Shu Yamada
Graduate School of Business Sciences, University of Tsukuba, Bunkyo,Tokyo, Japan
Ladda ner artikel
Ingår i: 10th QMOD Conference. Quality Management and Organiqatinal Development. Our Dreams of Excellence; 18-20 June; 2007 in Helsingborg; Sweden
Linköping Electronic Conference Proceedings 100:100, s.
Publicerad: 2008-02-15
ISBN:
ISSN: 1650-3686 (tryckt), 1650-3740 (online)
Applying design of experiments techniques is a key issue in digital engineering in terms of quality management. Because computer simulation plays an important role in modern technology development and it is used for validating developing technology; optimizing technology to adapt it for use in the real world and so forth. Another reason for the popularity of computer simulation is the advantages it offers in terms of speed and cost. Therefore; computer simulation is an alternative tool that is increasingly being used in the place of physical experiments. One example is the use an automobile crash simulation instead of carrying out physical experiments. Since physical experiments take time and are expensive to perform; using computer simulation is preferable. Many companies consequently mare promoting the use of computer simulation company-wide because of the advantages.
Physical experiments have been used to develop new technology for a long time. Once a hypothesis regarding a phenomenon has been formed; it is confirmed by physical experiments. Design of experiments is a methodology for conducting experiments effectively; it has been beneficial tool for developing new technology. It originated from the works by Fisher; R. A. at the beginning of 20th century (Fisher (1966)). Therefore; there is a need to apply design of experiments to computer simulation experiments in addition to physical experiments because of the recent popularity of computer simulations.
This paper discusses an application of design of experiments in computer simulation as a grammar for technology development. Specifically; the grammar described in this paper includes techniques for validating a tentative model involving response and factors; by screening the many factors and approximating the response function using the active factors.
Computer simulation; design of experiments; radial basis function; supersaturated design; uniform design
Inga referenser tillgängliga
