Sunteți pe pagina 1din 22

LSS SIX SIGMA EXAM 1

Study online at quizlet.com/_154tzq

1. 1st Quartile Point in your rank ordered sequence, where 13. Affinity Useful tool to help analyze and solve unfamiliar,
25% of the observed data fall below this value Diagram new or complex problems by discovering or
organizing patterns of thought (structural and
2. 3rd Quartile Point in your rank ordered sequence, where
mathematical problems are not included).
75% of the observed data fall below this value
14. Alpha Risk Risk of incorrectly concluding that the population
3. 4 Benefits of 1. Shorter lead time
average is within your calculated confidence
a generic 2. Less cost
interval when, in reality, it is not
pull system 3. Lower wip
that is 4. Health and safety benefits as less WIP on the 15. alpha risk determines how much area under a distribution
working floor. is in the critical region, type 1 risk
well 16. Alpha Risk Risk to reject a true hypothesis
4. 4M Version Machine, Material, Method, and Manpower (AKA Type
of the Cause I)
and Effect 17. Analysis of Almost all experiments involve exploring,
Diagram
Variance investigating, and comparing the sources of
5. 5M and E Machine, Material, Method, Manpower, (ANOVA) observed variations. This is an advanced method
version of Measurement, and Environment that allows you to categorize and quantify all the
the cause various sources of variation.
and effect 18. Analyze Use data and tools to understand the cause-and-
diagram
effect relationships in the process or system
6. 9 Identified by Garvin and Besterfield: 19. ANOVA Analyze variances, test for equality of variances,
Demensions performance, features, conformance, reliability,
(analysis and determine whether there is a valid
of Quality durability, service, response, aesthetics, and
of relationship between variables
reputation
variance)
7. 80-20 Rule The Vital few versus the trivial many; In a 20. Attainable Can the level and range of the specification be
process, a few key variables are the cause of
reached?
most performance problems or defects.
21. Attribute Data that can be counted
8. Accuracy Accuracy describes how centered your
data
measurement system's variation is with the
actual variation of the process or characteristic. 22. Attribute This type of data is all around you: Data
Data (AKA observations fall into discrete, named value
9. Accuracy Getting an unbiased true value; The accuracy
Category categories. No Mathematical operations can be
level of an instrument when compared to a
Data) performed on the raw data. You can count the
standard can only be less than or equal to the
number of occurrences you see of each category.
standard to which it is compared
Telephone area codes are attribute data.
10. Adjusted Compares the width of the specification to the S, M, L, XL, XXL clothing sizes are attribute data.
Long-term long-term width of the process and accounts "Pass" or "fail" judgments pronounced on just-
Capability for off-centering of the process from the assembled products are attribute data.
Index (Ppk) specification. "Good" or "bad" assessments of the output from
a process are attribute data.
11. Adjusted min(Cpu, Cpl) = min( ((USL-X_)/3o), ((x_-
When studying the performance of automobiles,
Short Term LSL)/3ost) )
manufacturer names like Ford, Chevy, and GM
Capacity
are attribute data. // Can I meaningfully add or
Index (Cpk)
subtract values of this data? If no, it is this type of
12. Affinity The outcome of an exercise whereby members data.
Diagram of the team write their brainstorming ideas
23. Attribute P Chart, U Chart
onto small, separate pieces of paper or sticky-
data
back notes. Then the team silently groups the
control
individual idea notes into natural categories.
charts
This modified brainstorming technique has the
advantage of gathering all team members'
input without the inhibition of criticism, and it
encourages the natural emergence of
important groupings or categories of ideas.
24. Audit A measurement system study where 33. beta risk Beta risk is defined as the risk of accepting
you compare your measurements to a the null hypothesis when, in fact, the
known, correct standard. For example, alternate hypothesis is true. In other words,
you may compare what your computer stating no difference exists when there is an
system says you have in inventory (your actual difference. A statistical test should be
measurement system) to what is actually capable of detecting differences that are
in your physical inventory warehouse. important to you, and beta risk is the
Any differences between the two reflect probability (such as 0.10 or 10%) that it will
variation in your measurement system. not. Beta risk is determined by an
organization or individual and is based on
25. average loss
the nature of the decision being made. Beta
function(smallest)
risk depends on the magnitude of the
difference between sample means and is
managed by increasing test sample size. In
general, a beta risk of 10% is considered
acceptable in decision making.
...
34. Beta Risk The risk of a hypothesis not being rejected
26. Balanced A method or process for managers to (AKA Type II) when it is false. The risk when the null
Scorecard focus on business performance metrics, hypothesis is not rejected and it should be.
strategy management, financial metrics,
stakeholder interests
35. Between Statisticians using advanced techniques look
Group at the data and numerically compute the
27. Baseline Original performance Variation variation between the centers of variation for
28. Basic Stats The basic and descriptive statistics, such each of the different X4 conditions
as averages, ranges, variance, and so on, 36. Bimodal The distribution has two modes (two peaks
used routinely in Six Sigma analysis on the curve).
29. Behavior Charts Takes into account the order in which 37. Blocking When you know the source of nuisance
the measured data is observed. Time or variation that is not part of your selected
order are critical factors, especially when experimental factors, you can purposely
you're trying to figure out the causes include this nuisance effect in all your
behind variation and changes in process experimental runs. In this way, you
behavior; form the foundation of guarantee that there will be no bias on only a
detecting and finding the root cause of portion of your experimental settings. EX:
non-normal behavior. Performing the runs at the same time each
30. Believable Have you bought into the specification day.
setting? Can you and your coworker 38. Box and Best when comparing distributions (like
peers strive to meet the specification? Whisker Plots putting two people back-to-back to see who is
31. Benchmarking Normally undertaken by a company or (AKA box taller) allow you to directly compare two or
organization for the purpose of finding a plots) more variation distributions
leader in an area felt to be deficient, 39. Box-Behnken ...
developing methods of measuring
performance, and identifying gap
40. Breakthrough Y = f(X) + E
between the present and desired Equation
performance; typically ignored because 41. Breakthrough A rate of improvement at or near 70 percent
of the perceived barriers to sharing Improvement over baseline performance of the as-is
internal company information; process characteristic.
Sequence: Understand your own
42. Business These tools enable you to model and map
processes, identify improvement criteria,
Process processes, simulate how they'll work, and
measure competitive performance, and
Analysis (BPA) analyze the results
implement significant improvements
43. Business Understanding, controlling and improving
32. Best approach Select those which integrate most
Process an organization's processes to create value
when selecting efficiently with the entire quality system
Management for all stakeholders
quality measuring
devices
44. Business These tools enable you to connect with the 54. Checklists and Basic quality tools most applicable for a
Process information systems in your business and Flowcharts work team to use when there is a desire
Management provide process measurement and control to follow procedures and work
(BPM) functions instructions more closely
45. Calculate the 9-1 = 8. 55. chi-squared
range from
the following
data set:
1,3,5,5,6,7,9
46. Calibration Ensuring traceability of all calibrations to a
department standard laboratory, maintaining and
duties adequate record system, suspending
measuring equipment from use when
conditions warrant, identifying equipment
with a label indicating calibration status
47. Capability The long term performance level after a
process has attained statistical control. 56. Chi-square to make decisions and construct
distribution confidence intervals by summing the
48. Capability How well the voice of your process or
square of normal random variables
characteristic matches up with the voice of
your customer, or how well your process 57. Coded Design A quick, shorthand way to create a
performs in meeting customer expectations. Matrix complete table of an experiment's unique
run combinations
49. Capability and Analyze the tradeoffs between product
complexity complexity and process capability, and 58. Coefficient of R^2; What R2 tells you is how much of the
analysis define the proper configuration of each to Determination total observed variation is determined or
achieve desired outcomes. explained by your linear model. You want
this number to be 80 percent or higher
50. C&E (cause- For the outcomes of any process, define all
and-effect) the contributors, weight their effects, and 59. Coefficient of The standard deviation as a percent of
matrix determine the significant contributors to the Variation the mean; = (Standard Deviation / Mean)
outputs. * 100
51. Central Limit When you repeatedly calculate statistics like 60. Common-Cause "You can act to reduce this type of
Theorem the average of a sample) for a process or Variation variation, but you can't eliminate it. It's
characteristic, the repeated sample statistics natural, and it's part of every system. It's
have variation themselves. This sampling in there and it's not going away! (ex: coin
variation follows a normal distribu- tion toss, time between waves at the beach)
centered on the variation of the process or Those sources of variability in a process
characteristic itself. Further, the width of the that are truly random; that is, inherent in
sampling distribution depends on how the process itself."
many measurements you take. The results of 61. Commonly Lack of business impact for the company
this allow you to predict the bounds of the reported
future and to quantify the risks of the past. problem
52. Characterizing When you have screened out the associated with
Experiments unimportant variables, your experiments Six Sigma
focus _____ and quantifying the effect of the projects
remaining critical few. These _______ reveal 62. Communications A plan to handle stakeholders that could
what form and what magnitude the critical Plan resist change rusulting in improvement
factors take in the Y = f(X) equation for your project failures
process or system.
