JMP Wish List

What inspired this wish list request?  Dealing with "one-sided" specification limits which have second limit just for sanity check purposes and not being able to set it as natural/boundary limit in JMP.     What is the improvement you would like to see?  Let set Spec Limit lower/upper limits as natural/boundary limits.   It is possible to have limits which cannot be exceeded (for example due to physical reasons) but they are still set (for example for sanity check reasons). In JMP if we are using Spec Limit column properties, this would mean that we have to delete such limits as JMP cannot set then as natural limit (they should be ignored in capability calculations and values beyond those handled as missing values).    One suggestion how it could look like   Why is this idea important?  Makes it easier to perform capability analysis as you wouldn't have to always remove the specification limit from column properties. ... View more

What inspired this wish list request? Part of my job is analyzing data from OQ/PQ validations for injection molding. We utilize control charts to assess stability and capability of CTQ dimensions per cavity. The information i need for this is the control chart, histogram with spec limits, capability summary, and assessment of normality. The legacy control chart i like the best because it includes a normal quantile plot and allows for preference control for shading zones and tests. The new IMR chart does not have preference control or normality test so i need to combine information from other sources which is messy and time consuming.     What is the improvement you would like to see? I would like to the see two items. First, the capabilities of the legacy control charts preference settings for the new control charts as the legacy will be discontinued in JMP 19. Second, there should be a selection to view the normal quantile plot on the histogram and/or the goodness of fit tests (anderson/darling or shapiro-wilk).     Why is this idea important? It is imperative that your data is normal when assessing capability. Including normal quantile plot or the anderson-darling/shapiro-wilk test will provide supporting evidence in a single area. The preference control will help make analysis quicker. Certain items arent necessary to show and deselecting them everytime is time consuming.     ... View more

Hey @mia_stephens , these go along with my EWMA and CUSUM posts   What inspired this wish list request? I'm using the multivariate platform more but understanding what is driving out of control points is not an easy task and there are different features in the MDMVCC vs the norma Multivariate control chart platforms that would be nice to have live in the MDVCC.     What is the improvement you would like to see?  Be able to save out T^2 values and T^2 formula Have options to highlight out of control rows Have an option to add an out of control row indicator column Add a parallel plot option with mean curve (like Multivariate) Add ability to show the average of the selected groups or selected out of control points Show the mean value (red in this example) Be able to add groupings that will show up on parallel plot from the score plot (currently can do group A and B, add in up to n items) Be able to have stages like control chart builder Have red bars on mean contribution charts that at least one of the selected samples was out of control on the univariate chart. (vs current meaning that the average is out of control) Add in the change point detection option we have in Multivariate control charts     Why is this idea important? These options will allow determining what is driving out of control situation to be an easier task as well as allow staying in one platform vs having to jump between MDMVCC and MVCC platforms.   ... View more

What inspired this wish list request? I have been using the EWMA platform more recently and believe these improvements will make the user experience more efficient and less frustrating.     What is the improvement you would like to see?  Add stages option like control chart builder Allow local data filter and column switcher to work Shift lines Have different colors for up shift and down shifts lines Have ability to have the shifts work like CUSUM vs just when the trend line cross the mean/target line Have platform ignore excluded rows as a red arrow drop down menu item Make it so the chart size doesn’t reset when you change a setting Have an option to show UCL and LCL values on the EWMA chart (use constant limit numbers) Have an option to show UCL and LCL values on I-MR/X bar chart Add ability to have a “head start” EWMA(-1) value This will allow if you have rolling data for the trend line to not keep restarting at the target value and thus changing the nature of the control chart Add in being able to save limits in a tall table format Add in an ARL profiler that’s would have ARL(0) and ARL (Delta) be optimized for values of Lambda and K sigma Add K Sigma to the parameter platform vs being buried in the red drop down menu items     Why is this idea important? This will remove inefficiencies in the platform and allow more capabilities for the end user.     ... View more

