Development of Physical Simulations Using Artificial Intelligence for Implementation in BIM Methodology

Abstract

This article is associated with an Educational Innovation Project (IE23.0303) conducted at the Polytechnic University of Madrid. The objectives consist of obtaining optimized free-form surfaces through the development of physical simulations using evolutionary computing, a branch of artificial intelligence. The text presents the three main methods used, allowing students to digitally emulate surfaces that are reminiscent of those popularized by architects such as Gaudí or Frei Otto. In an initial phase, students learned to parametrize the geometry of their projects using Grasshopper. Subsequently, physical simulations were introduced to achieve force equilibrium, enabling the generation of structurally efficient surfaces through graphical methods without the need for numerical calculations. Students utilized artificial intelligence to fine-tune predefined parameters, achieving outcomes that were beyond the reach of traditional analog approaches. Finally, students successfully incorporated the surfaces generated through simulations into BIM, thereby defining the architectural project and automating document generation. The process highlights the importance of geometry and algorithmic thinking in the education of architecture students. At the same time, it aims to raise awareness among students about the need to act in an environmentally friendly manner by designing solutions that reduce material consumption and CO2 emissions.