Six Sigma Terminology.pdf

Glossary of Six Sigma Terminology

Acceptance sampling plan. A speciﬁc plan that indicates the sampling sizes and the

associated acceptance or non-acceptance criteria to be used. There are two types: attrib-

utes sampling and variables sampling. In attributes sampling, the presence or absence of a

characteristic is noted in each of the units inspected. In variables sampling, the numerical

magnitude of a characteristic is measured and recorded for each inspected unit; this

involves reference to a continuous scale of some kind.

Action plan. The detail plan to implement the actions needed to achieve strategic goals

and objectives (similar, but not as comprehensive as a project plan).

Afﬁnity chart (diagram). Brainstorming tool used to gather large quantities of informa-

tion from many people; ideas usually are displayed visually, then categorized into similar

columns; columns are named giving an overall grouping of ideas. In other words, it is a

management and planning tool used to organize ideas into natural groupings in a way

that stimulates new, creative ideas. Also known as the “KJ” method.

) risk . The maximum probability of saying a process or lot is unacceptable

when, in fact, it is acceptable ( see Producer’s risk).

Analysis of means (ANOM). A statistical procedure for troubleshooting industrial

processes and analyzing the results of experimental designs with factors at ﬁxed levels. It

provides a graphical display of data. Ellis R. Ott developed the procedure in 1967 because

he observed that non-statisticians had difﬁculty understanding analysis of variance.

Analysis of means is easier for quality practitioners to use, because it is an extension of the

control chart. In 1973, Edward G. Schilling further extended the concept, enabling analysis

of means to be used with normal distributions and attributes data where the normal

approximation to the binomial distribution does not apply. This is referred to as analysis

of means for treatment effects.

Analysis of results. The effect of each factor on the response of the system is determined.

Using simple statistical techniques, the largest effects are isolated and a prediction equa-

tion is formulated to predict the behavior of the system more accurately.

Analysis of variance (ANOVA). A basic statistical technique for analyzing experimental

data. It subdivides the total variation of a data set into meaningful component parts asso-

ciated with speciﬁc sources of variation in order to test a hypothesis on the parameters of

the model or to estimate variance components. There are three models: ﬁxed, random

and mixed.

Analytical thinking. Breaking down a problem or situation into discrete parts to under-

stand how each part contributes to the whole.

Analyze. The third component of the DMAIC phase where process detail is scrutinized

for improvement opportunities. Note that: 1. data is investigated and veriﬁed to prove

suspected root causes and substantiate the problem statement, and 2. process analysis

includes reviewing process maps for value-added/non-value-added activities.

Alpha (

Glossary of Six Sigma Terminology

Appraisal costs. Costs incurred to determine the degree of conformance to quality

requirements.

Assignable causes (of variation). Signiﬁcant, identiﬁable changes in the relationships of

materials, methods, machines, measurement, mother nature and people (5M&P).

Attribute data. Go/no-go information. The control charts based on attribute data include

fraction defective chart, number of affected units chart, count chart, count per-unit chart,

quality score chart and demerit chart.

Availability. The ability of a product to be in a state to perform its designated function

under stated conditions at a given time. Availability can be expressed by the ratio:

[uptime]/[uptime + downtime].

Average. See Mean.

Average chart. A control chart in which the subgroup average, X-bar, is used to evaluate

the stability of the process level.

Bathtub curve. Also called life-history curve. A graphic demonstration of the relationship

of life of a product versus the probable failure rate. Includes three portions: early or

infant failure (break-in), a stable rate during normal use and wear out.

Bayes theorem. A theorem of statistics relating conditional probabilities.

Bell-shaped curve. A curve or distribution showing a central peak and tapering off

smoothly and symmetrically to “tails” on either side. A normal (Gaussian) curve is an

example.

Benchmarking. An improvement process in which a company measures its performance

against that of best-in-class companies (or others who are good performers), determines

how those companies achieved their performance levels and uses the information to

improve its own performance. The areas that can be benchmarked include strategies,

operations, processes and procedures.