What inspired this wish list request? I have been using the CUSUM platform more and have found that there are some gaps between it and other Control Chart platforms that can make the user experience frustrating.     What is the improvement you would like to see?  Have ARL show Shewhart ARL like the EWMA platform Have options to highlight out of control rows in the red drop down menu Have an option to add an out of control row indicator column in the red drop down menu Add stages option like control chart builder Add in I-MR/X Bar charts like in EWMA platform Add option to show k*sigma (from CUSUM) band on this chart.  This will help show when points are adding or subtracting from the CUSUM totals Be able to use Local data filter and column switcher Be able to have CUSUM reset after a certain number of runs are within the K*sigma band This would allow the CUSUM control charts to quickly recover back to showing in control  like we get with a EWMA chart Another option could be having the CUSUM reset after a "flag value" kinda like the EWMA will reset after a missing value or excluded row.  So for example, 999.9 could be the reset value for positive sums and -999.9 could be the reset value for negative values.  These could be set as a column property or as a entry from the red drop down menu. Have platform ignore excluded rows as a menu option red drop down menu (in case you forgot to click it in the set up menu) On ARL profiler, need to show what the ARL(Delta) is Would be great to be able to have the ARL (0) show as well, maybe as a second Y axis in the profiler, like when you fit multiple Y’s in Fit Model. Also would be great to be able to use the desirability function to find the best setting for a given ARL(0) and ARL (delta) for a set k value. Need to change the behavior of the ARL Profiler so that only the parameter of h or k is change when one of the values are updated in the Parameter control box. Right now if you change one it will change both, which is annoying if you are trying to update the parameter table to match the ARL profiler so that you can see all the ARL performance on the ARL table. Update the ARL profiler so that the values you can enter always return an ARL value   Why is this idea important?  The CUSUM control chart is getting more use but could use some additional updates to make it as functional as the Control Chart builder platform.  Also, since the correct ARL value is critical for setting up the charts correctly more tools that get away from a guess and check workflow would be great. ... View more

What inspired this wish list request?  Making small mistakes when launching JMP Platforms and I already have implemented this to my platforms.   What is the improvement you would like to see?  I would like to see "Keep dialog open" checkbox in all of the JMP platforms which have a launcher and close it when you press "OK". Other option would be to provide some sort of "Preview" for the platforms but this would be much more complicated to implement to not slow down platforms too much.     Why is this idea important?  Making very small mistake in JMP launcher forces you to find the launcher from menus again and pressing Recall or you can use View/Recall Last Platform (hidden inside menus and has very bad shortcut by default as it forces you to remove your hand from mouse). ... View more

What inspired this wish list request?  Unlike other statistical software (Minitab), it seems that JMP doesn't provide detailed formulas for analyses. This lack of details and transparency can make it difficult for users to understand the underlying calculations.     What is the improvement you would like to see?  Provide more formulas for analyses where it makes sense to include them in the support documentation and in cases where formulas are quite complex, simplified formulas for simplified examples so the user can understand the concept.   Why is this idea important?  understand the underlying calculations, it is quite often to get request to explain the analysis, how do you derive this conclusion and what kind of analysis and fundamental concept formulas are being used in the analysis.   ... View more

Hello, I know there has been similar posts and requests on this topic, sorry about that. Some applications (e.g. in spectroscopy analysis) require doing signal deconvolution using a simple combination of multiple gaussians (see capture attached). I have seen alternatives suggested here (by using the distribution plateform, or with PLS). Does it mean that the multiple gaussians fit method will never be implemeted in JMP (or in a custom add-in)? Thanks Best Regards    ... View more

What inspired this wish list request? Please describe the current issue that needs improvement or the problem to be solved that is not easy or possible right now, with an example use case.    Often you conduct comparison and MSA over hundreds of characteristics or columns. In the MSA output, f y/x, variability chart, matched pairs add an output similar to the "Process Screening" platform. So you can review across many columns the specific parametric data, and focus on those that are significant.    As is, in these platforms, you cannot screen results. It would be very helpful to be able to screen these results.   1 more suggestion, in response screening, allow you to factor in spec limits and percent of tolerance shifts, in addition to fit / Rsq significance.   What is the improvement you would like to see? Please describe the idea for improving JMP. Please include mock-ups, wireframes, screenshots, scripts, other documents or examples from other software that help describe the change you would like to see.      Why is this idea important? Please describe the value to you and/or other users if the idea is implemented (for example, ease of use, must have,…).     ... View more

