Predicting the Performance of Shear Wall Structures Using the Confidence Nets Model

The recent increase in earthquake engineering employs machine learning technologies to develop prediction models for structural behavior. A reinforced concrete shear wall is one of the most important structural parts of a structure for resisting lateral loads. However, predicting the behaviors of shear wall members and their influence on the structure has always been difficult. Recent studies suggest the use of artificial intelligence (AI) models in this field and considerable amount of attention in the earthquake engineering community, as they can map the complicated relationship between the anticipated damage and the input variables and have shown promising results. In this paper, we aim to push the accomplishments of AI models further by providing more reliable predictions supported by an estimate of a confidence score. Moreover, the proposed model is 185 times faster than the standard finite element analysis method. The model’s predictive performance is also compared with the FEM method.