Predictive Approaches to Architectural Expansion: Colleges of Medicine and Science at The University of Duhok as a Case Study.

This study examines predictive approaches for identifying suitable locations for expanding educational buildings at the University of Duhok by reviewing the history, types, and methods of expansion. To address the research question: How can existing educational buildings be expanded multiple times in the future? An analysis of 10 previous studies on physical building expansion reveals a gap in growth strategies for public institutions, such as universities. The research problem focuses on the location of expansions that align with long-term growth guided by five key criteria: available land, continuity with existing structures, access points, circulation systems, and functional relationships. The study aims to determine suitable expansion locations, identify missing criteria in past expansions, and forecast future needs using statistical and architectural tools, while also exploring the role of AI. A mixed-methods approach combines qualitative and quantitative techniques, applying AutoCAD 2024, Julius AI, SPSS, Photoshop, and a Python AI. As a result, it shows that all past expansions only met 44% of key architectural criteria. Additionally, most expansions did not comply with statistical standards. The AI learned three criteria from human design: inside boundary (92%), having open-ended potential for future expansion (81%), and maintaining functional relationships (83%). However, it failed to meet three other architectural criteria. Overall, AI and human expansion plans showed 62% agreement. In conclusion, the existing buildings' past expansions show clear shortcomings in architectural quality. While human-designed predictions performed better, AI demonstrated strong potential in complex cases and, with further refinement, could support more adaptive and intelligent planning.