skip to content

Primary supervisor: Dr Ying Jin; secondary supervisor: Dr William Fawcett

Ran Xiao


Research overview:

Architectural typology is an analytical concept that embodies the classification of designs according to certain criteria. Patterns often emerge from typological analyses of examples of similar buildings and are a central part to design learning. Christopher Alexander promoted the well-known pattern language approach as a way of record and communicate design knowledge. However, applying existing patterns to new situations effectively needs certain design heuristics. Ran’s research created a machine learning prototype that re-apply existing design patterns to new context, using residential layouts as a study model. The heuristic used in the prototype blends contextual information with geometry. Inspired by early research at the Martin Centre on architectural morphology, he also invented new computable representation for the prototype. He also worked closely with practitioners to understand how computational design tools may be better designed for architects, using Human-Computer Interaction research techniques. 



Ran Xiao is a registered architect in the UK. After graduating from the MPhil in Architecture and Urban Design programme at Cambridge in 2014, he worked at a leading London practice and worked on cultural, residential, and urban design projects internationally. His experience in practice has informed his doctoral research questions. Since 2017, he has been researching on the application of machine learning in architecture with a focus on computable representation, building morphology, design heuristics and residential typologies.