Bayesian Estimation of Ammunition Demand Based on Multinomial Distribution

In view of the small sample size combat ammunition trial data and difficulty forecasting demand for ammunition, a Bayesian inference method based on multinomial distribution is proposed. Firstly, considering different damage grades hitting targets, results are approximated as distribution, model established, which provides theoretical basis multigrade under condition samples. Secondly, conjugate Dirichlet selected prior Dempster–Shafer evidence theory (D-S theory) introduced to fuse multisource previous information. made through Markov chain Monte Carlo Gibbs sampling, at obtained by referring cumulative probability. The study result shows that highly maneuverable can be used predict