Semi-Supervised Clustering for Architectural Modularisation

Modular construction allows for a faster, safer, better controlled, and more productive process, yielding quality results with low risk controlled costs. However, despite the potential advantages of this methodology, its adoption has remained slow due to reasonably high degree standardisation repetition that projects require, inexorably clashing unique building designs created meet clients’ needs. The present article proposes performing modularisation process after design is complete, reaping most benefits modular while preserving vision building. This objective achieved by implementing semi-supervised methodology reliant on clustering individual rooms subsequent user validation obtained clusters identify base modules representative each cluster. proposed applied in case study an existing apartment complex, which was previously performed manually—thus serving as baseline. acquired display 99.6% reduction process’ duration, maintaining 96.4% Normalised Mutual Information Score 93.3% Adjusted Score, justifying continuous development assessment future works.