Demand Prediction of LRU Parts with Backorder for SRU

For the problem that the demand of vari-indenture Recoverable Parts did not submit to the Poisson distribution, put forward the Negative binomial distribution to improve the forecasting accuracy. Used the fill rate to estimate the supply degree of Recoverable Parts, the restricted total security funds and the lowest fill rate as the constraint conditions, searching for the fill rate maximization as the objective function, established the demand forecasting model of Recoverable Parts, through the marginal analysis to solve it. The example proves that the model has good prediction effect.