One-Factor-at-a-Time versus Designed Experiments

Article excerpt

Many engineers and scientists perform one-factor-at-a-time (OFAT) experiments. They will continue to do so until they understand the advantages of designed experiments over OFAT experiments, and until they learn to recognize OFAT experiments so they can avoid them. A very effective way to illustrate the advantages of designed experiments, and to show ways in which OFAT experiments present themselves in real life, is to introduce real examples of OFAT experiments and then demonstrate why a designed experiment would have been better. Three engineering examples of OFAT experiments are presented, as well as designed experiments that would have been better. The three examples have been successfully used in an industrial workshop and can also be used in academic courses.

KEY WORDS: Teaching statistics.


Engineers and scientists often perform one-factor-at-a-time (OFAT) experiments, which vary only one factor or variable at a time while keeping others fixed. However, statistically designed experiments that vary several factors simultaneously are more efficient when studying two or more factors.

That is what statisticians know. But in industry, they need to be able to convince adult, practicing engineers that what they have been doing for years can be improved upon. This is particularly true because engineers usually have higher standing in the company than statisticians from the Quality Assurance Department, and hence may be inclined to discount the statisticians' advice unless they understand it. Also, engineers need to learn to recognize OFAT experiments in order to avoid them. When teaching an academic course, it is important to convince engineering and science students that designed experiments are relevant to their applications, and to give statistics students (some of whom will work in industry) a better understanding of practical considerations.

In teaching a three-day design of experiments workshop for engineers in industry, the author has found it extremely helpful to use examples of real engineering OFAT experiments, and to compare the OFATs to designed experiments to illustrate why the latter would have been better. It is important to describe the disadvantages of OFAT experimentation early in an industrial workshop, so students do not drop out of the course to do "more important" things. College textbooks, which present a fair amount of statistics up front, often introduce OFAT experiments later: Box, Hunter, and Hunter (1978, pp. 312 and 510); Montgomery (1997, p. 201); and Mason, Gunst, and Hess (1989, p. 101).

The first morning of the three-day design of experiments industrial workshop is an overview. The overview starts with a brief description of what designed experiments are, what they are used for, and how the rest of the industry uses them. The core of the overview is a complete, real example ([2.sup.3] with center points) that is used to introduce the basic concepts, including description of the process, planning the experiment, conducting the experiment, analyzing the data with main effect and interaction plots, and reaching conclusions and implementing recommendations. The example is followed by a section on "Why DOE Works--or Why it is Possible to Study Several Factors Simultaneously and Still Get Useful Information." The overview ends with a section on the advantages of designed experiments over OFAT experiments, which will be described in this article.

The student reaction to the overview is very positive. The material is stripped down to bare essentials, and is illustrated by real-life examples they can relate to. In the author's experience, this goes a long way toward convincing engineers (and managers) to use designed experiments.

Section 2 describes advantages of designed experiments over OFAT experiments, and Section 3 gives three examples that illustrate these advantages. Section 4 is a summary. The OFAT examples can be used in both academic and industrial design of experiments courses. …