Maintenance of industrial machinery is a critical aspect of ensuring smooth operations and reducing downtime in manufacturing and production environments. Traditional maintenance approaches, such as time-based or condition-based maintenance, often lack precision and can lead to unnecessary downtime or unexpected failures. This paper presents a comprehensive solution for machine maintenance monitoring that leverages the Internet of Things (IoT) technology and machine learning techniques. The proposed system integrates NodeMCU, an opensource IoT development board, with a suite of sensors capable of monitoring critical parameters, including temperature, sound, and vibration, on milling machines. Sensor data are collected in real time and transmitted to the ThingSpeak IoT platform for storage and analysis. The platform facilitates the aggregation and processing of data streams, enabling seamless integration with machine learning models. Machine learning models, including Anomaly detection algorithms, are employed to analyze sensor data and predict maintenance needs and anomalies. This predictive maintenance system harnesses historical data for model training and adapts to changing machine conditions. Maintenance alerts and notifications are generated when deviations from normal behavior are detected, empowering maintenance teams to proactively address potential issues. A user-friendly dashboard provides real-time insights into machine health, allowing operators and maintenance personnel to monitor key parameters and respond promptly to maintenance alerts.