抄録
This paper proposes the pararel distributed interactive genetic algorithm (PDIGA) which is an interactive genetic algorithm (IGA) based on a parallel distribution model. IGA is an optimization method using a genetic algorithm (GA) where the evaluation of the individuals is performed by a user, not by a computer. By extending IGA to a parallel distribution model, the design solution from other users can be efficiently incorporated in its search, and it is thought that it leads to new way-of-thinking support. We performed comparison between an IGA system and a PDIGA system. The migration individual obtained better evaluation than a mutation individual. Moreover, it turns out PDIGA system can be used as a consensus-building tool.
