Training

Design of Experiments

Course Description: Design of experiments (DOE) is a structured approach to efficiently characterize, improve and optimize a process by systematically changing several factors (e.g. temperature and pressure) simultaneously and analyzing the results. Designed experiments may be used during manufacturing or process development to: solve problems during a manufacturing crisis, optimize process performance to improve yield, reduce experimental cost, time and resources, identify the key factors with greatest impact on process performance, manufacture products that consistently meet specifications and are robust to variability (Taguchi), analyze results easily and unambiguously, and efficiently bring new manufacturing equipment on line.

Engineers should be able to plan a DOE and analyze the results after completing the course.

Course Duration: Three days

Table of Contents

Part 1. Overview
  1. What is DOE?
  2. Example of DOE
  3. Advantages of DOE
Part 2. Preliminary Planning
  1. Experimental Planning
  2. Selecting Responses
  3. Selecting Factors
Part 3. Full Factorials
  1. Full Factorial Designs
  2. Data Formatting in JMP
  3. Main Effects and Interactions
  4. Graphical Analysis
  5. Regression Models
  6. Example of Full Factorial
Part 4. Practical Considerations
  1. Randomization
  2. Sampling and Replication
  3. Homogeneity – Blocking
  4. Practical Considerations
Part 5. Fractional Factorials
  1. Fractional Factorial Designs
  2. Example of Fractional Factorial
Part 6. Optimization and Robustness
  1. Response Surface Designs
  2. Example of Central Composite Design
  3. Robust Design Experiments