63. Confidence A 90% confidence interval means that
53. Checklists Often used for attribute or counted Interval given the sample data, there is a 90%
information chance that the true population mean is
contained in the interval. As sample size
increases the width of the interval
decreases.
64. confidence 75. control
interval for a limits
sample equaltion

76. Control A method of assuring the long-term


Plan sustainability of the fix to the problem
65. Confidence x_ plus/minus Z(o/sq(n)); Z is the sigma value
77. Correlation Tells you how closely the relationship between
Interval for corresponding to the desired level of confidence
the two variables follows a linear pattern; does
more than 30 you want to have; o is the calculated standard
not tell you is how much a given change in one
data points deviation from your sample, n is the number of
variable will change a related variable; Don't
(for means) data points in your sample // 1o = +-68%, 2o = +-95%,
assume there is a causal link when there is
3o = +-99.7%
correlation.
66. Confidence A function of n-1 based on the chi-square
78. Correlation The word used to quantify how closely related
interval for distribution
two characteristics are to each other; whenever
population
you can naturally fit a drawn line to a set of
variance
plotted points you know that the
67. Confidence Use the central limit theorem to tell you how much characteristics are correlated.
Intervals stock you can place in any of your measurements or
79. Correlation Always between -1 and 1, the sign of r tells you
statistical conclusions
Coefficient the direction of the relationship between the
68. Conformance An affirmative indication that a product or service variables; r = positive = variable relationship is
has met the requirements. positive = if the value of one variable is
69. Continuous Refer to control charts that display performance of increased, the other variable also increases / r
Control process input or output measurements that are = negative = the variable relationship is
Charts continuous data - data where decimal subdivisions negative = if the value of one variable is
have meaning; when used to monitor output CTQs increased, the value of the other variable
it is referred to as Statistical Process Monitoring decreases / the absolute value of r tells you
(SPM) how strong the relationship is (the closer it gets
to -1 or 1 the stronger (the closer it gets to the
70. Continuous Data observations can take on numberical value
line)
Data (AKA and are not confine3d to nominal categories. Data
80. CT (critical Identify, organize, and display parts of the
Variable values can be meaningfully added and subtracted.
Data) Ex: dollars, meteres, seconds, amps, survey to...) tree process according to areas of critical
response: 1 = disagree, 2 = neutral, 3 = agree importance
81. CTQ Indentify the customers, identify the
71. Continuous Averages and ranges (X - R), Averages and
data control standard deviations (X - S), Individual values and element customer's needs, identify the customers basic
charts moving ranges (/ - MR) sequence requirements, add additional levels as needed,
validate the requirements with the customer
72. Continuous Normal, Log-normal, Exponential, Extreme Value, F,
82. CTQ CTQ Tree Diagram
Probability Student t, Chi-square, and Weibull
Distributions Structure

73. Control Establish plans and procedures to ensure the 83. CTQ Tree Technique where improvement team
improvements are sustained investigates their customer base by moving
from general to more specific customer needs
74. Control A graphical tracking of a process input or an output
84. Cultural Status of the product (a BMW)
Chart over time. These tracked measurements are visually
compared to decision limits calculated from Needs
probabilities of the actual process performance. 85. Cumulative Graphical display of the total percentage of
Distribution results below a certain measurement value
Function
86. Cumulative the frequencies are cumulative, each class 100. Define: Bias The difference between the observed value
Frequency frequency is added to the total of all previous and the actual value.
Graph (AKA class frequencies 101. Define: The operation or step that inserts the largest
Ogive)
Bottleneck amount of time delay into a process.
87. Curvature The assumption of 2k experiments is that the 102. Define: The variation due to random shifts in factors
effects of your experimental factors is linear.
Common that are always present in the system.
Although this is often a good first
Cause
approximation, there are many times when a
Variation
line doesn't fit your process or system. For
those cases, you need to design your 103. Define: An operation or step with a process time too
experiment to reveal the curved nature of Constraint long to meet customer demand. There can
reality. This is usually done by including be multiple constraints.
more than two levels for each of your 104. Define: The difference in the accuracy values
experimental factors. Linearity through the expected operating range e.g.
88. Curve Fitting The is to develop an approximate equation temperature.
that describes the system or process 105. Define: Mean The sum of the data divided by the sample
statistical behavior as much as possible; lets size
you know quantitatively what effect one
106. Define: What data is arranged into ascending or
variable has on another. You also know
Median descending order, the median is the central
which variables are significant influencers
data point. If there's an even number of
and which ones are just in the noise. Finally,
data points, then average the middle two.
you know how much of the system behavior
your equation does not explain. 107. Define: Mode Most frequently observed data value in a
sample or population. Can have more than
89. Customer Persons, groups of people, companies,
one mode.
systems, and downstream processes that
receive the output of the process 108. Define: PLT is the time a single work piece takes to
Process Lead travel through a process.
90. Customer The largest portion of unhappy customers
Time
data from utilize a product that was not intended for
unhappy their use (getting information from 109. Define: Range The difference between the maximum and
customers unintended end users). the minimum data values.
flaw 110. Define: Sampling is using a smaller group to
91. Customer Life Acquisition, Retention, Attrition, Defection, Sampling represent the whole population. It is often
Cycle and Reaquisition not physically possible to count all of the
members of a population.
92. CuSum Cumulative SUm
111. Define: The act of extracting an item or items from
93. Data Mining The labor of digging and sorting through
Sampling the population or process to measure.
data for clues to where the improvement
event
gems may lie
112. Define: How accurately the measurement system
94. Defect When a process or characteristic doesn't
Stability performs over time.
perform within its specifications. Or in other
words, it produces a non-compliant condition 113. Define: Time Any time delay in a process
Trap
95. Defector A customer that has been lost
114. degrees of df= n(a) +n(b) -2
96. Defects per DPO X 1,000,000
freedom
Million
Opportunities 115. Deploy Assigning, training, and equiping the staff
(DPMO) 116. Describe two 1. Repeatability - one operator measures the
97. Defects per # of defects observed on a unit / # of sources of same characteristic of the same part
Opportunity opportunities on a unit measurement repeatedly.
variation 2. Reproduceability - multiple operators
98. Defects per Number of Defects Observed / Number of
measure the same charateristic of the same
Ubiquitous Defects Inspected
part.
Unit (DPU)
117. design alias confused experiment or confounded
99. Define Set the context and objectives for the project
118. Designing Fewer people are directly involved in 127. Distributed "The statistical term used to describe the
Domain designing, which focuses on improving the (AKA: relative likelihood of observing values for a
designs of new processes Probability variable factor"
Distribution
119. Design of An efficient, structured, and proven
and
Experiments approach to investigating a process or
Probability
(DOE) system to understand and optimize its
Density
performance.
Function)
120. Determinism The principle that you can create a desired
128. Distributions Lognormal, Weibull, and Exponential
outcome by configuring and controlling the
bounded at Distributions
inputs in a specific manner.
the left by
121. Deterministic In some rare cases, the exact details of the Y zero
= f(X) equation relating the Xs to the Y is
129. DMAIC "Every Six Sigma project follows a
known without having to do any curve
standardized and systematic method
fitting - either from a very mature
known as DMAIC (Define-Measure-
understanding of the physics of the process
Analyze-Improve-Control), a formalized
or system, or from some other source of
problem-solving process. The DMAIC
knowledge; you know with certainty that
process can improve any type of process in
setting the input Xs to certain values will
any organization to improve its efficiency
always lead to the exact same value for the
and effectiveness."
output Y, even when the process is
repeated. 130. DMAIC Define, Measure, Analyse, Improve, Control.
