Assessment and Integration of Sustainability and Circularity Metricswithin Generative Bridge Design

Given the construction sector’s large environmental impact, analysing and optimising sustainability of a structure becomes increasingly important. The growing number of Life-Cycle Assessment (LCA) tools for buildings is however neither directly applicable nor transferable to bridges. Furthermore, circularity is hardly ever measured, let alone enforced in bridge design. We derived and implemented a software tool that enables automated computational evaluation of the environmental impact and circularity of bridges. The tool is applied within an innovative performance-based design space exploration and multi-objective optimisation framework. A Conditional Variational Autoencoder is trained on synthetically generated bridge alternatives to enable designers to make informed decisions towards more sustainable and circular yet reliable bridge structures. The study proves the framework with integrated LCA and circularity measure valuable for the conceptual design phase and simultaneously identifies challenges for its broader adoption within bridge design.