Clayton Miller is an Assistant Professor at NUS, part of BUDS Lab, a scientific research group that leverages data sources from built and urban environments to improve energy efficiency and conservation, comfort, safety, and satisfaction of humans. He holds a Doctor of Sciences from the ETH Zürich, an MSc. (Building) from the National University of Singapore (NUS), and a BSc. Masters of Architectural Engineering (MAE) from the University of Nebraska - Lincoln (UNL).
ArchDaily had the chance to interview Miller and find out his point of view on how programming and data science can help in improving architecture and construction.
Fabián Dejtiar (FD): What led you to tackle data science in architecture and construction?
Clayton Miller (CM): My career in the building industry has spanned design, construction, and operations with roles in both the academic and industrial fields. I have had many memorable experiences where large amounts of data were being under-utilized. Ten years ago, I started independently exploring how to leverage these data in more robust and scalable ways. I began learning Python programming on my own, with the encouragement of my computer and science-oriented friends. Actually, I got hooked on learning these skills when I started transforming large and messy data for colleagues, executing projects more quickly and easily. I felt like I had superpowers when compared to those using more simplified tools. Ever since I have been giving workshops to building industry experts and, through my role at NUS, I have started teaching these skills in several courses. Most recently my team has launched an online EDx course called Data Science for Construction, Architecture, and Engineering that has 12,000 + participants from over 140 countries.
FD: How important is data science in architecture today? Do you think it can help build better cities?
CM: Architects, especially in universities, are starting to see the power that programming and data science hold, especially when it comes to automating and expanding their design capabilities. Many designers are in fact rethinking their design process and integrating generative techniques that take their ideas, autonomously, in several different directions, allowing them to create more options to analyze and consider.
These techniques can help the designer balance, more comprehensively, challenges such as building performance, passive design, carbon neutrality, and the wellness and comfort of the occupants whhen they opt for an iterative design process.
Having the ability to integrate more data into the process helps us build better cities that can achieve all these criteria simultaneously.
- Related Project:
NUS School of Design & Environment
FD: Part of knowing what to do with data is knowing what questions to ask. How do you realize what are these interrogations? What does your work process look like?
CM: Knowing what question to ask is a fundamental part of the data science process — arguably the most important with respect to the impact that analysis can have! This is the reason why I think that it is important for each profession to learn enough coding for prototyping their ideas without the help of a programmer.
Architects, for example, have been trained deeply in their field and understand the context of what data science process could be used for in terms of creating value.
Giving them the skills to prototype those techniques to a certain level is really important. Once this kind of process becomes valuable for a company, then they can start bringing in larger teams of computer scientists and machine learning experts to put the techniques into production.
FD: Do you believe universities are doing enough to address technology in architecture? or is it more promoted in the professional field?
CM: Universities are just starting to see the value of teaching more technical skills such as coding to all majors and programs. A few years ago, NUS started leading the way by urging every program in the university to develop new curriculums that include classes in computational thinking. In our program, for example, I was responsible to ensure that programming skills are part of this effort. The industry is just starting to understand what these technologies could do for them, and the demand for such skills is starting to grow.
FD: Any advice for those who are interested in following this path?
CM: Of course, I would highly recommend that those interested check out the introductory-level EDx online course to get an overview of how to use Python for various applications in design, construction, and operations. This course is free to audit or can be taken on a verified track that results in a certificate of completion. Python is only one of the tools moving this skill set forward in the built environment — designers can also dive deeply into the world of Rhino/Grasshopper, Revit automation using coding, Microsoft Power BI, and tools like Tableau to expand their skill set.
