ARCHIVES

The evolution of Industry 4.0 has significantly transformed industrial monitoring systems by integrating artificial intelligence and data-driven decision-making. This paper presents “IndusMind”, a software-based smart factory monitoring and predictive maintenance system designed to identify potential machine failures in real time without relying on physical sensors. The system utilizes simulated operational data, including temperature, vibration, and power consumption, to model machine behavior and evaluate performance conditions. A machine learning approach based on Logistic Regression is employed to classify machine states into normal and failure categories. The model is trained using synthetic datasets and integrated into a Python-based prediction engine, which is connected to a Node.js backend API. A React-based frontend dashboard provides real-time visualization of machine parameters, prediction results, health scores, and actionable maintenance insights.To enhance usability and system security, role-based access control is implemented, allowing different levels of access for operators, maintenance engineers, and managers. The system also includes features such as simulation-driven data generation, prediction control mechanisms, alert logging, and machine-wise analytical views. Unlike traditional industrial systems, the proposed solution operates entirely at the software level, making it cost-effective and suitable for academic and prototype environments. The results demonstrate that the system can effectively predict machine conditions, provide meaningful insights, and support decision-making processes. This approach highlights the potential of integrating with modern technologies to develop scalable and intelligent industrial monitoring solutions.