Building Information Modeling Steered by Evolutionary Computing

Abstract

In recent years, there has been a surge in the use of building information modeling (BIM) within construction. Much of the previous published research focuses on the impact of BIM during the detailed design, construction, and postconstruction stages. This paper combines the search and exploration powers of evolutionary computing with BIM during the conceptual stage of design. The main contribution of this paper is a methodology to integrate solutions generated by the interactive and visual clustering genetic algorithm (IVCGA) into a BIM environment to enhance the design information and allow the solutions to be viewed in greater detail as building information models. These models can then be utilized or modified by individual members of the design team to aid in the selection of concept designs that appropriately meet client/design requirements. An Autodesk Revit plug-in application has been developed that enables the IVCGA solution to be transformed into a building information model in a common repository. Using this approach during the conceptual design stage can potentially enhance multidisciplinary decision-making capacity by allowing a wide range of concept designs to be assessed in high detail with more accurate costing prior to committing a large sum of funding to the project