122. Diagnosis and Diagnosis: Your process is operating at its 131. DOE (Design of Systematically investigate the process or
Prescrption entitlement level of variation. Prescription: Expirements) product variables that affect product
for: Cp=Cpk Continue to monitor the capability of your quality
and Pp=Ppk process. Redesign your process to improve 132. Does the Yes, lack of clarity of the operational
its entitlement level of performance.
operational definition may show up in the MSA,
123. Diagnosis and "Diagnosis: Overall, your process or definition especially around reproducebility, as two
Prescrption characteristic is centered within its influence the operators may be measuring the
for: Cp=Cpk specifications. Prescription: As needed, focus outcome of characteristic differently.
and Pp=Ppk on reducing the long-term variation in your the
process or characteristic while maintaining Measurement
on-center performance." System
Analysis, MSA.
124. Diagnosis and Diagnosis: Your process or characterstic
Prescrption suffers from a consistent offset in its center 133. Domains of Thinking, Processing, Designing, Managing
for: Cp=Pp location. Prescription: Focus on correcting Activity
and Cpk=Ppk the set point of your process or characteristic 134. DPU DPU = -ln(RTY)
until it is centered
calculation
125. Difference binary data is only 2 outcomes, e.g. pass or for RTY
between fail 135. Entitlement The best demonstrated performance for an
discrete and discrete data is based on counts, only a finite
Level existing configuration of a process or
binary data number of values is possible. The values
system. It is an empirical demonstration of
can't be subdivided.e.g. half a sheep.
the level of improvement that can
126. Discrete Hypergeometric, binomial, and Poisson potentially be reached.
(attribute 136. Error In the field of statistics, this term doesn't
based)
mean something is wrong. The term simply
Distributions
means a calculated deviation from a
comparison value.
137. estimate 95%
confidence
interval
138. EVOP Evolutionary Operations 151. First Time The proportion going through the process
Yield (FTY) correctly the first time captures the harsh
139. EWMA Exponentially weighted moving average
reality of the effectiveness of the process,
140. Excel That ubiquitous spreadsheet program that including inspection and rework. ((In - Scrap
comes on nearly every PC is also a powerful - Rework))/ In)
computational and display program, and it's
152. Fishbone Brainstorming, Prioritizing, Action plan
used extensively in Six Sigma statistical
(AKA Cause development; problem statement is at far
analysis applications. You have to program
and Effect right; it takes time to perform and prioritize
this tool to perform the calculations, but
Session, AKA and it does not develop an action plan;
there are also a plethora of software
Ishikawa Separate a problem into smaller
companies that sell add-ons and extensions
Diagram) components, show how various causes
for statistics.
Parts interact, display many possible causes in a
141. Expand The scope of the initiative to include graphical manner
additional organizational units
153. Fishbone Create a high-level C&E in the form of a tree
142. Experimental Block what you can and randomize against Diagram structure, with categories for each major type
Safegaurds what you can't block. of contributor. A method for capturing
143. Experiments Instead of just letting the Xs of the process potential causes and inputs to a process.
you are studying take on whatever values 154. Flowcharting A tool frequently used in the audit. Can
they do, you purposely set and control the reveal wasted manpower or resources,
values that the Xs take on. You actively excess handling, or extra processing. Can
control and modify the process being find holes, gaps, or weak areas in the control
studied. system; There can be multiple paths through
144. Explain the Sampling reduces cost and time while the flowchart, there can only be a single
benefits of allowing more meaningful data to be process flow start point, flow charts can have
sampling collected. it often simplifies and increases the parallel processes, a flow chart can have
accuracy of the measurement process as multiple end points; Asking why we do it this
there aren't as many measurements to take way, asking what would make it perfect,
compared to measuring a population. analyzing each step in detail, a comparison
with processes different than your own
145. Explain the Upper limit on the amount of inventory
function of a within a predefined physical work area or 155. FMEA (failure For any activity or item, define the potential
WIP cap process. mode effects failure modes, including the likelihood of
analysis) occurrence, and the ability to detect and
146. Explain what The total observed variation is equal to the
characterize the effects of those failures.
is included in actual product variation plus the variation
the total due to the measurement system. 156. Focus Small groups with a specific topic
observed Groups
variation in a 157. Fractional 2k full factorial experiments can be adapted
process Factorial to more efficiently search through a large
147. Exploratory Methods used to explore data before Experiments number of experimental factors. What you
Analysis applying more traditional statistical analysis give up in increasing the number of
tools experimental factors is analysis accuracy.
These type of experiments teach how and
148. Exponential The distribution used to describe the time
where to adapt your experiment to get the
between failures which occur independently
most out of your search efforts.
and at a constant rate. The mean and
standard deviation are equal. 158. From left to 1. Output (Y).
right, what 2. Stratification factors (X Variables).
149. Factors (AKA: The input characteristics, or variables you
are the 3 3. Measurements.
conditions, purposely control during the experiment;
stratification
variables, the Xs in the Six Sigma Y = f(X) equation
headings.
inputs)
159. General term The probability of occurrence of an event of
150. F To test for equality of variances from two
for the interest (termed as a success) with n trials
distribution normal populations
binomial and with f failures, then the number of
distribution occurrences follows this type of distribution
160. Give 3 1. Number of Pizza toppings 169. How many males (1.96 * 6 / 0.5) ^2 = 554
examples of 2. Pass / Fails would you need to
attribute data 3. Number of sheep in a field. sample to gauge
their average height?
161. Give 3 1. Christmas shutdown.
Standard Deviation
examples of 2. Server crashes
6cm, Confidence
circumstances 3. Broken part
95%, Precision
that may
0.5cm.
result in a
special case 170. How many long form - 10 products, 3 operators,
variation operators, parts and 3 replicates or measures.
measures are
162. Give two 1. So each individual 'counts' things the
required to conduct
reasons why same way.
a long form gauge R
its critical to 2. So we can plan how to measure
& R.
establish effectively.
operational 171. Hypergeometric Used when there are two possible
definitions distribution outcomes on each trial, but the
probability of success on each trial
163. Hard Savings Reduce expenses and result in a financial
differs because there is sampling
improvement
without replacement; can model
164. Hidden Among Six Sigma practitioners, this reality of discrete data when the population
Factory / fixing the results of propagated error. size is small compared to sample size
Hidden ground of all organizations. It is the factory
172. Identify the seven 1. Transportation.
Operation that fixes problems, corrects mistakes, and
forms of Waste - 2. Inventory.
other- wise wastes both time and money - a
TIMWOOD. 3. Motion.
company's two most precious commodities.
4. Waiting.
Corrective and non-value-added work
5. Overproduction.
applied to produce a unit of output
6. Overprocessing.
generally not properly recognized as
7. Defects/Rework
unnecessary and a form of waste of time,
resources, materials, and cost. 173. If there are 100 items 100 / 5 = 20 hours
of WIP in a process
165. Hierarchy of Basic, Expected, Desired, Unanticipated
and the exit rate is 5
Customer
units per hour what
Expectations
is the process lead
166. Histogram Displays the distribution of a sample (not a time.
(Relative population), often the readings are
174. Implement Projects and improve performance,
frequency grouped into uniform intervals
yielding financial results
chart or
graph) 175. Improve Develop the modifications that lead
to a validated improvement in the
167. House of Side walls = indicated by customer needs
process or system
Quality and customer competitive assessment;
foundation = the technical competitive 176. Indicating Gage Shows the amount of variation in size
assessment; ceiling = design features from the specification
168. How do By reducing the variation in process lead 177. Initialize Establish goals and installing
generic pull time, inventories can be reduced, and so infrastracture
systems reducing process lead time itself. 178. Inputs Materials, information, and other
control lead
resources provided by the suppliers
time
that are consumed or transformed
in the process.
179. Interaction Sometimes, one variable by itself doesn't 188. latin square A specific repeated-measures design with
effects plots have a major impact on an output. But when design built-in counterbalancing, cannot be used
you combine it with other variables, it has a when estimations of interactions are desired
significant influence. This is called an 189. Levels In your experiment, you choose two or more
interaction effect. For example, adding eggs
values for each of the experimental factors.
by themselves to a cake batter doesn't
These values are called the ___ for that factor.
immediately impact the cake's texture. But
Planning your experiment includes deciding
adding eggs together with oven heat to the
how many ___ you need to use for each
batter produces a yummy dessert.
factor.
