An Integrated Smart Manufacturing Framework Based on Industrial Monitoring and Artificial Intelligence

Industry 4.0 is transforming traditional production environments and requires operational systems to be more intelligent, responsive, and capable of supporting complex autonomous decision-making. In this context, the present work introduces a new conceptual framework designed to enhance current industrial monitoring systems by integrating artificial intelligence technologies. The study begins with an analysis of the existing system, focusing particularly on the quality workflow. This process is modelled using a BPMN diagram to clearly represent the interactions between system stakeholders and the flow of information. Based on this analysis, a prediction module was developed by combining real-time data collection from sensors in the production line with neural network models that are able to implicitly learn complex and non-linear relationships between defects or failures and their root causes. After validating the performance of the predictive model, a new intelligent framework is proposed. It includes improved BPMN representations that define the new architecture, interactions, and data flows of the proposed system. The resulting solution is capable of processing larger and more diverse datasets in real time, generating more relevant insights, and supporting proactive decision-making. By integrating industrial methods and artificial intelligence, this system contributes to reducing non-quality costs and improving overall industrial performance through early defect prediction.