A Nonparametric Repetitive Sampling DEWMA Control Chart Based on Linear Prediction

IEEE Access

Ambreen Shafqat¹  Zhensheng Huang¹  Muhammad Aslam²  Muhammad Shujaat Nawaz³ 

[1] Department of Statistics and Financial Mathematics, School of Sciences, Nanjing University of Science and Technology, Nanjing, China;Department of Statistics, Faculty of Science, King Abdulaziz University, Jeddah, Saudi Arabia;Higher Education Department, Government of the Punjab, Lahore, Pakistan;

关键词: Binomial distribution;    control chart;    DEWMA;    EWMA;    linear prediction;    repetitive sampling;   

DOI  :  10.1109/ACCESS.2020.2989132

来源: DOAJ

【 摘 要 】

Distribution-free control charts can be useful in statistical process control (SPC) when only limited or no information about the distribution of the data of the process is available. In this paper, a linear prediction related double exponentially weighted moving average (DEWMA) sign control chart using a repetitive sampling scheme (RSNPDEBLP) has been considered for a binomially distributed process variable to improve the efficiency of detecting small drifts in its place of small changes. The proposed RSNPDEBLP control chart is assessed in average run length (ARL) for the various values of sample sizes. The efficiency of the proposed RSNPDEBLP control chart is compared with the existing EWMA and DEWMA sign control charts using single sampling and repetitive sampling schemes in terms of ARLs. When there are small changes in the process after the stabilization period, the proposed control chart is used to control small trends rather than small shifts.

【 授权许可】

Unknown