Paper Authors

Abstract

NOTE: The first page of text has been automatically extracted and included below in lieu of an abstract

Session 3266

Teaching the Taguchi Method of Experimental Design: Design and Testing of Concrete Mixes

Deborah J. Hochstein, Azmi Bin Ahmad, Robert E. Magowan The University of Memphis

Abstract The primary objective of this project was to demonstrate the Taguchi Method of experimental design for a graduate-level course entitled, “Advanced Statistical Quality Control”, at The University of Memphis. The nature of the project enabled students to participate in the entire process, from start to finish. This particular topic, design and testing of concrete mixes, was selected for two reasons. First, it involved several factors, both controllable and uncontrollable. Second, there is historical evidence of the constituents and proportions of a good mix. If the best design mix predicted by the experiment is consistent with the historical recommendation, it lends credibility to the Taguchi Method in the eyes of these students. If the students complete this exercise with a knowledge of the basic skills required for experimental design using the Taguchi method, and confidence in the method’s efficacy, then the pedagogical objective will have been met.

Introduction Engineering science is well founded on the basic laws of nature which have been proven over time using the scientific method which requires that a hypothesis be proven by experimentation. At an introductory level, almost without exception, students are taught that the best way to conduct an experiment is to hold constant all variables except one, and to vary the remaining variable in order to reveal the dependence of the experiment’s outcome on that variable. While this method of experimentation, which requires a full factorial design, has served the scientific community well in academic environments, it has proven to be ineffective in the manufacturing environment. Performing experiments by varying one quantity at a time is far too costly in both time and money for most manufacturing firms. In addition, this method does not enable the engineer to observe the interaction between the quantities being varied. Dr. Taguchi, in the 1950’s, devised a partial factorial method of experimental design that requires far fewer trials than the traditional full factorial scientific method. His method combines engineering techniques with statistical methods in such a way that rapid improvements in quality and cost reduction occur when optimizing product designs and manufacturing processes. “Ford Motor Company was one of the first companies in the United States to recognize the value of Taguchi’s approach to quality. Ford brought Dr. Taguchi to Dearborn, Michigan, to teach its suppliers these techniques in 1981.” (Magowan, 1991). “The quality of Japanese automobiles is attributable largely to the widespread application of the Taguchi Method.” (Roy, 1990).

It is imperative that engineering students who plan to enter a manufacturing environment be provided with an opportunity to study this technique as an alternative to the traditional univariable approach. Application of Taguchi’s experimental design methodology early in the product’s design phase will facilitate the evaluation and comparison of design configurations and material selection. Used in the early phases of facility design, this methodology can improve process yields and reduce variability. “The use of experimental design in these areas can result

• Citation

• Format
  • APA
  • APA - LaTeX bibitem
  • MLA
  • MLA - LaTeX bibitem
  • Bibtex
  • EndNote - RIS

Magowan, R. E., & Ahmad, A. B., & Hochstein, D. (1997, June), Teaching The Taguchi Method Of Experimental Design: Design And Testing Of Concrete Mixes Paper presented at 1997 Annual Conference, Milwaukee, Wisconsin. 10.18260/1-2--6825