Algorithmic Generation of Building Typology for Office Building Design

Abstract

Numerous office building design optimizations are in international research to reduce energy consumption, optimize costs and provide optimal comfort. However, there is a lack of knowledge on the effects of geometry and space organization. This study deals with space organization problems and searches for all possible optimal building space structure configurations in terms of energy and comfort parameters using a mathematical algorithmic method. The methodology is based on the formulation of feasible architectural rules and their translation into an algorithm that can generate 2D floor plans satisfying all boundary conditions. In the framework of an exemplary modeling procedure, a 4-story office building geometry generation was carried out, resulting in 17-floor plan versions and 7 different building geometries. The resulting building shapes were classified by energy-related geometry parameters (envelope surface/useful area) for the future step of the research, where the cases will be compared with the help of building simulations. With the help of the method, it was possible to significantly narrow the search space, but future improvements are needed for faster work for wider applicability.