Kuei, C. (2004), "Statistical Methods for Six Sigma in R&D and Manufacturing", International Journal of Quality & Reliability Management, Vol. 21 No. 5, pp. 591-592. https://doi.org/10.1108/02656710410536590
Emerald Group Publishing Limited
Copyright © 2004, Emerald Group Publishing Limited
The beautiful thing about six sigma is that the level of process effectiveness is always quantifiable. You can always measure your firm's process capability, understand the extent of variability, find the causes of variability, initiate plans for improvement, and offer a statistical sense of control in a capricious operations environment. As one scrapes through the wholeness of this six sigma way, one may wonder: what color is my belt?
As true or prospective total quality leaders, we may oftentimes ask ourselves: do we have the solid, sound technical and statistical backgrounds to follow through the circle of continuous improvement?
Dr Joglekar's new book, Statistical Methods for Six Sigma in R&D and Manufacturing, is written to help quality and technical professionals reconnect with statistical quality management. It seems to me that this book has five main parts, each made up of chapters emphasized on a specific aspect of statistical quality management. They are highlighted as follows:
Part I: Basic and intermediate statistics. This part focuses on fundamental statistics topics that concern TQ leaders and quality engineers, the data they collect from a single population or multiple populations, analyses and interpretations of the data, the capricious operations universe, and many things about R&D and manufacturing studies. One can get additional insights into the data with a thorough review on the first three chapters of the book.
Part II: Control charts and process capability. This part is devoted to process control and capability assessments. Chapter 4 deals with basic variable and attribute control charts. This is followed by a discussion of key success factors in implementing control charts. Chapter 5 explains basic process capability concepts, and discusses ways to estimate capability and performance indices and their associated confidence interval. Once the stage is set, the rationale behind the six sigma goal is discussed and presented. This chapter concludes with planning for improvement. In view of the success in explaining things in the entire book, there is, however, a notable absence of materials on the presentation of these improvement efforts. Examples in point or cases in practice should be considered in the future edition of the book to enrich this fine chapter.
Part III: Advanced SPC. This part, or chapter 6, introduces five additional control charts: risk‐based charts, modified limit (or acceptance control) charts, moving average charts, short‐run charts, and charts for non‐normal distributions. The just‐in‐time system with multiple products and short production runs, for example, is used to explain how to establish short run individual and moving range charts. The discussions and examples are very easy to follow.
Part IV: Variability reduction and quality planning. In this part, with several diverse examples, key tools, and structured procedures, Dr Joglekar provides a general introduction on how to economically and correctly reduce system variability. The interest in chapter 7 centers on how to determine the total system variability and the contribution of each cause to the total variability. How to collect data in a structured manner is also described to facilitate the data collection and analysis process. In chapter 8, a pizza manufacturing process aimed at producing millions of 12‐ounce pizzas with 34 grams of pepperoni on each pizza is presented to illustrate how to make the most of variability reduction tools and analyses. Multi‐vari charts, for example, can be used for getting insights into the principle source of variability concerning the pepperoni weight. Variance component analyses can also be used to estimate and assess the contribution of each source of the variation. Implications for quality management and planning on an economic basis are also included in this section. In this context Dr Joglekar also discusses economic losses as a function of the Cp index.
Part V: Measurement systems. The observed variance is normally composed of product‐to‐product variations and measurement variations. In the presence of measurement uncertainties, correct decisions have to be made to obtain true process capability or valid statistical conclusions. The emphasis in chapter 9 is therefore on subjects such as the statistical properties of measurement systems, stability and bias studies, repeatability and reproducibility studies, and the assessment and improvement of measurement systems.