Simulation-based engineering and science

The work of this pioneering group at SBI is characterized by a close coupling of computation with experiment. It has a strong focus on the development of the computational infrastructure for systems biology and on the experimental techniques that will be needed to produce the data that will interact with the computation.

With regard to where will the systems biologists of the future come from, Dr. Kitano believes the best educational background may be an undergraduate degree in physics, masters in computer science, and PhD in molecular biology. A computer science background would work if the student is willing to learn the experimental side.

This group has now been funded for 10 years. It was initially funded on a noncompetitive, nonrenewable 5 year grant, which has been renewed once. This is different from current funding models in the United States. A source of stable, sizeable funding has undoubtedly played a role in the world-class success of this research.

Hamahashi S., S. Onami, and H. Kitano. 2005. Detection of nuclei in 4D Nomarski DIC microscope images of early Caenorhabditis elegans embryos using local image entropy and object tracking. BMC Bioinformatics (May) 24;6(1):125.

Hucka, M., A. Finney, H.M. Sauro, H. Bolouri, J.C. Doyle, H. Kitano, et al. 2003. The Systems Biology Markup Language (SBML): A medium for representation and exchange of biochemical network models. Bioinformatics 19:524-531.

Kitano, H. 2007. A robustness-based approach to systems-oriented drug design. Nature Reviews Drug Discovery 6:202–210 (March). Doi :10.1038/nrd2195.

———. 2000. Perspectives on systems biology. New Generation Computing Journal. 18:199–216. Ohmsha, Ltd., and Springer-Verlag, 99-216.

———. 2002a. Computational systems biology. Nature 420:206–210.

———. 2002b. Systems biology: A brief overview. Science 295:1662–1664.

———. 2004a. Biological robustness. Nature Review Genetics 5:826–837.

———. 2004b. Cancer as a robust system: Implications for anticancer therapy. Nature Reviews Cancer 4(3):227–235.

———. 2006. Computational cellular dynamics: A network-physics integral. Nature Reviews Molecular Cell Biology 7:163.

———. 2007. Towards a theory of biological robustness. Molecular Systems Biology 3:137. Doi:10.1038/msb4100179. Published online 18 September.

Kitano, H., .and K. Oda. 2006. Robustness trade-offs and host–microbial symbiosis in the immune system. Mol. Syst. Biol. 2(1): msb4100039-E1 ( Jan. 17),.

Kitano, H., et al. 2005. Using process diagrams for the graphical representation of biological networks. Nature Biotechnology 23(8):961–966.

Kyoda, K., and H. Kitano. 1999. Simulation of genetic interaction for Drosophila leg formation. Pacific Symposium on Biocomputingf99, Hawaii, 77–89.

Moriya H., Y. Shimizu-Yoshida, and H. Kitano. 2006. In vivo robustness analysis of cell division cycle genes in S. cerevisiae PLoS Genet. DOI: 10.1371/journal.pgen.0020111.

Morohashi, M., A. Winn, M. Borisuk, H. Bolouri, J. Doyle, and H. Kitano. 2002. Robustness as a measure of plausibility in models of biochemical networks. Journal of Theoretical Biology 216:19–30.

Oda, K., and H. Kitano. 2006. A comprehensive map of the toll-like receptor signaling network. Mol. Syst. Biol. msb4100057 (Apr. 18).

Oda, K., Y. Matsuoka, A. Funahashi, and H. Kitano. 2005. A comprehensive pathway map of epidermal growth factor receptor signaling. Molecular Systems Biology msb4100014., E1–17.

Site: Toyota Central R&D Labs, Inc.

