Design of Experiments (DOE)

We need to “Improve the Way We Do Business” to survive in the modern global economy. DOE is a discovery tool to better comprehend the state of the business systems/processed at our organizations and optimize their performance in accordance to business goals. The DOE methodology addresses planning, execution, analysis and interpretation of a set of planned experiments designed to provide the road to improvement and ultimately optimization of the business system/process or product design under consideration. This methodology has been successfully applied countless times to solve quality problems and improve process improvement in basic sciences, economics, engineering and industrial sciences. Effective application of DOE has proven to: improve quality, reduce costs, reduce product/process development time, accelerate the pace of the learning process, reduce variation, and achieve consistent and on-target performance.

DOE allows the experimenters to realistically look at the impact of several factors and their interaction. This is far superior to the one factor at a time approach to experimentation. This leads to a quicker more efficient experimentation strategy. DOE strategies include statistical analysis (Analysis of Variance - ANOVA) to determine the significance of the results assuring that decisions made, based on the result of the experimentation, are sound. This course includes guidance on the application of ANOVA.

Genichi Taguchi made many contributions to Engineering and Quality, most notable were his techniques on applications of DOE. His techniques simplified the Experimental Designs. He introduced strategies such as: Robust Design, Parameter Design and the Signal to Noise Ratio which provide more information on the optimization of product and process designs. This course introduces the use of Taguchi concepts and techniques.

What Will You Learn

Upon completion of this course participants will be able to:
  • Define the DOE terminology and describe the methodology
  • Explain and employ the Taguchi methods for quality engineering
  • Evaluate quality losses due to variation (loss function)
  • Apply design of experiments to practical situations
  • Select the most economical and efficient test strategy during the planning phase
  • Carry out the selected test strategy and analyze and interpret the data generated
  • Conduct confirmation experiments to validate the conclusions drawn from the completed design of experiments
  • Describe and employ the parameter design methodology to make a product or process resistant to various environmental factors that change continuously with customer use
  • Describe and employ tolerance design concepts to achieve quality and cost requirements by strategically adjusting tolerances on appropriate specifications
  • Apply commercially available templates to DOE efforts

Is This Course For You

Product and Process Engineers involved in the optimization of specifications and related product or process performance. Quality professionals involved with problem solving relate to quality issue. Technical Managers who assist their staff in optimization and problem solving activities will be better prepared to support their teams.

Materials Provided

This data is not available at this time

Course Requirements

This data is not available at this time


I. Introductions
  • Instructor’ experience and qualifications
  • Participant’s experience and DOE history
  • Identifying participant issues that DOE could address
II. The Need to Improve the Way We Do Business
  • Consistently meeting customer requirements in today’s world
  • Quality System requirements
  • Quality Costs
  • Lean
III. The Economics of Reducing Variation
  • Conventional cost versus specification limits
  • Taguchi model for cost versus specification limits: the “Loss Function”
IV. The Design of Experiments Process
  • Introduction to the DOE process
  • DOE process step by step
  • Contrasting initial participant’s past experience with what they now see as DOE
V. Test Strategies (120 minutes)
  • Efficient test strategies
  • Recommended experiment design approach
  • Orthogonal array selection and utilization
  • Participants create lists of potential factors for use in experimental designs addressing issues identified at start of program
VI. Conducting Tests (90 minutes)
  • Testing logistics
  • Statistical considerations for conducting experiments
  • Participants identify data gathering methodologies for their real issues and possible errors
VII. Analysis and Interpretation Methods for Experiments
  • DOE process final phase
  • Observation method
  • Ranking method
  • Column effects method
  • Plotting methods
  • Participants complete exercises using simulated data
  • Analysis of Variance (ANOVA)
  • Participants complete exercises on ANOVA using simulated data
  • Templates for data collection and analysis
VIII. Simulated Design of Experiments
  • Experiment setup
  • Conducting the DOE
  • Evaluation
IX. Confirmation Experiment
  • Introduction to confirmation experiment
  • Capability estimates
  • Confirmation experiment decisions
X. Parameter Design
  • Introduction to parameter design
  • Signal-to-Noise ratios
  • Parameter design strategy
  • Case study of parameter design
  • Participants complete exercises on Signal –To-Noise ratios using simulated data
XI. Tolerance Design
  • Introduction to tolerance design
  • Tolerance design using the loss function
  • Tolerance design example