Six Sigma demystified

A recent article in the Financial Times suggests there is evidence that the rate of adoption of Six Sigma has started to slow1. This will come as a surprise to those who have only just started to hear the term used in relation to their industry. The term "Six Sigma" is used to describe processes that operate at the highest quality levels with minimum variation and producing less than four defects per million opportunities. This low defect rate is achieved through the application of statistical techniques and other quality improvement tools in a formalised way. The "Sigma" referred to in Six Sigma is the standard deviation of a quality characteristic of a business or manufacturing process against a target value. These characteristics can relate to the final product (drug content in a pharmaceutical or layer thickness on a semiconductor wafer) or to business processes (time to process an invoice or the stock levels of stationery). Fig. 1 shows data from a pharmaceutical process where content uniformity is the characteristic of interest. This process is on-target and producing about three defects per thousand. The statistic that is quoted as a measure of process performance is Cp. This is defined as (Upper Specification Limit – Lower Specification Limit)/(6 x sigma) for normally distributed data. In Fig. 1, Cp equals 1. As processes are rarely centred on their target value, a second statistic, Cpk, is used to measure the degree to which the process is off-target. For normally distributed data the Cpk statistic is defined as (Upper Specification Limit – Process Mean)/(3 x sigma) or (Process Mean - Lower Specification Limit)/(3 x sigma), whichever is the smaller.

Measuring performance Six Sigma processes achieve a Cp of at least 2 and a Cpk of at least 1.5. This means that the specification range represents 12 times the standard deviation of the process. A Six Sigma process is illustrated in Fig. 2. The upper and lower specification limits are usually defined to meet customer requirements; these may be internal or external to the organisation. Motorola coined the term Six Sigma over a decade ago, defining the target capabilities for processes, the standard set of quality tools to use in pursuit of the goal and the way in which they should be applied. Many manufacturing companies use Cp and Cpk as key measures in production, but few use them to measure the day-to-day performance of their business processes. There are two key elements to Six Sigma. The first is the application of statistical and quality improvement tools and techniques to analyse, optimise and then monitor the performance of a process (Fig. 3). The second element is the application of the tools in a directed way using a methodology. The most commonly used methodology is called DMAIC. The DMAIC acronym comes from the five main steps involved: Decide Measure Analyse Improve and Control (Fig. 4). The names given to the steps in the methodology highlight the main goal of each. Within each step there are a number of objectives, activities and deliverables, along with the toolbox of statistical and quality techniques to help achieve the goal. The model is fairly loose, giving improvement teams room to adapt it to their needs. The use of statistical methods is key to the success of Six Sigma but the teaching of applied statistics to scientists, engineers and technicians in higher education remains inadequate. It is often left to employers to provide training in even the basics.

Support for companies Companies such as mine specialise in providing support in this area and we have found that a just-in-time training programme built around the methodical application of statistical tools can deliver dividends in terms of successful development, training effectiveness and risk management. In support of the programme, companies need a statistical software package. The latest of these empower people to use statistics in a way they never could 25 years ago and one or two are now providing support for the application of methodologies such as Six Sigma. Six Sigma has now spread far and wide, and has been adopted in many service industries as well as in manufacturing. While some of the big names, such as General Electric, were early adopters, now service giants like American Express and Starwood Hotels also have major programmes. As you might expect, a whole industry that grown-up to support Six Sigma programmes including consultancies, recruitment agencies and publications. Having said that, many companies have little or no desire to invest large amounts of money on corporate initiatives that often wither on the vine. Those of us who have been around for a while will remember how Total Quality Management (TQM) was going to transform the way in which companies worked, but how many companies do you know where TQM is still practised? Although some in the Six Sigma industry would argue against this, one of the advantages of the approach is that corporate initiatives are not, in my mind, a prerequisite. Six Sigma as a goal can be pursued at a department or facility level through a programme of training and facilitation in the tools, and a re-alignment of how individuals view their work. The institutionalised and long-lasting use of statistical methods and quality tools is where the real benefits will be seen. Trumpeting yet another corporate initiative which costs a lot of money and runs out of steam will not only give a limited return on investment but might actually be detrimental by dmotivating those involved. Six Sigma may well go the way of TQM and fade into last year's annual report, but the tools and techniques that underlie both are here to stay.