Tasarım Sürecinde Üretken Yapay Zeka Asistanlarının Rolü: Mimarlık Öğrencileriyle Ampirik Bir Araştırma

Abstract

Uzun süre yalnızca akademik çalışmalar ile sınırlı kalmış olan üretken tasarım asistanları, makine öğrenmesi tabanlı yapay zeka teknikleri sayesinde ana akım mimari pratik için de erişilebilir olmuştur. Bu çalışma, gelecekte daha da yaygınlaşacağı düşünülen bu üretken tasarım asistanlarının (GDA) mimari tasarım sürecine entegrasyonunu araştırmaktadır. Araştırma, üçüncü ve dördüncü sınıf mimarlık öğrencilerine odaklanarak, bu araçların tasarım sürecine nasıl entegre edildiklerini ArchiGAN ve HouseGAN araçları üzerinden incelemektedir. Araştırma kapsamında gerçekleştirilen çevrimiçi atölye çalışmasında, 12 katılımcının tasarım süreci kayıtları, tasarım çıktılarının bağımsız bir jüri tarafından değerlendirilmesi ile elde edile başarı puanları, ve son olarak atölye sonrası öğrenci anketleri ile toplanan geri bildirimler çalışmanın nicel ve nitel verilerini oluşturmaktadır. Araştırma, öğrencilerin GDA kullanımlarında üç ana stratejinin ortaya çıktığını göstermiştir: (1) Tasarım süreci boyunca sürekli kullanım, (2) seçici problem çözme kullanımı ve (3)başlangıçta fikir oluşturma kullanımı ardından geleneksel yöntemlere geçiş. Araştırmada, GDAlara aşırı bağımlılığın, tasarımcının yorumlayıcı ve geliştirici katkısını potansiyel olarak sınırlayabileceği gözlenmiştir. Anket çalışması ise, farklı GDA’ların tasarım sürecine farklı aşamalarda katkı sağladığını göstermektedir. ArchiGAN, keşif ve fikir oluşturma aşamasında yardımcı olurken, HouseGAN tasarım problemlerini yeniden tanımlama ve tasarım iterasyonu konusunda destekleyici gözükmektedir. Sonuç olarak, çalışma, mimari tasarım sürecinde GDAların dönüştürücü potansiyelini ve sürece entegrasyonlarında karşılaşılabilecek zorlukları göstermektedir. Araştırma, dengeli bir GDA entegrasyonunun gerekliliğini ortaya koymakta ve gelecekteki araştırmalar için, mimarlık eğitiminde GDA' ların uzun vadeli etkilerine odaklanılmasını önermektedir.

Keywords

• Yapay Zeka
• Üretken Tasarım
• Tasarım Asistanı
• Mimari Tasarım Pedagojisi
• Tasarım Bilişi

The Role of AI Design Assistance on the Architectural Design Process: An Empirical Research with Novice Designers

Abstract

This study explores the integration of Generative Design Assistants (GDAs), specifically machine learning based tools, in the architectural design process. It investigates how these tools, once confined to experimental realms, are now influencing mainstream architectural practice, particularly among novice architects. The research focuses on third and fourth-year architecture students, examining how they adapt to and integrate these advanced AI tools into their design workflows. Through an empirical online workshop, the study collected data of design process recordings, design output success scores of students by an independent jury, and post-experiment surveys. This approach provided insights into the timing, frequency, and sequence of GDA usage, as well as the influence of specific GDA features on design success. The research reveals that three primary strategies emerged in students' GDA usage: continuous use throughout the design process, selective problem-solving use, and initial ideation use followed by traditional methods. However, an over-reliance on GDAs was noted to potentially limit the designer’s interpretive and developmental input. The survey shows that different GDAs have distinct strengths and impacts on the design process. In terms of selected GDAs for the experiment, ArchiGAN aids in discovery and ideation, while HouseGAN excels in reframing design problems. In conclusion, the study underscores the transformative potential and challenges of GDAs in architectural design and highlights the need for balanced GDA integration. The research outputs show that future research should focus on the long-term implications of GDAs in architectural education. This research aims to guide the effective integration of AI in architecture, enhancing the human designer's role rather than overshadowing it.

