Six Sigma For Dummies

Organizations across the globe have been implementing Six Sigma programs over the last dozen or so years. More recently Lean Six Sigma has made its way into the business world. The U.S. military has adopted it and some executive branch agencies are training their employees on the methodology. For many, though, six sigma remains a mystery. For others, it's seen as a training exercise. So here are some simple facts – and common myths – about what it is and is not.

Methodology
Six Sigma is a universally applicable approach for improving efficiency, results, and customer satisfaction in any organization. In most organizations, problems arise and solutions are quickly implemented. For example, if processors make errors, a common solution is to implement an error checking step. But what happens when the error checkers make errors? Or what happens when work volumes increase? The obvious solution is to increase staff. If this is not an option, then new procedures are implemented, existing staff will be expected to work harder, and performance, along with customer satisfaction, tend to decline. The six sigma methodology, DMAIC, imposes a discipline that requires an analysis of process and data to understand root causes. Through the use of fishbone diagrams or a (Cause & Effect Matrix), as well as more advanced statistical techniques, symptoms are traced back to their underlying causes. Solutions become self-evident. 

Tools
The observed problem – errors, delays, bottlenecks, risk, high cost, low productivity – is really only a symptom of a deeper root cause. Addressing the symptom is ineffective (because the problem does not go away), and costly (because you’ll need to continually fix the problem). Six sigma provides a comprehensive set of tools and techniques for isolating root causes, including process mapping, cause & effect diagram, fish bone diagram, data sampling, regression analysis, hypothesis testing, and advanced statistical analysis. But the power of Six Sigma is the framework itself: Define the problem, gather and analyze data to link observed problem (symptom) to root causes, implement solutions that address root causes not symptoms, and manage root causes so the problem does not recur. 

Comprehensive
Six Sigma is different from TQM and BPR. While these two approaches have merit in certain situations, they fundamentally lack the rigorous data analysis which links cause to effect. Process re-engineering focuses on changing the process; i.e. how work is done. Six Sigma is far more comprehensive, combining process analysis with rigorous analysis of process and input data to permanently reduce variation in outputs.

The term sigma comes from the normal distribution (aka the Gauss curve or the bell curve). The normal distribution is used to describe many different types of numerical information. Using just two numbers (theaverage, which is the typical value or in statistical terms the most likely value to occur at random; and thestandard deviation aka Sigma, which quantifies the amount by which any data point could vary from the average) we can make statements about likely values (average) in a set of numbers and quantify how much the likely value could vary. For example, we can talk about the average (most likely) cost to process a transaction. Yet this cost will vary. The standard deviation tells us by how much it can vary. This gives us a range of possible values which can be helpful in decision-making.

With these two values (average and standard deviation), we can estimate where any data point is likely to fall. In the curve below, for example, we know that there is a 68.26% chance of a value falling between -1 and 1 standard deviation (sigma) from the average. Similarly, we know that there is a 95.44% chance of any data value falling between -2 and +2 sigmas from the average, a 99.7%chance of any data value falling between -3 and +3 sigmas and so on. Put another way, in any normally distributed set of numbers, 99.7% of the values will fall between -3 and +3 sigmas from the average. So if 6-month old babies in the U.S. weigh 13 lbs, on average, with a standard deviation of 1 lb, we can say that most 6-month old babies (i.e. 99.7%) will weigh between 10 and 16 lbs.

Management Philosophy
The lasting power of six sigma is the discipline and mindset it brings to execute flawlessly. As a management philosophy that continually strives for “zero defects”, six sigma engenders process discipline, employee engagement, metrics that foster excellence, and continuous innovation.