Beneﬁt-cost analysis. Collection of the dollar value of beneﬁts derived from an initiative

and the associated costs incurred and computing the ratio of beneﬁts to cost.

) risk. The maximum probability of saying a process or lot is acceptable when, in

fact, it should be rejected ( see Consumer’s risk).

Bias (in measurement) . A characteristic of measurement that refers to a systematic differ-

ence. That systematic difference is an error which leads to a difference between the aver-

age result of a population of measurements and the true, accepted value of the quantity

being measured.

Binomial distribution (probability distribution) . Given that a trial can have only two possi-

ble outcomes (yes/no, pass/fail, heads/tails), of which, one outcome has probability p and

Beta (

Glossary of Six Sigma Terminolgoy

the other probability q (p + q = 1), the probability that the outcome represented by p

occurs r times in n trials is given by the binomial distribution.

Black belt. A team leader, trained in the DMAIC process and facilitation skills, responsi-

ble for guiding an improvement project to completion. They are technically oriented

individuals held in high regard by their peers. They should be actively involved in the

organizational change and development process. Candidates may come from a wide

range of disciplines and need not be formally trained statisticians or engineers. Six sigma

technical leaders work to extract actionable knowledge from an organization’s informa-

tion warehouse. Successful candidates should understand one or more operating systems,

spreadsheets, database managers, presentation programs and word processors. As part of

their training, they will be required to become proﬁcient in the use of one or more

advanced statistical analysis software packages.

Block diagram. A diagram that shows the operation, interrelationships, and interdepen-

dencies of components in a system. Boxes, or blocks (hence the name) represent the com-

ponents; connecting lines between the blocks represent interfaces. There are two types of

block diagrams: a functional block diagram, which shows a system’s subsystems and

lower-level products, their interrelationships, and interfaces with other systems; and a

reliability block diagram, which is similar to the functional block diagram, except it is

modiﬁed to emphasize those aspects inﬂuencing reliability. Also known as a boundary

diagram.

Brainstorming. A problem-solving tool that teams use to generate as many ideas as possi-

ble related to a particular subject. Team members begin by offering all their ideas; the

ideas are not discussed or reviewed until after the brainstorming session. Another way of

deﬁning brainstorming, is a rapid pooling of ideas by a team of people before any discus-

sion or judgment takes place.

Breakthrough. A method of solving chronic problems that results from the effective exe-

cution of a strategy designed to reach the next level of quality. Such change often

requires a paradigm shift within the organization.

Glossary of Six Sigma Terminology

Calibration (of instrument) . Adjusting an instrument using a reference standard to reduce

the difference between the average reading of the instrument and the true value of the

standard being measured, i.e. to reduce measurement bias.

Capability (of process) . The uniformity of product which a process is capable of produc-

ing. Can be expressed numerically using C p , C r , C pk , and Z max/3 when the data is normally

distributed.

Capability ratio (C p ). Is equal to the speciﬁcation tolerance width divided by the process

capability.

Cause and effect diagram. A visually effective way of identifying and recording the pos-

sible causes of a problem and the relationships between them as they are suggested by

the team. A pictorial diagram showing all the cause (process inputs) and effect (resulting

problem being investigated) relationships among the factors that affect the process. In

essence, it is a tool for analyzing process variables. It is also referred to as the Ishikawa dia-

gram, because Kaoru Ishikawa developed it and the ﬁshbone diagram, because the com-

plete diagram resembles a ﬁsh skeleton (causes are the bones of the ﬁsh and the effect are

shown as the head of the ﬁsh). Sometimes it is also called the feather diagram, because it

resembles a feather. The diagram illustrates the main causes and subcauses leading to an

effect (symptom). The cause and effect diagram is one of the seven tools of quality.

c-chart. For attributes data: a control chart of the number of defects found in a subgroup

of ﬁxed size. The c-chart is used where each unit typically has a number of defects.

