Stock-constrained optimization of partially disassembled trusses

Abstract

Reuse of structural components is a relatively unexplored area of research, with a lot of potential environmental impact. Structural reuse significantly reduces new material use, carbon emissions, and construction waste. This thesis shows a novel way of reusing structural components through partial disassembly of trusses into triangular components. A computational approach for quickly designing trusses with both new and recycled components is presented. The algorithm aggregates components row by row to fill a target area defined by the user, this is done by cutting the reused components where necessary and adding new material members and triangles where appropriate to prevent voids. In case the reusable inventory has a variety of component sizes, multiple designs can be generated. The workflow uses a genetic algorithm to explore and optimize these different designs, taking into account the user’s stock input and target dimensions. Three case studies, reusing realistic trusses, illustrate the algorithm’s applicability to existing truss inventories that could be reused.