Smart Manufacturing Assessment on Cybersecurity Risks using Bayesian and Hybrid Fuzzy Logic Networks

Smart manufacturing systems integrate advanced technologies, including IoT, AI, and cloud computing, to enhance productivity and efficiency. However, these interconnected systems expose industries to significant cybersecurity risks. Existing risk assessment methods often lack dynamic adaptability and struggle with uncertainty in cyber threat analysis, leading to inaccurate risk evaluations. To address these challenges, this study proposes a Bayesian-Fuzzy Inference for Risk Assessment (BFIRA) framework, which combines Bayesian networks for probabilistic reasoning and fuzzy logic for handling uncertainties in cybersecurity risk evaluation. The BFIRA model enhances risk prediction accuracy by incorporating expert knowledge and real-time threat data, improving decision-making in smart manufacturing environments. The proposed method enables a more precise, adaptable, and context-aware cybersecurity assessment, helping manufacturers prioritize mitigation strategies effectively. Findings demonstrate that BFIRA outperforms traditional models in terms of accuracy, reliability, and adaptability in dynamic cyber risk scenarios.