针对发动机平面度的二维 Fused LASSO 多元统计控制图

摘要

A new multivariate statistical process control ( MSPC) chart methodology is developed based on the two-dimensional Fused LASSO ( least absolute shrinkage and selection operator ) for quality control in which the mean shifts have some natural ordering in two-dimensional plane .Because of the two-dimension-al Fused LASSO penalties on the difference between successive variables , it is a desirable tool to identify the natural ordering among variables .The numeric simulation study shows that the two-dimensional Fused LASSO control chart can more quickly and efficiently detect mean shift of such structure , especially where the shift value is large .Finally an application to the engine flatness data shows that the two-dimensional Fused LASSO control chart is applicable and useful .%针对具有二维平面自然顺序关系的多元数据，提出了一种基于二维Fused LASSO回归模型的多元统计过程控制图。由于二维Fused LASSO多元回归模型对相邻系数的差异进行惩罚，因此它对检测出变量间的自然顺序关系非常有效。通过数值仿真实验可以看到，二维Fused LASSO控制图能更加快速有效地检测出具有二维平面自然顺序的多元过程中均值发生的变化，且在偏移量越大的情况下优势更加明显。最后通过对发动机平面度数据的实例验证，给出了二维Fused LASSO控制图的步骤和适用性。