Developing ensemble machine learning for estimating and parametrically assessing the moment capacity of ferrocement members

Currently, ferrocement members are being increasingly used in civil engineering projects. However, the literature still lacks investigations on the potential of using machine learning techniques in predicting the properties and behavior of such members. This study fills the gap in the literature by providing a detailed study that investigates the application of ensemble machine learning approaches in estimating and parametrically assessing the moment capacity of ferrocement members. Additionally, a notable contribution of this study is integrating feature importance and partial dependence analysis. These analytical methods allow for a detailed parametric assessment, aiming to understand the influence of each parameter on the moment capacity. As part of the study, several key input parameters are considered through utilizing a comprehensive database derived from tests on ferrocement. These parameters include the width and depth of specimens, cube compressive strength of mortar, and wire mesh’s tensile strength and volume fraction. The ensemble machine learning approach offers potential advantages in terms of accuracy and reliability over traditional methods. This research underscores the importance of advanced predictive tools in civil engineering and their potential to enhance the understanding and prediction of moment capacity in ferrocement members. The findings from this study can serve as a foundation for future research, aiming to further refine and optimize the prediction models for various construction materials and methods.