Securing Smart Manufacturing With Explainable Clustering for Intrusion Detection

Abstract Artificial Intelligence (AI) and Machine Learning (ML) are transforming manufacturing processes by optimizing efficiency, enhancing quality, and enabling predictive maintenance in the era of Industry 4.0 paradigm. However, this increased reliance on AI-driven systems in smart factories amplifies cybersecurity vulnerabilities, including data breaches and system intrusions. In these critical environments, accurate intrusion detection and the interpretability of model decisions are essential for safeguarding operations. Explainable AI (XAI) has proven instrumental in cybersecurity, particularly in intrusion detection systems (IDS), by regarding AI models’ decision-making transparent for end-users. Drawing from advancements in XAI for cybersecurity, this paper introduces a novel explainable clustering approach that leverages XAI methods to intrusion detection within manufacturing, thereby enhancing transparency, trustworthiness, and operational resilience.