Cell (of frequency distribution and/or histogram) . For a sample based on a continuous vari-

able, a cell is an interval into which individual data points are grouped. Also, a cell is a

layout of workstations and/or various machines for different operations (usually in a U-

shape) in which multitasking operators proceed, with a part, from machine to machine,

to perform a series of sequential steps to produce a whole product or major subassembly.

Center line. For control charts: the horizontal line marking the center of the chart, usu-

ally indicating the grand average of the quantity being charted.

Central limit theorem. If samples of a population with size n are drawn, and the values

of x-bar are calculated for each sample group, and the distribution of x-bar is found, the

distribution’s shape is found to approach a normal distribution for sufﬁciently large n.

This theorem allows one to use the assumption of a normal distribution when dealing

with x-bar. “Sufﬁciently large” depends on the population’s distribution and what range

of x-bar is being considered; for practical purposes, the easiest approach may be to take a

number of samples of a desired size and see if their means are normally distributed. If not,

the sample size should be increased.

Central tendency. A measure of the point about which a group of values is clustered;

some measures of central tendency are mean, mode and median. Another way of saying

it is the propensity of data collected on a process to concentrate around a value situated

somewhere midway between the lowest and highest value.

Glossary of Six Sigma Terminolgoy

Champion. An individual who has accountability and responsibility for many processes

or who is involved in making strategic-level decisions for the organization. The champion

ensures ongoing dedication of project resources and monitors strategic alignment (also

referred to as a sponsor).

Change agent. The person who takes the lead in transforming a company into a quality

organization by providing guidance during the planning phase, facilitating implementa-

tion and supporting those who pioneer the changes.

Characteristic. A dimension or parameter of a part that can be measured and monitored

for control and capability. A property that helps to identify or to differentiate between

entities and that can be described or measured to determine conformance or nonconfor-

mance to requirements.

Charter. A documented statement ofﬁcially initiating the formation of a committee,

team, project or other effort in which a clearly stated purpose and approval is conferred.

It is the speciﬁc team document deﬁning the context, speciﬁcs and plans of an improve-

ment project; it includes business case; problem and goal statements; constraints and

assumptions; roles; preliminary plan; and scope. Periodic reviews with the sponsor ensure

alignment with business strategies; review, revise, reﬁne periodically throughout the

DMAIC process based on data.

Check sheet. A simple data-recording device. The check sheet is custom-designed for the

particular use, allowing ease in interpreting the results and to reduce the likelihood of

errors in recording data.. The check sheet is one of the seven tools of quality. Check sheets

are often confused with data sheets and checklists. A sheet for the recording of data on a

process or its product. (These may be: forms, tables, or worksheets facilitating data collec-

tion and compilation; allows for collection of stratiﬁed data. See also Stratiﬁcation.) The

check sheet is designed to remind the user to record each piece of information required for

a particular study, and to reduce the likelihood of errors in recording data. The data from

the check sheet can be typed into a computer for analysis when the data collection is com-

plete. A check sheet is also a simple data-recording device.

2 ). As used for goodness-of-ﬁt: a measure of how well a set of data ﬁts a pro-

posed distribution, such as the normal distribution. The data is placed into classes and the

observed frequency (0) is compared to the expected frequency (E) for each class of the pro-

posed distribution. The result for each class is added to obtain a chi-square value. This is

compared to a critical chi-square value from a standard table for a given

(alpha) risk and

degrees of freedom. If the calculated value is smaller than the critical value, we can con-

clude that the data follows the proposed distribution at the chosen level of signiﬁcance.

Coaching. A continuous improvement technique by which people receive one-on-one

learning through demonstration and practice, and that is characterized by immediate feed-

back and correction.

Coded plan matrix. The levels of each factor within the plan matrix are represented by a

code. The codes can be “1’ and ‘2’ or ‘–’ and ‘+’. The use of ‘–’ and ‘+’ is preferred as these

simplify the use of the matrix when calculating the effect of each factor. Taguchi, on the

other hand prefers the “1” and “2” designation.

Chi-square (

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: