Optimizing Robotic Manufacturing in Industry 4.0: A Hybrid Fuzzy Neural Bayesian Belief Networks

Abstract

In the era of Industry 4.0, robotic manufacturing systems must adapt to dynamic and uncertain environments, where optimal decision-making is crucial for operational efficiency. This paper presents a novel hybrid decision-making framework that integrates Fuzzy Neural Reinforcement Learning (RL), the Best-Worst Method (BWM), Levenshtein Distance, and Bayesian Belief Networks (BBN) to optimize robotic manufacturing processes. By combining these methodologies, the framework effectively handles uncertainty, enhances decision-making, and accelerates learning in complex manufacturing scenarios. A comprehensive system formulation is provided, along with the development of an optimization algorithm that integrates these components. Numerical simulations demonstrate the framework’s performance, highlighting its efficacy in reducing operational costs, improving production quality, and strengthening adaptive capabilities. The results show that the proposed model outperforms traditional approaches across diverse manufacturing scenarios.