Song Jingwen

ABSTRACT |  A strong dichotomy exists between the increased architectural design agency offered by digital tools today and the affordances given by many construction contexts, especially building environments in developing countries with limited available means.

This creative practice research project questions whether computational design tools can be used differently in contemporary architecture practice so that onsite affordances can be increased in parallel with the expanding virtual design solution space. It postulates that by incorporating incertitude as a productive and constructive component in alternative computation-driven design and materialisation methods (rather than as negatives), the locally available solution space for built architecture can be dramatically expanded and onsite ability and agency increased.

The study develops a methodology that aims at procedurally managing serious slippage from aleatory occurrences during materialisation and transforming it into a practical opportunity for non-standard project realisation. In doing so the case is made for the use of more democratic epistemic models and more intelligent structures of approximation than (common) deterministic approaches in digital design would allow for.

The field of study is ‘Post-Digital Architectural Design Practice,’ which is characterised by virtually unlimited computational ability, precision and unprecedented access to specialised and complex processes, wherein the challenge lies in the translation of its expanded agency into professional application. The employed methodologies are ‘Participatory Action Research’ and ‘Reflective Practice,’ applied on an existing and developing body of creative works of increasing scale and complexity, built in volatile construction contexts. Their study identifies and challenges preconceived notions behind their creation and frames these in dialectic space for further development until higher-order knowledge is extracted in contribution to the overall field of study.

The demonstrator employs protean design diagrams capable of absorbing serendipity throughout the project crystallisation process. Named ‘vibrant objectiles,’ these diagrams consist of holistic, computation-driven, associative design models that are procedural in nature and cover both project design and delivery. They are built from a select number of interdependent components that are optimised for local implementation. These are placed in and communicate across a hierarchy of scale and impact, enabling the emergence of larger-scale complexity. They relate to aspects of project materiality, material system and/or materialisation, and can incorporate site-specific idiosyncrasies. They have potentially high levels of volatility or uncertainty, termed ‘vibrancy’. This vibrancy is managed throughout the project development, rather than locked down. From early on, the extensibility of potential deviations is harnessed within the developing objectile that incorporates continuous feedback from rigorous prototyping. Until project completion, uncertainty is given room to feed back into the system, giving rigour and animus to the whole. Hence, rather than seeing the act of design materialisation as the literal translation of digitally designed objects, the proposed method embraces fluctuations as the design develops into its final singular site-specific solution.

Evaluation of the potential and efficacy of the proposed method challenges the role practising architects are often given and the extent of their authority within professional legal frameworks. For the proposed method to successfully bridge the dissociation between theory and practice, the architects’ mode of operation needs to change, and the extent of their actions to dramatically expand. However, once these changes are made possible, the discussed projects demonstrate that the increase of local onsite affordance can be substantial.