitate the integration of manufacturing software applications such as manufacturing simulation, production scheduling, manufacturing process planning, workflow, business process re-engineering, product realization process modeling, and project management. Specifically, research will include the design of ontologies using first-order logic, the formal characterization of their logical and mathematical properties, and the application of this work to support systems integration. Work is in progress on the characterization of the ontologies for the Process Specification Language, which is nearing completion as an international standard. This work forms the foundation for future activity in developing the technologies to allow systems to become self-integrating.
Return to top
Software Agents to Interoperability Among Product Design, Manufacturing Engineering ,and Supply Chain Management Software Systems 50.82.61.B6234
Feng, Shaw C. (301) 975-3551
sfeng@cme.nist.gov
In general, current design, manufacturability analysis, shop floor management, and manufacturing resource management systems aid engineers in developing detailed design, process plans, shop floor control programs, job schedules, and resource acquisition plan. In recent years, there is a rapid increase of the demand for new software to help engineers specify product concepts, analyze manufacturability, and acquire adequate resources in the early design stages increases. In responding to this demand, agent-based framework for design, manufacturing, and supply chain management software interoperability is necessary. The goal of this research is to develop a flexible and ready-to-connect software architecture for integrating conceptual design and early manufacturing analysis (conceptual process planning). Based on functional requirements, conceptual design defines a basic product structure without specifying detailed information on components and parts. To evaluate the conceptual design, conceptual process planning assesses manufacturability and estimates product cost. Our goals are to (1) develop formal specifications for a multi-agent framework which will enable software reuse and information sharing, including the application of ontological principles and formal languages (e.g. first order based languages such as KIF) to manufacturing specifications and standards and mechanisms that can hide the complexities of these formal methods from the end-user without sacrificing the semantic power, (2) explore novel software integration architectures which will allow users to rapidly integrate components into computer-aided conceptual design, process planning, cost estimating systems and manufacturing execution systems, (3) develop prototype information models using a selected modeling tool to support agent communication, (4) develop software components to test the specifications and the agent framework, and (4) link selected computer-aided design (CAD), computer-aided process planning (CAPP), cost estimating, and supply chain management systems into this framework. (5) link selected computer-aided design, computer-aided process planning, and cost estimating systems into this framework.
Intelligent and Distributed Engineering Product Development 50.82.61.B3913
Sriram, Ram (301) 975-3507
sriram@cme.nist.gov
Design of complex engineering systems is increasingly becoming a collaborative task among designers or design teams that are physically, geographically and temporally distributed. The complexity of modern products means that a single designer can no longer manage the complete effort. Designers are no longer merely exchanging geometric data, but more general knowledge about design and design process, including specifications, design rules, constraints, rationale, and more. As design becomes increasingly knowledge-intensive and collaborative, the need for intelligent CAD tools to support the representation and use of knowledge among distributed designers becomes more critical.
This program focuses on the development of next-generation computational tools for intelligent and distributed engineering product development. Areas of interest include distributed design tools (Web/Internet/intranet-based tools to support knowledge exchange, sharing, communication), knowledge-based design tools (capture, use and reuse of design knowledge) and design integration tools (flow of knowledge across disciplines and throughout the product lifecycle).
Product Engineering 50.82.61.B3914
Sriram, Ram (301) 975-3507
sriram@cme.nist.gov
The scope of this program spans a full cross-section of activities involved with product engineering and product development, from the early conceptual design stages to final production. The two main thrusts within this program are (1) the development of knowledge representations and information protocols for interoperability, and (2) the development of enabling technologies, to support product engineering in industry.
The information protocols thrust is developing representations and information models for product engineering. Work in this area covers various types of product engineering knowledge, including extended design models for the representation of design artifact knowledge, geometric representations (parametric information, constraints, and features), assembly models and assembly level tolerancing, layered manufacturing data, engineering knowledge bases and others. A main objective of this thrust is to support interoperability at various stages in the product development process, from design to manufacturing and beyond.
The thrust in enabling technologies for product engineering ranges from research to applied development, and spans a broad cross-section of technology areas. These activities cover different phases of the product development process, design artifact modeling to assembly design to process planning, etc., and also cut across different kinds of engineering activities such as synthesis, analysis, optimization, and so on. This thrust includes the development of new prototype tools and interfaces, as well as the use of virtual environments, physics-based design, and commercial engineering software tools.
