Cover Image

PAPERBACK
$45.00



View/Hide Left Panel

Cognitive Manufacturing

ELIZABETH HOEGEMAN
Cummins Inc.

J. RHETT MAYOR
Georgia Institute of Technology

Consider the degree of computer-enabled technology penetration in everyday life, with self-parking cars and smartphones that present locale-specific information through augmented reality displays. Given this increased use of computer-enabled decision making, is it plausible to consider the near-term realization of science fiction notions of autonomous production systems with “machines making machines"?

Manufacturing as an industry has been pervasively impacted by the rapid adoption of information technology (IT). Modern manufacturing systems execute highly sophisticated IT-enabled operations and control infrastructure that track production metrics, quality metrics, and component status in real time. The state of practice in the field exhibits the characteristics associated with “smart” systems, as distributed processors feature embedded low-level logic systems that trigger alerts in response to single value break points, or level-based go/no-go indicators, and report these alerts to supervisory human operators through IT-enabled communication channels. Decisions about how to respond to such alerts are made by human operators based on their knowledge of the process and reasoned judgment. That is, the cognitive process is performed by human intelligence and remains the primary function of the operator.

Cognitive manufacturing is an evolutionary step in computer-enabled production system control that pushes beyond smart technologies, in which the intelligence and reasoning are retained by the human user, and endows the manufacturing system with capabilities of perception and judgment to enable the autonomous operation of the system based on embedded cognitive reasoning, reliant only on high-level supervisory control.



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 27
Cognitive Manufacturing Elizabeth Hoegeman Cummins Inc. J. Rhett Mayor Georgia Institute of Technology Consider the degree of computer-enabled technology penetration in everyday life, with self-parking cars and smartphones that present locale-specific informa- tion through augmented reality displays. Given this increased use of computer- enabled decision making, is it plausible to consider the near-term realization of science fiction notions of autonomous production systems with “machines making machines”? Manufacturing as an industry has been pervasively impacted by the rapid adoption of information technology (IT). Modern manufacturing systems execute highly sophisticated IT-enabled operations and control infrastructure that track production metrics, quality metrics, and component status in real time. The state of practice in the field exhibits the characteristics associated with “smart” systems, as distributed processors feature embedded low-level logic systems that trigger alerts in response to single value break points, or level-based go/no-go indicators, and report these alerts to supervisory human operators through IT-enabled com- munication channels. Decisions about how to respond to such alerts are made by human operators based on their knowledge of the process and reasoned judgment. That is, the cognitive process is performed by human intelligence and remains the primary function of the operator. Cognitive manufacturing is an evolutionary step in computer-enabled produc- tion system control that pushes beyond smart technologies, in which the intelli- gence and reasoning are retained by the human user, and endows the manufacturing system with capabilities of perception and judgment to enable the autonomous operation of the system based on embedded cognitive reasoning, reliant only on high-level supervisory control. 27

OCR for page 27
28 FRONTIERS OF ENGINEERING Cognitive manufacturing systems perceive changes in the production pro- cess and “know” how to respond to these dynamic fluctuations by adapting the production to stay within target ranges of production cost and rate, and, as are increasingly important, sustainability indices such as energy intensity and carbon footprint. The embedded cognitive capability can be accomplished through the development of cognitive reasoning engines, or distributed intelligence agents, deployed throughout the production system at three hierarchical levels: (1) the manufacturing process level, (2) the manufacturing system or factory level, and (3) the supply chain or production system logistical level. The speakers in this session introduce and explore cognitive manufacturing as an emerging frontier of engineering science that integrates domain knowl- edge from industrial and systems engineering, manufacturing process science, computer learning, information technology, adaptive control theory, biologically inspired system design, and environmentally cognizant design and sustainability. The presentations cover the deployment of computer-enabled cognitive reasoning at the three levels of production systems and the application of computer-enabled cognitive manufacturing systems to achieve sustainable production systems and mass sustainability. The first speaker, Dragan Djurdjanovic (University of Texas at Austin), discussed the development of distributed anomaly detection agents to recognize and address unprecedented faults (i.e., those that the system could not have been programmed to recognize). He illustrated the application of such an approach to dramatically reduce downtimes—and significant costs—associated with fault remediation. In the second presentation Chris Will (Apriso/Dassault Systèmes) traced the emergence of business process management technologies to accelerate process improvement, standardization, and excellence programs by translating process modeling results into an executable form that limits or eliminates the need to code or customize a core application. Next, Steve Ellet (Chainalytics) demon- strated that, by using sophisticated new modeling techniques and tools, companies are making better, faster, fact-based decisions that require fewer resources to make and move their products to market. The session’s final speaker, Steven Skerlos (University of Michigan), explained how cognitive systems can advance the state of the art in sustainable manufacturing, stressing the importance of integrating sustainability objectives into the product design and describing the application of various life cycle assessment methods to clarify the link between manufacturing systems and their environmental and social consequences.