Measurement Techniques for Six Sigma

Sigma is a symbol used in statistics to measure standard deviation. The Six Sigma business management strategy was originally developed in the mid-1980s by Motorola, Inc. as a total quality management tool to reduce product defects. Under Six Sigma training, a manager learns to identify business process problems using a variety of statistical analysis methodologies that measure business processes or product short-comings in order to find solutions to eliminate them.

ANOVA

The analysis of variance, or ANOVA, is an analysis technique used to compare two or more populations of data to evaluate whether differences exist and, if so, the degree of differentiation. ANOVA may be used to gain information about process performance variances within time periods or between similar work groups within the same organization. ANOVA is generally used within the define-measure-analyze-improve-control phase, DMAIC, of the Six Sigma process improvement framework.

Histogram

Histogram, also called a frequency plot, is a measurement graph that uses a pictorial to visualize and study data. The height of each chart on a histogram reveals the number of times a rank-ordered data value falls within a range. It is used within Six Sigma as a graphical model of analyzing and measuring the frequency of data in categories or subgroups.

Pareto Charts

"Pareto chart is a vertical bar graph that illustrates problems in a prioritized order to determine which problems should be tacked first," writes Paul Keller in his book "Six Sigma Demystified". Pareto's analysis is the idea that 80 percent of results are caused by 20 percent of the potential causes. Six Sigma uses these charts to graphically track the magnitude of an effect per cause. It then sorts this information from the greatest to the least effect to determine which causes, or outputs, create the greatest impact. The chart is designed to distinguish the vital few (20 percent) from the trivial few (80 percent).

FMEA

FMEA, failure modes and effects analysis, is a measurement technique in Six Sigma that identifies high-risk processes and features based on the impact of failure and the potential for a failure to occur. In the Six Sigma analyze stage it is a funneling tool used to identify, prioritize, anticipate and prevent failures.

Time Series Plots

Time series is a statistical methodology that gathers data at a sequence of consistent and successive time intervals of change for a process or product. The time-series plot is a graphical chart that sets out data gathered at both continuous and discrete data points. Continuous data is collected from a process that requires a continuous stream of outputs. Discrete data is collected at regularly spaced intervals, such as hourly or every week.

Further References:

• "What Is Six Sigma?"; Pete Pande; 2001
• "Six Sigma Demystified"; Paul Keller; 2011
• Motorola University: What is Six Sigma?