Knowledge Discovery in Large-Scale Knowledge/Data Bases 50.82.61.B5204
Sriram, Ram (301) 975-3507
sriram@cme.nist.gov
Large-scale knowledge/data bases in many applications are being increasingly used to support capture of information from distributed sources, and sharing and exchange of knowledge among geographically and temporarily distributed users. Providing distributed database access is well within today's state of the art. A key technological barrier to using large-scale knowledge/data bases is not the issue of distributed access , but how to get the "right" information out, both when the desired information is explicitly represented somewhere in the knowledge/data base (knowledge/data retrieval) as well as when it is implicitly but not explicitly represented (knowledge discovery). This issue cuts across many domains, including product development and manufacturing in engineering industry, medical and health information in the healthcare industry, and intelligence gathering for national security. The complexity of knowledge and related interactions in these domains is pressing the limits of what current knowledge-based infrastructures can provide. The goal of this work is to explore the use of computational techniques, such as information modeling, formal ontologies, data mining algorithms, to aid in automating knowledge discovery in large-scale knowledge/data bases in engineering and related fields.
Return to top
Novel Computational Techniques Supporting Engineering Synthesis and Analysis 50.82.61.B3915
Sriram, Ram (301) 975-3507
sriram@cme.nist.gov
The goal of this program is to explore the applicability of various revolutionary computational techniques to complex problems such as engineering design. Examples include approaches such as DNA computing, quantum computing, neural networks, and immune system-based optimization. DNA computers are inherently parallel computational systems and can potentially solve very complex problems in time periods that are orders of magnitude less than it would take the fastest digital computer available now. This program will explore appropriate DNA encodings for design, identifying the appropriate enzymes, and experimenting with design generation. This program is also interested in identifying the various design principles of biological systems and how these principles can be applied to novel product engineering techniques, such as self-replicating and fault-tolerant devices, design synthesis and optimization based on neural networks and the immune system. In the area of quantum computing, this program will probe the applicability and the use of quantum computing techniques for large-scale engineering analysis. The ultimate goal is to develop standard representations for encoding design and manufacturing knowledge in these emerging fields.
Return to top
Semantic and Process Representation 50.82.61.B1602
Ray, Steven Robert (301) 975-3524
steven.ray@nist.gov
Research focuses on automated assistance in process development - for integrating manufacturing software applications. The primary stand used for process representation is , (ISO 18629), also known as the Process Specification Language (PSL), based on first-order logic and situational calculus. This ontology establishes formal, definitions that are combined to support a lexicon, grammar, and structural models in order to define the process language. Although PSL supports the exchange of process information in a manufacturing context, it also has broad applicability for systems integration, in general, as well as synthesizing and managing of organizational processes. In addition, it provides a rich means of expressing all process information to downstream manufacturing systems such as scheduling, production management, and control. Future directions in which this work could include automated consistency checking of processes against requirements and other other processes, the engineering of new formalisms for defining interoperability specifications, methods for testing conformance, and automated methods for describing and discovering services available on the Worldwide Web, plus application of the work performed to date.
Simulation-Based Learning Using Video Game Technology Simulation Models 50.82.61.B5898
Chuck Mclean. (301) 975-3511
charles.mclean@nist.gov
Video game technology has primarily been used for entertainment rather than educational purposes in past years. More recently, industry, government, and academia has recognized that video game technology promises to provide a more engaging learning experience than traditional classroom or computer-aided instruction methods. Rapid growth in the use of video games for education and training is expected in the near future.
Video game engines provide integrated environments for creating virtual worlds. Engines typically have capabilities for creating three-dimensional graphics, sound, animated characters, intelligent character behaviors, and models of various physical phenomena. They may also provide support for the creation of user level game modifications (mods), Web-based software distribution, and distributed multi-player game participation over the Internet. But, new functionality will be needed to make this software suitable for training applications. Key game system elements will include: graphics and sound, game engine management, character animation, physics, artificial intelligence, game editor and development tools, application data import and export, multi-player operations and server support, software distribution and security mechanisms, and learning system support.
