A comparison of weighted CUSUM procedures for monitoring process proportions with varying sample sizes

Abstract

The conventional binomial CUSUM chart is an efficient tool for monitoring changes in fraction nonconforming when the sample size is fixed. However, it is less efficient in the case with varying sample sizes as the mean number of the nonconforming items is changing over time in this case. In order to deal with the varying mean of the binomial observations, this paper extends the idea of weighted CUSUM (WCUSUM) technique for monitoring fraction nonconforming under varying sample sizes. The comparison results with the traditional binomial CUSUM method and combined Shewhart-CUSUM method favour the proposed method. An example is used to illustrate the application of the proposed method.

Keywords:

• fraction nonconforming
• cumulative sum
• likelihood ratio
• statistical process control

Acknowledgments

The authors would like to thank the editor and referees for their valuable suggestions.