Optimal double sampling control chart based on gauges

Abstract

Control charts based on inspection with gauges are an interesting alternative to monitor quantitative variables when measurement is complex or expensive. The main advantage over variable control charts lies in greater operational simplicity and economy. However, gauge-based control charts are usually less powerful in order to detect shifts in the process than variables control charts, when the same sample size is used. To improve its performance, the use of double sampling on gauge-based monitoring is proposed. The new chart shows better performance than $\overline{X},$ $S$ and $\overline{X}-S$ charts, for the same average sample size, keeping low sampling cost.

Keywords:

• adaptive control chart
• double sampling
• gauge
• optimal design
• statistical process control

Additional information

Notes on contributors

Jaime Mosquera

Jaime Mosquera is an assistant professor at the Statistical School at Universidad del Valle (Cali, Colombia). He is a student of the doctoral program in Statistics and Optimization of Universidad Politécnica de Valencia (UPV), Spain. He holds an MS in Data Analysis Engineering from UPV and a BSc in Statistics from Universidad del Valle. He is a member of the Colombian Statistical Society. His research interests are in the areas of Statistical Process Control, Industrial Statistic, and Statistical Modeling.

Francisco Aparisi

Francisco Aparisi is a full professor at Universidad Politécnica de Valencia (UPV), Spain. He holds a PhD in Statistics and Operational Research by UPV. His major research interest is in Statistical Process Control and optimization by Genetic Algorithms.