41-1, Aza Yokomichi, Oaza Nagakute,

Nagakute-cho, Aichi-gun, Aichi-ken, 480-1192, Japan

http://www.tytlabs.co.jp/eindex.html
Date Visited: December 6, 2007
WTEC Attendees: P. Cummings (report author), A. Arsenlis, C. Cooper, L. Petzold, G. Karniadakis, D. Nelson
Hosts: Dr. Satoshi Yamazaki, Research Advisor and Manager
Technology and Systems Laboratory

E-mail: syamazaki@mosk.tytlabs.co.jp

Dr. Shiaki Hyodo, Research Manager
Computational Physics Laboratory, Materials Department

Email: e0668@mosk.tytlabs.co.jp

Dr. Hitoshi Washizu, Researcher
Tribology Laboratory, Mechanical Engineering Department

Email: washizu@mosk.tytlabs.co.jp

Dr. Tomoyuki Kinjo, Researcher
Computational Physics Laboratory, Materials Department

Email: e1308@mosk.tytlabs.co.jp

Dr. Seiji Kajita, Researcher
Tribology Laboratory, Mechanical Engineering Department
BACKGROUND

The Toyota Central R&D Labs (TCRDL), Inc., was established in 1960 for the purpose of carrying out basic research for Toyota Group companies and thus contributing to company growth and the advancement of science and technology. The major product of the Toyota Group companies is automobiles, at which Toyota is singularly successful. In the first quarter of 2007, Toyota became the leading producer of cars worldwide, surpassing General Motors for the first time. Other Toyota products, overshadowed by the automobile segment, include sewing machines (http://www.sewtoyota.com) and prefabricated houses (http://www.toyota.co.jp/en/more_than_cars/housing/index.html).

The research activities of the TCRDL focus on issues related to automobile production, such as resource and energy conservation, environmental preservation, enhancement of comfort and safety, and advanced information processing. There are five R&D divisions at TCRDL. In the areas of energy and the environment, TCRDL is researching new catalysts; analyzing combustion mechanisms in internal combustions engines; and developing next-generation batteries, solar cells, and fuel cells. In the safety/human engineering area, TCRDL is studying human body dynamics to determine the kind of injuries people are likely to suffer in auto collisions in order to help design safer automobiles. In the mechanical engineering area, TCRDL contributes to the development of vehicle technology that improves the performance, efficiency, and active safety of automobiles through research on control, kinetic and vibration analyses, and tribology. In the systems engineering and electronics area, TCRDL carries out research on high-speed digital mobile telecommunication technologies for man-machine interfaces and display devices; on power control devices for electric vehicles; and on information processing technologies for the improvement of design, manufacturing, and logistics processes. In the materials area, TCRDL is developing functional materials, including metallic composites, organic-inorganic molecular composites, and high-performance cell materials, as well as developing plastics and rubber recycling technologies. Funding for these activities is drawn from the Toyota Group corporations, and projects are selected from the technical needs of the Toyota Group and from proposals from the TCRDL staff.

R&D ACTIVITIES

Simulation-based engineering and science activities support all five divisions of the TCRDL. The role of SBES at the laboratory is in the development and application of simulation tools to aid in understanding phenomena that are observed by experimentalists at the facility. Some highly publicized simulation packages at TCRDL include the following:

19. 
    THUMS: Total HUman Model for Safety is a suite of detailed finite element models of the human body that can be subjected to impacts to assess injury thresholds. Models included detailed meshing of musculoskelature and internal organs.
    
20. 
    EVAS: Engine Vibration Analysis System is an engine design tool that is helping to develop lighter and quieter engines by providing highly accurate vibration predictions.
    
21. 
    Aerodynamic Noise Simulator COSMOS-V is used to calculate temporally fluctuating airflow and predict noise such as wind noise and wind-throb, as well as vibrations caused by pressure fluctuations.
    
22. 
    Mill-Plan: Software designed to determine the optimal number of machining processes and the tool form and cutting conditions for each process by closely examining the machining procedures and evaluating their overall efficiency.
    

It was evident that there is a growing interest in modeling human behavior and interaction with machines, and social network interactions to understand disruptions to Toyota’s production lines and/or to predict the behavior of markets towards disruptive technologies such as the hybrid automobile. The discussion in this area brought up the issue of uncertainty, and the view of uncertainty that was offered by the TCRDL staff was one in which uncertainty was an emergent property of complex interacting systems whose individual dynamics are well understood but whose synergistic behavior may become chaotic. As pointed out by the TCRDL researchers, the introduction of a hybrid automobile more than a decade ago (the Prius) was a very nonlinear, highly uncertain event that has paid off handsomely for Toyota.

