论文信息 - A note on the average run length of cumulative sum control charts for count data

A note on the average run length of cumulative sum control charts for count data

This note presents preliminary results on the average run length (ARL) of a Cumulative Sum Control Chart (CUSUM) used to monitor defects or counts, when the count distribution is Neyman Type-A. The ARL for this distribution is contrasted to the more customary Poisson model for counts or defects. We show that the ARL curve of a Neyman Type-A CUSUM always lies below the ARL curve of a Poisson CUSUM.

Douglas C. Montgomery | John S. Gardiner | David J. Friedman

[1] E. S. Page. CONTINUOUS INSPECTION SCHEMES , 1954 .

[2] C. Stapper. The effects of wafer to wafer defect density variations on integrated circuit defect and fault distributions , 1985 .

[3] James M. Lucas,et al. Counted Data CUSUM's , 1985 .

[4] J. Edward Jackson,et al. All Count Distributions Are Not Alike , 1972 .

[5] Douglas C. Montgomery,et al. Introduction to Statistical Quality Control , 1986 .

[6] D. A. Evans,et al. An approach to the probability distribution of cusum run length , 1972 .