What is Design of Experiment

Design of experiment is very useful to identify the new causes who are impacting on effects. This Article will help you to understand the definition of DOE.

What is Design of Experiment

Design of Experiments (DOE)

Design of experiments (DOE) is a systematic method to find the relationship between factors affecting in a process and the output of that process and of varying all the factors of interest and observing impact of these all factors for the desired output. It is also helping us to identify the relationship between cause and effect.

Also, it is used to find cause-and-effect relationships. This information is needed to manage process inputs in order to optimize the output. It is a very important element for Lean Six Sigma methodology.


OFAT Experiment- “One factor at a time”-One factor instead of multiple factors.

Process:- The set activities that can be execute to convert the input into output with the help of resources.

Factors:- It is a variable that potentially effects on response

Level:- It can be a values, that a factors can be taken on.

Effects:- Identify the dependency or the influences the mean response of the interaction between factors or main factors

Main Effect:- When the variable is moving from low to high

Interaction Effect:- At times of the observation when we identify that the response is not only sum total of main effect of both the factors, it is called interaction effect.

Treatment:- The combination of one or more factors

Design Space:- The range of values between the factors

Design points:-The values of the factors in which the experiments are conducted (e.g- A factor has two levels as -1 for low level and +1 for high level

Randomization:- Factor should be taken without any pattern means allow them to vary randomly

Replication:- Replication is the repeated experiment

Confounding:- When we do not conduct adequate number of experiments then the response we observed may be effect of combination of one or more factors, such observed effects are called confounding.

Blocking:- Grouping of the experiments