Uncertainty-based reliability analysis of cutters via improved Bayesian prior distribution

The analysis of product reliability is crucial across various applications. Traditional probabilistic statistical methods often exhibit significant inaccuracies, particularly for estimating distribution function parameters from small sample data. An improved Bayesian prior determination method using error bootstrap is proposed to address these issues. This method extends the sampling range by considering small samples and expanding virtual samples, thereby reducing parameter uncertainty and enhancing the accuracy of Bayesian priors. The proposed method demonstrates advantages in parameter estimation under small sample conditions. Application to machining cutters parameter estimation has shown improved accuracy and reliability assessment precision. This study contributes to enhancing product reliability, increasing equipment utilization, and maximizing economic benefits.