A control chart is proposed for monitoring the capability of a process when sensitivity data are all that are available on a particular measure of interest, such as the threshold height at which a dropped bottle will break. Sensitivity data are data that are collected as either pass or fail of a sample at a certain level exposure to a control variable, such as drop height. Methodology is presented for selecting a certain quantile of the distribution of the measure of interest and monitoring that quantile with a standard p chart. The trade-offs between increasing the power of the control chart and decreasing the number of measurements are explored.
- Binomial distribution
- Destructive testing
- Power of the test
- Statistical process control
ASJC Scopus subject areas
- Safety, Risk, Reliability and Quality
- Industrial and Manufacturing Engineering