Assessment of Progressive Collapse Resistance in Reinforced Concrete Structures Using Machine Learning Models

Ensuring structural integrity in buildings and infrastructure under extreme loading conditions represents a pivotal challenge in modern civil engineering. Exposure to natural disasters, accidental impacts, and deliberate attacks can result in the application of unprecedented stresses, which may ultimately lead to progressive collapse and catastrophic failures. While traditional analytical methods are reliable, they often prove inadequate in meeting the increasing demand for rapid and accurate assessments in complex scenarios. However, recent advances in computational tools, particularly machine learning (ML), offer a new approach to address these challenges. In this study, nonlinear static analyses of 250 reinforced concrete systems are conducted within pushdown procedures. Load factor and vertical drift capacities of systems are obtained and accepted as target outputs for ML based predictions. By leveraging data-driven models, it becomes possible to predict structural behavior under extreme conditions with greater precision and efficiency. This study builds on this emerging field, aiming to provide novel insights into collapse mechanisms and robustness through advanced machine learning techniques.