Design of Experiments (DOE), one of the most valuable Six Sigma tools, is a technique that helps find critical relationships often hidden inside overwhelming amounts of data. Design of Experiments is a structured approach that can identify the factors within a process that contribute to particular effects. DOE can then be used to create meaningful tests that verify possible improvement ideas.
Experiments are Not Just for Science
Most of us remember having fun in high school science class because we were allowed to experiment – which is a wonderful activity for a curious mind. As adults, we became aware that science and medicine were the fields in which experimentation thrived – but science and medicine are not the only ones.
Since the basic principles of cause and effect operate everywhere, experiments can be designed and conducted for any process in any field, be it healthcare or manufacturing or service. Design of Experiments is a structured method for determining the relationships between factors that affect a process and the variable outputs of that process.
DOE also serves to verify if a cause and effect relationship really does exist, and to identify the causes of variation.
How Design of Experiments Works
Design of Experiments within Six Sigma uses sophisticated statistical techniques to understand and control variation. It is this understanding that improves the predictability of business processes.
Experimental methods are first used to quantify previously undefined factors, as well as the interactions between factors. This is done by creating strategic experiments where controlled changes of factors are used to determine which factors have the biggest impact on quality characteristics.
By systematically observing the experiments and statistically measuring the results, critical data can be gathered and analyzed to understand the relative importance of different factors to overall process variability.
The Basic Concepts of DOE
The basic concepts of Design of Experiments are factors, levels, and responses. A factor is simply an independent variable – nothing more, nothing less. In a planned experiment, it is the factors that are deliberately varied in a predetermined manner.
A level is a state of the factor that is deliberately varied. Levels can be discrete (present/absent) or numeric. Experimentation is typically done at two, or occasionally three levels for every factor; each separate level constituting an experimental run. The results of the experimental runs are measured at each run of each factor-level combination. It should be noted that the response can also be discrete or numerical.
A good experimental design ensures multiple factors are varied in an intelligent and controlled sequence. By designing in this way, response data can then be collected intelligently.
The Importance of Randomization
Combining all factors and their levels can easily become too large and expensive of a project, so teams must make informed deductions as to which factors will generate the most useful information that will provide confident results.
To accomplish this, the sequence of runs in the experiment must be randomized. Randomization is crucial to give all external factors an equal chance to affect every run of the experiment.
Non-randomized experiments are at risk of external factors acting in a systematic manner, which can add much unwanted noise to the response. Multiple sets of experimental runs, called replication, will provide even more data and greater confidence in evaluating the results. If an organization’s budget allows, conducting more replications offers the best results.
Successfully designed experiments will show the relationship between the change in level of each of the factors and the change in response. Once these relationships are understood, they can be used to find practical and efficient solutions to process improvement and variation reduction.
Design of Experiments is a crucial part of the Six Sigma methodology that allows organizations to see into the heart of their business processes and what really drives them.
