RELIABILITY INFERENCE VIA QUANTUM BAYESIAN NETWORKS WITH EXPONENTIAL DISTRIBUTION MODELING

This study explores the application of Quantum Bayesian Networks (QBNs) to reliability analysis, focusing on systems with variable modeled using exponential distribution. A four-node Bayesian structure is proposed, integrating both discrete and continuous variables. The continuous node is encoded using quantum amplitude encoding, while probabilistic dependencies are implemented using controlled quantum gates. Failure probability is estimated via Iterative Amplitude Estimation (IAE) and compared to a classical Monte Carlo simulation. Results show that the quantum method provides more precise estimates with narrower confidence intervals, despite a slight difference in the mean values. The findings highlight the potential of QBNs to enhance reliability assessments by capturing complex conditional structures with high precision and fewer computational resources. This proof-of-concept contributes to advancing quantum methods within the domain of operations research and reliability engineering.