Glass-box model representation of seismic failure mode prediction for conventional RC shear walls

The recent surge in earthquake engineering is the use of machine learning methods to develop predictive models for structural behavior. Complex black-box models are typically used for decision making to achieve high accuracy; however, as important as high accuracy, it is essential for engineers to understand how the model makes the decision and verify that the model is physically meaningful. With this motivation, this study proposes a glass-box (interpretable) classification model to predict the seismic failure mode of conventional reinforced concrete shear (structural) walls. Reported experimental damage information of 176 conventional shear walls tested under reverse cyclic loading were designated as class-types, whereas key design properties (e.g. compressive strength of concrete, axial load ratio, and web reinforcement ratio) of shear walls were used as the basic classification features. The trade-off between model complexity and model interpretability was discussed using eight Machine Learning (ML) methods. The results showed that the Decision Tree method was a more convenient classifier with higher interpretability with a high classification accuracy than its counterparts. Also, to enhance the practicality of the model, a feature reduction was conducted to reduce the complexity of the proposed classifier with higher classification performance, and the most relevant features were identified, namely: compressive strength of concrete, wall aspect ratio, transverse boundary, and web reinforcement ratio. The ability of the final DT model to predict the failure modes was validated with a classification rate of around 90%. The proposed model aims to provide engineers interpretable, robust, and rapid prediction in seismic performance assessment.