180. Interval Data is like ordinal except we can say the 190. Leverage The ability to apply the critical few Xs that
intervals between each value are equally
have the greatest impact on your desired Y
split. The most common example is
temperature in degrees Fahrenheit. The 191. Lifetime the best measurement of the value of a
difference between 29 and 30 degrees is the worth customer
same magnitude as the difference between 192. likert scale scale that uses numbers (5 strongly agree, 4
78 and 79. With attitudinal scales and the agree...), focuses on measuring the direction
Likert questions you usually see on a survey, of an attitude
these are rarely interval, although many
193. Linearity This is good when the centering and the
points on the scale likely are of equal
magnitude of the measurement system
intervals.
variation is consistent across its range of
181. JMP From the prestigious SAS Institute, it is a operation. This is poor in a measurement
(pronounced professional statistics package in its features system when its centering or magnitude of
"jump") and capabilities. Its greatest advantage is in variation changes across its range of
its multiplatform support. Runs on Windows, variation.
Macintosh, and Linux.
194. Long-term Compares the width of the specification to
182. Kano Model Includes Dissatisfiers, Delighters, Satisfiers; Capability the long-term width of the process.
improvement teams most often select from Inex (Pp)
among satisfiers and delighters (not
195. Long-term Used if you are using a long-term standard
dissatisfiers). Analyze customer needs and
Sigma Score deviation (more aligned with the real world),
requirements, determine the basic satisifers
shows performance degradation by shift,
for the customer, meet the latent
drift, and trend influences
requirements of the customers. Cannot
determine when teh customer's expectations 196. Lower A value designating a lower limit below which
have risen. Specification the process or characteristic performance is
Limit (LSL) unacceptable.
183. (Kano Model) Unexpected features.
- Delighters 197. Low Sigma Low sigma (z) score means that a significant
are: (Z) Score vs part of the tail of the distribution is extending
High Sigma past the specification limit, the higher the
184. (Kano Model) Basic requirements.
(Z) Score sigma score, the fewer the defects. A process
- Dissatisfiers
or characteristic gets a good sigma score
are:
when the variation distribution is safely away
185. (Kano Model) Performance requirements, the more the from the edge of the specification cliff
- Satisfiers better.
198. Main Possibility that the team will expand the
are:
disadvantage project boundaries
186. Latent Delighters, features or services that go of presenting
Requirement beyond the immediate expectations of the a team with
customer. an initial
187. latin square a square matrix of n rows and columns, a project
technique to control for order effects without lasting more
having all possible orders; a limited set of than 160
orders constructed to ensure that each days
condition appears at each ordinal position
and each condition precedes and follows
each condition one time
199. Main Effect The quantitative influence a single 211. Measurement- Diagnosis: Marginal measurement system.
experiment factor has on the response Y. to-Observed Contribution of the measurement system
YOu will have this for each factor in your Variation Ratio to the overall observed variation is
experiment. For example, how much effect Values and beginning to cloud results. There is a
does ice cream flavor - going from "Vanilla" Interpretation: significant risk of making a wrong decision
to "Strawberry" - have on the resulting filled 0.1 <= o from the measurements. Prescription: Use
weight of the carton? Measure / o with caution only if no better measurement
Observed <= alternative exists. Begin to improve the
200. Main effects An extremely easy and powerful way to
0.3 Diagnosis measurement system by training
plots explore the principle effect of a variable and
and operators, standardizing measurement
its different levels on a critical output.
Prescription procedures, and investigating new
201. Managing A small number of business and quality measurement equipment.
Domain leaders are responsible for managing the
212. Measurement- Diagnosis: Good measurement system.
overall Six Sigma initiative
to-Observed Contribution of the measurement system
202. Market Targeted subsets of the market Variation Ratio to the overall observed variation is small
Segmentation Values and enough to enable good decisions from the
203. Master (AKA The working gage for accuracy prior to Interpretation: measurements. Prescription: Use
Reference measuring the product o Measure / o measurement system as it is. Look for
Gage) Observed <= opportunities to simplify or make th
0.1 Diagnosis emeasurement system less expensive or
204. Matrix (AKA Identify the impact that various process
and more efficient.
Prioritization input variables could have on key process
Prescription
Matrix) output variables
Diagrams 213. Measurement- Unacceptable measurement system.
to-Observed Guessing is probably just as precise. Do not
205. Mean Average
Variation Ratio base important decisions off of information
206. Mean (AKA The most common measure of central Values and from a measurement system in this
Average) tendency. May not have actually occurred Interpretation: condition. Prescription: Measurement
within your measurements, it is the value o Measure / o system needs to be corrected before any
most likely to occur next in a sequence or Observed >= valid information can be derived from the
population of data. Add up each of the 0.3 Diagnosis system. Investigate causes of gross
measurements and divide the result by the and inconsistency.
number of measurements; most common Prescription
and familiar
214. Median The point along the scale of measure
207. Measles Identify locational data; useful to show where half the data are below and half are
Charts visually where x occurs above. The preferred measure of variation
208. Measurable Can you measure the characteristic's location when your collected data contains
performance against the specification? If outliers, or extreme data points well
not, there will be a lot of debate between outside the range of other data; used
you and your customer as to whether the when data contains outliers
specification has been met or not. 215. Minitab The undisputed leader in statistical
209. Measure Get the baseline performance and analysis software. Is taught extensively in
capability of the process or system being colleges and universities, used extensively
improved in major corporations, and fostered
universally by Six Sigma consultancies. In
210. Measurement The analysis of the measurement system to its 14th release as of this writing, it is
Systems determine the accuracy and precision of the packed full of features and is completely
Analysis data obtained from the measurement. capable of stellar performance for you on
every statistics tool discussed earlier in this
chapter. It runs only on a PC.
216. Mode "The value observed most frequently and is 227. Name 3 examples of Ordering, sales & marketing, financial
associated with the highest peak of a Business Non Value reporting
distribution. Drawback: Many distributions Add - BNVA
don't have a single clear peak and some have 228. Name 3 examples of Inspecting, transporting, signoffs
more than one peak of roughly the same
Non Value Add -
height. In these cases, the presentation of this
NVA
single metric by itself does little to deepen
your knowledge of the variation; problematic, 229. Name 3 examples of Postcoding the data, printing the DM
seldom used" Value Add - VA piece, inserting the DM piece

217. Mode measure of the central location for the 230. Name 3 Improve 1. Generate Solutions.
nominal scale, a very low level statistic activities 2. Select Solution.
3. Develop and execute a pilot plan.
218. Modeling Define and design processes, including the
flow of work or material, the timing of activities, 231. Name 3 things to 1. Give input on any check sheet
resources consumed, and points of decision, prepare an design.
inspection, and delivery. individual to 2. Has been trained an allowed to
collect data to practice.
219. Most Primary Customers
avoid bias 3. Have knowledge of the measuring
difficult
process.
area to
obtain 232. Name 3 types of 1. Self selection.
meaningful Bias 2. Self exclusion.
information 3. Missing key representatives.
during the 233. Name 4 Define 1. Identify process to be improved.
design activities 2. Identify customers needs and
phase requirements.
220. Multi-modal Histogram showing two or more distinct 3. Define the current performance
Distribution peaks, two or more values dominate the standards and measures.
variation, always dig deeper to discover what 4. Establish the project charter (that
factor or factors is causing the characteristic's includes problem statement and
schizophrenic behavior project objectives).

221. Multiple When you work to create an equation that 234. Name 4 general 1. Establish the purpose of the
Linear includes more than one variable rules for preparing meeting.
Regression and running 2. Publish appropriate pre work.
meetings 3. Come prepared, pre work complete.
222. Multi- Allow you to investigate the effect of several
4. Use agendas.
variable input variables at a time on a critical output.
studies 235. Name 4 Measure 1. Determine what to measure (x's
activities and y's).
223. Multivariate The analysis of data from multiple
2. Clearly defined process output
analysis measurements on various tems or subjects.
measures (y's).
The output is a graphical picture of the
3. Manage measurement.
various relationships.
