Predictive Models of the Steel Beam Capacity Using Artificial Intelligence – Preliminary Studies

Artificial intelligence based models, namely using machine learning techniques, have been widely used in different engineering fields, providing accurate, easy‐to‐apply and fast assessment tools to predict the mechanical behaviour. These models not only can be used to increase the inference about the mechanical phenomena of a certain problem but also, for example, allow to quantify the capacity of steel beams, as studied in this work. For that purpose, different models were developed and trained using machine learning algorithms, and it is demonstrated that the capacity predicted by these models surpass the accuracy given by existing analytical models, when using the results obtained with the finite element method as the baseline.