Assessing the advisory potential of a performance optimisation-based workflow in early-stage architectural design

Sustainable construction is paramount in the current context of a declared climate emergency. However, due to the condensed, output-focused early design stages, there is a gap in introducing sustainable strategies in the inception of a project when they have the potential of generating the largest impact. As the strategic decisions made at this stage influence a building’s performance over its entire lifespan, how can architects ensure that the right characteristics are embedded to maximise a proposal’s sustainable profile?

This research explores the potential of plugging this gap through a generative, simulation-and-optimisation based workflow. This acts as a design advisor to architects- providing proactive support for decision-making and training the users’ intuition in developing context-specific, sustainable proposals. The design advisor workflow is implemented in the Rhino environment and employs the visual programming language Grasshopper, the environmental plug-ins Honeybee and GH_Wind and the Steady State Island Evolutionary algorithm to automatically generate optimised design variants.

The workflow is assessed in a case study environment which replicates the early-stage design process of a commercial office in central Manchester.
The research shows that the workflow exhibits adequate advisory capabilities for decision support, albeit the input process poses a significant barrier for the inexperienced user. As shown through the case study, for well-defined problems, the computer can generate sufficiently optimised solutions the computer can generate sufficiently optimised solutions which set out a formal direction for sustainable design.

However, the case study highlights that in its current format, the workflow requires the user to provide significant input information and interpret outputs, which constitute significant barriers for adoption.
This paper proposes several recommendations to be incorporated in the technical implementation of the design advisor workflow which can lower these barriers and increase uptake in practice.