Approximate distribution of demerit statistic-A bounding approach

Authors:
Fengming M. Chang;Long-Hui Chen;Yueh-Li Chen;Chien-Yu Huang
Affiliations:
Department of Information Science and Applications, Asia University, Taichung, Taiwan, ROC;Department of Logistics Management, Shu-Te University, Kaohsiung, Taiwan, ROC;Department of Industrial Engineering and Management, Cheng-Shiu University, Kaohsiung, Taiwan, ROC;Department of Information Management, Shu-Te University, Kaohsiung, Taiwan, ROC
Venue:
Computational Statistics & Data Analysis
Year:
2008

Citing 5
Cited 0

Average run lengths and run length probability distributions for cuscore charts to control normal mean

Computational Statistics & Data Analysis
Zero-inflated Poisson model in statistical process control

Computational Statistics & Data Analysis
Distribution strategies for the contextualized mobile Internet

Electronic Commerce Research and Applications
Using recurrent neural networks to detect changes in autocorrelated processes for quality monitoring

Computers and Industrial Engineering
An accurate algorithm to compute the run length probability distribution, and its convolutions, for a Cusum chart to control normal mean

Computational Statistics & Data Analysis

Quantified Score

Hi-index	0.03

Visualization

Abstract

The traditional classical demerit control chart is used to plot the demerit statistic, a weighted sum of the number of defects in each category, on a control chart. The approximate normal method is usually used to obtain control limits though the distribution that depends on the values of the weights and the parameters of the Poisson distribution which may not always be normal. [Jones, L.A., Woodall, W.H., Conerly, M.D., 1999. Exact properties of demerit control charts. Journal of Quality Technology 31 (2), 207-216] used the characteristic function approach to determine the distribution of the demerit statistic. Unfortunately, the process that they used to determine the distribution needs complex integral evaluation via mathematical software packages or using the approximate truncated infinite series. Moreover, the characteristic function does not provide an accurate result easily. In this paper, a bounding approach is proposed to determine the approximate distribution of the demerit statistic. It is easy to implement and also the approximate error can be controlled to meet the desired accuracy. In addition, an example is demonstrated to illustrate the proposed method. The results indicate that the proposed approach is efficient and accurate. Finally, the performance among the approximate normal method, the characteristic function approach, and the proposed bounding approach are discussed.