Generative Design for Building Information Modeling

Guilherme Henrique Aciaioli de Sousa Barreto
Thesis to obtain the Master of Science Degree in
Information Systems and Computer Engineering
Supervisor: Prof. Ant´onio Paulo Teles de Menezes Correia Leit˜ao

Generative Design (GD) is an algorithmic-based approach to design that allows the generation of forms and shapes through algorithms. It has been vastly explored with the Computer-Aided Design (CAD) paradigm, but not to the same extent with the Building Information Modeling (BIM) paradigm.

Therefore, this thesis by Guilherme Henrique Aciaioli de Sousa Barreto proposes a solution that allows the exploration of GD using the BIM paradigm, taking full advantage of its capabilities. The solution is an extension of Rosetta, a GD environment that supports a wide range of back-ends, namely, OpenGL and CAD applications.

Author expands Rosetta’s abstraction layer with modeling operations capable of producing models on a BIM application, ArchiCAD, having into consideration portability between the already supported back-ends.

Furthermore, alongside the development of the solution, another BIM application was also being added to Rosetta, Revit, allowing to test portability between the two back-ends.

Finally, author evaluates the solution in terms of its adequacy, portability, performance, and support of ArchiCAD-specific operations.