考虑了装备使用时间、行驶里程和配备时间等影响备件消耗的多种因素,依据装备备件的消耗特点,在分析偏最小二乘回归方法原理的基础上,运用方法对小样本数据条件下装备备件的消耗数量进行预测.应用示例表明,偏最小二乘回归方法比传统多元回归分析法、逐步多元回归分析法和删除多元回归分析法具有更高的预测精度. According to the consumption characteristics of equipment spare parts, based on the analysis of the principle of partial least squares regression method, the method of using equipment spare parts for small sample data .The results show that partial least squares regression method has higher prediction accuracy than traditional multiple regression analysis, stepwise multiple regression analysis and deletion multiple regression analysis.