Monitoring the process location by using new ranked set sampling-based memory control charts. Issue 3 (3rd May 2020)

• Record Type: Journal Article
• Authors: Nawaz, Tahir; Han, Dong
• Is Part Of: Quality technology & quantitative management. Volume 17:Issue 3(2020)
• Page Start: 255
• Page End: 284
• Publication Date: 2020-05-03
• Subjects: Exponentially weighted -- homogeneous weights -- median ranked -- neoteric ranked set sampling -- process monitoring; Quality control -- Periodicals; Quality control -- Statistical methods -- Periodicals; Industrial management -- Periodicals; Industrial management; Management -- Research -- Methodology -- Periodicals; Qualitative research -- Periodicals; Management; Quality control; Quality control -- Statistical methods; Periodicals; 658.00721
• DOI: 10.1080/16843703.2019.1572288
• Languages: English
• ISSNs: 1684-3703

Abstract

ABSTRACT: The process monitoring and construction of control charts involves the measurement of the quality characteristic under certain time and cost constraints. Ranked set sampling (RSS) is a very useful and inexpensive method of obtaining a more representative sample when the actual quantification of sampling units is expensive or destructive, while the ranking of the observations is easier. In this paper, RSS and its variation, extreme RSS, median RSS and neoteric RSS are employed to construct new memory type homogeneously weighted moving average (HWMA) control charts to monitor the process location. The HWMA chart assigns a particular weight to the most recent sample, while all the previous samples are assigned an equal proportion of the remaining weight. The run length properties of the proposed control charts are studied by using extensive Monte Carlo simulations. The performance of the proposed charts is compared with the simple random sampling-based exponentially weighted moving average (EWMA) and HWMA besides RSS-based EWMA counterparts. The comprehensive comparisons established the better shift detection ability of the proposed control charts. To demonstrate the practical implementation of the proposed charts, a real industrial dataset is used which also confirms the better mean shift detection ability of the proposed charts.