Most of the simulation work is performed on stand-alone workstations. However, if more computing power is needed, TCRDL has an NEC SX-5 on site and has mechanisms in place to buy computer time at the Earth Simulator and at other locations. In terms of CPU time, TCRDL spends the most time in performing materials science simulations to uncover material structure-property relationships, followed in decreasing order by time spent on structural simulations of vibrations and acoustics, magneto-electric simulations in support of LIDAR (LIght Detection and Ranging) applications, and CFD/combustion simulations of internal combustion engines.

Impressive simulation results were presented in TCRDL’s effort to predict the viscosity of synthetic lubricants. The results showed that molecular dynamics simulations of the viscosity of known molecules with small simulation volumes could reproduce the trends observed in the laboratory, but the absolute value of the viscosity predicted by the simulation was orders of magnitude off because of the difference in the film thicknesses and shear rates of the experiment versus the simulation. Large-scale molecular dynamics simulations were conducted on Japan’s NAREGI national project supercomputer system; in those simulations, the film thickness and shear rate were comparable to the experimental measurements, and the two came into agreement.

Publication of the research conducted at TCRDL is promoted when possible, and the laboratory publishes a quarterly journal entitled R&D Review of Toyota CRDL that it makes freely available on the Internet. The TCRDL also sponsors small (~20 invited speakers) closed conferences for its scientists to interact with leading academics and learn about the latest research; for example, a workshop in the area of SBES entitled "Decision Making and Uncertainty in Nonlinear Complex Systems" was held in Copenhagen, Denmark, in November 2006.

The researchers at TCRDL identified several grand challenge problems for future SBES activities. The first is to perform first-principles molecular dynamics simulations of fuel cells to understand the mechanisms of hydrogen atom transport within those systems. The second, in the area of catalysis, is to understand the nature of carbon-oxygen bonding and resonances in the vicinity of catalyzing surfaces.

The depth and breadth of SBES activities at TCRDL are impressive, and the research atmosphere is comparable to, if not better than, other corporate R&D centers around the world. Although most of the lab’s simulation requirements are satisfied with OTS hardware, its researchers have access (as part of a national policy on industrial utilization of national high-performance computing facilities) to the largest national computing facilities in Japan, if they are required to obtain meaningful results. This interaction between government and corporate facilities in SBES appears to be beneficial to both parties in Japan. The cost of computer time at the national supercomputing facilities was not discussed.

Site: Tsinghua University Department of Engineering Mechanics

School of Aerospace

Beijing, 100084, P.R. China

http://hy.tsinghua.edu.cn/English/about/page9.asp
Date: December 3, 2007
WTEC Attendees: S. Kim (report author), S. Glotzer, M. Head-Gordon, J. Warren, P. Westmoreland, G. Hane
Hosts: Prof. Quanshui Zheng, Yangtze Chair, Professor, and Chairman, Department of Engineering Mechanics
Email: zhengqs@tsinghua.edu.cn

Prof. Zhuo Zhuang, Deputy Dean, School of Aerospace

Email: zhuangz@tsinghua.edu.cn

Prof. Gexue Ren

Prof. Xiong Zhang, Director of Institute of Dynamics and Control

Dr. Baohua Ji, Associate Professor

Email: bhji@tsinghua.edu.cn

Dr. Bin Liu, Associate Professor

Email: liubin@tsinghua.edu.cn

Background

The Department of Engineering Mechanics at Tsinghua University, believed to be the largest such department in the world, is ranked #1 in China in its field. Research projects in the department span the entire spectrum of activities in mechanics, including significant SBES activities involving global collaborations. Thanks to its reputation, the department has several high-profile SBES collaborations with multinational companies. The department has 89 faculty, 240 undergraduate students, 260 graduate students, and 19 postdoctoral fellows. The annual research budget is 20 million RMB¹⁰, of which 50% is from the National Natural Science Foundation of China (NSF-China; http://www.nsfc.gov.cn/Portal0/default106.htm) and 30% is from industry-sponsored projects. The professors have numerous awards and are active on editorial boards of the top international journals in Mechanics. From 1999 to 2006, approximately 100 PhD theses in all fields (natural and social sciences, engineering, and so on) were selected for China’s National Distinguished PhD thesis award; 16 of these were awarded in Mechanics, and Tsinghua’s Department of Engineering Mechanics received eight of these coveted awards.

