فصلنامه مطالعات هنر

فصلنامه مطالعات هنر

ارزیابی خلاقیت هوش انسانی و ماشینی در روند طراحی معماری

نوع مقاله : مقاله پژوهشی

نویسندگان
1 دانشجوی دکتری معماری،گروه معماری،دانشکده فنی مهندسی،واحد چالوس،دانشگاه آزاد اسلامی،چالوس،ایران
2 دانشیار گروه معماری، دانشگاه صنعتی جندی‌شاپور دزفول، ایران
10.22083/ssa.2025.536182.1085
چکیده
ابزارهای هوش مصنوعی با قابلیت تطبیق‌پذیری بالا، توانایی تعامل با مؤلفه‌های مختلف محیطی همچون نور، صدا، گرما، باد، و تغییرات وضعیت انسان را بدون دخالت مستقیم انسانی دارند. این ابزارها می‌توانند راه‌حل‌هایی متنوع و بهینه‌شده براساس داده‌های ورودی و الزامات طراحی ارائه دهند. با این حال، تفاوت‌های بنیادین میان هوش مصنوعی و هوش انسانی همچنان قابل توجه است. ماشین‌ها عمدتاً در انجام وظایف محاسباتی سنگین، پردازش داده‌های حجیم و تحلیل‌های الگوریتمی بسیار توانمندند، اما ساختار انعطاف‌پذیر ذهن انسان، قدرت یادگیری تجربی، و توانایی استدلال انتزاعی و خلاقیت مستقل را در اختیار ندارند. در این پژوهش، برای مقایسه ویژگی‌های خلاقیت در تولیدات انسانی و ماشینی، هفت اثر برجسته معماری با استفاده از هوش مصنوعی MidJourney بازطراحی شده‌اند. سپس با تدوین پرامپت‌های اختصاصی و تحلیل داده‌های بازتولیدی، میزان خلاقیت در آثار ماشین‌محور و انسانی مقایسه شده است. ارزیابی با مشارکت ۱۰۵ دانشجوی معماری آشنا با مفاهیم پایه‌ای هوش مصنوعی، از طریق نظرسنجی اینترنتی انجام شد. برای تحلیل داده‌ها از مقیاس لیکرت پنج‌درجه‌ای و روش دلفی فازی استفاده شده است. افزون بر آن، با انجام مصاحبه‌های عمیق با پنج عضو هیئت علمی متخصص در حوزه هوش مصنوعی و یادگیری ماشین، امکان الگوریتم‌سازی مفهوم خلاقیت در هوش مصنوعی بررسی شده است. نتایج نشان می‌دهد که اگرچه هوش مصنوعی قادر به بازتولید جلوه‌هایی از خلاقیت بصری است، اما از نظر عمق، تنوع، و اصالت، فاصله محسوسی با خلاقیت انسانی دارد. این مطالعه بستری نظری و تجربی برای بهره‌برداری مؤثر و هم‌افزای هوش مصنوعی در طراحی معماری فراهم می‌سازد.
کلیدواژه‌ها

موضوعات


عنوان مقاله English

Evaluating the creativity of human and machine intelligence in the architectural design process

نویسندگان English

Mehrdad Shabkhiz fordoei 1
Behzad Vasigh 2
1 Phd student of Architecture, Department of Architecture, College of Technical Engineering, Chalos Branch, Islamic Azad University, Chalos, Iran
2 Associate Professor, Department of Architecture, Jundi-Shapur University of Technology, Dezful, Iran
چکیده English

Introduction:

Artificial Intelligence (AI) has made remarkable strides in recent years, particularly in its ability to adapt to diverse environmental conditions such as light, sound, heat, and wind. This autonomous adaptability enables AI systems to function with minimal human intervention and to respond intelligently to shifts in both environmental stimuli and human emotional or cognitive states. In the field of architecture, AI presents novel opportunities by optimizing the design process through the analysis of large datasets and the generation of data-driven solutions. These advancements allow for increased efficiency, precision, and innovation in the early stages of design. However, despite these capabilities, fundamental differences remain between machine-based computation and the deeply human qualities of creativity, intuition, abstraction, and experiential learning that underlie the architectural design process.



Methodology:



To investigate the creative capacity of AI in architectural design, this study focused on a comparative evaluation of human-generated and machine-generated designs. The generative AI platform MidJourney (Pro Plan) was utilized to reinterpret seven iconic architectural works. These buildings were selected for their historical significance and diverse architectural styles, including classical, modernist, and deconstructivist examples. Tailored prompts were crafted for each case to guide the AI in reflecting the original design’s aesthetic values, spatial logic, and symbolic content.

کلیدواژه‌ها English

Creativity
human intelligence
artificial intelligence
architectural design
design process