Towards a Theory of Anticipatory Adaptive Assemblages
Doctoral Thesis in Civil and Architectural Engineering
kth royal institute of technology
This thesis by Magnus Larsson is an investigation into how meta-heuristic multi-objective optimisation processes (genetic algorithms driven by evolutionary solvers) can bring about materials-related advantages in architectural performance.
It redefines the architect’s and engineer’s role from being designers of a singular space or structure to being designers of entire species of spaces, and discusses a particular method – anticipatory adaptive assemblages (AAA) – that allows such processes to produce many generations of design iterations that eventually yield individuals optimised for a set of predefined objectives.
This includes in particular the optimisation of building materials, with a certain focus on timber structures. The thesis provides a theoretical foundation (assemblage theory) that connects an ontology, a methodology, an epistemology, and an axiology to the computational operations used, elevating the domain beyond simplistic notions of parametricism.
It leverages contemporary generative design methods to introduce a range of novel concepts and tools such as auxiliary loads, material phase transition (MPT) diagrams, generative life cycle assessments (GLCA), parametric epistemic things (PET), presilience, and postponism.
Finally, it provides a case study that shows how this assortment of contrivances, and AAA theory at large, can be used not just for theoretical musings, but to produce actual architectural schemes based on more precise data analyses than is typically the case in today’s built environment.
A concluding discussion establishes that the use of more advanced and complex optimisation strategies is not just a possibility but a necessary obligation for an architecture, engineering, and construction (AEC) industry that – if the manufacturing of building materials are added to the construction and operation of buildings – is responsible for between 35% and 40% of both global final energy use and worldwide energy-related CO2 emissions.
Claiming that our knowledge of materials, including the auxiliary loads that they carry (such as their global warming potential) can be used to design and engineer architectural assemblages capable of replacing energy-consuming with energy-producing buildings, it suggests that Le Corbusier’s famous dictum that buildings are ‘machines for living in’ should be replaced with the notion that all buildings are potential power plants for living in. Risks associated with the development of AAAs are discussed, and future studies proposed.