4. Measure key process inputs and
224. Name 2 Systematic e.g. taking data at certain intervals, outputs.
types of and Random sampling. 236. Name 8 six sigma process mapping, value analysis
sampling
tools that are likely pareto analysis, C&E fishbone
225. Name 3 1. Develop a prioritised list of potential sources to be used during a diagrams, control charts, run charts,
Analyse of variation. project check sheets, scatter diagrams.
activities 2. Identify potential root causes. 237. Name three Internal / External data
3. collect data to validate sources.
methods for Customer complaints
226. Name 3 1. Plan and implement solution rollout. gathering VOC. Research - direct data collection
Control 2. Implement statistical process control to meet 238. Negatively If the relationship indicates that an
activities CTQ's.
Correlated increase in one characteristic
3. Document project and process
translates into a decrease in the other,
improvements.
and vice versa.
239. Nominal Data that consists of names or categories 248. % of time 50%
only. Basically refers to categorically discrete spent by the
data such as name of your school, type of team in
car you drive or name of a book. This one is define and
easy to remember because nominal sounds measure
like name (they have the same Latin root). stages
240. Non-normal Variation above and beyond the expected 249. One-Sided A specification limit that designates only a
Behavioral normal level. Specification single transition point from acceptable to
Chart unacceptable performance. For example, if
you care only that the characteristic or
241. Normal "A process or characteristic has variation
process perfor- mance not exceed a certain
Behavioral that bounces around a central, horizontal
upper value, that is a one-sided specification.
Chart level on the behavior chart. Most of the
observed variation will be clustered close to 250. Opportunity "A specific characteristic that could either
this central level. Also, every now and then, turn out as a defect or as a success;
there will be excursions that are farther Individual characteristics that are critical to
away from the center. The variation will be the performance of the system;
completely random over time, without characteristics that have a specification"
patterns or trends." 251. Optimization After characterizing your process or system,
242. Normal Distribution type that is bell shaped, most of Experiments the final step is to conduct experiments that
Distribution the observed values of the characteristic are teach you what the best settings are for the
close to a central point with fewer and fewer input variables to meet your desired
appearing as you get farther away from the outcome goal. Your goal may be to maximize
central tendency or to minimize the value of the output. Or it
may be to hit a certain target level. More
243. Normal The distribution characterized by the
often, your goal is simply to minimize the
Distribution smooth, bellshaped curve; synonymous
amount of variation in the output Y.
(bell curve) with Gaussian distribution.
_____________ find the best settings of the Xs to
244. Normally Predictable meet your Y goal.
distributed
252. Ordinal Data that is arranged in order (and
process
differences between values cannot be
245. Nuisance Factors that you haven't foreseen that have determined or are meaningless). Refers to
Factors the potential to blur the clarity of your quantities that have a natural ordering. The
analysis and insights ranking of favorite sports, the order of
246. Objective Desired performance people's place in a line, the order of runners
finishing a race or more often the choice on a
247. Observational Revolve around analyzing the variation in
rating scale from 1 to 5. You cannot state with
Studies the observed critical output or outputs and
certainty whether the intervals between
investigating which input variables it is
each value are equal. For example, we often
linked to. What you are looking for are
using rating scales (Likert questions). On a 10
potential sources of the variation. Simply
point scale, the difference between a 9 and a
investigate the variation and data as it
10 is not necessarily the same difference as
happens naturally - whatever the values
the difference between a 6 and a 7. This is
may be. Experiments provide greater
also an easy one to remember, ordinal
insight but sometimes, it is not possible or
sounds like order.
ethical to perform a more powerful
experiment. For example, it wouldn't be 253. Ordinal Data Attribute data that can be logically placed in
right to purposely overcrowd a (AKA Rank an order from smallest to greatest or in an
kindergarten classroom with 75 students to Order Data) order of time.
see what the effect on learning would be. 254. Organization Can show organizational deficiencies
Instead, education researchers gather Chart
naturally occurring data on classroom size
255. Outlier A recorded observation that is well outside
the range of variation of the rest of the data
256. Outputs The products or services produced by the
process and used by the customer
257. Output Changes in the number of incoming calls at a 269. Poisson Used to model rates. The probability of exactly
Variation call center per hour. If the output metric varies Distribution r events occuring can be computed using this
in a non-random way, it is safe to say that some (get a Poisson Table); has a mean equal to the
combination of special cause factors has variance; use ex: can tell you the # of defects
affected the volume of incoming calls. per disk drive or the # of defects per
automobile; widely used in queuing theory
258. Parameter Population value
270. Poka-Yoke Mistake-proofing is an action taken to remove
259. Parato Commonly used to display category
or significantly lower the opportunity for an
Chart occurences, constructed to evaluate the
error or to make the error so obvious that
criticality or financial impact of each category,
allowing it to reach the customer is almost
lists categories from left to right in order of
impossible.
highest to lowest with "other" at the end
271. Poka-Yoke Japanese term that means "mistake-proof";
260. Pareto Used to prioritize problems
making processes so easy and simple to follow
Diagram
that even a child can perform them
261. Pareto A great diagnostic tool for detecting problems
272. Population A term that means any full set of something,
Diagram with a measurement system
like all the people in your hometown, or all the
(AKA 80-20
pens you've produced, or all the invoices sent
Rule)
out over the last year.
262. Perceived What the customer thinks is desired (a new
273. Positively The relationship indicates that an increase in
Needs car)
Correlated one characteristic translates into an increase
263. Perceptual A specialized matrix diagram that captures the in the other.
Map perceptions of a customer; items of customer
274. Potential A form of hard savings, but require some
importance vs customer's corresponding
Savings action or decision to become realized. An
satisfaction level
example is a project that optimizes the design
264. Plackett they are not fractional factorial and they are of a new product. Until the re-design is
and main effects only implemented, the savings are only potential.
Burman
275. Practical Generally a system or chronic problem that is
non-
Problem impacting the success of a process or function.
geometric
design 276. Practical The solution is not complex, expensive, or
Solution irrational and is readily implementable.
265. Plackett- An advanced method for efficiently screening
Burman dozens of potential Xs. Although they don't 277. Precision Getting consistent results repeatedly, the
experiment reveal all the detailed knowledge provided by agreement or closeness of measurements on
designs a 2k factorial design, these experiments quickly the same item
identify which experimental variables are 278. Precision How widely spread the variation of your
active in your system or process. You then
measurement system is compared to the
follow these screening studies up with more
actual variation of the process or characteristic
detailed characterization experiments.
you are measuring.
266. Plans Use the outputs of simulation and analysis to 279. Problem Details the issue that the team wants to
define how data will be collected and how the
Statement improve and a reference to the baseline
processes will be controlled and audited.
measurement
267. Plots and Histograms, Pareto charts, control charts 280. Problem Significantly clarifies the current situation by
Charts
Statement specifically identifying what needs to improve,
268. Poissen the level of the problem, where it is occurring,
mean and and the financial impact of the problem. It also
variance serves as a great communications tool, helping
to get buy-in and support from others. When
problem statements are well written, people
readily grasp and understand what it is that
you're trying to accomplish.
281. Problem Only 10% of the complaints are recorded 290. Process Intended to be a collection of all the critical-
with Management to-quality outputs, or CTQs, for a process, a
Complaint Summary department, a division, or even up to an
Cards entire company. The summary rolls up to
whatever level an organization needs for
282. Process The set of actions and activities that transform the
monitoring, reviewing, and taking action to
inputs into the outputs
assure acceptable process and business
283. Process Determination of the capability of a process to performance. Each time a Six Sigma project
Capability perform to expectations The output is a numerically is completed, that project's CTQs are added
Analysis defined index of capability to the summary, building a complete
284. Process The determination that the process can meet the process management system
Capability product specifications as intended; Recognizing the 291. Process Used to visualize the process being
Analysis nature of process variability the process capability Mapping described, check current processes for
target is usually tighter than product specifications duplication or redundancy (unnecessary
285. Process Uses a tool called a process control plan to monitor complexity), and to assist in work
Control inputs. The objective of the process control plan is to simplification
create systematic feedback loops and actions to 292. Process Uses a tool called a process management
assure the process has inherent, automatic control. Monitoring summary to monitor outputs. The objective
With a good process control plan, you can change of the process management summary is to
out people, equipment, materials, and production enable the visibility, review, and action for
rates without significantly altering the performance all critical process outputs in an organization
quality of the process
293. Process Enable you to design, simulate, and optimize
286. Process The companion to the process management optimization work processes. These include tools for
Control Plan summary. The process control plan is focused on the tools creating process and work flow diagrams,
Xs, the inputs to the process. The inputs, by building cause-and-effect matrices,
definition in the formula of Y = f(X), are the critical Xs constructing fishbone diagrams, developing
that are determined from the Six Sigma project. But SIPOC (Suppliers-Inputs-Process-Outputs-
it is okay to place process outputs, the Y (the CTQs), Customers) diagrams, assessing process
on a process control plan, too. When done correctly, capabilities, and more. The goal of these
the process control plan creates a complete picture tools is help you see how work is performed
of all possible inputs, outputs, and activities for a and identify where the source of problems
single process. is.