What inspired this wish list request?    Add more options to error bars or indicators in graph builder for box plots. It would be great to be able to add error bars in multiples of sigma (not just 1). I.e. 4 (represents Cp 1.33).       What is the improvement you would like to see?      Why is this idea important?      ... View more

What inspired this wish list request? Please describe the current issue that needs improvement or the problem to be solved that is not easy or possible right now, with an example use case.    Please make Goal plot calculate correctly for unequal bilateral tolerances. It does not calculate percent correctly. You need to remove the nominal to make the goal plot work.   Also please add more features to goal plot, allow to modify individual data points, give different row labels, and modify the lines. Should give multiple ranges of PpK, allow to put multi-levels.   What is the improvement you would like to see? Please describe the idea for improving JMP. Please include mock-ups, wireframes, screenshots, scripts, other documents or examples from other software that help describe the change you would like to see.      Why is this idea important? Please describe the value to you and/or other users if the idea is implemented (for example, ease of use, must have,…).     ... View more

What inspired this wish list request?  There are some confusions between %R&R and %P/T in JMP variability chart output     What is the improvement you would like to see?  Replace the wording Gauge R&R in the first output, because when spec are entered, the third column is % of tolerance and the figure for the line Gauge R&R and the column % of tolerance is actually a P/T (and several Customer believe it is the %R&R which does not use the tolerance in the calculations, AIAG definitions).   Why is this idea important?  because confounding %R&R and %P/T is a real issue (in the first table of the journal, 2.17 is %P/T, not %R&R   ... View more

What inspired this wish list request? I am trying compare the Within Sigma estimate of variation between groups, and I am unable to do so since I cannot generate confidence intervals for those estimates.   What is the improvement you would like to see?  The Capability platform now helpfully includes 95% Confidence Intervals for the various Cpx and Ppx estimates, but I can’t get the same type of information for the Within Sigma estimate.  Ultimately, I would like to be able to easily obtain confidence intervals for the Overall and Within Sigma estimates in Process Capability in all platforms where that is used.  (Distribution, Control Chart Builder, Process Capability, etc.)  It would also be beneficial to generate CIs for the Stability Index to provide a quick assessment of how this ratio compares to 1.   As an editorial comment, it is harder than I would like it to be to generate the confidence interval information for the Overall Sigma Std Deviation, but I can do it in the Distribution Platform using the Confidence Intervals option.  (Would like it to be an option for the summary statistics as well).  I can also get there in Fit Y by X under the Means and Std Deviations option in JMP 17, but you have to right click the table and add the columns to the report – and not a lot of people know you can even do that.   Why is this idea important? Confidence intervals are important so that we can understand the uncertainty associated with any statistical estimate.  Without this information readily available, people will often make incorrect assumptions - especially when sample sizes are small.     ... View more

What inspired this wish list request?    Sometimes, one doesn't have access to the raw data, but rather a summary of it, commonly provided as mean and standard deviation numbers - it would be great to be able to visualise such data in graph builder, but currently the only way to plot ranges associated with an average value is to have the raw data (individual data points) so that JMP can do the stdev calculation (among others).   What is the improvement you would like to see?    Functionality in graph builder to accept a column with average values and a column with range (eg, std deviation) associated with each value.   Why is this idea important?    There must be many users who don't always have access to source/raw data? ... View more

