Reliability-Box of Systems Under Model Parameter Uncertainty Based on Evidential Variable and Evidential Network

Advanced engineering systems possess a large number of components with complicated failure dependencies. To accurately assess the system reliability, degradation models should be known in advance and model parameters estimated via quantity historical time-to-failure data. In real-world situations, due to limited data, lack knowledge, vague judgments from experts, components’ are, however, inevitably encountered epistemic uncertainty oftentimes quantified as evidential variables. this article, upper lower bounds termed reliability-box, are when elicited experts by first place, constrained optimization is leveraged reliability-box each component giving variable component’s parameters. Next, based on structure, network constructed propagate level level. Therefore, focal elements i.e., reliability bounds, can assessed belief plausibility functions mass function leaf node network. The effectiveness proposed methods demonstrated rolling chip cutting detection module.