Control of architectural drawing criteria with artificial intelligence

The relationship between architecture and artificial intelligence is undergoing a significant transformation today. Artificial intelligence technology is impacting many aspects of architecture, such as design, construction, and sustainability. This relationship explores how AI is used in architecture and the advantages it brings to the sector. Particularly in the design stages of architectural housing plans, the relationship with artificial intelligence is becoming increasingly important. Architectural design encompasses not just aesthetics but also functionality, energy efficiency, and sustainability factors. AI offers significant advantages in these areas. It can be used for design optimization, personalization, speed and efficiency, energy efficiency, community and spatial planning, structural resilience, cost analysis, and data analysis. AI has great potential to optimize design processes and suggest the most efficient plans. Additionally, it can offer personalized plans based on customer preferences, speed up architectural drawing processes, and enhance energy efficiency. Artificial intelligence can also analyze factors like structural resilience, material selection, and construction costs. This thesis aims to evaluate the compliance of architectural housing plan images with pre-project drawing standards. This evaluation is based on the minimum requirements for architectural work stemming from the basic principles of the Union of Chambers of Turkish Engineers and Architects. The study uses 171 training and 26 test data obtained from project drawings of students from the Department of Architecture at Düzce University, Faculty of Art, Design, and Architecture. This study evaluates the compliance of housing project plans with pre-project drawing standards using artificial intelligence technology developed with DNN and Resnet-50 models. Criteria from Architectural Project Drawing and Presentation Standards such as viewpoint, dimensioning, room name and m², module and construction axes, and furnishings are analyzed for compliance with drawing standards. The results of the study show that artificial intelligence is generally successful in many criteria. However, significant deficiencies were identified in the viewpoint criterion. These deficiencies require careful examination of AI performance during the project development stage. The study also emphasizes the importance of using AI tools in architectural education and how these tools can help students adapt to industry innovations and increase design efficiency. It also discusses the potential for municipalities to use AI-supported control systems to make inspection processes more effective and objective. The study aims to provide an academic perspective on the advantages of using AI-supported control systems. Keywords:Architectural Design, Drawing Standards Control, Artificial Intelligence,Artificial Neural Networks.