AI-assisted design exploration for post-disaster housing: reassessing Bam reconstruction
Solmaz Arzhangi, Mohadeseh Ramezanzadeh Lak, Sayed Arash Hosseini Sabzevari, Fereshteh AslaniPurpose
The purpose of this study is reassessing Bam reconstruction after earthquake using Artificial Intelligence. Post-disaster permanent housing plays a critical role in ensuring the long-term recovery and well-being of affected communities. However, architectural design processes in disaster responses are often rushed, leading to solutions that may overlook cultural, environmental and functional needs.
Design/methodology/approach
This study examines the potential role of artificial intelligence (AI) in improving post-disaster housing design by reassessing the reconstruction that followed the 2003 Bam earthquake in Iran. Using a process-oriented perspective, existing post-disaster permanent houses were evaluated against AI-generated architectural alternatives produced through Midjourney.
Findings
Key indicators and evaluation criteria were identified through a literature review and supported by field surveys conducted in Bam. An evaluation flowchart was then developed to systematically assess design performance in terms of safety, sustainability, cultural compatibility, functionality, accessibility and affordability.
Practical implications
The findings suggest that generative AI can support architects and planners by rapidly generating diverse design alternatives and expanding the range of solutions considered during the early stages of post-disaster reconstruction. However, the results also demonstrate that expert judgment remains essential for evaluating and adapting AI-generated proposals to local conditions, reconstruction constraints and long-term resilience objectives.
Originality/value
This study not only demonstrates the application of generative AI in post-disaster housing design using the 2003 Bam earthquake as a case study, but also proposes a rapid evaluation framework for assessing AI-generated architectural alternatives as a complementary decision-support tool in post-disaster reconstruction.