Comparison of Design of Experiments via Traditional and Taguchi Method

Abstract

This paper presents a case study on comparison of Design of Experiments (DOE) via traditional and Taguchi methods in terms of efficiency. First, a three-level, four-parameter, full factorial DOE was conducted for finding the effects of machining parameters on the surface roughness (arithmetic average) of parts produced by turning operation. The results were analyzed applying average response, Taguchi's S/N ratio, and Pareto ANOVA. Subsequently, the same data was analyzed applying Taguchi's L9 orthogonal array. The comparison of two results revealed that despite an 88.9% savings of experimental runs with the Taguchi method, both methods produced similar results.