This work presents an evolutionary approach that enables the automatic design of modular residential homes in mass customized production. Given a set of modular placement rules for the design, the formal problem can be viewed as a two-dimensional single large object placement problem with fixed dimensions and additional positioning constraints. This formulation results in the search of a floor plan layout constrained by dimensional and positional restrictions over a combinatorial size search space. A genetic algorithm strategy for the automation of floor plan design was implemented and shown to deliver the required design solution layouts. The layouts, when embedded into a proper graphical interface system will allow for future owners to acquire houses fitting their exact needs at affordable prices, improving the quality of living while simultaneously complying with a language of design.

--

• Ciências da Computação e da Informação - Ciências Naturais