Teaching Daylight Simulations – Improving Modeling Workflows For Simulation Novices

Abstract

This paper is a follow up study on previously published research that focused on improving the quality and accuracy of simulations conducted by daylight simulation novices. The previous work identified common mistakes that simulation beginners make and proposed a set of simulation guidelines meant to guide users on how to avoid making these mistakes. Based on these earlier findings this paper proposes a set of simulation exercises to teach daylight modeling to simulation novices. Exercises in this teaching toolkit range from simple static metrics such as daylight factor or daylight illuminance, to more advanced dynamic metrics such as daylight autonomy. The proposed teaching toolkit was tested in a semester-long class on daylighting with an enrollment of 17 students during Spring 2012 at the Massachusetts Institute of Technology. The paper evaluates the quality and accuracy of the resulting student models. The study also analyzes whether the remaining errors are reasonable and what impact they have on different daylighting metrics. The latter point should help to clarify whether simulation novices are capable of accurately modeling advanced, dynamic metrics such as daylight autonomy or useful daylight index. Dynamic daylight metrics, although commonly acknowledged to be superior to static metrics, are frequently perceived to be ‘too complicated’ for nonexperts to model. Such perceptions typically support the use of static metrics over dynamic metrics in green building rating systems.