NATIONAL SCIENCE FOUNDATION
TOKYO REGIONAL OFFICE
The National Science Foundation's (NSF) Tokyo Regional Office periodically receives and disseminates reports on research developments in Japan that are related to the Foundation's mission. It also provide occasional reports on developments in other East Asian Countries.
These reports present information for the use of NSF program officers and policy makers; they are not statements of NSF policy..
Special Scientific Report #02-03 (March 15, 2002)
SIMULATION UNCERTAINTY IN MULT
IDISCIPLINARY DESIGN OPTIMIZATIONThe following report was prepared by John Renaud of the Department of Aerospace and Mechanical Engineering at the University of Notre Dame, who was in Japan from May 11-23, 2001, supported by a short-term invitational fellowship from the Japan Society for the Promotion of Science (JSPS). Professor Matsataka Yoshimura of the Department of Precision Engineering at Kyoto University served as his host. Prof. Renaud may be reached at Renaud.2@nd.edu.
Summary
The primary goal of this JSPS Short-Term Invitation Fellowship project was to explore and develop cooperative research opportunities in simulation-based multidisciplinary design optimization with both Prof. Masataka Yoshimura of Kyoto University and Professor Hiroshi Yamakawa of Waseda University in Tokyo. The goal of developing a coopeartive research effort is to enhance the research being conducted as part of NSF grant DMI-98-12857: Simulation Uncertainty in Multidisciplinary Design. The goals were successfully achieved during the Fellowship period.
Acknowledgements and General Comments
First, I would like to express my sincere gratitude to the Japan Society for Promotion of Science for making it possible to undertake this visit to Japan. The financial support provided by the JSPS is gratefully acknowledged.
My deepest gratitude and thanks go to Professor Masataka Yoshimura of Kyoto University who was my primary host for this visit. His very kind, friendly and professional mannerisms made this visit very pleasant and memorable. His meticulous attention to the details of my itinerary and professional planning of the visit to various places in Japan made my trip extremely successful and enlightening. His kindness to me and my wife was exceptional. The hospitality of Prof. Yoshimura and his students was impeccable, and I am very thankful to them for this.
I would like to take this opportunity to thank Prof. Hiroshi Yamakawa of Waseda University in Tokyo for hosting my visit of his research laboratory. I would like to express my gratitude to Prof. Yamakawa for his gracious hospitality, for free exchange of ideas and scientific information, and for the precious time he spent with us.
Academic and Research Activities
Prof. Masataka Yoshimura of the
Department of Precision Engineering at Kyoto University hosted my two week stay in Japan under the JSPS Short-Term Invitation Fellowship. I feel very fortunate to have been able to work with this preeminent expert in the field of concurrent optimization for product design and manufacturing. During my visit, Prof. Yoshimura and I had an opportunity to discuss and review his recent work on the concurrent design of structural systems using operators acting upon modules. In Prof. Yoshimura’s method the complex systems being designed with the use of simulation analysis tools are decomposed into modules, and then the modules’ specifications are changed via operators, which act to expand the design space. By using simple modules to express the complex system, several stages of design and the attendant manufacturing variables can be treated within the modules concurently and in parallel, therby reducing the design cycle time.I also reviewed Prof. Yoshimura’s research involving evolutional operators as part of a living systems approach for multiobjective optimization of structural systems under multiple environmental conditions. This evolutional approach provides for global optimization capabilities. Our discussions and exchange have allowed us to initiate a collaborative research effort that will focus on adapting Prof. Yoshimura’s approach of using operators acting on modules for concurrent design within the decision based collaborative optimization framework being developed at Notre Dame. While at Kyoto University, I had an opportunity to present a seminar detailing recent advances in simulation-based design optimization and overview a decision based collaborative optimization being developed at Notre Dame as part of NSF sponsored research.
The JSPS Short-Term Invitation Fellowship provided the needed catalyst and support for the initiation of collaborative research in design optimization. Subsequent to my visit to Japan the University of Notre Dame was proud to host Prof. Yoshimura for a two day visit to Notre Dame in August of 2002. Prof. Yoshimura’s presented a research seminar during his visit. Dr. Yoshimura and I are currently in the process of developing a proposal to the National Science Foundation’s program for International Opportunities for Scientists and Engineers. This program will provide resources to insure that our collaboration and research exchange continues.
Figure 1 Seminar at Kyoto University
Prof. Hiroshi Yamakawa of the Department of Mechanical Engineering in the School of Science and Engineering of Waseda University in Tokyo hosted my visit to his campus during the first part of my stay in Japan. Prof. Yamakawa is a renowned expert working in the field of structural optimization. Dr. Yamakawa’s recent efforts have focussed on introducing the tools of soft- computing (i.e., genetic algorithms, fuzzy numbers and multi-agent architectures) in structural systems design and optimization. Simulation-based design optimization has its roots in structural design, where since the 1970’s researchers have been invetigating and developing methods for the concurrent optimization of structural and control systems. One of Prof. Yamakawa’s recent research efforst has focussed on simulation-based design under uncertainty. His approach of using fuzzy numbers as intermediate variables in robust design optimization holds great promise for managing uncertainty in simulation-based design optimization processes. I had an opportunity to meet with Prof. Yamakawa and his students and was able to discuss and exchange research ideas in the area of structural optimization.
We were able to develop collaborative research plans that will focus on adapting Prof. Yamakawa’s approach of using using fuzzy numbers as intermediate variables for managing uncertainty in simulation-based design. We have proposed to implement Dr. Yamakawa’s robust design optimization strategy within a decision based collaborative optimization framework being developed at Notre Dame. I was given the opportunity to present a seminar at Waseda Univerity detailing recent advances in simulation-based design optimization. I also had the opportunity to overview recent advances in trust region model management for approximate optimization that have been developed at Notre Dame.
The JSPS Short-Term Invitational Fellowship has allowed me to initiate independent reserach collaborations with both Prof. Yoshimura and Prof. Yamakawa directed toward developing advanced simulation-based design optimization strategies. Developing these types of international collaborations significantly benefitted from the substantial face-to-face interaction, which the JSPS Short-Term Invitational Fellowship provided. Prof. Yamakawa was scheduled to visit the University of Notre Dame on September 13 and 14 of this year, but the tragic events of September 11, prevented his travel to Notre Dame. I am planning on hosting the visit of Dr. Yamakawa to Notre Dame in September of 2003.
Figure 2 Seminar at Kyoto University
Closure
My visit to Japan was rewarding both intellectually and personally. It was a great opportunity for me to work with Prof. Yoshimura and to meet with other outstanding Japanese educators, researchers, engineers and research administrators. I was able to strengthen existing research ties with Prof. Yoshimura and Prof. Yamakawa. Additionally I was able to meet with graduate student researchers and exchange ideas with Japan’s next generation of researchers. The research being conducted by my Japanes colleagues is of the highest caliber and of immediate practical relevance. I am deeply impressed by the dedicated professionalism of my Japanese colleagues and am especially obliged to each individual I met on this for sharing their valuable time, scientific data, and information freely. I have certainly learned a tremendous amount in technical and social terms from my Japanese colleagues and I am looking forward to developing a more fruitful and intellectual relationship with the many Japanese colleagues I met on this trip.