We are excited to share today, PhD BT candidate, Keith Lee's work. Here are some of his thoughts- @keith_j_l2 My Master's degree in structural engineering at McGill University was focused on bolted steel connection design. I was amazed at how such a simple system creates an innumerable number of possible outcomes. This problem led me to develop computational workflows to better understand the relationship between variables, and introduced me to the concept of the design space and the work of Professor Caitlin Mueller. I think in general, the openness of MIT is what helps me strive. I can take courses across disciplines without special permissions or approvals; I can ask an expert down the hall to grab a coffee and pick their brain. I'm surrounded daily by people who work across (seemingly) unrelated disciplines that give and take expertise freely. My current work is focused on quantifying the gap between what a structure wants (demand) and what is being built (supply), and the material and carbon cost associated with this difference. The hard part is that we can't easily borrow concepts from Econ 101 to model this difference: demand isn't measured by a quantity of something, but rather a combination of local/global geometry and induced forces/stresses in a structural element. The trick is to find suitable abstractions of these complex systems to find meaningful ways of characterizing multi-dimensional information. Shown in the pictures is one useful abstraction for a specific problem: how different are the force transfer requirements at the connections of a spatial truss structure? The last two images are of ongoing work attempting to sample variations in load demand at the scale of full buildings. We see again a significant variation in load demand, as well as a statistical lifecycle analysis of the potential carbon cost of not meeting this variation with an equally varied range of supply.