A departmental overview and background context pertaining to SBES was presented by the Department Chair Prof. Quanshui Zheng. This was followed by highlights of research from the department’s SBES portfolio, presented by the lead investigators: multiscale simulations of materials (Z. Zhuang); multibody dynamics with applications (G. Ren); coarse-grained MD simulations with applications to modeling of protease inhibitors (B. Ji); and bead-spring and bead-rod polymeric chain dynamics (B. Liu).

Multiscale simulations, in particular bridging continuum and molecular scales via combination of finite element and molecular dynamics simulations was an overriding theme that emerged across all research examples presented in the overview. The department’s strong roots in solid mechanics were also evident in the presentations. The broader societal impact of the research applications included development of stronger filaments of carbon nanotubes from multiscale understanding of fracture mechanisms, multibody dynamics simulations of the operation of telescopes, coarse-grained models of protease inhibitors for HIV drug development, and acceleration of large-scale simulations of polymeric chains.

In November 2007, the department acquired a 2 TFlops 32-node cluster. Each node has 2 quad-core processors, which are interconnected by a 10 G infiniband.

After the presentations, there was an open, albeit brief, discussion of future SBES opportunities and challenges facing the department, which touched on the following points:

• 
  The research program features many international collaborations in SBES.
  
• 
  The department has a computational dynamics course using FEAP from Prof. Taylor of UC Berkeley. With the emergence of multicore architectures, mechanics students will have to take advantage of courses from other departments (e.g., computer science) to receive their training in parallel computing. Currently, there is relatively little emphasis in formal training of students on data management.
  
• 
  Despite the department’s significant research activities and excellent international reputation, relatively little funding currently comes in the form of the large-scale “973” and “863” national funds for high-visibility projects; this impacts the budget available for computational and data infrastructure. Furthermore, NSFC does not fund code development.
  
• 
  Only about 20% of undergraduates go directly to careers in industry; most go on to graduate studies. Graduate students go on to a balance between industry and academic careers. Training in SBES-related activities is highly valued by the students because of resulting career opportunities.
  
• 
  Industry interactions with industry R&D centers within China and also on a global scale with leading multinational companies, as noted earlier, are very strong,. The research sponsored by Medtronic, Inc., in the simulation of cardiac muscles was cited as an illustrative example.
  
• 
  The faculty values the accessibility of open source software in furthering their research goals in SBES-related fields. The WTEC team’s hosts also noted that this aspect is counter-balanced by researchers’ desire to maintain a competitive edge by developing their own codes.
  

Ji, B.H., and H.J. Gao. 2004. Mechanical properties of nanostructure of biological materials. Journal of the Mechanics and Physics of Solids 52:1963-1990.

Gao, H.J., B.H. Ji, I.L. Jäger. E. Arzt, and P. Fratzl. 2003. Materials become insensitive to flaws at nanoscale: Lessons from nature. Proc. Natl. Aca. Sci. USA 100:5597-5600.

Liu, B., Y. Huang, H, Jiang, S. Qu, and K.C. Hwang. 2004. The atomic-scale finite element method. Computer methods in applied mechanics and engineering 193(17-20):1849-1864.

Wang, L.F., Q.S. Zheng, J.Z. Liu, and Q. Jiang. 2005. Size dependence of the thin-shell model for carbon nanotubes. Phys. Rev. Lett. 95:105501.

Zheng, Q.S., B. Jiang, S.P. Liu, Y.X. Weng, L. Lu, Q.K. Xue, J. Zhu, Q. Jiang, S. Wang, and L.M. Peng. 2008. Self-retracting motion of graphite microflakes. Phys. Rev. Lett. 100:067205.