What inspired this wish list request? While using the summary table, I have to find work arounds for interacting with the raw data table with selections that area easy to make in the PS summary table.  For example, if I wanted to just select the columns that have out of spec conditions there is no easy way to do that.     What is the improvement you would like to see?  Add the ability in the red arrow drop down to select the columns in the associated data table from the highlighted "Column" rows in the PS summary table  In the select where options in the red triangle , add in "not missing" or add in an "Invert selection" option Add in the ability to lock the highlighted "Column" rows so that if you change anything on a local data filter, the same "Column" rows will remained highlighted after the filtering is finished.  Right now it moves around (and I not sure what logic it uses) Add in the options for "Process Screener" under the red arrow options of "Launch for Selected".  This goes along with second item above (the item above is how you would selection items when you then want to use the red arrow option of "Remove Selected Items", this suggestion allows you to remove unselected items) When the sorting order of the PS summary table is changed, have the "Charts as selected" and "Charts for selected" order update to match the new order. Allow selecting rows in the data table with out of spec and/or Alarms to be selected vis a red arrow button for the highlighted "Column" rows in the PS Summary table. On the "Charts as selected" and "Charts for selected" have preferences for the Control limits and Spec limit formatting that are platform specific Be able to sort the PS summary table by more than one column Have the ability so that on "Charts as selected" the axis settings will not get updated for the X axis as you select other "Column" rows     Why is this idea important? This will improve the user efficiency and experience as well as allow better presentation and exploring of the outputs from the platform.    Thanks for considering Steve   ... View more

What inspired this wish list request?  Simplicity and efficiency of process screening platform compared to measurement system analysis.   In my earlier positions I was used to running measurement system analysis using more or less complicated macros (minitab) or more or less complicated scripts (jmp). JMP does offer easy way run MSA but to get usable report you have to do more work and if you wish to analyse results further, it is even more work.    What is the improvement you would like to see? I would like to have MSA platform which is similar to Process Screening. User would determine on launch of the platform what to calculate and how to get limits (use limits table/use limit columns in same table/use column properties) and so on... Then when user presses OK, it will create similar table as we can see in Process Screening, lots of information, fairly easy to sort, we can select rows and perform extra actions, we can add more columns to result table and so on...    Why is this idea important? Makes the calculation of MSA / gage easier for multiple parameters and makes analysing results from that single result table possible.       I can provide some ideas and mock-ups what could be included but it will take me multiple hours of work so I would rather first hear is there any point in this request. ... View more

What inspired this wish list request? I use the tree map as a the base graph to report out a dashboard of summary statistics for multiple areas in my plant.  The current squarify option leaves some "boxes" almost useless.   What is the improvement you would like to see? I'd like to see a "forced square" option in the Layout drop down menu that would make the boxes all the same size, which could mean that there are blank boxes if there are not enough categories to fit a even row/column matrix.  Ideally every box would look like chem mill stacked or heat query stacked.  You can see on dry times stacked that there is one long and skinny box that is hard to see.         Why is this idea important? Dashboards are becoming key tools for conveying operational status quickly so charts that are easy to see all the elements are vital.  This is also critical as we move to JMP live as dashboard style graphics are a main reason for the move.     ... View more

What inspired this wish list request?      Desirability is a highly effective concept for making a multi-criteria choice. For the moment, JMP only allows this concept to be used on the basis of a model. It would be very useful to be able to use it on a finite list of options (e.g. finding the best possible flat to rent from a list of offers).   What is the improvement you would like to see?    the manage limits tool allows you to manage specs limits efficiently. It would be very useful if it could also manage responses limits. Ideally, it should also be able to save the desirabilities one by one as well as the overall desirability.       Why is this idea important?    Stephane George (JMP system engineer) created a script for that purpose. Its addin meets the need perfectly well.     ... View more

In many case, there is the needs to generate random data with different shape (Normal, Weibull, Lognormal,...) and parameters. For example: one dataset (1000 data)  normal distribution (Average = 10 and Standard Deviation = 1.1) and one data set (2500 data) Weibull distribution (Shape Paremeter = 1, Scale Parameter = 0.1, Threshold = 0.1). This need is linked to some simulations (capability, ppm, correlation,...)  we need to do with not real/concrete/process data in our hands.      What is the improvement you would like to see?  I would like to have a button in some menu (Table or Analysis) -> push the button -> open the menu where I can select the type of Random data Distribution and set number of data to be generated and the parameter (see the attched file)   Why is this idea important? Please describe the value to you and/or other users if the idea is implemented (for example, ease of use, must have,…). - Have the possibility to do some analysis and evaluation without real/concrete data; - Easy menu for data generation; - Flexibility on different data distribution needs;     ... View more