Hi folks, I am culling the 2016 info a bit to keep suggestions for 2017 and ask you to give some more.
Here are suggestions so far, July 27, 2016:
1. Building Formulas - Jerry Coooper jerry.cooper based on his 2016 US Summit presentation with/about Exxon.
2. Building Dashboards - Suggested by Daniel Valente danielvalente covering:
* Building a custom dashboard using the Dashboard Builder.
* Adding a title and company logo to your dashboard.
* Saving the dashboard as an Add-In for sharing with others who have JMP.
* Saving the dashboard as interactive HTML to share with others who don't have or have access to JMP.
* Combine windows to create a dashboard.
* Using filters and graph filters to enrich your dashboard.
* Advanced customization of the dashboard to add tabs, change the layout, use summary mode.
* Connecting to a data source through the query builder and building a dashboard automatically with fresh data with the post-query script.
This list is just a quick brainstorm of the use cases to fill the hour. I think that is more than enough to cover in a 60 min presentation.
3. Market Basket - Mary Loveless FlyGirl is working with Matt Flynn on a presentation for Summit that might evolve into a Mastering. Spoke with Mary Aug. 29 and she thinks there is not enough here for a full session.
4. Advanced Methods for Warranty - Scott Wise scwise thinks there might be new features added in JMP 13 that might lead to a new Advanced Mastering demo. Erich is doing an Advanced Mastering Reliability Using JMP Pro in October 2016 after JMP 13 is released. JMP Pro gives you the ability to examine the reliability relationship between a system's components, understand how component reliability contributes to the success or failure of a system, and simulate and forecast the repair needs and costs of components.
5. We are breaking Reports and Journals into two different webcasts. Journals could take the whole 45 minutes and there are new report options. With 13, is there still two much about reports for a single Mastering session? Mary will consider this.
6. This webcast covers: reliability block diagrams, repairable systems simulation and other tools exclusive to JMP Pro.
Erich, erich.gundlach do you plan to tied this to warranty or is that a separate modeling demo based on outgorwth of data you might get during your demo?
7. Customizing JMP for Your Organization - Scott Wise scwise may consider doing a how-to based on the popularity of Brady's Add-In, and which goes beyond the Add-In. A reminder that we still have a few SEs mentioning the Six Sigma add-in we used to have. Scott, might that be a good example
8. DSD - If Jones/Nachtsheim add-in gets built into JMP 13, Scott Wise scwise and Tom Donnelly think there may be a new or upgraded demo there. Some talk about going beyond what Reliasoft offers organizations like GE, aircraft, military, etc. I am not sure there is enough for a full webcast, though. Tom tom.donnelly , perhaps incorporate it into your DSD and eliminate some of the background to make time for the new stuff?
9. Six Sigma - Scott Wise scwise will see if any interest in case studies using DMAIC are requested from or useful for customers. SEE #4 ABOVE.
10. Consumer Research - Peter Hersch phersh suggested this. Jian does and Advanced Mastering. With JMP 13, do we have need for one Mastering and a deeper dive for Advanced Mastering? Scott Wise scwise also said Advanced Consumer Research topic may be good to revisit with the added text mining capabilities coming in JMP 13.
11. Defect Reduction - Peter Bartell peter.bartell@jmp.com suggests broadening Mike Anderson's michael.anderson Defect Reduction. Mike is glad to do it, but has put a call out for data to use. Peter Bartell and Ian Ian@JMP said: I see managing defectively on products that have a two-dimensional structure as very valuable, and something we have all the capabilities for (unlike some other software . . . ). So, FWIW, I would see this as a great addition. Chuck chuck.boiler or ian Ian@JMP, can you help with data?
On Dec. 20, 2017 Peter closed the issue after looking at the 2016 Defect Reduction videos. Peter wrote:
I re-watched Mike’s videos. Here’s my take. I think his first video is a good example of spatial defects analysis in a planar product/process system…regardless of shape of the original object being evaluated, the underlying process creating a ‘defect’, or the type of defect.
I don’t think it’s a big stretch at all for a competent engineer working in a roll to roll coating process environment to take what Mike has shown and ‘make it their own’ with their process and data. I don’t think, at least from my seat, it’s worth the effort to create ‘new’ data tables or case studies for people that might work in this roll to roll coating process environment.
Here’s why: The coating process and product engineers I worked with at Kodak and some of our suppliers for roll to roll raw materials (like Felix Schoeller for paper, or Alcoa for aluminum) have lots of technology and Jidoka-like systems in place for these problems. Hence even in our Mid Atlantic sales patch, for say W. L. Gore or DuPont, which do lots of roll to roll coating work, I think it would be relatively straightforward for them so envision how Mike’s demo could work for them.
I think the biggest challenge quite frankly would be getting defect data OUT of the data warehousing systems these companies all have in place…lots of machine vision type systems ‘capture’ this data in real time. The trick for a engineer/JMP user would be getting this data out of the what I’ll call the ‘source system’ into JMP…but that’s a Query Builder’ish track…not what Mike was after in his demo.
