Design of Experiments Training | DOE Training for Non-Engineers

Design of Experiments Training, DOE Training for Non-Engineers

Why TONEX’s Design of Experiments Training? DOE Training Course Description

The design of experiments training, DOE training for non-engineers course is designed to teach you how to analyze root causes of failures using applied statistics. DOE also will provide you the ability of problem solving, designing experiments, conducting them, and analyzing the results. Through this hands-on course, we will teach you how to make analyzing complex, multilevel systems simpler without loosing the accuracy of analysis.

Since statistics and mathematical analysis is not usually taught to non-engineering majors at school, professionals with no engineering background might find it hard to conduct such analysis at work. The design of experiments for non-engineers seminar is ideal for those non-engineer individuals who need to use root cause analysis, problem-solving, optimization, and/or designing products/processes in their job, while they don’t have a very strong math background.

To give you a brief description of Design of Experiments (DOE), it is a powerful tool that helps you evaluating the effects of multiple input factors on a target response variable. The advantage of using DOE over the conventional One-Factor-At-A-Time (OFAT) method of studying various factors’ impacts on an experiment is that you can change several factors at the same time and then observe and analyze their effect on the results simultaneously. With DOE, you can study a combination of full factorial or fractional (partial) factorials at the same time.

There are four main problem categories that DOE can be used in: Comparative, Screening/Characterizing, Modeling, and Optimizing. During this seminar, you will get introduced to all of these problems and learn how to deal with them in DOE. During this 2-day workshop, we will teach you how to design and experiment, how to manipulate the input factors to get the desired results, how to set up the process windows, and how to use robust designs. The design of experiments for non-engineers course will cover both the theory of designing effective experiments and a lot of hands-on practices on simulated and live processes.

Audience

The design of experiments for non-engineers training is a 2-day course designed for the individuals who need to apply statistics in their work, yet have no engineering background. The workshop will benefit quality managers and technicians, scientists, research and development personnel, marketing and business analysts, and testing and manufacturing professionals.

Training Objectives

Upon the completion of this seminar, attendees will be able to:

• Design, execute, and interpret effective experiments
• Identify root causes of a failure
• Identify the effect of various input factors on the response variables (output)
• Develop effective solutions and improve the process to eliminate the current problems
• Select appropriate design options and tactics to optimize and accelerate the process of getting the desired results
• Apply DOE to improve the efficiency of the day-to-day operations and project management
• Understand the principals of DOE
• Understand the principals of experimentation
• Design the experiments step by step
• Understand and execute different features of DOE based on the situation
• Understand basic statistical principals
• Conduct and apply t-test
• Assume properly for t-test and evaluate their assumptions
• Properly anticipate how many observations they need for an experiment
• Choose the proper multiple comparison procedure for their situation
• Allocate their observations among their k treatment groups
• Understand and perform F-test
• Understand the principals of blocking
• Handling Randomized Complete Block Design with missing data
• Exploit Latin Square Designs
• Apply Graeco-Latin Square Design
• Crossover Designs and their special clinical applications Balanced Incomplete Block Designs (BIBD)

Course Outline

Introduction To Design of Experiments

• Historical background of DOE
• The Basics of DOE
• Planning, Executing, and Analyzing
• Preparation steps

Components of Experimental Design

• Factors
• Levels
• Response

Purpose of Experimentation

• Alternatives comparison
• Effect of input factors on output
• Optimal Process Output
• How to reduce variability
• How to get to the desired output
• How to increase the robustness of process or product
• Balancing tradeoffs

Requirements Prior To DOE Execution

• Problem and objectives statements
• How to know DOE is the right method
• Selecting desired response variables and the factors affecting them
• Actual and surrogate responses
• Experiment logistics
• Test set-up and data collection planning
• Selecting and evaluating a gage

 Comparative Experiments

• Simple comparative experiments
• Determining sample size
• Power determination

Single Factor Experiments, ANOVA, In A Completely Randomized Design (CRD)

• Single factor experiments in multilevel systems
• Determining sample size
• Multiple comparisons
• Dunnett test and its optimum allocation
• One-way random effects models
• Linear test

Blocking

• Principals of blocking
• Various blocking scenarios
• RCBD
• How to deal with RCBD’s missing data?
• The Latin Square Design
• Replicated Latin Squares
• How to deal with a system with more than 2 blocking factors?
• Crossover Designs
• Incomplete block designs

Basic Factorial Designs

• Factorial designs with 2 treatment factors
• Factorial design example

The 2^k Factorial Design

• Estimated effects and the sum of squares from the contrasts
• Un-replicated 2^k Factorial Designs
• Transformations
• The simplest scenario

Confounding And Blocking in 2^k Factorial Designs

• Blocking in an un-replicated design
• The 2^3 design
• Blocking in replicated designs
• Split-Plot
• Blocking in 2^k Factorial designs
• Examples to practice

The Two-level Fractional Factorial Designs

• Fractional Factorial Designs
• Analyzing a Fractional Factorial Design step by step
• Foldover Designs
• Plackett-Burman Designs

The Three-level And Mixed-Level Factorials and Fractional Factorials

• 3^k designs in 3^p blocks
• Mixed factorials

Response Surface Designs

• Multiple responses
• Response surface designs
• Mixture experiments
• Experiments with computer models

Robust Designs

• Robust designs parameters
• Crossed array design
• Combined array design

Experiments with Random Factors

• Random effects models
• Two factor factorial and random factors
• The two factor mixed models
• Fining expected mean squares
• F tests approximation

Nested and Split Plot

• The two-stage nested design
• The general m-stage nested design
• The split-plot designs
• The split-split-plot design
• The strip-plot designs

Design of Experiments Training | DOE Training for Non-Engineers