In this ever-changing world the Total Quality Management (TQM) is an important way to manage the organization. It involves management, engineering system and empowerment of all people of the company.
This paper has the objective of improving the solution of problems that exists in the production environment and must be solved to get better process or products. Everything is done quickly and the root cause must be eliminated as soon as possible.
There are a lot of decisions based on common sense, feeling, and old experience or just based on reasons. By the way we cannot forget that the interaction between factors in a study is usual.
The decisions based on a small sample during a test in the middle of the production line can be done, but some effects not expected must be avoided. The not expected effects are caused by one or another factor not considered at the beginning.
To support those decisions in a proper time and cost, considering the important factors, is necessary to design some tests, or experiments. Statistical methods can greatly increase the efficiency of these experiments and often strengthens the conclusion so obtained.
By the statistical design of experiments, we refer to the process of planning the experiments so that appropriate data will be collected, which may be analyzed by statistical method resulting in valid and objective conclusions.
Don't be overzealous in the use of complex, sophisticated statistical techniques. Relatively simple design and analysis method are almost always best.
In this paper we will introduce to persons one way to implement in few stages a procedure to execute DOE in the shop floor and to present cases of study in the aluminum tube production process.