Font Size
Topic DOE Optimizer with 2 Response Variables

DOE Optimizer with 2 Response Variables

Home Forums General Forums Methodology DOE Optimizer with 2 Response Variables

This topic contains 6 replies, has 3 voices, and was last updated by  Chuck White 3 months ago.

Viewing 7 posts - 1 through 7 (of 7 total)
  • Author
  • #706474 Reply

    I’m working in Minitab 17 and work at an injection molder. I’m studying a molding process molding 30% glass-filled nylon 6/6. I used the Assistant to create a screening study which narrowed 6 potential variables down to a significant 3. The problem I have is that I’m measuring 2 response variables which are opposed to each other, knit line strength and porosity. If I decrease porosity, my knit line strength goes down. If I increase my knit line strength, my porosity goes way up. Porosity likes a cooler barrel temp and slower injection & hold velocities. Knit line strength likes hotter barrel temps and faster injection velocity. I’ve been crawling over Minitab’s website help all day but I can’t find a way to get the model to fit 2 response variables. I’ve done contour plots but I need to find an happy medium that will optimize both response variables without sacrificing either of them. I don’t know if using the Assistant has tied my hands to a relatively simple structure but I’m really getting frustrated. Any help would be greatly appreciated.

    #706478 Reply

    I don’t know that much about Minitab but, assuming it isn’t a typo, your statement “I can’t find a way to get the model to fit 2 response variables” would indicate you need to back up and start again.

    The way one addresses response variables is one at a time. Therefore, if you have two response variables you construct two models, one for knit line strength and one for porosity and you take these two models use them for purposes of prediction.

    Since you only have two models I would recommend you build the models based, not on the actual levels of the independent variables, but on their scaled values where each independent variable has been separately scaled from -1 to 1 corresponding to the minimum and maximum values of range that variable took in the design**.

    If you do this you can look at the equations for the two responses and directly compare the term coefficients. This allows you to see not only which variables are having the most impact on the responses separately but also where the conflict with respect to optimization is occurring.

    In your case it sounds like even if you do this you will not be able to optimize both of the responses so what you will need to do is work with the models to identify the best trade-off settings where neither one is at an absolute optimum but both are acceptable.

    The need to accept trade-offs with respect to absolute optimums is common and the more product characteristics you want to optimize the more likely you are to have to look for independent variable settings that result in acceptable trade-offs in product properties.

    ** the way you do this scaling is to take the minimum and maximum value for each independent variable and compute the following

    A = (max + min)/2
    B = (max – min)/2

    Scaled X = (X-A)/B

    #706488 Reply

    Robert, thank you for your response. It was helpful and informative. I do have two independent models currently and will likely do as you suggest and optimize them separately and come to a happy medium.

    #706491 Reply

    Ieeeeee….don’t do that. If you try to optimize them separately you will spend a lot of time wringing your forehead for no reason and get basically nowhere. What you need to do is take both equations and optimize them together. In other words build a table with all of the X variables and the two Y responses and look at how the two Y’s change SIMULTANEOUSLY as you make a change in a specific X.

    If you first build your equations in the form I mentioned in my first post so the coefficients are for normalized (that’s the old term for scaling between -1 and 1) X’s then you should be able to look at the two equations together and almost eyeball which X is going to be the driver for the need for trade-offs.

    #706494 Reply

    David, your early suspicion was correct: the Minitab Assistant tied your hands. For multiple responses, you will want to use the more powerful functions available in the Stat > DOE menu. You won’t need to start from scratch, but you may need to define your DOE before you can analyze it (Stat > DOE > Factorial > Define Custom Factorial Design).

    As Robert pointed out, you will need to analyze each response separately at first, to create their models. Minitab will store the model for each of your responses, so you don’t need to work directly with the equations if you don’t want to. Once you have your separate models, you can use Minitab’s Response Optimizer (Stat > DOE > Factorial > Response Optimizer) to find the best combination of inputs. You can also add Importance and Weight ratings to each of your responses to get the best trade-off. The Response Optimizer is a very powerful tool that shows the effect of your inputs on your responses graphically, and allows you to make changes to the inputs to see the effect. It’s just not available through the Minitab Assistant.

    #706499 Reply

    Thank you both Robert and Chuck! Chuck, I have fitted both models separately and have the response calcs for both in separate folders. When I launch the Response Organizer, I see that it says that I can optimize up to 25 responses but only the response in that file is there. I don’t see an option to add another response. I thought if I manually enter the results from the other response in the data sheet, it might show up but it didn’t. Under view model, the current model was the only one available.

    I’m suspecting that (as you mentioned), I need to go back and create a custom factorial form scratch. I’m hoping that from there, I can designate two responses and then the optimizer can see them. I’ll give that a shot. Thanks again!

    #706500 Reply

    Yes, you are right. Minitab stores the model information for responses in the data worksheet (in the background, not in the table cells), so both responses have to be in the same worksheet, as part of the same DOE, to analyze them together with the Response Optimizer.

    Assuming the run order is the same in both of your worksheets, you should be able to copy the response column from one worksheet and past it into the other. You will have to re-analyze the pasted response to get the model for it in the combined worksheet.

Viewing 7 posts - 1 through 7 (of 7 total)

Register Now

  • Stop this in-your-face notice
  • Reserve your username
  • Follow people you like, learn from
  • Extend your profile
  • Gain reputation for your contributions
  • No annoying captchas across site
And much more! C'mon, register now.

Reply To: DOE Optimizer with 2 Response Variables
Your information:

5S and Lean eBooks

Six Sigma Online Certification: White, Yellow, Green and Black Belt

Six Sigma Statistical and Graphical Analysis with SigmaXL
Six Sigma Online Certification: White, Yellow, Green and Black Belt
Lean and Six Sigma Project Examples
GAGEpack for Quality Assurance
Find the Perfect Six Sigma Job

Login Form