In this research, a Genetic Algorithm (GA) has been developed and the well-known one-dimensional bin packing problem (BPP) has been implemented within the structural optimization process. The Objective Function formulation lies in a marked change of the paradigm in which the target function is represented by the amount of steel required by the factory instead of the structural cost (e.g. weight). The best design is obtained by varying the geometry properties of the members and the cross-section assignation ensuring optimal stock of existing elements. Finally, the structural cost and the Carbon emission are calculated for a spatial reticular dome. The mass of the waste with respect to the mass of the stock, Mwaste/Mstock, is evaluated by adopting both the cutting Stock approach and the traditional approach. The former leads to a waste saving that is almost twice that obtained from the latter. However, no significant differences in terms of carbon emission can be observed by comparing the two approaches.

The growing need for structural, energetic and architectonic refurbishment of the existing structures led to the emergence of alternative retrofitting techniques. Among them, steel exoskeletons stand out for their non-invasive and time-efficient nature, especially in cases where the relocation of the building’s activities is not allowed. In this research, the optimal exoskeleton configuration, in terms of number of exoskeletons and their position around the building, along with the sizing of their constituent elements, has been obtained through an optimization process. The optimization tool is based on a Genetic Algorithm with the aim of weight minimization, including as constraints a maximum allowable inter-storey drift for preserving the elastic behavior of the existing building, and the structural safety of the exoskeleton members. Two exoskeleton typologies were selected as case studies, employed for the retrofitting of an existing RC moment-resisting frame building. Furthermore, Life Cycle Assessment analyses were conducted on the optimal configurations considering two materials: steels with high and low percentages of recycled materials. The results yielded interesting insights into the structural performance of both exoskeleton typologies, as well as into the environmental characteristics of both studied materials.

Lowering environmental impacts has lately been a critical objective of structural optimisation due to the significant amount of greenhouse gas (GHG) emissions in the civil engineering sector. This work introduces a Life Cycle Assessment (LCA) based multi-objective optimisation framework for the optimal design of mixed steel-timber structures by varying the building design’s size, shape, and topology. The study’s novelty stems from the integration of an environmental objective function in the early design process, based on LCA methodology and standard environmental indicators, and the definition of a structural target function where a penalty-based approach is implemented for reducing structural complexity in situ. The structural cost and the Global Warming Potential (GWP) are the objective functions of the optimisation problem. The analysis outcomes reveal that minimising the number of connections as well as moving towards timber-steel solutions represents the key aspect to achieve a sustainable and effective design of spatial truss structures.