Collecting and analyzing empirical data are essential to learning and implementing lessons in structural engineering. Historically, the methods that have been used to analyze and draw conclusions from such data have been limited to statistical and machine learning. The models developed using these techniques are able to capture associative relationships between important variables. However, the intervention decisions geared toward enhancing the resilience of infrastructure should ideally be informed by an understanding of the causal mechanisms that drive their performance. This presentation will advocate for a paradigm shift in structural/earthquake engineering where the language, tools, and models that have been developed to draw causal conclusions from observational data are adopted. Several categories of data-driven structural/earthquake engineering problems that can benefit from causal insights will be examined. Example applications of causal analysis to structural/engineering problems will be highlighted, including case studies where machine learning models are used to establish causal relationships.
Henry Burton (Tue,) studied this question.