Determination of a new sampling system based on exponentially weighted moving average

Abstract

In this article, a new sampling system based on an exponentially weighted moving average statistic is proposed for the inspection of normally distributed quality characteristics. The proposed sampling system can be applicable for compliance testing. The performance measures of the proposed system are derived and analyzed. Tables are also constructed for determining the optimal parameters of both known and unknown standard deviation cases. An optimization problem is formulated as a nonlinear program in order to construct the tables in which the objective function used is the minimization of the average sample number and the constraints are related to the probability of acceptance at acceptable quality level and limiting quality level under the operating characteristic curve. The efficiency and advantages of the proposed sampling system are discussed over the existing variables sampling plans. Extensive tables are provided for industrial applications. Implementation of the proposed sampling system is discussed with the help of a real-time industrial example.

KEYWORDS:

• Consumer’s risk
• EWMA
• nonlinear programming
• producer’s risk
• sampling system
• variables sampling plan

SUBJECT CLASSIFICATIONS:

• 62D05
• 97K80
• 90C30

Acknowledgments

The author thanks the Editor-in-Chief and the anonymous reviewers for their valuable and constructive comments that led to an improved presentation of the article.