A multi-objective design tool for decarbonising buildings at the concept stage

Abstract

To enable the move to carbon-neutral building structures with net-zero operation requires a step change in how buildings are designed. Modern buildings are designed by multidisciplinary design teams each of which has competing performance objectives. Daisy is a flexible performance-based parametric design tool which harnesses computational design workflows to enable designer-led multi-objective optimisation, allowing for the design of buildings which simultaneously achieve high performance across multiple performance objectives. A case study of designs for a 63-storey commercial building is presented that demonstrates how early design space analysis facilitates better-performing buildings across architectural and engineering performance metrics and the potential of multi-objective optimisation for enabling the move towards carbon-neutral building structures. Beyond building structures, the design methodology established for Daisy can be applied to a broad range of civil engineering optimisation problems using a flexible and accessible design space formulation that can be adapted for project-specific design requirements. Designs generated using the Daisy methodology demonstrated that by increasing the spatial daylight autonomy of the benchmark building by 14.7%, the building can be moved from the 10% of designs with the highest energy use intensity to lower than 97% of other simulated designs. However, this design change significantly increases the building embodied carbon, with the new design having a building embodied carbon which is higher than 90% of other buildings simulated. Alternatively, by reducing the spatial daylight autonomy by 32.8%, the design can have an energy use intensity which is lower than 75% of other simulated designs, with a 4.4% reduction in energy use intensity, and a building embodied carbon which is lower than 60% of other simulated designs, enabling a 1.3% reduction in building embodied carbon.