Perhaps hundreds of thousands of game-based learning applications will be created in the years ahead. Pervasive use of this technology for training purposes will require that many new developers gain access to tools that are used to create learning applications. If game engine software is to be used effectively, the traditional business models of these game engine vendors may need to change. Currently software licenses for game development systems and game distribution are very expensive. Access to game development software is often restricted to a small group of companies selected by game engine or console manufacturers. Standard interfaces and open source software are needed to make this technology more available for general use.
There are few standards defined within the game industry that support training needs. A common development strategy, a standard architecture, and neutral data interfaces are needed. Repositories of game-compatible learning objects will need to be established using Extensible Markup Language (XML) and extensions to the Sharable Content Object Reference Model (SCORM). Our program is working with other government agencies, academia, and software industry to develop standards and open source solutions for simulation-based training using video game technology.
Our research interests focus on the development of solutions that enable open-architecture, cross-platform, open-source video game technology for educational applications. Research opportunities include: (1) specification of open architectures and neutral interfaces for game engines and component modules, (2) development of prototype open source game engines, (3) development of state-of-the-art open source implementations of artificial intelligence, physics, graphics visualization, and character animation modules, (4) creation of novel example learning applications using game technology, (5) creation of taxonomies, i.e., classification schemes, that can be used to organize repositories of learning objects, (6) specification of object structures using XML and SCORM standards, and (7) creation of reference data sets and testing mechanisms that can be used to verify software compliance with evolving standards.
Return to top
Computational Metrology 50.82.61.B5901
Craig Shakarji (301) 975-3545
craig.shakarji@nist.gov
Feng, Shaw C. (301) 975-3551
sfeng@cme.nist.gov
This research focuses on applying computational data analysis algorithms to coordinate metrology in order to aid in the dimensional tolerancing inspection process. Such computational metrology is important for establishing datums and assessing deviations. Current research involves extending previous work that involved least-squares and Chebyshev fitting algorithms for standard geometries. The new areas include 1) extending the fitting algorithms to include handling of a variety of complex surfaces, 2) quantifying uncertainties associated with data analysis software dependent on (among other things) sampling strategy, 3) investigating new fit objectives and algorithms, 4) applying and studying the effects of data filtering on algorithms and uncertainties, and 5) using simulation methods to evaluate uncertainties of task-specific measurements.
Reuse and Integration of Ontologies 50.82.61.B5902
Jones, Albert (301) 975-3554
jonesa@cme.nist.gov
The homeland security and intelligence communities are investing heavily in knowledge management technology not only to help capture what they know, but also to help them "know what they know," so that they will "connect the dots" when older intelligence suddenly becomes more relevant.
Ontologies, which define the terms used to describe and represent areas of knowledge, are critical to this kind of knowledge management. Ontologies also have applications in manufacturing, healthcare, and e-commerce.
The goal of this research is to break down the known barriers to ontology reuse and integration. The short-term objective is to apply formal methods to translate the soft, philosophical problems into hard, scientific ones, and then distribute the solutions to customers directly and via open standards.
Standards-Based Methods for Systems Integration 50.82.61.B5903
Morris, Katherine C.
(301) 975-8286
kcm@nist.gov
The increased automation of manufacturing processes and use of electronic commerce results in a growing need for standards to help in the integration of systems across organizations in a business supply chair. Standards based on traditional methods of integration have emerged. The use of these standards in the non-traditional environments of electronic commerce is imposing new requirements on the standards and standards development processes. Methods are needed to determine how to specify, augment, and test the use of these standards in today's environments. Traditional approaches to manufacturing systems integration are based on data exchange using formal definitions for information requirements in the form of standards. Integration requirements in the environment of electronic commerce go beyond the traditional role of standards for data exchange. The new environment differs in the following significant ways:
- new specifications are emerging much more rapidly than previously. Additionally, data integrity requirements are often independent of the specification of information structures
- the distinction between data models and implementation models is often missing in the new specifications while based on a data exchange paradigm, must operate in a transaction oriented environment
- finally, the proliferation of specifications without an accompanying development methodology makes testing difficult.
This project will research methods for rapidly prototyping and testing standards for system integration in the environment of electronic commerce involving multiple business partners. A key challenge for systems integrators is to develop robust standards quickly. A methodology for standards development including early testing of the standards will be key to producing standards of a high enough quality to meet demands of system integration across organizations.
melwebmaster@nist.gov
Last updated:
May. 06, 2008
Return to top |
