Firearm product small sample reliability evaluation method based on Bayesian maximum entropy

The invention provides a firearm product small sample reliability evaluation method based on Bayesian maximum entropy, and the method comprises the steps: S1, obtaining generalized knowledge about the reliability of a firearm product, extracting the feature information of a reliability parameter of the firearm product, and constructing a corresponding distribution feature constraint; s2, solving the maximum entropy prior distribution of the reliability parameters of the firearm product according to the distribution characteristic constraint; s3, acquiring life data from the small samples of the firearm products through a reliability projectile test; s4, constructing a likelihood function; s5, obtaining posterior distribution of reliability parameters of the firearm product through Bayesian inference; s6, updating the life distribution of the firearm product according to the posterior distribution; and S7, constructing a firearm product reliability function as a final result of firearm product small sample rel