See how to:
- Learn how to deploy JMP's interactive quality and SPC capabilities
- Identify routine (common-cause) and abnormal (special cause) process variation over time
- Measure how well a process is performing compared to its given specification limits
- Easily explore stability and capability metrics for a large number of processes across time
- Easily compare stability data Pre and Post a process change
- Learn the basics for using
- Control Chart Builder
- Process Capability
- Manage Spec Limits
- Process Screening
- Attribute and Variable Control Charts with Phases
- Pareto Plots
Questions answered by Clovis Weisbart @cweisbart and Mike Anderson @MikeD_Anderson during the live demo.
Q: Is the range chart a std dev chart?
A: A range chart looks at the running range of the data. You can convert it to an std dev chart in the control chart builder options.
Q: How is the within Sigma determined for the Cpk?
A: Cpk is a measure to show how many standard deviations the specification limits are from the center of the process. The Cpk indices are based on an estimate that attempts to capture only common cause variation. The Cpk indices are constructed using within-subgroup or between-and-within-subgroup estimates of sigma. In this way, they attempt to reflect the true process standard deviation. When a process is not stable, the different estimates of sigma can differ markedly. See the documentation for more details.
Q: Is there a look-up table option for Missing Value Codes?
A: You can define multiple missing value codes in column properties. You can do this for multiple columns using “Standardize Attributes.” You can create a lookup table manually using JSL.
Q: Does the name of the LSL, Target, and USL have to be the same in the spec limit table. For example, if I had a spec limit table with "Lower", "Upper", and "Target" column names, would this not work?
A: The Manage Spec Limits tool does expect those names. The Process Screening Platform does not.
Q: Can you modify/scale the subgroup axis to show only last 50 lots so that the Lot IDs in IR/MR charts are readable?
A: Yes. Just double click on the axis to access the axis properties. Or you can drag axes like you would with other charts. One way to do this would be to modify your first graph X-axis, use Edit > Copy Axis Settings, and then hold down control and use Edit > Paste Axis Settings.
Q: Can we define our own tests?
A: Yes, there is a custom test option. In Control chart Builder, Customize Tests is available in the right-click men for some chart types. It enables you to design custom tests and select or deselect multiple tests at once. After the option is selected, the Customize Tests window appears for designing the tests. Select a test description and enter the desired number (n) and label. You can save the settings to preferences and restore the default settings. Available only for Variables and Attribute chart types. See the documentation for more details.
In Process Screening, Choose Tests enables you to choose the tests that you want to include in the calculation of Alarm Rate and Any Alarm. To select multiple tests, press Alt and click the Process Screening red triangle to open a menu of all platform options. See the documentation for more details.
Customize Tests - Control Chart BuilderrCustomize Tests - Control Chart Builderr
Q: Can I create Levey-Jennings Control Charts with date (where multiple results and missing results for some dates are possible)
A: Yes, Levey - Jennings is an option in Control Chart Builder. You just provide the data and select the chart type. Here is a list of all supported Control Charts.
Q: Do you have any comments on using Ppk vs Cpk? That's been a point of confusion since the labeling changed a few versions ago. What used to be Cpk in JMP is now Ppk.
A: Yes, it is a bit complicated. Have a look at this. Here are more details on the differences.
Q: How would you import specs for a phased chart where the different phases have unique specs?
A: Currently, you are only able to use one set of specs per process.
Q: How do I generate Control Charts with spec limits and not calculate limits?
A: Control limits are generally always calculated. Spec limits are generally set based on customer requirements. You can set control limits using column properties.
Q: I noticed in your column properties list you have "Detection Limits" option. That is not available for me. Why?
A: It was introduced in JMP Pro 16, so you need JMP Pro Version 16 or later.
Q: Can JMP help me use historical data to create specification limits?
A: Yes, you can fit distributions to historical data and also use the Interactive Capability Plot to explore the impact of changing specification limits.
Interactive Capability PlotInteractive Capability Plot
Q: How can I plot control charts for two processes right next to each other? For example, I am using process A from 2020 to date and then in 2022 I made a change to the process and now it's process B. I want to plot these consecutively to show the switch in Xbar.
A: You can do that using a Phases and a Phase column value. Here is an example using Diameter.jmp from our sample data, creating one chart or separate charts to show the processes.
Q: What is the difference between a Control Limit and a Specification Limit?
A: Control Limits are there to monitor the variability of the process to see if the variation is expected. Specification Limits are about the capability of the product and are the tolerance levels defined by the customer or stakeholder.
Resources:
