Identification of Fatigue using Data Analytics and Machine Learning to Improve Worker’s Health

This study intends to enhance the well-being of construction workers by employing machine learning and data analytics models to detect fatigue at an early stage and develop efficient algorithms capable of identifying proactive fatigue by analyzing a comprehensive dataset that encompasses physiological and cognitive factors. In today's dynamic work environment, characterized by escalating job demands and performance expectations, factors like fatigue, prolonged work hours, and stress can significantly impact employees' physical and mental health. This research study focuses specifically on diagnosing fatigue within the construction industry, where such issues are particularly challenging and unpredictable. The proposed methodology involved the application of machine learning algorithms to analyze the collected data. The dataset used in this study likely comprised various physiological metrics (e.g., heart rate, body temperature) and cognitive assessments (e.g., reaction times, attention span). The results indicate that the Random Forest model has achieved an impressive accuracy of 95.7%, while the XGBoost model demonstrated a precision of 97% and a recall of 57%.