Enhanced Machine Learning Classification with Metaheuristics for Early Geotechnical Hazard Warnings

Geohazards represent a multifaceted array of engineering challenges warranting deeper inquiry and exploration. The complexities inherent in these phenomena demand thorough investigation. For instance, the seismic disturbances triggered by earthquakes manifest as seismic bumps within coal mines, soil liquefaction, and slope collapse, necessitating a comprehensive assessment of various unpredictable factors. Devising solutions for these intricate geotechnical issues involves evaluating numerous variables. Furthermore, the aftermath of natural disasters often leads to significant consequences and losses, underscoring the critical need for proactive measures. Implementing an early warning system emerges as a crucial solution capable of potentially saving numerous lives. Such systems can significantly enhance evacuation efficiency by providing alerts mere seconds to minutes before the onset of disasters, particularly in high-risk locations. Moreover, integrating these systems at the planning stages of projects can substantially minimize the life cycle loss of life, material resources, and effort. This study introduces a groundbreaking approach—an innovative classification system that seamlessly integrates swarm and metaheuristic intelligence designed to yield categorical outputs. Accompanying this system is a meticulously developed graphical user interface (GUI), serving as a valuable tool for engineers and researchers engaged in sophisticated data mining tasks. The efficacy of the proposed system is rigorously validated through its application to various geotechnical case studies. Comparisons with empirical methods and previous works underpin its credibility and robustness. Three pivotal contributions emerge from this investigation: the introduction of a pioneering nature-inspired classification system, the development of an intuitive GUI complementing the classification system, and the demonstration of the system's effectiveness in forecasting geotechnical hazards ahead of time—an invaluable asset in providing early warnings and mitigating impending impacts.