Predictive analysis of pharmaceutical equipment

he pharmaceutical industry demands high standards of equipment reliability to ensure product quality and operational efficiency. This study explores a predictive maintenance framework that integrates machine learning and real-time video analysis to monitor equipment health and prevent failures. The system comprises three main functionalities: training a machine learning model to predict the Remaining Useful Life (RUL) of equipment based on historical sensor data, manual input for RUL prediction, and real-time video monitoring to detect equipment malfunctions. A RandomForestRegressor is employed to model RUL prediction using sensor data such as pressure, temperature, and fan speed, while the YOLO object detection model analyzes video footage to identify anomalies and potential hazards. This approach enables early detection of issues, reduces unplanned downtime, and improves the overall reliability of pharmaceutical manufacturing equipment. The integration of sensor data, predictive algorithms, and visual monitoring forms a robust system aimed at enhancing operational continuity, maintaining product safety, and ensuring regulatory compliance in the pharmaceutical sector. Key Words: Streamlit, Predictive Maintenance, Model Training, Random Forest Regressor, Real-Time Video Monitoring, YOLO Object Detection, RUL Prediction, Machine Learning, Video Upload, Video Processing, Equipment Performance, Data Preprocessing, Model Deployment, Sensor Data, Video Annotation, Preventive Maintenance.