287. process has 3 294. Process Shift Due to rapid shifts in the process pattern
steps, what being plotted. Events that could prompt
is total such a change include a change in crew
defects per (machine settings) or a change in measuring
unit device or method (not gradual tool wear or
288. Processing Focuses on improving existing processes, and a large a reduction in defective level due to Kaizen
Domain number of employees are involved in this; here you techniques).
take actions and implement improvements that 295. Process Determined by examining the variation
enable your process to perform to its utmost Variation between part averages that are averaged
potential all the time, not just part of the time, within Establishment among inspectors
the limitations of its current design.
296. Progressive- Screening >> Characterizing >> Optimization
289. Process Ties together the activities of a company Iterative
Management Approach to
Six Sigma
Experiments
297. Project Short summary of strategic reasons for the
Business Case project
298. Project The team will be focused on the appropriate 312. Range The difference between the largest and the
Charter problem, solution will be aligned towards the smallest observed data values in a
organizations goals, the team champion will distribution (Max - Min). Simple, preferred
be supportive, provides consistent target for metric for sets of data with few (2 to 5)
the team members. Drawback: Greatly influenced by
outliers.
299. Project Life Practical Problem, Six Sigma Project, Statistical
Cycle Problem, Statistical Solution, Control Plan, 313. Ratio An extension of the interval level that
Practical Solution, Results includes an inherent zero starting point.
Data is interval data with a natural zero
300. Project The boundaries of the project
point. For example, time is ratio since 0 time is
Scope
meaningful. Degrees Kelvin has a 0 point
301. Provide a PLT is determined by the amount of WIP - (absolute 0) and the steps in both these
description since the bigger the queue, the longer it takes scales have the same degree of magnitude.
of the a single work piece to travel through a
314. Real Needs What the customer really wants
relationshop process. The speed in which WIP exits the
(transportation)
between process - Exit Rate, also effects the queue time
PLT, WIP and hence the overarching PLT. 315. Reasonable Is the specification based on a realistic
Exit Rate assessment of the customer's actual needs?
Does the specification relate directly to the
302. purpose of Analyse the data and discover the causes of
performance of the characteristic?
Analyse the problem.
316. Recorded Measureable, Continuous, Variable
303. purpose of Implement and control the process to make
values
Control sure defects don't occur.
reflecting
304. purpose of Define the problem and what the customer length,
Define requires. volume, and
305. purpose of Improve the process to remove causes of the time
Improve problem described in the 'problem 317. Recording Generally used for tally counts or attribute
statement'. Checksheets data, including machine, operator,
306. purpose of Measure the defects and process operation. characteristic, and so on; subjective data can
Measure be recorded on a checksheet; can record
variable, attribute, and locational data
307. Purpose of Keeping the project on track
Milestones 318. Regression Determining the strength of the relationship
between a response variable (Y) and one or
308. QFD The desires of the customer and how to make
more predictors (Xs).
a produce or process meet them; a customer
driven process which is implemented by the 319. Reliability Accelerated life testing, lifetime characteristics
organization; includes customer competitive and analysis, growth curves
assessment and technical results Survivability

309. Quality Conformance to standards or specifications. 320. Repeatability The part of measurement variation that
occurs when you repeat measurements with
310. Randomize This spreads out the otherwise concentrated
the same item, the same measurement
or confounding potential for nuisance effects
setup, and the same equipment, under the
evenly and fairly over all of the experimental
identical conditions. In a way, this can be
runs and preserves the clarity of your results;
thought of as the short-term part of
reduces the risk of extraneous variables
measurement system variation.
skewing the results of your analysis; The idea
is to guarantee that only the effect of your 321. Repeatability Determined by examining the variation
selected factors is purposely concentrated between the individual inspectors and
during your experiment. within their measurement readings

311. Random Theoretically means that each item in the lot


Selection had an equal chance to be selected
322. Replications If you repeat your experiment, you get 330. Results Tangible results measurable in metrics with
slightly different results. This shouldn't quantifiable financial or strategic value
surprise you. Variation, as always, is a 331. Robustness The ability of a process or system to perform
part of everything - including your
consistently in the face of variation. Taguchi
experiment. Repeating runs of your
and other experiment designs allow you to
experiment allows you to estimate how
investigate and optimize your process or
much of the observed variation in your
system so that it is as immune as possible to
process or system is explained by the
the ravages of variation.
derived Y = f(X) equation and how much
remains unexplained. 332. robustness system(concept) design then parameter
approach design, then tolerance design
323. Replications Processes and systems have variation.
Part of experimentation is repeating 333. Rolled You have to establish an extremely high
your whole experiment, or parts of it, to throughput probability of success for each individual
understand how much variation there yield component characteristic if you ever expect
actually is. your final products, services, and transactions
to be highly successful and defect-free. The
324. Reproducibility The part of measurement variation that
probability of a unit going through all process
occurs when you repeat measurements
steps or system characteristics with zero
with different items and different
defects.
measurement setups, under different
environmental conditions. Captures all 334. rolled Rolled throughput yield is the probability that
the long-term variation influences in throughput a single unit can pass through a series of
your measurement system. yield process steps free of defects.

325. Reproducibility Determined by examining the variation 335. Rolled The product of the yields of all process steps
between the average of the individual Throughput
inspectors for all parts measured Yield
Calculation
326. Resolution A comparison of the smallest increment
your measurement system can provide 336. Rolled The overall yield for a string of processes
to the characteristic you are trying to Throughput (multiply the first time yields for each step
measure. A good rule of thumb is to use Yield (RTY) together)
a measurement system with at least ten 337. Rolled RTY = e^(-DPU)
increments within the specification width Throughput
you are measuring or within the process Yield (RTY)
variation you are trying to observe. So, calculation
as an example, for measuring the for DPU
variation of a process that should be
338. R&R Study Reproducibility, Process Variation, and
completed between 9 and 10 minutes,
allows you Repeatability; can determine measurement
you want to use a measurement system
to error
with increments of no more than 0.1
determine
minutes.
339. RUMBA Reasonable, Understandable, Measurable,
327. Response The term used for the output of the
Believable, Attainable
process that you investigate in the
experiment; synonymous with the Y in 340. Run Order An axis for the order in which the
the Y = f(X) equation. The whole point of measurements occurred
the experiment is to figure out how the 341. Runs Every experiment is made up of a series of ___.
Xs combine together to effect the Y Each experimental ___ consists of a unique,
328. Response surface The purpose of many experiments is to predetermined set of values for each of the
methods (RSM) find out the best values to set the input factors. You then conduct the process or
and optimization variables at. A whole branch of the field system through one cycle with those input
of DOE focuses on designing and values, and the output is recorded. That is a
analyzing experiments to find the local ___ in an experiment.
or global optimal operation settings. 342. Sample Repeated measurements of a characteristic or
329. Responsibility A matrix developed during the six sigma a process; any subset of a population
Communications define stage that plots levels of
Chart commitment versus key stakeholders
343. Sample Data Any sample size if randomly selected can be 352. Short-term "Compares the width of a two-sided
suitable for audit purposes, since we are not Capability specification to the effective short-term width of
directly performing lot acceptance or Index (Cp) the process. = (USL-LSL) / 6sigmashortterm;
rejection; Coding and rounding affect both Value equal to 1 means that the voice of the
the mean and standard deviation customer is equal to the voice of the process,
value less than 1 means that the process is
344. Sample Data Collected from the process characteristic (an
wider than the specification, with defects
(AKA input or an output) in which you are
spilling out over the edges, value greater than 1
Subgroup interested. The process must be allowed to
means that the effective width of the process
Data) operate normally while you're taking a
variation is less than the required specification,
sample.
with fewer defects occurring"
345. sample size(n)
353. Short-term Used if you are using a short-term standard
for
Sigma deviation, short term variation performance
E(interval)and
Score represents the best variation performance that
std dev
you can expect out of your currently
configured process, idealistic measure of
346. Sampling validity, applicability, and known risks capability (also, easy to collect)
plans for both 354. Short- Always less than or equal to the calculated long
inspection Term term deviation
and auditing Standard
consider Deviation
three things Equation
347. Scatter Displays the relationship between variables; 355. Short-term Purely random; Short-term variation is caused
Diagram produces correlation coefficients Variation by the combined effect of all the little things
348. Scatter Plot Used to explore the relationship between (Common that are too hard to include in your
(Main Effects two characteristics; The key to creating a Cause understanding of the process.
