Logo
 
Quick Links
Courses
What Is DOE?

Statistical Design of Experiments is a powerful body of statistically based techniques that allow a scientist or engineer to efficiently characterize a design space and to establish the link between cause and effect for processes that have multiple input factors and associated interactions. Through judicious selection of sampling points within the design domain and statistical analysis of results, a person can simultaneously measure and identify those factors that impact a process and then can optimize their effects.

Any experimental scenario with identifiable causative factors and one or more measurable responses (R) can be a candidate for DOE and associated model development. The major difference between test programs that use DOE and more traditional test programs is the simultaneous, patterned variation of several contributing factors as opposed to one-factor-at-a-time processing.

Why Use DOE?

Multi The key methodology that makes all the disciplines work together to yield the best overall formulation is called Desirability. Desirability is the latest advance in optimization technology.

Smart D Technology
    DOE helps you:
  • Systematically analyze the nature and cause of variation using controlled testing;
  • Maximize the research benefits of testing by producing high-quality data;
  • Minimize the resources needed to perform a test program.
  • (DOE-based test programs typically cut cost and time expenses by 50%.);
Course Objectives
Orth
    Students will be able to:
  • Plan, execute, and analyze a two-level statistically designed experiment;
  • Understand and assess more complex two-level statistically designed experiments such as those presented by Air Force contractors;
  • Be cognizant of advanced DOE methodologies that go beyond the basic two-level designs;
Course Topics and Descriptions
The SynGenics two-day Basic DOE course presents all the critical elements of two-level factorial designs. Course topics include:
  1. Quick walk through a simple designed experiment;
  2. Important pre-test planning;
  3. The Pre-Test Planning Process;
  4. Identifying factors, levels, interactions, and responses;
  5. Hidden dangers (discussed throughout);
  6. How to use Orthogonal Arrays and Triangular Tables;
  7. DOE data-analysis methods;
  8. Model building and use;
  9. Incorporating Analysis of Variance (ANOVA) into DOE;
  10. Adding Significance and Confidence with DOE;
  11. Survey of advanced methods;

In-class group exercises walk each participant through the DOE process in progressive levels of complexity, as new material is covered. By course end, the participant will be well equipped to design and conduct a sophisticated two-level factorial experiment and evaluate the resulting data using a variety of statistical analysis techniques including ANOVA.

Course Cost

Learn to set up designed experiments and perform DOE analysis in two days. Cost is $700 per student.

Supplied Materials

Each participant will receive:

  1. A notebook of the course;
  2. CD containing several basic DOE Excel spreadsheets;
  3. An electronic (.pdf) copy of the National Institute of Standards and Technology (NIST) Engineering Statistics Handbook, (also available as a download on http://www.itl.nist.gov/div898/handbook/

About the Instructor

John C. Sparks has nearly 34 years of experience as a practicing scientist, engineer, and manager. He recently retired from the Air Force Research Laboratory where he was a key proponent of DOE. Mr. Sparks designed and presented several AFRL-wide workshops including DOE, SPC, Metrics, Myers-Briggs Type Indicator, and a variety of safety/security briefings. He brings over 15 years of AFRL training experience into the DOE classroom. Mr. Sparks joined SynGenics Corporation in September, 2007.

Tinker-Toy Catapult Diagram

Tinker Toy This is a catapault designed by SynGenics during one of the courses. Click the image to see a full size version of the catapult.

Statistical Design of Experiments is a powerful body of statistically based techniques that allow a scientist or engineer to efficiently characterize a design space

DOE-based test programs typically cut cost and time expenses by 50%.

Two-day course does quick walk through a simple designed experiment.

The instructor, John C. Sparks has nearly 34 years of experience as a practicing scientist, engineer, and manager.