基于信息熵的抽样特性曲线性能评价

Entropy based Performance Evaluation of Operation Characteristic Curve

抽样特性分析是质量检验中的一个重要组成部分,现行质量检验方法体系中,缺乏对质量特性曲线优劣的定量评价指标,只能进行抽样方案的单因素比较。本文提出了一种抽样特性曲线的评价指标——信息熵评价指标。将计数一次型抽样特性的信息熵评价方法进行了仿真实现,仿真结果表明了基于信息熵的抽样特性评价方法是有效可行的,它使得定量度量抽样特性的灵敏性成为可能,并可对抽样方案进行多变量的联合定量评价。

The operation characteristic analysis is one of the most important parts in Quality Inspection. However, the current quality inspection method system lacks a quantitative evaluation index and it only suits for one-way analysis of sample plan. In this paper, a feasible evaluation indexes for operation characteristic curve is proposed based on the concept of Shannon＇s entropy. The entropy-based evaluation method is implemented for the Single-Sampling-by-Attributes, and the simulation result shows that the method is feasible and efficient. Then the quantitative evaluation of operation characteristic is possible and the quantitative evaluation of muhivariable coalition can also be performed.