A comparative study of two proposed CCC-r charts for high quality processes and their application to an injection molding process

作者：Joekes Silvia; Smrekar Marcelo; Righetti Andrea F

来源：Quality Engineering, 2016, 28(4): 467-475.

DOI：10.1080/08982112.2016.1172092

• average run length (ARL)
• car parts plant
• CCC-r control charts
• high-quality processes
• negative binomial distribution

High-quality industrial processes, characterized by a low fraction of non-conforming items, require paying special attention to the statistical control methods employed since traditional Shewhart's control charts are no longer suitable. In this article, CCC-r charts are considered based on the cumulative count of conforming items inspected until r non-conforming items are observed. However, even though these charts have shown to be useful for high-quality processes, they are characterized by a biased average run length (ARL). In order to help engineers interested in this control methodology to select the best option, a computational study of statistical validation was performed to compare the two most outstanding procedures for the cases r = 2, 3, and 4. The performance was evaluated based on the ARL under control. The application of the CCC-r chart to a real process is shown with data from an automobile parts plant. Finally, analysis and discussion of the results are presented.

• 出版日期2016