DOE has a language of its own.
- In the world of designed experiments, the terminology is different than most of us are used to in business and industry.
- It is important to know and understand the language of DOE before delving into the techniques.
Major DOE terms include:
- Treatment Combination
- Experimental Run
- Factors are the independent variables of a process.
- Independent variables are the parameters or aspects of the process that we can set or change independently of the settings of another process variable.
- Factors can be related to people, equipment, methods, materials, and the environment.
- A level is a specific value or setting of a factor.
- Levels don’t have to be variable measurements. They can also be attributes.
- A treatment is a factor at a specified level.
- Treatment Combination
- A treatment combination is a set of factors and their levels.
- When conducting a DOE, processes are run with factors set at a specified set of levels.
- The responses are the outputs of the process. Process outputs are dependent variables. Outputs, or responses, can be related to quality, product performance, productivity, or safety.
- Responses are the results of all of the actions of the independent variables, the factors.
- Most DOEs allow us to study several responses at the same time.
- An experimental run occurs when we set a process at a specific treatment combination, run the process, and then collect the response data so it can be analyzed.
- If we change the level of a factor and we see a resulting change in the response, we can say that this factor has had an effect on the response.
- The effect is a calculated value of how much the response changes for a given change in the factor levels.
- Sometimes factors do not behave the same when they are looked at together as when they are alone; this is called an interaction.
- An interaction occurs when the levels of two or more factors are changed and produce a response that is different than the process would produce with the factors changed to those levels by themselves.
- When we run designed experiments, we will use experimental templates to set them up and to analyze them. We do not want to actually make the experimental runs in the order shown by the template; wherever possible, we want to randomize the experimental runs.
- Randomization of the run order is needed to minimize the impact of those variables outside of the experiment that we are not studying.
- Sometime we cannot totally randomize the experimental runs. Typically this is because it will be costly or will take a long time to complete the experiment.
- Blocking means to run all treatment combinations at one level before running all treatment combinations at the next level. Experimental runs within blocks must be randomized.
- Replication is making multiple experimental runs for each treatment combination. This is one approach to determining the common cause variation in the process so that we can test effects for statistical significance.
- A reflection is a new set of treatment combinations that are run at the opposite levels of the original set.