Comparing optimisation strategies of building design students and expert practitioners across subtasks

Optimisation techniques are increasingly used in building design. However, little is known about how optimisation strategies vary during a computational design session. Given that knowledge of optimisation is specialised, the experience level of a designer might significantly influence their strategy, since mental effort required to process information across optimisation subtasks relates to overall strategy. General differences between novice and expert designers are well documented, but little work has explored differences between students and experts while formulating and interacting with a design space using optimisation. In response, a study tasked graduate students from architecture and architectural engineering and expert practitioners to design an atrium using optimisation tools. Key optimisation events were identified to characterise strategies by the timing and frequency of these events. Eye tracking was used to approximate cognitive load via the Index of Cognitive Activity, fixation counts, and fixation durations. Experts exhibited more diverse design strategies and lower average values for the eye metrics, which could indicate an advantage in processing information due to expertise. The proxy metrics for cognitive load were relatively highest during initial parametric modelling and when reviewing optimisation results. These results can encourage emphasis on design space setup and interpretation of results while teaching optimisation.