## DOE, Design of Experiments Tutorial

DOE Definition

DOE is a mathematical methodology to plan and conduct experiments systematically. DOE helps you study the effect of changing multiple variables on an output. DOE changes input variables under controlled conditions in a way that the greatest amount of information is generated, with the smallest sample size and in the shortest time frame.

• Encourages brainstorming associated with the key factors affecting a certain response
• More efficient than the conventional method of doing experiment in which only one variable’s effect on the outcome could be studied
• Determines the optimum settings
• Is supported by several statistical software packages

DOE Main Steps

1. The design of the experiment
2. The collection of the data
3. The statistical analysis of the data
4. The conclusions
5. Recommendations

DOE Terminology

• Replication: repeating an experiment without changing any factors or conditions.
• This is necessary to find out the baseline, systematic error, or noise of the system
• Randomization: a statistical approach to minimize all the potential uncontrollable factors through randomly assigning those factors.
• This is necessary to average out of the extremes
• Blocking: a techniques to control potential block to block variability by breaking the experiment into homogeneous segments (blocks)
• Confounding: is used when multiple effects are tangled together into a parent effect and cannot be separated
• Factors: independent variables that are manipulated so that their effects on the output is investigated
• Responses: dependent variable measured to describe the output
• Treatment Combinations (run): experimental testing in which all the factors are regulated at a certain level
• Fixed Effects Model: when the experimenter specifically chooses the treatment levels, then results will only apply to those levels
• Random Effects Model: when the treatment levels are randomly selected among many possible treatment levels, then the results can be expanded to all treatment levels in the population

DOE Planning

• Define the goal and objectives of the experiment
• Define the questions that need to be answered
• Know the desired results
• Select a proper response (dependent variable)
• Choose the independent variables/factors (quantitative or qualitative) to be investigated
• Select an appropriate experimental design
• Relatively simple to run
• Economically reasonable
• Execute the experiment
• How accurately you collect data is a critical stage, so pay enough attention
• Analyze the data using the most appropriate statistical model
• Interpret the results

Simple Comparative Experiments

• Single Mean Hypothesis Test
• Difference in Means Hypothesis Test with Equal Variances
• Difference in Means Hypothesis Test with Unequal Variances
• Difference in Variances Hypothesis Test
• Paired Difference in Mean Hypothesis Test
• One Way Analysis of Variance

TONEX DOE Training Courses

TONEX offers several training courses in the field of DOE. While the courses’ content will cover all the basics and general concepts and techniques of DOE, they can be tailored based on your needs too.

DOE, Design of Experiments Tutorial