Glossary of Six Sigma Acronyms

ALPHA RISK - The probability of accepting the alternate hypothesis when, in reality, the null hypothesis is true. Also called Type I Error or Producer’s Risk.

ANOM (ANALYSIS OF MEANS): A graphical analysis approach to compare the means of several groups of size n.

ANOVA (ANALYSIS OF VARIANCE) A statistical procedure that can be used to determine the significant effects when there are more than two population means, such as in a factorial experiment.

BETA RISK - The probability of accepting the null hypothesis when, in reality, the alternate hypothesis is true. Also called Type II Error or Consumer’s Risk.

CL (CENTER LINE): A horizontal line on a control chart (usually solid) that represents the average or center of the process.

CP: A process capability index for short term process capability that shows the process capability potential, but does not consider process centering. CP may range from 0 to infinity with the larger value indicating greater potential capability. CP requires that the distribution of values follow the normal distribution.

CPK: A process capability index for short-term process capability, which is used to compare the natural tolerance of a process with the specification limits. CPK has a value equal to CP if the process is centered on the nominal. CPK is negative if the process mean is outside the specification limits. If CPK is larger than 1, the natural tolerance fall completely within the specification limits. CPK requires that the distribution of values follow the normal distribution.

CTQ’s: Critical to Quality….the things that are important to the customer.

DMAIC: Define, Measure, Analyze, Improve, and Control. The five phases of the Six Sigma breakthrough process.

DOE (DESIGN OF EXPERIMENTS): A set of experimental planning, procedures, methods, and analyses which permits one to objectively evaluate the effect of one or more variables on an output while negating the effects of extraneous variables.

FMEA (FAILURE MODE AND EFFECT ANALYSIS): An analytical approach directed toward problem prevention through prioritization of potential problems and their resolutions.

I: Individuals…individual data points in a sample.

KPIV: Key process input variable(s).

KPOV: Key process output variable(s).

LCL (LOWER CONTROL LIMIT): A horizontal line on a control chart (usually dotted) that represents the lower limits of natural process variation.

LSL (LOWER SPECIFICATION LIMIT): the lowest value of a product dimensional measurement that is acceptable.

MR: Moving range..the difference in consecutive values taken in time.

OCAP (OUT OF CONTROL ACTION PLAN): A documented response procedure that should be followed after detecting that the process has gone out of control.

OCC (OUT-OF-CONTROL CONDITION): A statistical process state which exists after a process mean has been shifted or the variation has changed from the “usual” situation.

OFAT: An experimental procedure in which each factor is studied one at a time. Usually this method is less desirable than taking combinations of all factors such as in a factorial experiment.

P VALUE: In a statistical hypothesis test, the P value is the probability of observing a test statistic at least as extreme as the value actually observed, assuming that the null hypothesis is true. If the P value is smaller than the significance level (typically, 0.05), the null hypothesis is rejected, and the test result is termed statistically significant.

PP: A process capability index for long term process capability that shows the process capability potential but does not consider process centering. CP may range from 0 to infinity with the larger value indicating greater potential capability. PP requires that the distribution of values follow the normal distribution.

PPK: A process capability index for long term process capability, which is used to compare the natural tolerance of a process with the specification limits. PPK has a value equal to PP if the process is centered on the nominal. PPK is negative if the process mean is outside the specification limits. If PPK is larger than 1, the natural tolerance fall completely within the specification limits. PPK requires that the distribution of values follow the normal distribution.

PPM: Defect rate expressed as parts per million, i.e. proportion defects x 1,000,000.

RPN (RISK PRIORITY NUMBER): A weighted number of occurance ranking times severity ranking times detection ranking in a FMEA. The higher the number, the higher the risk of a potential problem in the process.

R: The sample range (maximum – minimum).

R-bar: The average sample range.

R&R: Repeatability and reproducibility usually associated with gage or measurement studies.

S (SIGMA): The Greek letter (s) that is often used to designate the standard deviation of data.

SIX SIGMA - Sigma is a letter in the Greek alphabet. The term "sigma" is used to designate the distribution or spread about the mean (average) of any process or procedure. For a business or manufacturing process, the sigma value is a metric that indicates how well that process is performing. The higher the sigma value, the better. Sigma measures the capability of the process to perform defect-free-work. A defect is anything that results in customer dissatisfaction.

SOP (STANDARD OPERATING PROCEDURE): A set of documents which describe in detail how a procedure or method is to be done.

SPC (STATISTICAL PROCESS CONTROL): The graphical application (trend charts) of statistical techniques in the evaluation and control of processes. They can be used to assess process stability and to distinguish between common cause and special cause variation. The control limits typically set at + 3s from the centerline (mean).

UCL (UPPER CONTROL LIMIT): A horizontal line on a control chart (usually dotted) that represents the upper limits of natural process variation.

USL (UPPER SPECIFICATION LIMIT): the highest value of a product dimensional measurement that is acceptable.

X: Process input or factor.

X-bar: The sample average.

Y: Process output or response.

Z-SCORE: A standardized value formed by subtracting the mean and then dividing the difference by the standard deviation. This puts the data into standard normalized form with a mean of 0 and a standard deviation of 1.