Here is a follow up to the questions of the session:
1. How broad does JMP averaging for the moving range plot in the control chart builder? Can you influence that?
JMP is using all possible data for the average of the moving range plot. You can apply a local data filter to use specific values or you can use a phase variable to differentiate between different subgroup (for instance before and after process changes).
The moving range values itself are calculated by the difference of two consecutive individual values, which cannot be changed.
You can use X Bar and R Charts with a subgroup variable to use more than two values for the range plot.
Hope that helps, otherwise feel free to contact Florian or me here or via Mail: jonas.rinne@jmp.com; florian.vogt@jmp.com
2. Are there strategies to deal with control charts for processes with natural seasonal ups and downs? Does JMP has and specific functions to deal with that?
As far as I know there is no specific function in the Quality module to deal with that kind of long term variation. I would try to use smaller time batches (one month/one week) to make the influence of the long term variation smaller.
Learn how to use JMP's interactive quality and statistical process control capabilities to uncover routine (common causes) or abnormal (special causes) process variations over time.
You will learn how to use JMP to determine whether a process is stable and predictable and how to track the effect of adjusting for abnormal variations.
Introduction and measurement system analysis:
Control charts:
Specification limits, process screening and managing many specification limits on one data set:
Model-dispersed multivariate quality control chart:
