design of experiments with minitab

Design of Experiments (DOE) for Non-Standard Situations

Course Description
Most DOE courses give text book cases as examples. However, there are complexities that arise in real experiments across many industries which require special care or methods. This course  attempts to address this gap to allow the practitioner to know when a special case arises and what remedies may be available.

Examples of some of the complexities in real-world experimentation include:

  • Hard-to-vary factors (e.g. temperature)
  • Several variables at different process steps
  • Certain design level combinations that are not feasible
  • Higher order models
  • More factors than can comfortably be dealt with with standard methods
  • Factors at more than 2 levels
  • Botched experimental runs
  • Discrete responses
  • Covariates that can adjust outcomes or missing data in an experiment
  • How to maintain an optimum value

Who Should Attend
This course is designed for engineers, quality professionals, researchers, and managers who need to use DOE in real-world situations where traditional designs or methods may not be appropriate.

Learning Objectives
Through training, participants will learn the following:

  • Strip Plot and Strip Plot Designs to handle hard-to-vary factors and factors at multiple process steps
  • How to run smaller designs tham what are customarily found in most DOE software
  • How to develop Optimal designs to handle design constraints when a classical design will not work
  • How to handle botched runs, discrete responses, covariates, missing data and maintaining a process at an optimum.

Course Outline
Split Plot and Strip Plot Designs

  • Experiments with Hard-to-Vary Factors
  • Experiments with Factors at Various Process Steps

Optimal Experimental Designs and Advanced Diagnostics

  • Definitive Screening Designs
  • Selecting Designs
  • Minimum Run Res IV and V Designs
  • Max/Min Ratio
  • Design Evaluation Metrics
  • Handling Botched Runs
  • Advanced Diagnostics
  • D-Optimal Designs

Special Topics

  • Handling Discrete Responses
  • Using Covariates
  • Imputation Unclear Resolution IV Designs
  • EVOP Concept

Basic DOE and Advanced DOE or the equivalent

Course Format
10 hours
Instructor-led class training, with opportunities to practice learned skills using prepared data, live demonstrations, and data collected real time in class using computer simulations
Minitab or JMP Statistical Software

Follow QSG on LinkedIn!
Become a QSG Member today!

Always Keep Improving!