Site: University of Tokyo

Room 31A1, Eng. Bldg. #2, Hongo Campus, Bunkyo-ku

Tokyo 113-8656, Japan
Meeting Scheme: Workshop on R&D in Simulation-Based Engineering and Science
Sponsor: 21st Century Center of Excellence Program on Mechanical Systems Innovation

School of Engineering, The University of Tokyo

Prof. Nobuhide Kasagi, Leader of the 21COE Program,

Email: kasagi@thtlab.t.u-tokyo.ac.jp

Prof. Chisachi Kato, Institute of Industrial Science

Prof. Shigeo Maruyama, Dept. of Mechanical Engineering

Prof. Yoichiro Matsumoto, Dean, Dept. of Mechanical Engineering

Prof. Hiroshi Okuda, Research into Artifacts, Center for Engineering

Prof. Shinsuke Sakai, Dept. of Mechanical Engineering

Prof. Satoshi Watanabe, Dept. of Material Engineering

Prof. Shinobu Yoshimura, Dept. of Quantum Engineering
Visitors

Prof. Shunichi Koshimura, Disaster Control Research Center, Tohoku University

Prof. Michael A. Leschziner, Dept. of Aeronautics, Imperial College London

Prof. Masaru Zako, Dept. of Management of Industry and Technology, Osaka University

Professor Nobuhide Kasagi of the University of Tokyo’s Department of Mechanical Engineering generously organized an extremely informative workshop to introduce the visiting WTEC panelists to computational science in Japan and, in particular, to the 21st Century Center of Excellence (COE) Mechanical Systems Innovation program at the University of Tokyo that focuses on hyper-modeling/simulation. The WTEC team heard presentations from a number of the academic researchers working under this program.

The University of Tokyo (UT) is ranked number one in Japan; it was established in 1877 as the first national university in Japan. It offers courses in essentially all academic disciplines at both undergraduate and graduate levels and conducts research across the full spectrum of academic activity, including medicine. The University of Tokyo has a faculty of over 4,000 and a total enrollment of about 29,000, evenly divided between undergraduate and graduate students. There are currently more than 2,000 international students and over 2,500 foreign researchers at UT for both short and extended visits. The University of Tokyo is known for the excellence of its faculty and students; ever since its foundation many of its graduates have gone on to become leaders in government, business, and academia.

In June 2001, the Ministry of Education, Culture, Sports, Science & Technology (MEXT) issued a report on a new policy for structural reform of the national Universities in Japan and established a new program on Centers of Excellence (COE) for the period 2002–2007, the “21st Century COE Program.” The aims of this program are to form world-class research and education centers at universities in Japan, to cultivate creative human resources capable of improving research standards and leading the world, and to promote the building of universities of global appeal. Twenty-eight centers have been established at the University of Tokyo. Each is a five-year project. Therefore, projects begun in FY2002 have already ended; but some of them have developed into Global COE projects. The Global COE Program is a project that has inherited the fundamental concepts of the 21st Century COE Program and is given targeted support for forming internationally excellent education and research centers by MEXT. Applications for the Global COE Program will be accepted in each field from FY2007 to FY2011. In FY2007, six projects were adopted at the University of Tokyo.

The leader of the 21st Century COE Program at UT focusing on Mechanical Systems Innovation (MSI) is Professor Kasagi, who was the main organizer of the workshop. There are many researchers from different departments and institutes of UT participating in the 21st Century COE, including the Departments of Mechanical Engineering, Engineering Synthesis, Environmental & Ocean Engineering, Aeronautics & Astronautics, Quantum Engineering & Systems Science, Geosystems Engineering, Nuclear Engineering and Management, the Institute of Industrial Science, and the Center for Disease Biology and Integrative Medicine. There are three main research focus areas of the MSI program: (1) Energy, (2) Biomedicine, and (3) Hyper-modeling/simulation. The last one aims at advancing the art in multiscale/multiphysics modeling and the synergistic use of simulation and experiment. In addition, MSI has an educational aim in fostering creative young researchers. Professor Kasagi had invited some additional scientists from outside UT to enrich discussions at the workshop. In the following, we briefly describe the presentations given by the Japanese researchers.