基于粗糙集理论修正的后续备件指数平滑预测方法

针对后续备件需求预测误差大的问题,提出一种基于粗糙集理论修正的后续备件指数平滑预测方法。根据备件需求数据呈现的趋势,通过拟合确定指数平滑法的次数和平滑系数。从装备在使用过程中影响备件需求数据波动的因素出发,提出了不依赖于基本预测方法的改进预测思路。构建基于粗糙集理论的修正模型。结合算例,对比分析所提方法的优越性,结果表明修正方法可以显著提高预测精度,提出的改进方法不涉及基本预测方法内部特性且无需引入其他辅助方法,通用性较强。

Residual prediction method of subsequent spare parts based on exponential smoothing method and rough set theory

To address the problems of large error of the prediction method in predicting the subsequent spare part requirements, a prediction method of subsequent spare parts based on the exponential smoothing method and rough set theory is proposed. According to the trend of spare part requirements data orientation, the times and smoothing coefficient of exponential smoothing are determined by fitting. From the data fluctuation influence factors of spare parts demand in equipment using, the basic prediction method without depending on the proposed improved prediction is proposed. Then a correction model of rough set theory is constructed. Exploring the effectiveness of the model with example and analyzing the advantages of the proposed method, it shows that the correction method can significantly improve the prediction precision which is different from traditional ideas. This method has good generality without involving internal features of the basic prediction methods and introducing other auxiliary methods.