Level: Intermediate
Lee Jong-seon, CEO, IES Engineering Statistics Research Institute
Control charts used for process monitoring have the problem of generating false alarms due to incorrect control charts or control limit settings. These false alarms not only reduce engineers’ or workers’ unnecessary efforts to identify causes and their will to improve the process, but also lead to a decrease in trust in process monitoring. In order to apply control charts, the normality and independence of data must be reviewed in advance so that the average run length in the control state is 370, and the correct control limits must be calculated, but this is often not the case. We propose a method for implementing a method for reviewing control limits using the bootstrap method in JMP.
* To execute JMP Script, the attached Sample01.jmp file must be saved in the C:\Temp folder.
