| DOE helps to access through the vast amount of data | | | | The sequence of the experiment, which runs randomly, |
| collected as part of the Six Sigma project | | | | will allow all factors to affect all runs of the experiment. |
| implementation process and in designing experiments | | | | The drawback of a non-random run is the systematic |
| that are required to reduce variations and improve | | | | effect the external factors may have on the |
| efficiency. | | | | experiment. Replication can help provide greater |
| What is DOE? | | | | amounts of data and greater value to the results. |
| Design of experiments is a formal statistical tool that | | | | Successfully completed experiments will bring out the |
| helps to ensure that the testing phase of the project | | | | effect and the change in levels has on the responses. |
| produces data that would be beneficial for further | | | | This helps to understand the best solution for the |
| improvements to the process. The aim is to maximize | | | | process improvement and reduction in variation. |
| return on investment. | | | | Benefits of DOE |
| It is a systematic method to understand the cause and | | | | DOE helps to analyze the data to achieve quantifiable |
| effect relationship among the different factors that | | | | results to undertake experiments and make changes |
| affect the process and the variable results achieved | | | | to the processes to achieve the Six Sigma level. If Six |
| thereby. It even brings out the actual existence of any | | | | Sigma concepts and methodologies are not |
| relation in the cause and effect. Using the advanced | | | | implemented, precisely the expected results would not |
| statistical tools, it enables to understand the variations | | | | be achieved and it will hit the bottom line results of the |
| and to control them. | | | | organization adversely. |
| This helps to improve the predictability of the business | | | | Being a systematic way to approach the experiments |
| processes. | | | | to be undertaken, Six Sigma professionals can ensure |
| DOE is based upon factors, levels and responses. A | | | | the best suited tests, which are conducted and which |
| factor in this case is an independent variable, which is | | | | in turn lead to achieve the expected improvements. As |
| given varied values with the purpose to achieve varied | | | | the aim to set up processes is to achieve the same |
| results. The level is the state defined for that factor. | | | | quality of the products and services, the variations |
| The state varies and will bring out varied results. These | | | | have to be identified and eliminated. |
| results are the responses achieved from the | | | | DOE helps to understand the root causes of variations. |
| experiment at different levels. The results achieved | | | | Being well-equipped with the data on variations, Six |
| can be numerical or discrete values. | | | | Sigma professionals can work towards elimination of |
| In other words, Six Sigma professionals can select | | | | such variations. This ensures the success of the Six |
| those experiment designs which are best suited to | | | | Sigma implementation. |
| achieve the desired results from such experimentation. | | | | Design of experiments is thus an integral part of the |
| When selecting the factors and the state for | | | | Six Sigma implementation, irrespective of any specific |
| experiment, those factors that will generate pertinent | | | | industry. |
| data relative to the expected results will be selected. | | | | |