CQAS Webinars

 

Design of Experiments

 

Session 1 – Introduction to DOE and Simple Analysis

 

Topics include:

• Introduction
  • Why design experiments?
  • Better experimentation though blocking, randomization and replication
  • Strategy of Experimentation
• Simple Analysis with 1 variable
  • Hypothesis testing
  • 1-sample t-tests
  • 1-sample F-test
  • 1-way ANOVA

 

You will benefit by:

• Better understanding of how to set-up experiments
• Better understanding of hypothesis tests and when to use which one
• Ability to analyze and interpret results from a simple comparative experiment

 

Session 2 – 2k Full Factorial Experiments

 

Topics include:

• Introduction to factorial designs and the benefits over one-factor-at-a-time designs
• Introduction to factorial designs at 2-levels with multiple factors
• Main Effects and Interactions
• Analysis utilizing t-tests
• Statistical significance vs. practical significance
• Use of main effect and interaction plots to decide on optimal factor levels
• Model creation and the use of confirmation runs to assess model predictive ability
• Analysis with only one replicate
• Use of normal probability and half-effect plots
• Use of centerpoints for assessing lack of fit

 

You will benefit by:

• Understand the structure and advantage of factorial designs, especially 2k
• Ability to analyze and interpret results from 2k full factorial designs with or without replication
• Understand the benefits and drawbacks of using centerpoints for estimating error as well as checking model fit

 

Session 3 – Fractional Factorial Designs

 

Topics include:

• Need for and creation of fractional factorials
• Design confounding and alias structure
• Determining design resolution
• Analysis of fractional factorials and the effect of confounding on interpretation
• Introduction and justification for fold-over designs
• Single-factor vs. full fold-over designs
• Analysis and interpretation of fold-over designs
• Introduction to screening designs
• Analysis and interpretation of screening designs

 

You will benefit by:

• Ability to create useful fractional factorial designs
• Understand consequence of chosen design resolution
• Ability to fold designs over to eliminate certain confounding patterns
• Create designs to take a large pool of factors and eliminate the ones that are not active
• Ability to analyze and interpret results from the above designs

 

Session 4 – Response Surface Methods

 

Topics include:

• Introduction to response surface methods
• Central Composite Designs (CCD)
• Box-Behnkin designs
• 3-level designs
• Model for response surface designs
• Use of contour plots
• Types of surfaces
• Analysis and interpretation of results

 

You will benefit by:

• Ability to understand and create response surface designs to meet your needs
• Use of contour plots to analyze and interpret results

 

Who should attend

 

Engineers, managers, supervisors and other professionals who would like to get a better understanding of the creation, analysis and interpretation of designed experiments.

 

The software package Minitab will be used for this webinar. It is recommended that you have Minitab installed on your computer, so that you can participate in the hands-on exercises. If you don't have the software, you can download a 30-day trial version from the Minitab website at http://www.minitab.com/products/minitab/demo/.

 

Cost

 

$600 (Includes all four sessions).

Meet your instructor

 

Robert Parody is an Assistant Professor in CQAS in the College of Engineering at RIT. Previously, he worked in industry as a quality consultant and six sigma blackbelt improving processes and products. He has experience in industries such as pharmaceuticals, medical devices, filtration, and aluminum casting. Dr. Parody’s areas of expertise and interest include experimental design, response surface methods, mixture experiments, simulation, and quality control and improvement.