TY - JOUR
T1 - A Study on the Optimum Design Applying Multi-Agents Architecture
AU - Miyashita, Tomoyuki
AU - Yamakawa, Hiroshi
PY - 1999
Y1 - 1999
N2 - In recent years, multi-agent architectures which have been studied in the field of artificial intelligence might attract a great deal of attention to the ability of the cooperative properties for engineering. In this study, we proposed a new approach to find the improved design in the layout design and shape design problems of framed structures by applying the multi-agent architectures. We considered each agent as an element in Finite Element Analysis and proposed modification rules for the layout shape of the structure. Following this approach, each agent proposes modification plans based on his or her evaluate function. The proposed plans are picked up or rejected by communication among agents and the remained plans may improve a design created from an initially simple structure. We developed the simulation system running on the UNIX and considerations have been given to the simple truss structural design and some advantages of the proposed have been shown and confirmed.
AB - In recent years, multi-agent architectures which have been studied in the field of artificial intelligence might attract a great deal of attention to the ability of the cooperative properties for engineering. In this study, we proposed a new approach to find the improved design in the layout design and shape design problems of framed structures by applying the multi-agent architectures. We considered each agent as an element in Finite Element Analysis and proposed modification rules for the layout shape of the structure. Following this approach, each agent proposes modification plans based on his or her evaluate function. The proposed plans are picked up or rejected by communication among agents and the remained plans may improve a design created from an initially simple structure. We developed the simulation system running on the UNIX and considerations have been given to the simple truss structural design and some advantages of the proposed have been shown and confirmed.
KW - Artificial Intelligence
KW - Design Engineering
KW - Finite Element Method
KW - Layout Planning
KW - Multi-Agents System
KW - Optimum Design
UR - http://www.scopus.com/inward/record.url?scp=85024454903&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85024454903&partnerID=8YFLogxK
U2 - 10.1299/kikaic.65.4589
DO - 10.1299/kikaic.65.4589
M3 - Article
AN - SCOPUS:85024454903
SN - 0387-5024
VL - 65
SP - 4589
EP - 4596
JO - Nihon Kikai Gakkai Ronbunshu, C Hen/Transactions of the Japan Society of Mechanical Engineers, Part C
JF - Nihon Kikai Gakkai Ronbunshu, C Hen/Transactions of the Japan Society of Mechanical Engineers, Part C
IS - 640
ER -