TY  - CHAP
A1  - Parmee, Ian
A1  - Abraham, Johnson
T1  - User-centric Evolutionary Design Systems - the Visualisation of Emerging Multi-Objective Design Information
N2  - The paper describes further developments of the interactive evolutionary design concept relating to the emergence of mutually inclusive regions of high performance design solutions. These solutions are generated from cluster-oriented genetic algorithm (COGAs) output and relate to a number of objectives introduced during the preliminary design of military airframes. The data-mining of multi-objective COGA (moCOGA) output further defines these regions through the application of clustering algorithms, data reduction and variable attribute relevance analyses. A number of visual representations of the COGA output projected onto both variable and objective space are presented. The multi-objective output of the COGA is compared to output from a Strength Pareto Evolutionary Algorithm (SPEA-II) to illustrate the manner in which moCOGAs can generate good approximations to Pareto frontiers.
KW  - Konzipieren <Technik>
KW  - Bauwerk
KW  - Künstliche Intelligenz
Y1  - 2004
U6  - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:gbv:wim2-20111215-1094
ER  -