武器装备维修备件需求预测研究

Research on Demand Forecast of Weapon Equipment Maintenance Spare Parts

对于故障相对多发的装备部件,依靠历史故障数据可以预测备件需求。而针对未来较短期内,部件可能发生未曾发生过的故障,建立向量自回归(Vector Auto Regressive, VAR)模型预测特征参数值,通过比较特征参数值与故障阈值的大小,提前预知设备部件可能会发生故障,进而及早地制定相关备件计划,弥补因经验积累不足造成备件缺失的不足,解决装备的备件需求预测问题。

For equipment components with relatively frequent failures, the demand for spare parts can be predicted based on historical failure data. In the short-term future, components may fail in ways that have not occurred before. A Vector Auto Regressive (VAR) model is established to predict the value of characteristic parameters. By comparing the value of the characteristic parameter with the failure threshold, it is possible to predict in advance that the equipment component may fail. If a failure occurs, the relevant spare parts plan will be formulated as soon as possible to make up for the lack of spare parts due to insufficient experience accumulation, and solve the problem of equipment spare parts demand forecasting.