Two-Sided Charts for Monitoring Nonconforming Parts per Million

Abstract

Processes with a low fraction of nonconforming units are known as high-yield processes. These processes produce a small number of nonconforming parts per million. Traditional methods for monitoring the fraction of nonconforming units such as the binomial and geometric control charts with probability limits are not effective. In order to properly monitor these processes, we propose new two-sided geometric-based control charts. In this article we show how to design, analyze, and evaluate their performance. We conclude that these new charts outperform other geometric charts suggested in the literature.

KEYWORDS:

• cumulative count of conforming
• geometric control chart
• high-yield processes
• Markov chain
• run sum chart
• sequential monitoring

ACKNOWLEDGMENT

The work of Cesar A. Acosta-Mejia is supported by the Asociacion Mexicana de Cultura A.C.

Additional information

Notes on contributors

Cesar A. Acosta-Mejia

Cesar A. Acosta-Mejia is professor in the Department of Industrial Engineering at ITAM, Mexico City. He is a member of the Mexican System of Researchers (SNI). His research focuses on applications of the change point problem to quality and reliability engineering. He holds a MS in Statistics from The University of Texas at Dallas and a Ph.D. in Industrial Engineering from Texas A&M. He is a member of ASA and IIE.