DEVELOPMENT OF AN OPTIMAL LEAN SIX SIGMA MODEL

Abstract

Lean six sigma is a hybrid continuous improvement methodology that is not standardized and is not well understood. A review of literature found that the spectrum of lean six sigma approaches extends from those that are lean dominant to those that are six sigma dominant. This research illuminated the lean six sigma methodology by methodically assessing the literature via text mining and cluster analysis. Text mining was used to establish the degree to which lean six sigma models, as described in articles published in the International Journal of Lean Six Sigma, are lean dominant versus six sigma dominant. The iterative cluster analysis was used to identify clusters of articles that were interpretable. A cluster of lean dominant lean six sigma articles was identified and statistically validated as being distinct from other models. It was determined that characteristics of a lean dominant lean six sigma include the text mining key words “waste”, “value”, and “kaizen.” The research also found that these lean dominant lean six sigma articles ascertain lean as the dominant philosophy and six sigma as a subordinate tool used in achieving the lean objectives. The findings of the research as well extrapolation of the literature informed a recommended lean six sigma model. The recommended model is lean dominant and consists of two subordinate methods – six sigma and statistical process control. The three synergistic approaches not only each serve in their own way to manifest process improvements, they also all contribute to organizational learning which is considered a chief contributor to competitive advantage.