Epistemic Uncertainty Quantification in Product Degradation Modeling Considering Manufacturing Defects

Manufacturing defects significantly impact the early-stage reliability of products, leading to accelerated degradation and unexpected infant failures. Traditional reliability assessment methods primarily focus on statistical failure data, often neglecting the epistemic uncertainty introduced by product interior defects originated in manufacturing. Thus, an extended degradation model that incorporates product manufacturing defects into the reliability evaluation framework is proposed. First, by introducing defect-sensitive factors and utilizing an uncertainty theory-based degradation equation, the quantitative model characterizing the influence of manufacturing defects on product performance over time is established. Second, within the framework of belief reliability, a modeling method for reliability and failure rate considering manufacturing defects is proposed. Finally, a case study of an aerospace actuator component is provided to verify the effectiveness of the proposed approach in capturing the degradation behavior and predicting infant failure risks.