What is Six Sigma?

by Abhiram Patnaik, MBA class of 2007, IIIT Allahabad

  • A Philosophy (Make fewer mistakes in what we do)
  • A Statistical Measurement (Helps gauge adequacy of product process & services)
  • A Metric (A measuring system)
  • A Business Strategy (Good quality reduces cost)

Six Sigma has many facets: It is a metric, benchmark, vision, philosophy, method, tool, symbol, and a goal. It is a goal to achieve less than 4 defects per million opportunities. It can be described as the ultimate in optimization. In six sigma, the emphasis is on reducing defects in the product or services to less than 4 per million opportunities, reducing cycle time with aggressive goals such as 30-50% reduction per year and reducing costs to dramatically impact the bottom line. Six sigma provides specific methods to recreate the process so that defects and errors never arise in the first place. Needless to say, this leads to multifold increase in the profit as well as the customer base of an organization.

In statistics the Greek letter ‘sigma' is used to measure the amount of variation in data. Every process has measurable characteristics and the measurements follow a frequency distribution. Every distribution has a center point called ‘Mean' or ‘Average' (half the measurements are above and half below). The curve is like a ‘bell shape' getting closer and closer to zero but never quite reaching the line. The variation (fatness) of the curve is measured by the ‘standard deviation'- the distance between the mean and the point either side where the curve changes from convex to concave. This distance (standard deviation) is known as sigma.

Sigma can be broadly considered as a method of measuring the deviation trend of a data from its normal average value. The interesting thing about the normal distribution is that 68% of all measurements fall within one sigma either side of the mean. This is both mathematically proven and a practically experienced result. Similarly 99.74% of all measurements fall within 3 sigma. To give an example if someone measure shoe sizes of the entire population, the plotted measurements will have an average (Mean) and standard deviation (Sigma). Almost 100% of all people have shoe sizes from (Mean +3 Sigma) to (Mean – 3 Sigma). So if someone makes shoes with just this range of sizes he/she can satisfy 99.74% of all your customers.

Three Sigma- The current accepted standard

Most of the organizations operate at the three sigma level. This appears decent as only 0.26% of the output will fail to meet the customer requirements. But the case is different as today's services, manufacturing and operations are complex with many process steps and numerous parts. Each step of the process may deliver at 99+ % success, but as each step relies on the previous one, failures multiply and the outcome has more defects and failures than expected. In reality three sigma often fails customer requirements not 0.26% but 7% of the time.