Plot) scatter plot is in the capturing of the Variation)
measurement data. To investigate the (ST)
relationship between two characteristics, 356. Sigma Exactly how many standard deviations can fit
you need to capture measurements from Score (AKA between the mean and a specifiecation limit of
the two characteristics simultaneously. So at Sigma any process or specification; Z = |SL - x_| / o
each measurement time, you have to take Value or
simultaneous measurements for each of the just Sigma)
characteristics you are interested in. If you
357. Simple When you work to create an approximate
are interested in exploring the relationship
Linear equation for a system that has a single output Y
between characteristics X and Y, at each
Regression and a single input X
point of measurement, you have to collect
and record values for X and Y 358. Simulation Simulate the flow of work and material through
a process based on the model, and analyze the
349. S chart upper
results of the simulation for overall
control limit
effectiveness and efficiency. Find defects,
errors, bottlenecks, variation, and non-value-
added elements.
350. Screening At this first stage, experiments are designed 359. SIPOC A high level view with only 4-7 steps displayed,
Experiments to handle a large number of factors or provides team members an understanding of
variables. When you first start investigating the process; suppliers, inputs, process, outputs,
a process or system, you identify all the and customers
possible Xs that may be influencing the
output Y. The whole point of ________ is to
quickly verify which of these factors has a
significant effect on the output.
351. Sensitivity Reading to one decimal greater than the
reported dimension; the ability to detect
differences in measurement
360. Six Sigma A problem-solving methodology. In fact, it's 371. Soft Savings Financial benefits that may occur as a
the most effective problem-solving result of a Six Sigma project but are not
methodology available for improving accountable as a direct result of the
business and organizational performance. A project; Calculated using a rational
proven and proscriptive set of analytical assessment of the expected benefits and
tools, project control techniques, reporting a probability analysis of their likelihood.
methods, and management techniques that 372. Special-Cause You can reduce and eliminate this type of
combine to form breakthrough
Variation variation (ex: mailman gets a flat tire which
improvements in problem solving and
makes him late), best to reduce this type
business performance.
of variation first, not stable or predictible;
361. Six Sigma The prescriptive rollout of the Six Sigma Those non-random causes of variation
deployment methodology across an organization, with that can be detected by the use of control
assigned practices, roles, and procedures charts and good process documentation.
according to generally accepted standards. 373. Special Cause The output has varied to an extent that is
362. A Six Sigma When the key outcomes of a business or Variation inconsistent with what you would expect
improvement work process are improved dramatically, (Assignable from purely normal, short-term, natural -
often by 70 percent or more. Cause or random influences. You know that
Variation) something non-random has occurred
363. Six Sigma Initialize, Deploy, Implement, Expand,
and, therefore, you know that you can
Lifecycle Sustain
find the cause and solve the problem.
364. Six Sigma Uses Six Sigma methods and tools to
374. Specification Any value designating acceptable from
Organization improve performance: Continuously lower
Limit (SL) unacceptable performance.
costs, grow revenue, improve customer
satisfaction, increase capacity and capability, 375. Specifications Represents what the customer requires in
reduce complexity, lower cycle time, and order to be satisfied; that value separating
minimize defects and errors. acceptable from unacceptable
performance. This definition holds for all
365. Six Sigma The statistical term for a process that
process or characteristic performance
performance produces fewer than 3.4 defects (or errors)
measures.
per million opportunities for defects.
376. Stable Measurement system is one that stays
366. Six Sigma A well defined effort that states the problem
centered and free of offset changes.
Project in quantifiable terms with known
expectations. 377. Stakeholders Parties affected by the planned changes
of a project
367. Six Sigma Has a financial impact to EBIT (Earnings
Project Before Income Tax) or NPBIT (Net Profit 378. Standard The most commonly used measure of
Before Income Tax) or a significant strategic Deviation dispersion. For each number: ((# -
value // Produces results that significantly Mean)^2 + (# - Mean)^2 + ...)/ Sum of #s ;
exceed the amount of effort required to then square this // Note: IF it is a full
obtain the improvement // Solves a problem population divide variance by all
that is not easily or quickly solvable using numbers, if it is a sample divide by N-1
traditional methods // Improves 379. Standardization Reduces the nuber of characteristics or
performance by greater than 70 percent
features of a system
over existing performance levels
380. State 2 of the 1. Develop strategies for the management
368. Six Sigma Hard, Soft, and Potential
aims of a of different stakeholders.
Savings
stakeholder 2. Develop a risk and contingency plan to
Categories
management maximise the positive effect for each
369. Skewed Not symetrical and not normal plan stakeholder.
370. Skewed Variation shape that is not symmetrical, 381. Stated Needs What the customers say they want (a car)
Distribution either one side or the other of the 382. Statistic A number derived from sample data that
distribution extends out farther than the
describes the data in some useful way;
other side
Sample Value
383. Statistical For the vast majority of cases, the exact 391. Steps of the 1. Determine the Y: what specifically needs to
relationship between the Xs and the Y or Ys is Project be improved.
not known. This is due to the complexity of the Definition 2. Identify the associated processes and their
system and the human inability to address all Process physical locations.
the factors that truly exert influence on the 3. Determine the baseline performance for
output. Because of this natural limitation, each Y chosen.
repetitions of the system with the same input 4. Identify the cost and impact of the
values will not always produce the same output problem.
performance 5. Write the problem statement.
6. Write the objective statement.
384. Statistical These tools enable you to analyze data collected
7. Identify and recruit candidates for the
Analysis either from the real-world performance of a
project team.
product or process, or as the output of a
8. Obtain approvals and launch.
simulation or experiment. These include basic
statistics tools, and tools for analyzing variance, 392. Strength of The steepness of the slope of the fitted lines
conducting regressions, perform- ing Design of Correlation determines how strong an effect the input
Experiments (DOE), and building control charts, has on the output, steep slopes mean strong
plots, tables, and graphs. The goal of these tools effect
is to help you turn data into knowledge such 393. Superstitious Confusions of cause versus correlation
that you can make informed decisions.
Delusions
385. Statistical Mode, Mean, and Median 394. Suppliers Systems, people, organizations, or other
Measures
sources of the materials, information, or
of
other resources that are consumed or
Variation
transformed in the process
Location
395. The Create a high-level process map with a few
386. Statistical Data-oriented problem that is addressed with
Suppliers- key details about each of the key
Problem facts and data analysis methods.
Inputs- contributing elements.
387. Statistical Involves the use of statistical techniques to Process-
Process monitor and control the variation in processes. Outputs-
Control SPC is used first to stabilize out-of-control Customers
(SPC) processes. But it is also used as a follow-on, to (SIPOC)
monitor the consistency of product and service 396. Sustain Re-alignment, re-training, and evolution
processes.
397. t More conservative, because your smaller
388. Statistical Data-driven solution with known confidence/risk
sample size lessens the accuracy of your
Solution levels versus an "I think" solution.
calculated value for o. In fact, for each
389. Statistics The distilling of numbers, data, and desired confidence level, this is adjusted
measurements into knowledge and insight. If depending on how many data points are in
you understand a little bit about statistics, you your sample
can create a data-leveraging environment in 398. Taguchi A quality improvement tool. Understand
which you gain the utmost value from the
effects of process on output.
information you have.
