BP组合预测方法在维修保障费用预测中的应用

通过对时间序列预测方法和回归分析原理的分析,建立了适用于武器装备维修保障费用的BP网络组合预测的时间序列多输出模型,并结合某型装备多年的维修保障费用,进行了预测数据仿真,得出的结果与实际的数据相比误差较小,具有较大的理论价值和实际的军事价值,有利于提高维修保障费用的效益,有利于实现装备的优良效费比。

Application of BP Neural Network Combining Forecast to Maintenance and Support Expense

The BP(Back-Propagation)neural network Joint fofecast and timing multi-output model is established on the basis of the theory and method of timing sequence forecast and recursive analysis together with their transform.Data simulation is performed to forecast and propitious to reduce maintenance and support expense and enhance benefit in expense-efficiency,combining with the maintenance and support expense of certain weapon equipment.