Doe keys

Solo disponible en BuenasTareas
  • Páginas : 13 (3149 palabras )
  • Descarga(s) : 0
  • Publicado : 23 de mayo de 2011
Leer documento completo
Vista previa del texto
Mark J. Anderson and Shari L. Kraber
Consultants, Stat-Ease, Inc., Minneapolis, MN (e-mail:

This paper identifies eight keys to success in applying statistical tools for design of experiments (DOE). Quality managers who grasp these keys will be better able to support use of DOE in their organization. Ultimately this willlead to breakthrough improvements in product quality and process efficiency.



Design of experiments (DOE) provides a powerful means to achieve breakthrough improvements in product quality and process efficiency. This leads to increased market share, decreased costs, and big gains in profit. So why don’t more manufacturers use design of experiments (DOE)? In some cases it’ssimple ignorance, but even when companies provide proper training, experimenters resist DOE because it requires planning, discipline and the use of statistics. Fear of statistics is widespread, even among highly educated scientists and managers. Quality professionals can play a big role in helping their colleagues overcome these barriers.



Using DOE successfully depends onunderstanding eight fundamental concepts. To illustrate these keys to success, we’ll follow two case studies: • Increasing the life of a bearing • Reducing shrinkage of plastic parts from an injection molding process The focus will be on 2-level factorial design, where each input variable is varied at high (+) and low (−) levels. These designs are very simple, yet extremely powerful. For example,Figure 1 shows the results for a 2-level design on 3 factors affecting bearing life. Note the large increase at the upper right corner of the cube. In this case two factors interact to produce an unexpected breakthrough in product quality. One-factor-at-a-time (OFAT) experimentation will never reveal interactions like this.















Figure 1: Two-Level Design for 3 Factors (Response: Bearing Life)

Full factorials are much more efficient than OFAT, because they make use of multivariate design. It’s simply of matter of parallel processing (with factorial design) versus serial processing (with OFAT). Furthermore, you don’t need to run the full number of 2-level combinations,particularly when you get to 5 or more factors. By making use of fractional designs, the 2-level approach can be extended to many factors. Therefore, these DOEs are ideal for screening many factors to identify the vital few that significantly effect your response. The injection molding case will demonstrate the use of fractional 2-level design. 2.1 Key #1: Set Good Objectives

The first decisionbefore designing an experiment is “what is the objective, or purpose, of this study?” The focus of the study may be to screen out the factors that are not critical to the process, or it may be to optimize a few critical factors. A well-defined objective leads the experimenter to the correct DOE. For example, in the initial stage of process development or troubleshooting, the appropriate designchoice is a fractional two-level factorial. This DOE screens a large number of factors in a minimal number of runs. However, if the process is already close to optimum conditions, then a response surface design may be most appropriate. It will explore a few factors over many levels. If you do not identify the objectives of a study, you may pay the consequences: • trying to study too many or too fewfactors • not measuring the correct responses • arriving at conclusions that are already known

In essence, vague objectives lead to lost time and money, as well as feelings of frustration for all involved. Identifying the objective up-front builds a common understanding of the project and expectations for the outcome. In our case study of the injection molder, management wants to reduce...
tracking img