Keywords

• Artificial Intelligence
• Generative Design
• Design Assistant
• Architectural Design Pedagogy
• Design Cognition

References

1. Anderson, C. M. (2001). Swarm Intelligence: From Natural to Artificial Systems. Eric Bonabeau , Marco Dorigo , Guy Theraulaz. The Quarterly Review of Biology, 76(2), 268–269. https://doi.org/10.1086/393972.
2. As, I., & Basu, P. (2021). The Routledge companion to artificial intelligence in architecture. Routledge.
3. Bank, M., Sandor, V., Schinegger, K., & Rutzinger, S. (2022). Learning Spatiality - a GAN method for designing architectural models through labelled sections. In Proceedings of the 40th Conference on Education and Research in Computer Aided Architectural Design in Europe (eCAADe 2022) (Vol. 2). https://doi.org/10.52842/conf.ecaade.2022.2.611.
4. Başarir, L. (2022). Modelling AI in architectural education. Gazi University Journal of Science, 35(4), 1260–1278. https://doi.org/10.35378/gujs.967981
5. Carpo, M. (2023). Beyond digital: Design and Automation at the End of Modernity. MIT Press.
6. Carta, S. (2021). Self-Organizing floor plans. Harvard Data Science Review. https://doi.org/10.1162/99608f92.e5f9a0c7
7. Ceylan, S. (2021). Artificial Intelligence in Architecture: an Educational perspective. In International Conference on Computer Supported Education, CSEDU - Proceedings (1st ed.). https://doi.org/10.5220/0010444501000107
8. Chaillou, S. (2020). ArchiGAN: Artificial Intelligence x Architecture. In Architectural Intelligence (pp. 117–127). https://doi.org/10.1007/978-981-15-6568-7_8

