Date

2021-2-15

Author

Sönmez, Ayça

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

299
views

389
downloads

Cite This

The developments in recent technologies through Industry 4.0 lead to the integration of digital design and manufacturing processes. While manufacturing is becoming an essential input of the design, it is usually considered at the last stages of the design process. This misconception creates a gap between digital design and fabrication, resulting in fabricated outcome being different from initial design in the context of architectural tectonics. This thesis aims to provide a basis to bridge digital design and fabrication processes by training the design process through fabrication constraints. In order to provide a holistic design approach, a case study is presented, in which a wall design is considered through two stages. In the first stage, a top-down design process is obtained without considering fabrication. In the second stage, an artificial intelligence (AI) based generative design method trained by material and fabrication data is followed. Through top-down approach, problems with fabrication techniques are observed. On the other hand, AI based approach is adopted to study form-finding using different technologies such as generative design and cloud computing. The results indicate that design space explorations and generative design simulations can provide a significant variety of manufacturable design alternatives. While the results can be grouped based on their visual similarity, different aspects of design alternatives show that manufacturing has a large impact on transforming the initial design. All of the obtained design alternatives are used to construct a dataset that includes different design parameters. The dataset from the AI based method is used to train a decision tree classifier learning algorithm in order to provide decision making tasks in choosing the manufacturing method. The data obtained from the top-down method is used as a test data set in order to compare real-life cases and algorithm-predicted outcomes.

Subject Keywords

Generative Design, Computer-Aided Fabrication, Arcihtecture 4.0, Artificial Intelligence, Digital Tectonics

URI

https://hdl.handle.net/11511/89640

Collections

Graduate School of Natural and Applied Sciences, Thesis