Bioinspired Modularity in Evolutionary Computation and a Rule-Based Logic Design Solutions for Shared Office Space

Bioinspired Modularity in Evolutionary Computation and a Rule-Based Logic Design Solutions for Shared Office Space

Kim, Yujin

:

Evolutionary computation is a population-based problem solver that ischaracterized by a stochastic optimization in order to solve both a singleobjective and multiple objectives. Previous evolutionary computationalresearches provided various design options and improved optimization throughbeing evolved with fitness criteria, especially when multiple design objectivesconflict with one another. In this paper, a rule-based algorithm was combinedwith the evolutionary computational process to propose an assembly logic of themodules and to improve an architectural building design in order to adapt toenvironmental changes. Two algorithms - a rule based and generative algorithmproceeded simultaneously and provided various options as well as optimization inreal time. For the experiment set-up, existing buildings were divided into eachmodule; the modules were reinterpreted and reassembled with the logic driven byEvolutionary Developmental Biology. The conclusion is that when a rule basedlogic is combined with a developmental algorithm with a modular system, it ismore efficient for the design process to be analyzed, evaluated, and optimized.The ultimate outcome provides various options in a short amount of time.

Evolutionary computation is a population-based problem solver that ischaracterized by a stochastic optimization in order to solve both a singleobjective and multiple objectives. Previous evolutionary computationalresearches provided various design options and improved optimization throughbeing evolved with fitness criteria, especially when multiple design objectivesconflict with one another. In this paper, a rule-based algorithm was combinedwith the evolutionary computational process to propose an assembly logic of themodules and to improve an architectural building design in order to adapt toenvironmental changes. Two algorithms - a rule based and generative algorithmproceeded simultaneously and provided various options as well as optimization inreal time. For the experiment set-up, existing buildings were divided into eachmodule; the modules were reinterpreted and reassembled with the logic driven byEvolutionary Developmental Biology. The conclusion is that when a rule basedlogic is combined with a developmental algorithm with a modular system, it ismore efficient for the design process to be analyzed, evaluated, and optimized.The ultimate outcome provides various options in a short amount of time.

Palavras-chave:

DOI: 10.5151/proceedings-ecaadesigradi2019_138

Referências bibliográficas

• [1]

Como citar:

Kim, Yujin; "Bioinspired Modularity in Evolutionary Computation and a Rule-Based Logic Design Solutions for Shared Office Space", p-341-348. In: Proceedings of 37 eCAADe and XXIII SIGraDi Joint Conference, “Architecture in the Age of the 4Th Industrial Revolution”, Porto 2019, Sousa, José Pedro; Henriques, Gonçalo Castro; Xavier, João Pedro (eds.). São Paulo: Blucher, 2019.
ISSN 23186968, DOI 10.5151/proceedings-ecaadesigradi2019_138