12 August 11, 2016 from Nick Shelton nick.shelton : For 2017 Mastering JMP webcast, I would like to do a presentation on control charts and using the new process screening feature (JMP13) if no one else isn’t already doing those topics. Chuck has offered to mentor the presentation. Nick is going to look at the new Tracking and Trending Mfg. Metrics by Byron to see if he can complement that session, drill into one of the topics as a full webcast, and/or use different data.
13. Do we have a good business problem for non linear platform? Sue Walsh Aug. 17 had a request for webcast or materials on this topic from Arved Harding, Eastman Chemical. When I asked him what he was looking for he wrote:
Good afternoon Gail, I just made a presentation to our group of statisticians on using JMP’s non-linear fitting platform. I think it went well.
I used a nonlinear sample data set, Algae Unstacked, for my example to make it go quick. I use nonlinear modeling in many ways and love the capability JMP has added with the Fit Curve options now available and expanding in JMP 13. I think this capability has literally jmped your product into another world compared to most software. (Pun intended) JMP’s drive toward visualization paired with appropriate statistical techniques make this capability very value adding. And the ability to compare various nonlinear models for the same data and compare parameters for different grouping variables, such as subject, are wonderful additions to the nonlinear model building worlds.
I think it would be appropriate for JMP to host a webcast on this topic, and good timing would be soon after the JMP 13 release to subtly show the new features. .
Here’s some thoughts.
Often times the response is a curve rather than a data point or an average, that is, measurements may be taken over something like time or temperature on the same person or sample. Nonlinear models are sometimes the best choice because they:
- Fit the data better than other models such as polynomial models
- Are consistent with first principles in the appropriate sciences
- Are extrapolatable beyond the factor space. (This can help with scale up issues from pilot plant to manufacturing.)
My recent examples of using nonlinear models are related to weathering studies where samples are measured over time while being exposed to weather conditions like temperature, UV light and humidity. This type of study can be thought of as a repeated measures problem, but I often find fitting a nonlinear curve to the data and using the parameters from that curve for further study can be very helpful. For example, imagine a designed experiment where for each run several points are collected over time. If a nonlinear model is used, then the parameters can be responses and a model may be developed for each response (parameter). Then using a desirability function the “optimum” curve can be found. This can lead to a factor setting that generates a curve that has never been seen before which can lead to new discoveries.
A similar example in weathering is using a control every time a study is done. Then every study produces a control curve. One can monitor the parameters over time in a control chart individually or even in a Multivariate control chart. (FYI I am writing this problem up to be published in a chapter in a book soon.)
My point is that nonlinear modeling and the parameters from that can be very useful information and JMP provides some great tools for this.
I don’t know how hard it would be to get permission to use my examples and data but I don’t think it is necessary to move forward with a webcast and teach people the value of JMP’s nonlinear platform. In Sue’s paper and presentation entitled “When a Linear Model Just Won’t Do: Fitting Nonlinear Models in JMP” she describes some scenarios when nonlinear models are appropriate and does a great job of illustrating JMP’s capabilities. I think a presentation on that paper alone can be very valuable to users and potential users.
14. Bill Worley is interested in a Mastering (not Advanced like Brady's) session on Gen Reg using categorical data. Will build it upon a presentation he gave at P&G JMP Day Sept. 1, 2016.
15. Ruth is considering doing a session on Journals.
WHY WE PRESENT LIVE MASTERING
Mastering sessions are part of Customer Care activities to help keep customers learning and accounts growing. They are designed for people who have attended Getting Started with JMP, perhaps are still using the New User Welcome Kit, have some on-the-job experience using JMP for real problems, and want to go the next step to learn more. Advanced Mastering is for experienced JMP users with some background in the topic of the day. We don’t answer basic questions at the Advanced Mastering sessions and so far, have gotten the correct audience.
FREQUENCY
Each topic is presented live 3x/year. Some people present 2 topics, each 2x/year.
ATTENDANCE
Attendance numbers are attached. They are pretty much as expected – data prep and graphical exploration are popular. DOE basics and Multivariate Analysis are sure draws. Advanced Mastering numbers are lower – this audience is specialized so that is what we expect.
For 2017, please:
1. Based on your knowledge about what beginning customers need, suggest any new topics that we can cover in one hour (or in two one-hour parts). Kemal has already volunteered to do a Neural Nets session (single-layer using JMP; multiple-layer using Pro).
2. If you suggest a topic, let me know if you will be willing to present it. I will schedule all new webcasts for the end of 1Q or beginning of 2Q to give you time to prepare. Each topic will be presented live 3x/year.
3. If you are willing to present 2 topics (or a Part 1 and Part 2 of the same topic), I will schedule each topic/part 2x/year for a total of 4 live presentations from you.
4. Let me know if you are willing to present your topic from last year.
Message was edited by: Gail Massari
Message was edited by: Gail Massari