9. Chaillou, S. (2022). Artificial Intelligence and Architecture: From Research to Practice. Birkhäuser.
10. Chu, H., Li, D., Acuna, D., Kar, A., Shugrina, M., Wei, X., & Fidler, S. (2019). Neural turtle graphics for modeling city road layouts. In In Proceedings of the IEEE/CVF international conference on computer vision (pp. 4522-4530).
11. Cudzik, J., Nyka, L., & Szczepański, J. (2024). Artificial intelligence in architectural education-green campus development research. Global Journal of Engineering Education, 26(1).
12. Danhaive, R., & Mueller, C. (2021). Design subspace learning: Structural design space exploration using performance-conditioned generative modeling. Automation in Construction, 127, 103664. https://doi.org/10.1016/j.autcon.2021.103664
13. Del Campo, M., Manninger, S., & Carlson, A. (2019). Imaginary Plans. In Ubiquity and Autonomy (Vol. 39). ACADIA Conference.
14. Edirne, J., & Öztürk, M. (2024). Student-Artificial Intelligence Interaction in Architectural Design Education: Artificial Intelligence Workshop in Design. In International Research in Architecture Sciences (Vol. 2, p. 30). Eğitim Yayınevi.
15. Eroğlu, R., & Gül, L. F. (2022). Architectural form explorations through generative adversarial networks - predicting the potentials of StyleGAN. In 40th Conference on Education and Research in Computer Aided Architectural Design in Europe, eCAADe 2022 (Vol. 2). https://doi.org/10.52842/conf.ecaade.2022.2.575
16. Furtado, C. L. G. M. (2008). Cedric Price’s Generator and the Frazers’ systems research. Technoetic Arts, 6(1), 55–72. https://doi.org/10.1386/tear.6.1.55_1
17. Goodfellow, I. J., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., & Bengio, Y. (2014). Generative adversarial networks. arXiv (Cornell University). https://doi.org/10.48550/arxiv.1406.2661
18. Holland, J. H. (1992). Adaptation in natural and artificial systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence. MIT Press.
19. Kasabov, N. K. (1996). Foundations of neural networks, fuzzy systems, and knowledge engineering. Marcel Alencar.
20. Kelly, T., Guerrero, P., Steed, A., Wonka, P., & Mitra, N. J. (2018). FrankenGAN. ACM Transactions on Graphics, 37(6), 1–14. https://doi.org/10.1145/3272127.3275065
21. Lindenmayer, A. (1968). Mathematical models for cellular interactions in development I. Filaments with one-sided inputs. Journal of Theoretical Biology, 18(3), 280–299. https://doi.org/10.1016/0022- 5193(68)90079-9
22. Nauata, N., Chang, K., Cheng, C., Mori, G., & Furukawa, Y. (2020). House-GAN: Relational Generative Adversarial Networks for Graph-Constrained House Layout Generation. In Lecture Notes in Computer Science (pp. 162–177). https://doi.org/10.1007/978-3-030-58452-8_10
23. Nauata, N., Hosseini, S., Chang, K. H., Chu, H., Cheng, C. Y., & Furukawa, Y. (2021). House-gan++: Generative adversarial layout refinement network towards intelligent computational agent for professional architects. In IEEE/CVF Conference on Computer Vision and Pattern Recognition.
24. Negroponte, N. (1969). Toward a theory of architecture machines. Journal of Architectural Education, 23(2), 9–12. https://doi.org/10.1080/00472239.1969.11102296
25. Negroponte, N. (1970). The architecture machine: Toward a More Human Environment. MIT Press (MA).
26. Newton, D. W. (2019). Deep generative learning for the generation and analysis of architectural plans with small datasets. In Architecture in the Age of the 4th Industrial Revolution - Proceedings of the 37th eCAADe and 23rd SIGraDi Conference. https://doi.org/10.5151/proceedings-ecaadesigradi2019_135
27. Özman, G. Ö., & Selçuk, S. A. (2023). Generating mass housing plans through GANs - a case in TOKI, Turkey. APJ, 28(3). https://doi.org/10.54729/2789-8547.1197
28. Quintana, M., Schiavon, S., Tham, K. W., & Miller, C. (2020). Balancing thermal comfort datasets: We GAN, but should we? In Proceedings of the 7th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation (Pp. 120-129). https://doi.org/10.1145/3408308.3427612
29. Rodrigues, R. C., Koga, R. R., Hirota, E. H., & Duarte, R. B. (2022). Mapping space allocation with artificial intelligence - an approach towards mass customized housing units. In 40th Conference on Education and Research in Computer Aided Architectural Design in Europe, eCAADe 2022 (Vol. 2). https://doi.org/10.52842/conf.ecaade.2022.2.631
30. Sadek, M. (2023). Artificial Intelligence as a pedagogical tool for architectural education: What does the empirical evidence tell us? MSA Engineering Journal, 2(2), 133–148. https://doi.org/10.21608/msaeng.2023.291867
31. Singh, V., & Gu, N. (2012). Towards an integrated generative design framework. Design Studies, 33(2), 185–207. https://doi.org/10.1016/j.destud.2011.06.001
32. Sohl-Dickstein, J., Weiss, E. A., Maheswaranathan, N., & Ganguli, S. (2015). Deep Unsupervised Learning using Nonequilibrium Thermodynamics. arXiv (Cornell University). http://export.arxiv.org/pdf/1503.03585v8
33. Steinfeld, K. (2019). Gan Loci. In 39th Conference of the Association for Computer Aided Design in Architecture: Ubiquity and Autonomy (pp. 392-403).
34. Stiny, G., & Gips, J. (1971). Shape Grammars and the Generative Specification of Painting and Sculpture. In IFIP Congress (2), Vol. 2, No. 3, pp. 125– 135, 1460–1465. http://www.shapegrammar.org/ifip/SGBestPapers72.pdf
35. Tong, H., Türel, A., Şenkal, H., Ergun, S., Güzelci, O. Z., & Alaçam, S. (2023). Can AI function as a new mode of sketching. International Journal of Emerging Technologies in Learning (Ijet), 18(18), 234–248. https://doi.org/10.3991/ijet.v18i18.42603
36. Uzun, C., Çolakoğlu, M. B., & İnceoğlu, A. (2020). GAN as a generative architectural plan layout tool: A case study for training DCGAN with Palladian Plans and evaluation of DCGAN outputs. İTÜ Dergisi A, 17(2), 185–198. https://doi.org/10.5505/itujfa.2020.54037
37. Wang, S., Zeng, W., Chen, X., Ye, Y., Qiao, Y., & Fu, C. (2023). ActFloor-GAN: Activity-Guided Adversarial Networks for Human-Centric Floorplan Design. IEEE Transactions on Visualization and Computer Graphics, 29(3), 1610–1624. https://doi.org/10.1109/tvcg.2021.3126478
38. Wolfram, S. (1983). Statistical mechanics of cellular automata. Reviews of Modern Physics, 55(3), 601–644. https://doi.org/10.1103/revmodphys.55.601