Conceptual structural design today relies heavily on the intuition and experience of the structural engineer, often includes an investigation of similar reference projects, and is mostly a time-consuming and demanding task that is characterized by many iteration steps. This research hypothesizes that conceptual structural design as we know it today can be greatly assisted and improved by using modern machine learning (ML) methods.
To that end, we research a multi-step ML approach to improve the design and optimization process of bridges using the case study of network tied-arch bridges. The idea is to utilize pre-processed data of existing bridges (1) and to then identify clusters of continuous and categorical bridge parameters (2). From those clusters, a prior bridge design model is trained (3). Based on the parameter predictions of the prior model a parametric structural model is generated using Grasshopper (4) and subsequently optimized in a numerical fashion (5).