TY - GEN
T1 - A genetic algorithm based method for product family design optimization
AU - D'Souza, Brayan S.
AU - Simpson, Timothy W.
N1 - Funding Information:
This work was supported in part by the National Science Foundation under Grant No. DMI-0133923.
PY - 2002
Y1 - 2002
N2 - Increased commonality in a family of products can simplify manufacturing and reduce the associated costs and lead-times. There is a tradeoff, however, between commonality and individual product performance within a product family, and in this paper we introduce a genetic algorithm based method to help find an acceptable balance between commonality in the product family and desired performance of the individual products in the family. The method uses Design of Experiments to help screen unimportant factors and identify factors of interest to the product family and a multiobjective genetic algorithm, the non-dominated sorting genetic algorithm, to optimize the performance of the products in the resulting family. To demonstrate implementation of the proposed method, the design of a family of three General Aviation Aircraft is presented along with a product variety tradeoff study to determine the extent of the tradeoff between commonality and individual product performance within the aircraft family. The efficiency and effectiveness of the proposed method is illustrated by comparing the family of aircraft against individually optimized designs and designs obtained from an alternate gradient-based multiobjective optimization method.
AB - Increased commonality in a family of products can simplify manufacturing and reduce the associated costs and lead-times. There is a tradeoff, however, between commonality and individual product performance within a product family, and in this paper we introduce a genetic algorithm based method to help find an acceptable balance between commonality in the product family and desired performance of the individual products in the family. The method uses Design of Experiments to help screen unimportant factors and identify factors of interest to the product family and a multiobjective genetic algorithm, the non-dominated sorting genetic algorithm, to optimize the performance of the products in the resulting family. To demonstrate implementation of the proposed method, the design of a family of three General Aviation Aircraft is presented along with a product variety tradeoff study to determine the extent of the tradeoff between commonality and individual product performance within the aircraft family. The efficiency and effectiveness of the proposed method is illustrated by comparing the family of aircraft against individually optimized designs and designs obtained from an alternate gradient-based multiobjective optimization method.
UR - http://www.scopus.com/inward/record.url?scp=0036974612&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0036974612&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:0036974612
SN - 0791836215
SN - 9780791836217
T3 - ASME 2002 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, IDETC/CIE2002
SP - 681
EP - 690
BT - ASME 2002 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, IDETC/CIE2002
T2 - ASME 2002 International Design Engineering Technical Conferences and Computers and Information in Engineering Conference, IDETC/CIE2002
Y2 - 29 September 2002 through 2 October 2002
ER -