抄録
Under the trends of multi-disciplinary design optimization, modeling techniques and computational algorithms that are configured by following complicatedness of problems have been promoted and gathered much attentions. However, their applicability has been limited within continuously defined problems, while layout optimization is typically complicated but combinatorial. Under this contrast, this paper proposes a hierarchical optimization method for layout design. In the method, a rectangular packing layout problem is hierarchically decomposed into the top-level layout and a series of sub-level layout, those partial layout problems are represented with sequence pair and optimized by genetic algorithms, and they are coordinated through mutual exchange of sets of Pareto optimal solutions explored in the lower levels and ranges of desirable aspect ratio. The validity and effectiveness of the method is investigated through its application to an example problem.
