基于偏最小二乘回归的飞机维修保障费用预测

Aircraft Maintenance Support Cost Forecasting Based on Partial Least-Square Regression

分析了影响飞机维修保障费用的参数,提出用偏最小二乘回归方法来预测飞机维修保障费用。该方法对变量进行主成分分析、典型相关分析和多元线性回归,在处理存在多重线性相关的小样本多元数据方面效果很好。实例证明,与传统普通多元线性回归方法相比,偏最小二乘回归在飞机维修保障费用预测中精度更高。

Parameters of aircraft maintenance support cost is analyzed, a new cost forecasting method based on partial least-square regression is proposed. The method can be used to analyze the principal elements and typical relativity in parameters, then to regress the principal elements, and it is effective and precise in analyzing multivariate data with few observations. The results calculated show that, compared with ordinary multiple linear regression, the PLSR method has higher accuracy in aircraft maintenance support cost.