Algorithmic Nesting of Spalted Poplar: Material-Informed Methods for Computational Wood-Thermal Design

Contemporary architectural practice faces increasing
material scarcity and demands for sustainable design processes that can
accommodate natural variability. This research investigates how
computational design methods can engage with the irregular thermal
and geometric properties of spalted poplar wood to produce
performative and expressive architectural assemblies. The research
addresses the question of how algorithmic processes can utilise the
materials' inherent heterogeneity, arising from fungal decay, density
variation, and thermal inconsistency, as an active design parameter.
Two complementary computational workflows are developed. The first
method explores regular geometries with irregular thermal properties
using infrared thermography and computer vision analysis to classify
and sort uniform elements for thermally adaptive façade applications.
The second method addresses irregular geometries with consistent
thermal behaviour, employing physics-based nesting and Agent-Based
Modelling to assemble segmented wood pieces into cohesive structures.
Through these methods, the study establishes a framework that links
empirical material data with algorithmic design operations. Physical
prototypes demonstrate how computational sorting and nesting can
translate biological variation into architectural performance. The
findings suggest that embracing material irregularity enables new
modes of computational craft, expanding sustainable applications for
non-standard timber resources.