Information transformation perspective on experimental design in Six Sigma

Abstract

Being at the core of the "Improve" phase of any Six Sigma initiative, design of experiments has been gaining wider attention in industry and business recently. It is important that nonstatisticians (e.g., Black Belts and Green Belts) appreciate the role of design of experiments in information transformation and utilization and are able to integrate appropriate experimentation in Six Sigma projects. A practical perspective is offered in this article on the relationship between data collection effort and the kind of information to be expected from a statistically designed experiment, with the reasoning summarized in two rules governing the handling of empirical information. This intuitive exposition of the working of orthogonal experiments would help demystify statistical procedures and facilitate intelligent deployment of experimental design in Six Sigma-where a premium is placed, on the understanding of the rationale of specific procedures as well as good planning and timely completion of proje cts.