399. Taguchi Loss The loss curve is often called this. As the
390. Stem and Graphical data method that can show the value
Function figure shows, the traditional perspective is a
Leaf Plot of all individual readings
hit or miss view, where unnecessary cost is
incurred only when specifications are
missed. The Six Sigma view is one in which
additional cost is incurred as performance
moves away from the target.
400. Tampering A process is a system of events, activities, and
with the feedback loops. A well designed process
Process exhibits inherent self control, while a poorly
designed process requires frequent external
control and adjustment to meet
requirements. This term describes such
adjustments.
401. Target length 120 days (4 months) 414. Total Defects TDPU = -ln(RTY)
of an initial Per Unit
project (TDPU)
Formula
402. Target (T) The single designated value you wish the
process or characteristic to perform at. (This 415. Traceability The accuracy of a measuring instrument
is an ideal. Variation prevents the process linked to the U.S. National Standards.
or characteristic from exactly hitting this 416. Traditional The proportion of correct items (conforming
every time.)
Yield to specifications) you get out of a process
403. t distribution to compensate for error in the estimated compared to the number of raw items you
standard deviation of small sample size put into it. (Out / In)
404. Team Composition of the Improvement Team 417. Truncated Cut off in some way and is not normal
Membership 418. t- test,
405. Terrorist A customer who has turned against you standard
and will spread ill-will regarding your deviation not
company and product known
406. Things to look Variation beyond expected limits, Trends, 419. Two Repeatability and Reproducibility; Together,
for in process Runs, Shifts Components the metrics of repeatability and
and of reproducibility capture all of your
characteristic Measurement measurement system's precision. In Six
behavior System Sigma the acronym is R&R. You see this
charts Precision acronym used to describe how precise a
measurement gauge is, as in its gauge R&R.
407. Thinking Focuses on improving the reasoning and
When you see R&R think of how good your
Domain efficiency of EVERY employee
measurement system is.)
408. Three ways a "1) The location of the central tendency of
420. two different
Sigma (Z) score the distribution, the mean, moves either
std dev's, F
can change closer or farther from the specification limit.
test
2) The width of the distribution, as defined
by the standard deviation, gets either
wider or narrower.
3) The location of the specification limit SL
moves either closer or farther from the
characteristic or process variation."
409. Time Series Specific tools for analyzing results of data
collected over time - trends,
decompositions, moving averages 421. Two 1) Repetitions of the measurement event
important produce different outcome results
410. Time-Series Graphical method employed by Six Sigma
facts that can (variability). Like in the coin-flipping
Chart practitioners to track the progress of
be experiment, not every repetition of the ten-
improvement
generalized flip series produced the expected five
411. Tolerance The analysis of margins and tolerances to to any heads. The same is true if you repeatedly
Analysis determine optimal design specifications measurement take a five-point average of the thickness of
412. Tollgates In DMAIC, business processes are situation paper coming out of a paper mill. 2) This
improved by following a structured resulting measurement, or sampling
method with set steps distribution, is normally distributed. The
variation is also centered on the expected
413. Tools that Boxplots, histograms, and stem and leaf
outcome. And the more repetitions you
display the plots
make, the closer and closer the sampling
same data
variation gets to a perfectly normal
reads on a
distribution.
continuous
scale 422. Two-Sided A pair of specification limits creating an
Specification interval of acceptable performance between
the two limits.
423. Type 1 The most effective form of process control. 437. Variance The sum of the squared
Corrective This is a control applied to the process that deviations of a group of
Action eliminates the error condition from ever measurements from their
occurring. This is the primary intent of the mean divided by the number
Poka-Yoke method. of measurements
424. Type 2 The second most effective form of process 438. Variation (Variability) Deviation from expectation;
Corrective control. This a control that detects when The property of a
Action an error occurs and stops the process or characteristic, process, or
shuts down the equipment so that the system to take on different
defect cannot move forward. This is the values when it is repeated.
detection application of the Poka-Yoke 439. Venn Diagram (AKA Set Diagrams that show all possible
method.
Diagram) logical relations between a
425. Types of Specification limit (SL), One-sided finite collection of sets
Specifications specification, two-sided specification, (aggregation of things).
upper specification limit, lower Typically circles.
specification limit, target. 440. Voice of the Business All the needs and expectations
426. Typical Six This consists of qualified people with the (VOB) of the business
Sigma Team expertise needed 441. Voice of the Customer All the needs and expectations
427. Understandable Is the specification clearly stated and (VOC) your customers have for your
defined so that there can be no argument products and services.
about its interpretation? 442. Weibull Scale Parameter The point at which 63.21% of all
428. Uniform Variation is evenly spread out across a values fall below
Distribution bounded range, you are just as likely to 443. What are the 1/10th of the customers
observe a value for a characterstic at one
resolution/discrimination specified tolerance.
end of the interval as you are at the other
requirements of a
or anywhere in between
measurement system
429. Unintended The customer uses the product in an 444. What does 2^k mean? The 2 stands for only 2 levels
Needs unintended manner
for each input factor; k
430. Unit The item you are working on represents the number of
factors in the expirement; for
431. upper
each k in your expirement you
confidence
select a high and low value that
limit (3 sigma)
bounds the scope of the
investigation; provides only
knowledge within the two
432. Upper A value designating an upper limit above variable settings
Specification which the process or characteristic
445. What does CTQ stand for? Critical to Quality
Limit (USL) performance is unacceptable
give an example 10 day lead time +/- 1 day
433. Upstream DFFS methods and tools enable you to
446. What does S.I.P.O.C. stand Suppliers, inputs, process,
anticipate the source of development,
for outputs, customers.
manufacturing or performance problems
before they occur so that you can design 447. What elements of the When did the problem start?
and plan in a way that allows you to avoid following problem What is occurring?
them statement are missing Where is the occurrence?
'There is a problem with Who does it affect?
434. Variable Data Data that are measured with some sort of
the number of How does it happen?
instrument, such as a micrometer, tape
undercooked pizzas'.
measure, or watch.
448. What is a K.A.N.O. model? The Kano model is a helpful
435. Variables A quality which can assume several (more
way to understanding
than two) values
different types of customer
436. Variance Theoretically useful, but lacks direct tie to needs.
reality. For each number: ((# - Mean)^2 +
449. What is an operational A clear, precise description of
(# - Mean)^2 + ...)/ Sum of #s
definition. the factor being measured.
450. What is a R.A.C.I. Assists in agreeing and confirming 461. Within Calculation to quantify the average width of
chart how will be responsible / Group variation of all the conditions
accountable for what and who will Variation
be consulted and informed. 462. X bar Average of the raw data
451. What is S.M.A.R.T. Specific, Measurable, Attainable, 463. Yield How much good stuff - performance within
Relevant, Timebound.
specifications - you get out
452. What is stratification. Capturing and use of characteristics 464. Z vs t When you have only a few data points in your
to sort data into different categories
sample, you're not able to get an accurate
also known as 'slicing the data'.
estimate of the population standard deviation .
453. What is the benefit of A swim-lane flowchart emphasises When your sample has anywhere from two to 30
a swim-lane 'who does what' and enables the data points, you have to use a different factor
flowchart over a study of handoffs between people instead of Z. Statisticians call this new factor for
standard process flow and/or work groups in a process. small samples t
chart
454. What is the purpose Parking lots are temporary holding
of a parking lot areas for suggestions, ideas or
topics that are not directly related to
the purpose of the meeting.
455. What is the purpose A SIPOC diagram helps a team and
of a S.I.P.O.C. it's sponsor(s) agree on project
boundaries and scope. It is a high
level view of the end to end process
and links inputs, process steps and
outputs to the customer.
456. What two things do A measure of location and a
you need to describe measure of spread
any type of
distribution in
scientific terms?
457. What type of control X bar and R
chart would you use
if you had 3
measurements of a
process taken
everyday for a 6
month period.
458. What type of control I-mR
chart would you use
if you had one
measurement taken
everyday for a month
459. When would you use When each part is measured by
a nested gauge R&R only one operator, such as
study destructive testing. We normally use
Crossed R&R.
460. Why is the VOC so We have to understand the needs
important of the particular customer segment
in which we are targeting so that we
can design our systems and
processes to fulfill those needs (and
only those needs).

S-ar putea să vă placă și