DOE-6: Case Study in Creating Full Factorial Design in Minitab: Optimization of Fatigue Strength

Поделиться
HTML-код
  • Опубликовано: 9 фев 2025
  • Hello friends, We are happy to release this sixth video on Design of Experiments (DOE)! In this video, Hemant Urdhwareshe has illustrated step by step, how to create a full factorial design in Minitab software. In the next video, Hemant will illustrate complete analysis of the design in Minitab! We are sure, you will find this video extremely useful for Quality Practitioners, Design and Process Engineers, Six Sigma Professionals, Design for Six Sigma practitioners.
    We recommend that viewers should see our previous videos on Design of Experiments before viewing this video. Here are links to our previous videos on DOE:
    Introduction to DOE: • DOE-1: Introduction to...
    Application of DOE to Spot Welding using Excel: • DOE-2: Application of ...
    Coded and Uncoded Values and Developing Regression Equation: • DOE-3: Design of Exper...
    Fractional Factorial Designs: • DOE-5: Fractional Fact...

Комментарии • 20

  • @thesanitizingcompany7101
    @thesanitizingcompany7101 2 года назад +2

    the best series on doe. Thank you for being alive

  • @vikashkumar-cr7ee
    @vikashkumar-cr7ee 3 года назад +3

    Could you please make a video on three-level and three or more factor DOE analysis

    • @instituteofqualityandrelia7902
      @instituteofqualityandrelia7902  3 года назад +2

      With three and four levels, you need to use General full factorial designs. Instead of three level designs, try adding a center point and then go to RSM if curvature is significant. About video, will see sometime in future.

  • @mohammedjunaidsiddiqui6992
    @mohammedjunaidsiddiqui6992 2 года назад +2

    Kindly explain; reason of choosing factorial design, why not other methods?
    Or in a simple way
    Which method (screening, factorial, response surface, Taguchi, mixture) is suitable when?

    • @instituteofqualityandrelia7902
      @instituteofqualityandrelia7902  2 года назад +2

      Thank you! Screening designs are used when a large number of factors are there and full factorial design is impractical because of the large number of trials. Response surface designs are required if response is a nonlinear function. Hope this clarifies the choice. Watch our videos on fractional factorial design, rsm etc. Here are links:
      ruclips.net/video/-D9sVhq-x2E/видео.html
      ruclips.net/video/MQhf1KFhYCw/видео.html

    • @mohammedjunaidsiddiqui6992
      @mohammedjunaidsiddiqui6992 2 года назад

      @@instituteofqualityandrelia7902 Thanks for clarification.

  • @abhijitwaychal2954
    @abhijitwaychal2954 3 года назад

    What is a need of choosing 3 blocks? what is a significance of blocks? Thanks.

    • @instituteofqualityandrelia7902
      @instituteofqualityandrelia7902  3 года назад +1

      Hello Abhijit,
      Blocks are to be used when a discrete uncontrollable factor exists. In this case it is for batch! Wanted to illustrate it in the example.

    • @abhijitwaychal2954
      @abhijitwaychal2954 3 года назад

      @@instituteofqualityandrelia7902 thanks for answering

  • @abhijitwaychal2954
    @abhijitwaychal2954 3 года назад

    If I have very small quantity of raw material and don't want to replicate experiments in this case can I go for center point experiments ?

    • @instituteofqualityandrelia7902
      @instituteofqualityandrelia7902  3 года назад +1

      It will depend on your design. If the quantity is sufficient for extra runs, yes. Replication is not mandatory. In case of small quantity, choose an appropriate fractional factorial design to minimize runs. For more details, send mail to hemant@world-class-quality.com

    • @abhijitwaychal2954
      @abhijitwaychal2954 3 года назад +1

      @@instituteofqualityandrelia7902 thank for your kind reply and guidance. I had impression that in order to get sum of square due to error and hance F value it is necessary to go for replication. Now the concept is clear. Thanks.

    • @instituteofqualityandrelia7902
      @instituteofqualityandrelia7902  3 года назад +1

      You are correct. You need to have at least one degree of freedom for error to perform ANOVA. So need to assess cost of additional quantity vs benefit. Engineering judgement helps in all cases. One can look at magnitude of effects and reduce the model appropriately.

    • @abhijitwaychal2954
      @abhijitwaychal2954 3 года назад

      @@instituteofqualityandrelia7902 thanks.

  • @TheKurze1
    @TheKurze1 3 года назад

    If samples are from same lot ..shall we go for replicates ?

    • @instituteofqualityandrelia7902
      @instituteofqualityandrelia7902  3 года назад

      Please let me know more information. Replicates are useful to improve power of the experiment, i.e. to be able to discern smaller effects.

  • @paolo7206
    @paolo7206 4 года назад

    Why the 3 blocks?

    • @instituteofqualityandrelia7902
      @instituteofqualityandrelia7902  4 года назад +1

      I suggest you watch the previous videos on the subject. You will find explanation. The three blocks are for three batches of material.

    • @paolo7206
      @paolo7206 4 года назад

      @@instituteofqualityandrelia7902 thanks